m:\70 a] PYKKM-: AM) TKACK >1KTAf,S
UIAKLKSTOY SOI II! CAKOL5NA
l.S. FAY IRONMKNTAI I»KOTE< TION
Office of Air Onalitv Planning and Standards
R«'K< trrh 'I rianjile S'ark. North Curolina 27711
-------
EPA-450/2-75-004
BENZO[a]PYRENE
AND TRACE METALS
IN
CHARLESTON, SOUTH CAROLINA
by
Carl Spangler
Environmental Protection Agency
and
Noel de Nevers
University of Utah
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air Quality Planning and Standards
Monitoring and Data Analysis Division
Research Triangle Park, N. C. 27711
June 1975
-------
This report has been reviewed by the Monitoring and Data Analysis Division, Office
of Air Quality Planning and Standards, Office of Air and Waste Management, Environ-
mental Protection Agency, and approved for publication. Mention of company or
product names does not constitute endorsement by EPA. Copies are available free
of charge to Federal employees, current contractors and grantees, and non-profit
organizations - as supplies permit - from the Air Pollution Technical Information
Center, Environmental Protection Agency, Research Triangle Park, North Carolina;
or may be obtained, for a nominal cost, from the National Technical Information
Service, 5285 Port Royal Road, Springfield, Virginia 22161.
Publication No. EPA-450/2-75-004
11
-------
PREFACE
Carl Spangler, who was a trace metal specialist for the Environmental Protection
Agency, initiated the sampling reported here to see if a correlation could be estab-
lished between trace metal and benzo[a]pyrene concentrations and lung cancer mor-
tality. Mr. Spangler's untimely death in November 1972 occurred just as the results
of the sampling program were beginning to come in. In December 1973, Mr. Spang-
ler's colleagues in EPA commissioned Noel de Nevers of the University of Utah, who
had spent a year at EPA and worked with Mr. Spangler, to write this report, sum-
marizing and analyzing the results of the sampling program.
111
-------
CONTENTS
Page
LIST OF FIGURES vi
LIST OF TABLES vii
ABSTRACT viii
INTRODUCTION 1
SAMPLING PROGRAM 3
EXPERIMENTAL RESULTS 4
ANALYSIS OF AIR SAMPLING DATA 7
ANALYSIS OF SOIL SAMPLING FOR BaP - CHARLESTON, S.C 10
ANALYSIS OF SOIL SAMPLING FOR BaP - WEST VIRGINIA 12
ANALYTICAL RELIABILITY 14
HISTORICAL PROBLEM 15
CONCLUSIONS 16
REFERENCES 17
APPENDIX A. STUDY OF BaP, Be, Cd, Cr, Cu, AND Ni IN AIR
AT CHARLESTON, S.C., NOVEMBER 1972 THROUGH
MARCH 1973 19
APPENDIX B. ANALYSIS OF BENZO | a|PYRENE IN SOIL SAMPLES 39
-------
LIST OF FIGURES
Figure Page
1 Observed BaP Concentrations (ppm) in Soil, in Charleston, S.C.
(Values Represent Highest Concentration Measured at Site from
Table 2.) 11
A-l Outline Map of Charleston, S.C., Showing Locations of Air
Sampling Sites 21
A-2 Weekly Average BaP Content of Air at Three Sampling Sites
in Charleston, S.C., November 1, 1972, to March 31, 1973. ... 24
A-3 Weekly Average Be content of Air at Three Sampling Sites
in Charleston, S.C., November 1, 1972, to March 31, 1973 .... 24
A-4 Weekly Average Cd Content of Air at Three Sampling Sites
in Charleston, S.C., November 1, 1972, to March 31, 1973. ... 25
A-5 Weekly Average Cr Content of Air at Three Sampling Sites
in Charleston, S.C., November 1, 1972, to March 31, 1973 .... 25
A-6 Weekly Average Cu Content of Air at Three Sampling Sites
in Charleston, S.C., November 1, 1972, to March 31, 1973 .... 26
A-7 Weekly Average Ni Content of Air at Three Sampling Sites
in Charleston, S .C., November 1, 1972, to March 31, 1973 .... 26
A-8 Average Weekly Temperatures at Charleston, S.C., from
November 1, 1972, to March 31, 1973 27
A~9 Wind Roses for Days During which BaP Content of Air was
Relatively High at Sampling Stations 29
A-10 Wind Roses for Days During which Be Content of Air was
Relatively High at Sampling Stations. 30
A-11 Wind Roses for Days During which Cd Content of Air was
Relatively High at Sampling Stations 31
A-12 Wind Roses for Days During which Cr Content of Air was
Relatively High at Sampling Sites 32
A-13 Wind Roses for Days During which Cu Content of Air was
Relatively High at Sampling Stations 33
A-14 Wind Roses for Days During which Ni Content of Air was
Relatively High at Sampling Stations 34
VI
-------
LIST OF TABLES
Table ' Page
1 Summary of Measurements of Trace Constituents in Ambient Air
at Charleston, S.C., November 1, 1972, through March 31, 1973 . ... 4
2 Analysis of BaP Content of Soil In and Around Charleston, S.C. . . 5
3 Analysis of BaP Content of Soil Samples from Coal Refuse Banks
in West Virginia - Kentucky 7
4 Comparison of Charleston, S.C. , Ambient Air Trace Constitutent
Values with National Values 10
5 Summary of Reported Soil Concentrations of BaP 13
A-l Average Airborne Concentrations of BaP and Five Metals at
Charleston, November 1, 1972, to March 31, 1973 22
A-2 Range of Values for Selected Air Pollutants During Winter Months
at Charleston and Selected Mid-Atlantic Cities 22
A-3 Relative Air Pollution Levels at Three Sampling Stations in
Charleston, November 1, 1972, to March 31, 1973 23
A-4 Wind Direction Observations Recorded at Hi-Vol Air Sampling
Sites and at Charleston Municipal Airport 28
A-5 Possible Source Directions for Air Pollutants at Three Hi-Vol
Air Sampling Sites in Charleston, S.C., November 1, 1972, to
March 31, 1973 35
t
•A-6 Weekly Average BaP and Metal Contents of Air at Three
Sampling Sites in Charleston, S.C., November 1, 1972,
to March 31, 1973 37
B-l Results of Sample Analysis for Soils Taken in West Virginia
(Coal Refuse Banks) and in Charleston, S.C 45
Vll
-------
ABSTRACT
Charleston, S.C. , along with some other cities in the Southeastern Coast and
Gulf Coast area, has an anomalously high incidence of deaths resulting from lung
cancer - about 50 percent higher than the national average. Benzo[a]pyrene (BaP)
and trace metals are widely suspected of being causative agents in lung cancer. A
survey of BaP and trace metals in the ambient air in Charleston reveals, however,
that the air concentrations are lower than the national averages, generally falling
in the 0.01 to 29 percentile among American cities.
To test the view that atmospheric concentrations of BaP can readily be inferred
from soil concentrations, soil samples were taken in Charleston at sites roughly corre-
sponding to the area in which air was subject to testing in the air sampling program.
There are few values from the United States with which to compare the Charleston
soil values, though the soil concentrations of BaP there are somewhat higher than ob-
served urban values in the Soviet Union, for example. Accordingly, from the limited
data available, the Charleston soil values of BaP do not appear extraordinarily high.
Thus, it seems safe to infer that the abnormally high death rate resulting from
lung cancer is not due to higher-than-normal exposure to the agents addressed.
In addition to the testing mentioned, because of the suspected high emission rate
of BaP from burning coal refuse banks, a group of samples were taken in the vicinity
of several refuse banks in West Virginia. Data from the analyses show that soils
there do not contain abnormally high amounts of BaP.
Vlll
-------
BENZO[a]PYRENE
AND TRACE METALS
IN
CHARLESTON, SOUTH CAROLINA
INTRODUCTION
Considerable, and apparently uncontroverted, evidence-*- shows that the incidence
of death from lung cancer (defined as malignant neoplasm of the bronchus,trachea,
and lung) is not uniform across the United States, but rather is significantly higher
than average in the Southeastern Coast and the Gulf Coast area. In the most recent
survey available (1959-1961), the national average death rate from lung cancer for
white males per 100,000 of population was 40.79. The cities with the highest death
rates for white males from this cause were:
Annual death rate for white males
City per 100,000 population (1959-1961)
Lake Charles, La. 62.31
Albany, Ga. 61.34
Charleston, S. C. 60.63
New Orleans, La. 59.74
Shreveport, La. 55.90
Galveston, Texas 55.88
Jersey City, N. J. 55.87
Jersey City, in seventh place, is the first city to appear on the list that is not
part of the "Southeastern Coast and Gulf Coast area." This anomaly in the incidence
of death from lung cancer has not been satisfactorily explained. An epidemiological
study^ of the Charleston area suggests that airborne particulate matter in the city
itself is a possible cause of this anomab; , at least for that locality. Similarity, med-
3 4
ical evidence has been adduced ' to suggest that generally, if not in these cities ,
trace metals and benzofajpyrene (BaP) are implicated in the causation of lung cancer.
-------
This background impelled the late Carl Spangler, who was trace-metal specialist
for the National Air Data Branch of the Environmental Protection Agency (EPA) , to
initiate a sampling program to see if a correlation could be established between trace
metal and BaP concentrations and lung cancer mortality. Charleston was chosen as the
sampling site because it was close to Mr. Spangler's office in Durham, N.C. , was
high on the lung cancer mortality list, and had a public health department that was
interested in the effects of air pollution and willing to cooperate in the program.
Mr. Spangler's untimely death in November 1972 occurred just as the results of
the sampling program were beginning to come in; thus, he had no part in the
analysis of the data. In December 1973, Mr. Spangler's colleagues in EPA commis-
sioned Noel de Nevers of the University of Utah, who had spent a year at EPA and
worked with Mr. Spangler, to write this report, summarizing and analyzing the
results of the sampling program. Chemical analysis was conducted under EPA con-
tract by Geological Resources, Inc. , Raleigh, N.C. , and by the Research Triangle
Institute, Research Triangle Park, N.C.
-------
SAMPLING PROGRAM
The program consisted of two parts: air sampling and soil sampling. The air
sampling was conducted at three sites chosen to represent (1) suburban air far
from major industry, (2) air in a downtown commercial district, and (3) air direct-
ly across the river from the industrial district of Charleston. At each site filters
from Hi-Vol samplers were regularly collected for the period November 1, 1972, to
March 31, 1973. The filters were collected by the Charleston County Health Depart-
ment and sent directly to Geological Resources, Inc. , for analysis. Samples from
the filters were analyzed for BaP, beryllium (Be), cadmium (Cd) , chromium (Cr) ,
copper (Cu) , and nickel (Ni) . The test methods and results are detailed in Appen-
dix A.
The soil sampling program was initiated largely as a result of the extensive work
by Shabad^ who showed that the concentration of BaP in soils bears a strong relation-
ship to proximity of suspected sources of BaP emissions and,hence, presumably is an
indicator of the average atmospheric BaP concentration in the air over that particular
section of soil. If it can be established that one can reliably infer pollutant concen-
trations in air from concentrations in soil, it will then be possible to make detailed
surveys of airborne pollutant levels and concentration gradients using soil sampling
techniques, which are more rapid and less expensive than air sampling methods.
This relationship has not yet been convincingly demonstrated. To test it, soil
samples roughly corresponding to the area whose air would be tested by the air samp-
ling program mentioned above were taken in Charleston. In addition, because of the
suspected high emission rate of BaP from burning coal refuse banks , a group of sam-
ples were taken in the vicinity of several refuse banks in West Virginia. These sam-
ples were subjected to the same analytical procedure as the results from Charleston.
The analytical procedures used and the concentrations found are detailed in Appen-
dix B and in Reference 7.
-------
EXPERIMENTAL RESULTS
The experimental results of the sampling program are shown in Table 1, which
summarizes the air sampling results; Table 2, which summarizes the soil sampling
results from Charleston; and Table 3, which summarizes the soil sampling results
from the burning coal refuse bank areas in West Virginia.
ANALYSIS OF AIR SAMPLING DATA
The significance of the Charleston data in Table 1 is most apparent when compar-
ed with pollutant values in other urban areas. In the following analysis, particular
Table 1. SUMMARY OF MEASUREMENTS OF TRACE CONSTITUENTS IN AMBIENT AIR
AT"CHARLESTON, S. C. , NOVEMBER 1, 1972, THROUGH MARCH 31, 1973
All results in nanograms/m3 (1 nanogram = 10 g = 10~3 microgram)
Location
Radio Station WTMA
(22 samples)
Minimum
Maximum
Average
Queen St. Fire Station
(22 samples)
Minimum
Maximum
Average
Mt. Pleasant Post Office
(22 samples)
Minimum
Maximum
Average
Total (66 samples)
Minimum
Maximum
Average
BaP
0.0028
1.2409
0.5711
0.1693
1.6787
0.7441
0.1995
1.9767
0.7448
0.0028
1.9767
0.6866
Be
0
0.1609
0.0373
0
0.1247
0.0287
0
0.1529
0.0363
0
0.1609
0.0341
Cd
0.1267
0.8232
0.2751
0.1590
0.6328
0.3024
0.1136
0.9245
0.2531
0.1136
0.9245
0.2768
Cr
0.0916
2.1474
0.7293
0.0797
1.3573
0.5870
0
0.9413
0.3589
0
2.1474
0.5584
Cu
16.0
155.3
62.9
8.9
253.4
59.0
11.4
139.8
47.2
8.9
253.4
56.4
Ni
0.1466
2.4883
0.8332
0.2610
1.6172
0.8306
Trace
1.5188
0.6359
Trace
2.4883
0.7666
-------
Table 2. ANALYSIS OF BaP CONTENT OF SOIL IN AND AROUND CHARLESTON, S.C.
Location (EPA sample no.)
Reported BaP concentration, ppm
by RTIa
by
100 yds south of U.S. Army-Charleston Depot,
Hydrocarbon storage, 100 yds east of N.
Rhett Rd. (C-19A-A)
50 yds east of Virginia Ave., 0.25 mile west
of West Virginia Co. Pulp Mill. (C-19C1-A)
100 yds east of Virginia Ave. in small public
park adjacent to major hydrocarbon storage
areas. (C-26A-A)
St. Johns and McMillin Aves., 100 yds from
all traffic. Vacant section outside Naval
Hospital area. (C-25C)
Chicora and English Sts., outside fence of
U.S. Navy "fuel farm." (C-32B)
Back of Esso Tank Storage on Greenleaf St.
about 300 yds from Cooper River. (C-390)
About 0.5 mile north of Airco Ferroalloys
plant along gravel access road and rail-
road switches. (C-39B-A)
Near Columbus and Immigration Sts., near
Port Authority docks, 200 yds from
highway. (C-46D-A)
Cleared block, 100 yds west of Concord St.,
100 yds south of Laurens St. (C-53B-A)
Vacant lot at Limehouse St. and Murray Blvd.
(C-52D)
Cleared city block (center) Barr, Wentworth,
Halsey, and Beaufain Sts. (C-52B)
Residential area at Third Ave. and Wagener
St. (C-45B)
Ashepoo Lane, 50 yds west of 1-26, 100 yds
south of Koppers, 100 yds in front of
Agrico Chemical. (C-39A1-A)
Vacant industry site, 50 yds north of
Koppers storage, 200 yds west of 1-26,
elevated, on Col. Nitrogen Road. (C-39A2-A)
0.216
1.920
3.634
1.003
0.150
33.528
3.520
27.709
1.095
0.953
97.612
2.1007
1.0569
4.6451
4.0113
3.6893
3.1108
5.4251
5.4859
2.4673
2.8488
18.3454
-------
Table 2 (continued). ANALYSIS OF BaP CONTENT OF SOIL
IN AND AROUND CHARLESTON, S. C.
Location (EPA sample no.)
Reported BaP concentration, ppm
by
by GRC
100 yds from Hagood oil-fired power plant,
60 yds north of Col. Nitrogen. (C-39A(3)-A)
0.5 mile northeast of Mobile Chemical, along-
side old, unused plant entrance road next
to swamp. (C-32C)
Wilbur St., in industrial park, opposite
Bird and Sons. (C-31A)
Corner lot at Firestone and Blossom Sts. 2
miles from south end of airport runway.
(C-24A)
Corner lot at Prince and Taylor Sts. 0.5
mile from south end of airport runway.
(C17B-A)
Ashley-Phosphate Road, 1.25 mile from north
end of airport runway. (C-3A)
Carteret and Vassar Sts., vacant lot in high
economic level residential area. (C-38A-A)
Burning Tree and Greentree Lane, vacant lot
in residential area. (C52C-A)
Riverside Beach Road, rural. (C-47D-A)
20 yds beyond end of Moultrie St. in
residential area; space heating with oil
heater is common. (C-62A-A)
Residential area at end of Flynn Street
at Ashley River. (C-23-B)
1.875
0.606
0.696
2.502
0.231
0.489
0.255
2.0669
1.18787
4.2829
7.1528
6.3879
4.1038
3.6123
4.2925
1.3308
Research Triangle Institute (RTI) reports considerable variation in results
depending on method used.8 The highest value is reported here.
Geological Resources (GR).
emphasis will be given to comparison of the Charleston survey data with the nation-
wide data reported in Air Quality Data for Qrganics, 1969 and 1970^ and Air Quality
Data for Metals, 1968 and 1969,9 both of which are EPA compilations.
Consider first the results for BaP. The averages at the three sampling locations
in Charleston were 0.57, 0.74, and 0. 74 nanograms per cubic meter (ng/m^) . For
-------
Table 3.
ANALYSIS OF BaP CONTENT OF SOIL SAMPLES FROM COAL REFUSE
BANKS IN WEST VIRGINIA - KENTUCKY
Location (EPA sample no. )
Cedar Grove, W. Va.c
25 ft. from upper south edge of culm bank, from steep hillside;
necessary to separate gravel and roots. (A-l-A)
60 ft. from bank, uphill (60° slope). (A-2-A)
100 ft. from 'bank, uphill. (A-3-A)
1/8 mile from bank, below stripped area. All from undisturbed
soil, washed-over spots avoided. (A-4-A-)
Hand-selected yellowish platy crystals (with some coal) from
bank blow hole edges at surface. Other holes flaming in high
wind, nearby small sample. (A-5)
Marnie, W. Va.e '
Control sample taken at small hillside cemetary, approximately
1 mile south of bank. 100 yds. from highway. (A-6)
50 yds. from bank between railroad and cliff. (A-7)
50 yds. from bank between railroad and cliff, and about 100
yds. from previous sample. (A-8)
Taken between stream and highway and 100 yds. from previous
sample. (A-9)
Huntington-Ashland area1"
Taken in residential area, 0.5 mile downhill from end of west
runway of Ashland-Huntington Airport. (B-l)
Taken 300 yds., same elevation, from end of runway. (B-2)
Taken at Neal , W. Va., in field adjacent to gas compressor
station and across Big Sandy River from Ashland Refinery,
West. (B-3)
0.5 mile south of Ashland Refinery on Kentucky side of
Big Sandy River. (B-4)
75 yds. west of center of Ashland Refinery, midway front each
of two main units, which were about 0.25 mile away. (B-5)
Taken just outside of fenceline, about 200 yds. from Semet
Solvay coke ovens at Ashland, Kentucky. (B-6)
Taken across railroad and highway in open field, about 0.5
mile southwest from Semet Solvay coke ovens. (B-7)
Taken in Riverside Community, near bank of Ohio River, about
2 miles south of Semet Solvay coke ovens. (B-8)
Reported l3aP co
by RTia
1.004
0.076
0.432
0.326
-
0.513
1.475
0.969
22.200
-
-
_
_
-
-
-
-
ncentration, ppin
by GRh
5.451
2.383
3.200
3.851
26.043
(6.051)d
-
-
-
-
1.600
1.452
3.707
2.184
4.261
2.501
4.529
4.725
Research Triangle Institute (RTI) reports considerable variation in results depending on method
used. The highest value is reported here.
Geological Resources, Inc. (GR).
cSamples obtained on May 31, 1972--a cloudy, windy day with temperature of 55"F; rain occurred
the previous day.
Duplicate sample.
p
Samples obtained on June 1, 1972--a windy, partly cloudy day with temperature of 60°F. Samples
were taken in and around a single burning coal refuse bank.
Samples obtained in the Huntington, W. Va., and Ashland, Ky., area on June 1, 1972--a vindy
partly cloudy day with temperature of 60°F.
-------
1970, annual averages are available for 164 cities^; these values ranged from a high
of 3.09 to a low of 0.30 ng/m^. On a percentile basis, if the average of the three
Charleston sampling stations is used, then the resultant average of 0.68 ng/m^
corresponds to the 14th percentile of the range of American cities. Stated another
way, 14 percent of the typical American cities listed will have lower concentrations
than Charleston and 86 percent will have higher concentrations than Charleston.
Moreover, this comparison somewhat overstates the concentration in Charleston
because it compares values taken in winter in Charleston with annual averages for
the other cities. Typically, concentrations of BaP are higher in winter because of
seasonally greater emissions from space heating. The maximum value measured
in Charleston (1.97 ng/m^) corresponds to the 66th percentile of the observed
annual averages; i.e. , 34 percent of American cities have annual average concentra-
tions higher than the maximum measured value in Charleston.
It can be concluded from this comparison that in terms of BaP concentrations in
the air, Charleston is in the cleanest 20 percent of American cities. Other cities
having 1970 BaP annual averages close to the winter average measured in Charleston
are Little Rock, Arkansas; Riverside, San Diego, and San Francisco, California;
Des Moines, Iowa; Guayanilla, Puerto Rico; and Burlington, Vermont. None of these
cities are centers of heavy industry or of major coal burning sources of BaP .
For trace metals, the evidence supports the same conclusion, namely that
Charleston has ambient air concentrations that are lower than national averages,
but the form of the available data makes it difficult to state the result as precisely
as for BaP. The reason for the difficulty is that many or all of the results reported
by EPA are reported as "below the limits of detection" or zero. For example, of the
736 reported values for beryllium (Be) in 1969, all but 9 were reported as zeros.
From the data in Reference 9, it seems clear that any value less than 0.1 ng/m for
Be was reported as a zero. The averages at all sites in Charleston were about
0.03 ng/m , so that if they were included in this tabulation they would also be
reported as zero. Thus, for Be there seems to be little to be said, other than that
the values obtained were not large enough to attract interest.
For the other four metals surveyed (Cd, Cr, Cu, and Ni) , however, most of the
values reported are non-zero. For cadmium, EPA° reported annual averages in 1969
for 183 cities. The reported values ranged from 3 to 105 ng/m^. All values less than
3 ng/m^ were reported as zeros; there were 26 such values. The measured average
at Charleston (Table 1) is 0.3 ng/m^; in the tabulation by EPA, it would be reported
-------
as a zero. Thus it can be said that measurements place Charleston in the lowest 13
percent of American cities in ambient Cd levels. However, because the value in
Charleston is only one-tenth of that which is the lowest reported by EPA, the value
in Charleston must correspond to a percentile much lower than 13 percent. If the
1969 values are plotted on log-normal paper, they form a reasonably straight
line over the range of 3 to 38 ng/m^. If extrapolated to the 0.3 ng/rn-^ average meas-
ured at Charleston, it corresponds to the 0.01 percentile; i.e. , only 1 out of 10,000
American cities would be expected to have cadmium concentrations this low. This
extrapolation is not certain; so the result should not be taken too literally; but it does
show that the measured values of cadmium in Charleston are extremely low.
Survey findings for chromium follow a pattern similar to the cadmium result.
EPA° has tabulated 1969 annual average Cr values for 184 cities. These values ranged
from 7 to 102 ng/m-*. All values less than 7 ng/m were reported as zeros; there
were 110 such values. The measured Cr value in Charleston averaged 0.5 ng/m^,
which is less than one-tenth of the detectable limit reported by EPA.° Thus, the re-
sults in Charleston must be in the lower 60 percent (110/184 = 60 percent) of the sam-
ple of American cities. If the 1969 annual average values (7 to 102 ng/m^) are plot-
ted on log-normal paper, they make a reasonably straight line. If extrapolated to
the average Cr value found in Charleston, it corresponds to about the second percen-
tile; i.e. , 98 percent of American cities would be expected to have higher chromium
concentrations than Charleston . >
For copper, which is more abundant in the air than the other metals sampled,
the problem of zero values does not exist; the 184 annual average values for 1969
reported in by EPA' ranged from 20 to 1280 ng/m^ with no zeros. The measured
average in Charleston was 56 ng/m^, which corresponds to the 29th percentile; 71
percent of American cities would, therefore, have higher ambient Cu concentrations
than Charleston.
Finally, EPA' reported 1969 annual average values for nickel for 183 cities; these
ranged from 7 to 173 ng/m^. All values less than 7 ng/m^ were reported as zeros;
46 of the 183 average values (25 percent) fall in this category. The measured average
value in Charleston was 0. 7 ng/nH, about one-tenth of the detectable limit reported
by EPA. Based on the results of the two surveys,'Charleston would rank in the low-
er 25 percent of American cities in ambient nickel concentrations. Moreover, if the
reported values are plotted on log-normal paper and extrapolated to the observed
value in Charleston, Charleston is placed at about the 0.02 percentile; i.e. , only 2
out of 10,000 American cities would be expected to have nickel concentrations this low.
9
-------
From these results (summarized in Table 4) , it can be concluded that with respect
to the trace constitutents studied, Charleston is one of the cleanest cities in the United
States for which sample data are available.
Table 4. COMPARISON OF CHARLESTON, S.C. AMBIENT AIR TRACE CONSTITUTENT
VALUES WITH NATIONAL VALUES
Consti tutent
Benzo[a]pyrene
Beryllium
Cadmi um
Chromium
Copper
Nickel
Average measured winter season
value in Charleston, S.C.
(Winter 1972-1973),
ng/m3
0.68
0.034
0.28
0.55
56
0.77
Corresponding percentile of
American urban areas (annual
average 1970a
14
-
0.01
2
29
0.02
Some data are for 1969.
ANALYSIS OF SOIL SAMPLING FOR BaP - CHARLESTON, S.C.
Figure 1 is a map of the Charleston area with values in parts-per-million (ppm)
resulting from the analysis of BaP in soil (Table 2) placed at the location where the
7
sample was taken. If the two laboratories that conducted soil analyses found con-
flicting results, the larger of the two reported values was plotted.
The values shown on Figure 1 may be grouped as follows:
0.6 to 7 ppm - 22 values
27 to 97 ppm - 3 values
The 22 values in the 0.6 to 7 ppm range appear to be randomly distributed
through the metropolitan area with no apparent gradients. There seems to be no
obvious way to draw iso-concentration lines on this figure.
The three high values are isolated. The 97 ppm result is near a wood-treating
plant, where the creosote used in processing could be expected to cause the high
concentration observed; the other two high values have no obvious explanation from
the reported locations.
10
-------
Figure 1. Observed BaP concentrations (ppm) in soil, in Charleston, S.C.
(Values represent highest concentration measured at site from Table 2).
11
-------
The values at the end of the runways of the airport are slightly higher than the
average of the community, but not strikingly so.
It would be interesting to compare these values with those for other cities, as was
done with the ambient air values, to see whether these arc high, low, or average
values. This is not easy to do because there are few data on soil concentrations of
BaP and most of the available data are from the Soviet Union, where the mix of air
pollutant emissions is probably not the same as that in the United States. (The sum-
mary indexes of Chemical Abstracts for Volumes 69-76 and 78 list a total of nine
references on BaP in soils; all nine are from the Soviet Union.)
Table 5 presents a summary of reported soil concentrations of BaP. (Note that
most of the Soviet sources are based on the reviev"- in Chemical Abstracts; except
for two of them, the original sources were not consulted.) From this table it can
be seen that in the Soviet Union concentrations for industrial areas are 1 to 220
ppm, for urban areas they are 0.1 to 1 ppm, and in forests they are about 0.001 ppm.
The only reported value for the United States is 0.04 to 1.3 ppm in a forest area .
If these values are compared with the ones found in Charleston , the Charleston
values seem to fall in the range of the other values, with no abnormally high values
(except the 97 ppm near the wood-treating plant, which is not as high as that at the
Soviet refinery as reported by Shabad^) . On the other hand, the general urban
values in Charleston are about an order of magnitude higher than those reported for
Moscow.
ANALYSIS OF SOIL SAMPLING FOR BaP - WEST VIRGINIA
The soil samples taken in the vicinity of burning coal banks in West Virginia
(summarized in Table 3) reveal values mostly in the range of 0. 5 to 5 ppm of BaP .
Two show values of 22 and 26 ppm. The general group of values taken near this
source are not much different from those taken in Charleston. This result is sur-
prising, in the light of published estimates" that burning coal refuse banks emit up
to 75 percent of the total emissions of BaP in the United States. In the light of these
low concentrations it can be concluded either that there is little fallout of BaP from
burning coal banks in the vicinity of the source or that the estimates of the emission
rates from this source are much too high.
12
-------
Table 5. SUMMARY OF REPORTED SOIL CONCENTRATIONS OF BaP
Reference
10
11
4
12
13
14
15
16
Country
USA
France
USSR
USSR
USSR
USSR
USSR
USSR
Location
Mixed forest,
Conn, and Mass.
Forests
New building areas
Old building areas
Oil refinery near
naphtha cracker
Near an airport runway
Forests
Various regions not
described
Moscow city
Near the Moscow freeway
Various rural soils
Close to an industrial
plant
In old city territory
far from industry
In village on dead end
street
200-300 meters from freeway
with negligible auto traffic
Industrial areas
Concentration, ppm
0.04 to 1.3
0.002 to 0.30
0.1
0.25
220
0.01 to 0.07
0.0001 to 0.0006
0.0004 to 0.07
0.268 to 0.364
0.158 to 0.0668
0.001
5.5 to 8.3
0.6
0.18 to 0.52
0.007 to 0.089
191
13
-------
ANALYTICAL RELIABILITY
Research Triangle Institute (Appendix B) found that there was considerable
difficulty in making a quantitative extraction of BaP from soil samples. This organiza-
tion showed that, for different extractants, the measured concentration could vary by
a factor of ten. Table 2 shows that apparent duplicate samples analyzed by two
different laboratories can yield results that differ by a factor of ten.
Accordingly, if soil sampling is to be undertaken in a serious way, consideration
should be given to the following recommendations:
1. Analytical methods should be developed to ensure that results are
meaningful.
2. Results in this report should be considered as order-of-magnitude estimates.
In view of the ten-fold apparent uncertainty in the results , it seems inappro-
priate for the analytical laboratories to report these results to five signifi-
cant figures.
14
-------
HISTORICAL PROBLEM
Death rates from lung cancer probably reflect the results of long-term exposure
to pollutants. Pollutant concentrations 30 years ago could have been much higher
than those observed through current air- and soil-sampling programs. Therefore,
mortality data for 1959-1961, which were used in this report, may not accurately
reflect the effects of pollutant concentrations occurring today.
A significant program to reduce pollutant emissions took effect in the period
1968-1972. As a result of this program, the annual average concentration of sus-
pended particulates measured at the Charleston County Health Department decreased
as follows :
Annual average suspended particulate
Year concentration, micrograms/m^
1968 74.4
1969 57.9
1970 55.3
1971 44.3
1972 49.7
1973 36.7
Jacobs and Langdoc^° linked this drop in particulate concentration with a drop in
the cardiovascular death rate. (The views in that article have been reviewed
19
critically by EPA .) Whether or not the .views in the article are correct, they do
show that pollutants are being looked at now to explain events that occurred years
ago, when the pollution situation was much different.
The above particulate data indicate as much as a two-fold reduction in total sus-
pended particulates in downtown Charleston in recent years. Even if all the measured
ambient air concentrations of BaP and metals were doubled, the resulting values
would still not be in the high end of the range for American cities and would not con-
stitute strong evidence for their involvement in lung cancer. Thus, without addition-
al historical data, the possible link between BaP and metals and lung cancer cannot
be properly tested; but, given the extremely low currently measured ambient concen-
trations of BaP and metals, this argument must be considered speculative.
15
-------
CONCLUSIONS
1. In the most recent years for which summary data are available (1959~1961) ,
there is an anomalously high incidence of deaths from lung cancer in the Southeastern
and Gulf Coast area.
2. Measured ambient air concentrations in Charleston, S.C. , one of the cities with an
anomalously high lung cancer death rate in 1959-1961, show that this city has a low
concentration of BaP and trace metals in the ambient air. It is one of the cleaner
American cities as regards these pollutants.
3. The concentration of BaP in the soil in the Charleston metropolitan area is about
an order of magnitude higher than that in Moscow in the Soviet Union. There are few
data from the United States with which to compare these soil concentrations. Based
on comparisons with Soviet data, the Charleston values seem large but not startling.
4. Thus, the measured concentrations of BaP and trace metals found in this study do
not indicate a level high enough to cause the death rate from lung cancer in Charleston
to be higher than the national average.
5. The methods for analyzing BaP in soils are in need of additional refinements.
6. Results in this report should be considered as order-of-magnitude estimates.
16
-------
REFERENCES
1. Duffy, E.A. and R.E. Carroll. United States Metropolitan Mortality, 1959-
1961. Public Health Service, U.S. Department of Health, Education, and
Welfare, Cincinnati, Ohio. Publication No. 999-AP-39. 1967.
2. Caston, J.C., J.F. Finklea, and S.H. Sandifer. Cancer of the Larynx and
Lung in Three Urban Counties in South Carolina. Southern Medical Journal.
65:753-756, 1972.
3. Dixon, J.R., D.B. Lowe, D.E. Richards, and H.E. Stokinger. The Role
of Metals in Chemical Carcinogenesis-Asbestos Cancers. Department of
Health Education and Welfare, Occupational Health Program, Cincinnati, Ohio.
1971.
4. Shabad, L.M. Studies in the USSR on the Distribution, Circulation and Fate
of Carcinogenic Hydrocarbons in the Human Environment and the Role of their
Deposition in Tissues in Carcinogenisis; A Review. Cancer Research.
27:1132-1137, 1967.
5. Shabad, L.M. et al. The Carcinogenic Hydrocarbon Benzo(a)pyrene in the
Soil. Journal of the National Cancer Institute. 47:1179-1191, 1971.
6. Fox, R.D., L.M. Scale, and R.M. Bradway. Organic Carcinogens in our
Atmosphere. (Presented at the APCA meeting in Atlantic City, N.J. June
27, 1971. Paper No. 71-53.)
7. Stahel, E.P. Letter reporting soil sample analyses by Geological Resources
Inc., Raleigh, N. C. Dated August 11, 1972. Copy in files of Office of Air
Quality Planning and Standards, Environmental Protection Agency, Research
Triangle Park, N. C.
8. Air Quality Data for Organics, 1969 and 1970. Environmental Protection
Agency, Research Triangle Park, N. C. Publication No. APTD-1465.
June 1973.
9. Air Quality Data for Metals, 1968 and 1969. Environmental Protection Agency,
Research Triangle Park, N. C. Publication No. APTD-1467. June 1973.
10. Blumer, M. Benzopyrenes in Soil. Science. 134:474-475, 1961.
11. Mallet, L. and M. Heros. Pollution des terres vegetales par les hydrocarbures
polybenzeniques du type benzo-3,4-pyrene. C.R. Acad. Sci. (D) (Paris).
.251:958-960, 1962.
12. Kogan, Yu.L., K.N. Fedorova, and N.V. Stasyuk. Levels of Carcinogenic
Hydrocarbons in the Soils of Daghestan. Biol. Nauki 1972. 15_(10): 113-116.
Reviewed in Chemical Abstracts. 78:56910.
17
-------
13. Shabad, L.M. et al. Carcinogenic Hydrocarbons in the Soils of the Soviet
Union. Kazan Med. Zh. 1971. 5_:6-ll, Reviewed in Chemical Abstracts.
76:58140.
14. Shcherbak, N.P. Fate of Benzo(a)pyrene in Soil. Vop Onkol. 15:75-9,
1959. Reviewed in Chemical Abstracts. 7^:48877.
15. Yanasheva, N.Y. Soil Contamination with 3,4-benzpyrene in the Vicinity of
Coke Chemical Plants. Gig. Naselennykh Mest 1967. 193-6. Reviewed in
Chemical Abstracts. 69:109610.
16. Shcherbak, N.P. On the Benz(a)pyrene Detection in the Soil. Vop Onkol.
13:77-80, 1967. Reviewed in Chemical Abstracts. 66; 98273.
17. Jacobs, C.F. Personal communication March 11, 1974.
18. Jacobs, C.F. and B.A. Langdoc. Cardiovaocular Deaths and Air Pollution
in Charleston, South Carolina. Health Service Reports. 8J7: 623-632, 1972.
19. Hammer, D.I. Trip Report; Air Pollution and Cardiovascular Deaths. Nov-
November 9, 1971. Letter in files of Epidemiology Branch, Environmental
Protection Agency, Research Triangle Park, N. C.
18
-------
APPENDIX A.
STUDY OF BaP, Be, Cd, Cr, Cu, AND Ni
IN AIR AT CHARLESTON, S. C,
NOVEMBER 1972 THROUGH MARCH 1973
Prepared by
Geological Resources, Inc.
Raleigh, North Carolina
November 1973
19
-------
INTRODUCTION
The project discussed in this report was conducted in Charleston, S. C. , from
November 1, 1972, until March 31, 1973, to monitor variations in the amounts of
selected pollutants emitted in different sections of the city. Particulate matter was
collected on glass-fiber filters by Hi-Vol samplers at three locations in Charleston—
Radio Station WTMA, the Queen City Fire Station, and the Mount Pleasant Post Office
(See Figure A-l) . Cooperating in the collection effort were the Air Pollution Control
Section of the Charleston County Health Department and the Office of Air and Water
Programs, U.S. Environmental Protection Agency. Geological Resources, Inc. ,
Raleigh, N. C. , analyzed the particulate matter each week for benzo[a)pyrene,
beryllium, cadmium, chromium, copper, and nickel. Results of the analyses were
used to compute, for each station, the weekly average concentration of each pollutant
in nanograms per cubic meter of air. During the 5-month period, which included
the winter heating season, the chief meteorological factors evaluated with respect to
pollution levels were temperature and wind direction variations.
ANALYSIS OF SAMPLES
Filters analyzed in the study were collected from three sampling sites during
the period from November 1, 1972, to March 31, 1973, or 22 weeks. Upon collection
by the Charleston County Health Department for each sampling period (approximately
1 week, with some variations) , the filters were sent directly to Geological Resources,
Inc. A separate composite sample for each of the three sampling stations was then
made using all of the filters collected during the sampling period at that station. A
total of 66 composite samples were analyzed for benzo[a]pyrene (BaP) , beryllium (Be) ,
cadmium (Cd) , chromium (Cr) , copper (Cu) , and nickel (Ni) for a total of 396
separate analyses.
The composite samples for analysis were prepared in accordance with procedures
outlined in Analysis of Selected Elements in Atmospheric Particulate Matter by Atomic
Absorption. ^ Metals were determined, using standard procedures , with a Perkin-
Elmer Model 303 Atomic Absorption Spectrophotometer. Benzo[a]pyrene was
determined through techniques of thin-layer chromatography and spectrofluorometry,
again with standard procedures of preparation.
Results of analyses for BaP and metals are given in the addendum.
20
-------
URBAN. LOW DENSITY
URBAN. HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
1. WTMA
2 QUEEN STREET FIRE
STATION
3. MT. PLEASANT POST
OFFICE
4. CHARLESTON AIR-
PORT (WIND ROSE
FOR STUDY PERIOD)
Figure A-1. Outline map of Charleston, S.C., showing locations of air
sampling sites.
21
-------
DISCUSSION OF DATA
Air Pollution Levels at Three Sampling Sites iri Charleston
Table A-l shows average amounts of benzo[a|pyrene and five metals in air during
the study period at the three sampling sites. Comparison of analyses made during
the study with published data on air pollution at other urban areas along the mid-
Atlantic coast during the same time of year (Table A-2) indicates that the amounts of
Be, Cd, and Cr in the air at Charleston are somewhat less than measured quantities
at the other cities. BaP, Cu, and Ni pollution in the air at Charleston is on about
the same level as in other similar urban areas.
Table A-l. AVERAGE AIRBORNE CONCENTRATIONS Or
BaP AND FIVE METALS AT CHARLESTON,
NOVEMBER 1, 1972 to MARCH 31, 1973
(ng/m3)
Pollutant
BaP
Be
Cd
Cr
Cu
Ni
WTMA
0.5711
0.0373
0.2751
0.7293
62.9
0.8332
Queen Street
Fire Station
0.7441
0.0287
0.3024
0.5870
59.0
0.8306
Mount Pleasant
Post Office
0.7448
0.0363
0.2531
0.3589
47.2
0.6359
Table A-2. RANGE OF VALUES FOR SELECTED AIR POLLUTANTS DURING
WINTER MONTHS AT CHARLESTON AND SELECTED MID-ATLANTIC CITIES2
(ng/m3)
Pollutant
BaP
Be
Cd
Cr
Cu
Ni
Range of concentrations
0.003-2.22
0.000-0.161
0.114-0.924
0.000-2.15
8.9-253.4
Trace-2.48
Charleston
Charleston
Charleston
Charleston
Charleston
Charleston
1.4-3.63
0.00-2.0
0.00-20.0
130-180
(1.0-8.0
34.0-81.0
1-12
Norfolk, Va.
Norfolk, Va.
Baltimore, Md.
Baltimore, Md.
Greenville, S. C.)
Norfolk, Va.
Portsmouth, Va.
22
-------
In general, metal contamination in the air was highest at Radio Station VVTMA and
was lowest at the Mount Pleasant Post Office. On the other hand, BaP was lowest in
air at the radio station and highest at the post office. Air at the Queen Street Fire
Station had a pollution level generally intermediate in relation to the other two sites .
Table A-3 summarizes these observations.
Table A-3. RELATIVE AIR POLLUTION LEVELS AT THREE SAMPLING
STATIONS IN CHARLESTON, NOVEMBER 1, 1972, TO MARCH 31, 1973
Pollutant
BaP
Be
Cd
Cr
Cu
Ni
Highest
Post office
Radio station
Fire station
Radio station
Radio station
Radio station
Intermediate
Fire station
Post office
Radio station
Fire station
Fire station
Fire station
Lowest
Radio station
Fire station
Post office
Post office
Post office
Post office
As shown in Figure A-l, Radio Station WTMA, the site with the highest levels of
most metal air pollutants, is closer to the industrial section of Charleston; the Mount
l
Pleasant Post Office, with the lowest metal air pollutant values, is farthest from the
industrial section. Jacobs and Langdoc^ point out that practically all major sources
of industrial air pollution are located in this section of Charleston. Major industries
there include fertilizer plants, a kraft papermill, a ferroalloy plant, a chemical com-
plex, and military bases. Based on the principle that pollutants in air tend to become
diluted as they move with the wind out from their sources,^ it appears that activities in
the industrial section of Charleston result in release of some metal pollutants to the air.
The cause for the lower levels of BaP in the air at the radio station, compared
with the other two sampling sites, may be related to the station being located in an
area with lower traffic activity and fewer houses, both of which contribute BaP to the
atmosphere through fuel combustion . -*
Variation of Air Pollution Levels During Study
Variations in the level of pollutants in the air at each of the three sampling sta-
tions are shown in Figures A-2 through A-7. BaP, Be, Cd, and Ni levels in the air
tended .to increase at the beginning of the study period and to decrease toward the
end of the period. Average temperatures at Charleston showed a decline at the
beginning of the period and an increase at the end. Comparison of temperature
variations during the period of study (Figure A-8) with levels of BaP, Be, Cd, and
23
-------
2.5
2.0
I I I I, I I I
I I I I I I I I I I I
A RADIO STATION
O FIRE STATION
O POST OFFICE
234
NOVEMBER
56789
DECEMBER
10 11 12 13 1
TIME, week
JANUARY
4 15 16 17 18
FEBRUARY
19 20 21 22
MARCH
Figure A-2. Weekly average BaP content of air at three sampling sites in Charleston, S.C.,
November 1, 1972, to March 31, 1973.
0.20
CO
-a
a.
oT
00
0.15 —
0.10 —
0.05 —
RADIO STATION
O FIRE STATION
D POST OFFICE
2
3
4
5
VI
789
:EMBER
10 11 12 13 14
TIME, week I
JANUARY
15 16 17 18
FEBRUARY
19 20 21 22
!
MARCH
NOVEMBER
Figure A-3. Weekly average Be content of air at three sampling sites in Charleston, S.C.,
November 1, 1972, to March 31, 1973.
24
-------
A RADIO STATION
O FIRE STATION
D POST OFFICE
<-
NOVEMBER
DECEMBER
10 11 12 13 14 15 16 17 18 19 20 21 22
TIME, week
FEBRUARY
JANUARY
MARCH
Figure A-4. Weekly average Cd content of air at three sampling sites in Charleston, S.C.,
November 1, 1972 to March 31, 1973.
2.5
2.0
1.5
CO
I
a.
1.0
A RADIO STATION
O FIRE STATION
D POST OFFICE
234
NOVEMBER
56789
10
11 12 13 14 15 16 17 18 19 20 21
TIME, week
DECEMBER
JANUARY FEBRUARY
MARCH
2
Figure A-5. Weekly average Cr content of air at three sampling sites in Charleston, S.C.,
November 1, 1972 to March 31,1973.
Ni suggests a reasonably good (inverse) correlation of colder weather with higher
levels of these pollutants. BaP variations correlate best with temperature fluctua-
tions. Because combustion of fuels is a potential source of BaP^ and certain metals,
the probability is that increased combustion of fuels during cold weather leads to
higher concentrations of these pollutants in the air.
Although the burning of both solid and liquid fuels releases BaP and various
metals, this is not likely to be the only source of these pollutants at Charleston. A
25
-------
I I I I 1
• A RADIO STATION
O FIRE STATION
D POST OFFICE
1 234
NOVEMBER
56789
DECEMBER
10 11 12 13 14 IS 16 17 18 19 20 21 22
TIME, week
JANUARY FEBRUARY MARCH
Figure A-6. Weekly average Cu content of air at three sampling sites in Charleston, S.C.,
November 1, 1972 to March 31, 1973.
A RADIO STATION
O FIRE STATION
0 POST OFFICE
^
2
\r V^- — r <-r
MS/ I I I
1234
NOVEMBER
56789
DECEMBER
I
10 11 12 13 1
TIME, week
JANUARY
1
4 15 16 17 1
FEBRUARY
LJ
I I
8 19 20 21 2
MARCH
Figure A-7. Weekly average Ni content of air at three sampling sites in Charleston, S.C.,
November 1,1972 to March 31,1973.
26
-------
Ul
cc
LU
a
<
cc
1234
NOVEMBER
56789
DECEMBER
10 11 12 13 14 15 16 17 18 19 20 21 22
TIME, week
FEBRUARY
JANUARY
MARCH
Figure A-8. Average weekly temperatures at Charleston, S.C., from November"!, 1972
to March 31, 19737
comparison of BaP values and metal values in air at the Mount Pleasant Post Office
and Radio Station WTMA (Table A-l) indicates that the ratio of metals to BaP is not
constant. Thus, separate sources must account for some of the individual pollutants.
The levels of Cr and Cu in the air at Charleston during the study period exhibited
patterns different from the other air pollutants studied. Cr continued to increase in
the air until late February and early March before beginning a decline. Cu levels
showed a general decline throughout the period, with a low in January. Because Cu
and Cr levels in the air were higher at the radio station than at the other two sites ,
and because the radio station is closer to the industrial section of Charleston, it is
possible that the variations in levels of these substances reflect, to some degree,
variations in industrial processes and/or activities.
Air Pollution Level Variations and Wind
The purpose of combining wind data with air pollution in the study was to
determine whether certain wind directions tended to increase air pollution levels at
the three Hi-Vol air sampling sites. This aspect of the investigation has some limita-
tions . Wind speed effects are not integrated into the study even though high speed
winds will be less contaminated when passing over a pollution source than will low
speed winds moving in the same direction over the same source." Wind direction
data as recorded by the National Weather Service Office at the Charleston Municipal
27
-------
Airport do not agree with the directions observed at each of the sampling sites.
Table A-4 illustrates these differences during the first week of the study period.
Table A-4. WIND DIRECTION OBSERVATIONS RECORDED AT HI-VOL AIR SAMPLING
SITES AND AT CHARLESTON MUNICIPAL AIRPORT, 1972
Date
Nov. 1
Nov. 2
Nov. 3
Nov. 4
Nov. 5
Nov. 6
National Weather
Service observations9
Airport
100
180
230
260
50
80
Observations at air sampling sites
Radio
station
180
135
22K
225
45
135
Fire
station
135
180
225
225
45
225
Post
office
180
80
225
135
Local climatological data for Charleston.
Differences in wind directions recorded at the three sites probably resulted
from differences in time of observation: Hi-Vol air sampling sites were observed
only once each day, at the time air filters were changed, and this time varied from
site to site. Furthermore, winds observed at Hi-Vol air sampling sites located within
a few feet of the ground probably reflect local turbulence. ' On the other hand, wind
directions recorded at the airport are "resultant winds" based on several observations
each day. Because of the arrangement of weather instruments, these values are more
likely to reflect true wind directions over the entire city. For these reasons, it was
decided to use airport wind-direction data rather than sampling site wind data.
Examination of the curves shown in Figure A-2 through A-7 reveals that air
pollution levels for a given substance at the three Hi-Vol air sampling sites vary
rather similarly during the entire period of study. Detailed examination also shows,
however, that during a given week the pollution level may be relatively high at one
sampling site and relatively low at the other two. For example, note in Figure A-2
that during the first and fifteenth weeks of the study period BaP was relatively high
at the post office and relatively low at the other two sites. Plotting wind directions
during days in which air pollutant levels are relatively high less the days in which
pollution levels are relatively low at each of the three Hi-Vol air sampling sites has
resulted in the "pollution wind roses" shown in Figures A-9 through A-14.
28
-------
URBAN. LOW DENSITY
URBAN, HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
1. WTMA
4. CHARLESTON AIR-
PORT (WIND ROSE
FOR STUDY PERIOD)
Figure A-9. Wind roses for days during which BaP content of air was
relatively high at sampling stations.
29
-------
URBAN. LOW DENSITY
URBAN. HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
1. WTMA
2 QUEEN STREET FIRE
STATION
3. MT. PLEASANT POST
OFFICE
CHARLESTON AIR.
PORT (WIND ROSE
FORSTUDYPERIOO)
30
-------
URBAN. LOW DENSITY
URBAN. HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
I. WTMA
2. QUEEN STREET FIRE
STATION
3. MT. PLEASANT POST
OFFICE
4. CHARLESTON AIR
PORT (WIND ROSE
FOR STUDY PERIOD!
hnh ,
high at sampling stations.
in9 which Cd
of air was
31
-------
URBAN. LOW DENSITY
URBAN. HIGH DENSITY
. HUSTRIAL
SAMPLING SITE
1. WTMA
2. OUEEN STREET FIRE
STATION
4. CHARLESTON AIR
PORT (WIND ROSE
FORSTUOY PERIOD)
Figure A-12. Wind roses for days during which Cr content of air was relatively
high at sampling stations.
32
-------
URBAN. LOW DENSITY
URBAN. HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
1. WTMA
4. CHARLESTON AIR
PORT (WIND ROSE
FOR STUDY PERIOD)
Figure A-13. Wind roses for days during which Cu content of air was relatively
high at sampling stations.
33
-------
URBAN. LOW DENSITY
URBAN, HIGH DENSITY
INDUSTRIAL
SAMPLING SITE
1. WTMA
2. QUEEN STREET FIRE
STATION
4. CHARLESTON AIR-
PORT (WIND ROSE
FOR STUDY PERIODI
Figure A-14. Wind roses for days during which Ni content of air was relatively
high at sampling stations.
34
-------
Examination of the pollution wind roses suggests that the tentative conclusions
listed in Table A-5 may be made concerning general source directions for pollutants
at each of the three Hi-Vol air sampling sites.
Table A-5. POSSIBLE SOURCE DIRECTIONS FOR AIR
POLLUTANTS AT THREE HI-VOL AIR SAMPLING SITES IN
CHARLESTON, S.C., NOVEMBER 1, 1972, to MARCH 31, 1973
Sampling site
Radio station
Fire station
Post office
Pollutant
BaP
Cr, Ni
Be, Cd, Cu
BaP
Be, Cr
Ni
Cd, Cu
Cu
Ni
Cd
BaP
Cr
Wind direction
W, NW
W, E
Random
N
N, W
N, E
Random
N, W
N, W, S
N, E, S
Random
All directions low
Although this analysis of wind direction effects on levels of air pollution at the
different air sampling sites has limitations, in a general sense many of the pollution
wind roses (for example Cr and Ni) point toward Charleston's industrial section.
The positions of major traffic arteries and high-density housing areas, which are
also sources of air pollution, obviously increase the number of directions from which
winds could bring pollution to a given site.
SUMMARY
1. The amounts of Be, Cd, and Cr in the air at Charleston are somewhat less than
those quantities in urban areas along the mid-Atlantic coast. BaP, Cu, and Ni
pollution is on about the same level as in other similar urban areas.
2. Be, Cr, Cu, and Ni were found to be present in air in greater quantities closer
to the industrial section of the Charleston Peninsula, north of the city.
3. Higher amounts of BaP in air appear to be correlative with higher densities of
nearby traffic and housing.
4. BaP, Be, Cd, and Ni tend to increase in the air at Charleston during periods of
colder weather, suggesting that combustion of fuels contributes significantly to
the presence of the metals in the air. However, because the ratio of these metals
35
-------
to BaP is not constant from site to site, there must be separate sources for some
of the individual pollutants .
5. The levels of Cr and Cu in air during the study period exhibit patterns different
from the other air pollutants studied. Because their levels are highest in air
near Charleston's industrial section, it is possible that the variations in levels
of these substances reflect, to some degree, variations in industrial processes
and/or activities.
6. Analysis of wind direction data and pollution levels at the sampling sites during
the study period suggests that the industrial section of Charleston contributes
significantly to air pollution in that city but that other sources of air pollution,
not identified, also exist.
REFERENCES FOR APPENDIX A
1. Thompson, R.J., G.B. Morgan, andL.J. Purdue. Analysis of Selected Elements
in Atmospheric Particulate Matter by Atomic Absorption. Atomic Absorption
Newsletter. 9(3), 1970.
2. Air Quality for 1968 from the National Air Surveillance Network and Contributing
State and Local Networks. Office of Air Programs, Environmental Protection
Agency, Research Triangle Park, N. C. Publication No. APTD 0978. August
1972.
3. Jacobs, C.F. and B.A. Langdoc. Cardiovascular Deaths and Air Pollution in
Charleston, S. C. Health Services Reports . 8J7: 623-632, 1972.
4. Bosanquet, C.H. and J.A. Pearson. The Spread of Smoke and Gases from
Chimneys. Trans. Faraday Society. 32^:1249-1263,1936.
5. Shabad, L.M. et al. The Carcinogenic Hydrocarbon Benzo(a)pyrene in the
Soil. J. National Cancer Institute. 47_: 1179-1191, 1971.
6. Air Pollution Engineering Manual. Danielson , J .A. (ed.). Office of Air Quality
Planning and Standards, Environmental Protection Agency, Research Triangle
Park, N.C. Publication No. AP-40. May 1973.
7. Local Climatological Data for Charleston, S . C . National Oceanic and Atmospheric
Administration. U.S. Department of Commerce, Washington, D. C.
8. The Encyclopedia of Geochemistry and Environmental Sciences. Fairbridge, R.W.
(ed.). New York, Van Nostrand Reinhold Company, 1972.
9. Panofsky , H . A. Air Pollution Meteorology. Minerals Process . 1_0: 11-16, 60-61,
1969.
ADDENDUM
Weekly average benzo[a]pyrene, beryllium, cadmium, chromium, copper, and
nickel concentrations for Charleston are given in Table A-6.
36
-------
Table A-6. WEEKLY AVERAGE BaP AND METAL CONTENTS OF AIR AT THREE SAMPLING SITES
IN CHARLESTON, S.C., NOVEMBER 1, 1972, TO MARCH 31, 1973
(ng/m3)
Date and Site
11/1/72-11/8/72
Radio station
Fire station
Post office
11/8/72-11/15/72
Radio station
Fire station
Post Office
11/15/72-11/22/72
Radio station
Fire station
Post office
11/22/72-11/29/72
Radio station
Fire station
Post office
11/29/72-12/5/72
Radio station
Fire station
Post office
12/5/72-12/10/72
Radio station
Fire station
Post office
12/11/72-12/16/72
Radio station
Fire station
Post office
12/18/72-12/22/72
Radio station
Fire station
Post office
12/27/72-12/30/72
Radio Station
Fire station
Post office
1/1/73-1/6/73
Radio station
Fire station
Post office
1/7/73-1/14/73
Radio station
Fire station
Post office
BaP
0.2882
0.4450
1.6162
U.4692
0.5852
0.3462
0.2865
1.6787
0.8822
0.2087
0.4900
0.7457
1.1882
0.9222
1.1977
0.4496
1.0842
0.5055
0.3014
0.5598
0.4138
0.6973
1.2265
0.8368
2.2217
0.8390
1.9767
0.6850
0.7147
0.8390
0.9448
0.6413
0.5619
Be
0.0275
0
0
0.0022
0.0452
0.0018
0.0820
0.0574
0,0400
0.0816
0.1247
0.0863
0
0
0.0312
0
0
0
0
0
0
0.1609
0.0464
0.0665
0.0467
0.0162
0.1013
0.0355
0.0211
0.1529
0.0605
0
0.0114
Cd
0.2444
0.2419
0.9245
0.1778
0.2157
0.1237
0.2521
0.1850
0.1242
0.1865
0.1662
0.2158
0.2433
0.1594
0.1136
0.3636
0.3752
0.1688
0.4565
0.3016
0.1737
0.6438
0.5339
0.3162
0.8232
0.6328
0.7167
0.3831
0.4756
0.4905
0.1874
0.1859
0.1901
Cr
0.0916
0.2547
0
0.3174
0.2575
0.2199
0.3401
0.5105
0.2346
0.5130
0.3637
0.4058
0.1520
0.0797
0.0965
0.4990
0.5369
Trace
0.3701
0.2077
0.2004
0.4829
0.2205
0.1414
0.9354
0.3083
0.3321
0.4868
0.8667
0.6626
0.2418
0.8337
0.5388
Cu
55.0
253.4
139.8
145.7
33.4
30.7
110.2
8.9
38.0
28.0
119.5
75.1
138.4
40.4
41.5
117.6
73.1
55.9
120.9
56.3
36.1
103.0
58.0
47.4
155.3
84.4
89.2
38.1
59.7
195.0
28.4
14.0
14.9
Ni
0.1466
0.2610
0.5926
0.2753
0.3410
0.1328
0.3049
0.7019
Trace
0.6938
0.7586
0.3194
0.3650
0.5524
0.3863
0.4063
0.6274
0.4741
0.7279
0.7640
0.2605
2.2535
1 . 1 490
0.5575
2.4883
1.0872
1.1713
1.1963
1.6172
2.1662
1.0279
1.2699
0.8241
37
-------
Table A-6 (continued). WEEKLY AVERAGE BaP AND METAL CONTENTS OF AIR AT THREE
SAMPLING SITES IN CHARLESTON, S.C., NOVEMBER 1, 1972, TO MARCH 31, 1973
(ng/m3)
Date and Site
1/14/73-1/19/73
Radio station
Fire station
Post office
1/22/73-1/27/73
Radio station
Fire station
Post office
1/28/73-2/4/73
Radio station
Fire station
Post office
2/4/73-2/11/73
Radio station
Fire station
Post, office
2/11/73-2/17/73
Radio station
Fire station
Post office
2/20/73-2/24/73
Radio station
Fire station
Post office
2/26/73-3/3/73
Radio station
Fire station
Post office
3/4/73-3/10/73
Radio station
Fire station
Post office
3/12/73-3/17/73
Radio station
Fire station
Post office
3/19/73-3/24/73
Radio station
Fire station
Post office
3/26/73-3/31/73
Radio station
Fire station
Post office
BaP
1.2409
1.1366
0.6491
0.2835
1.0435
0.6684
0.1568
0.3759
0.6818
0.4581
0.8163
1.4852
0.8396
0.6832
0.6346
0.6925
1.8191
0.8601
0.2740
0.3484
0.1995
0.1329
0.2711
0.2658
0.3553
0.3914
0.5037
0.4509
0.2462
0.4263
0.0028
0.1693
0.2367
Be
0.0686
0.0408
0.0061
0.0661
0.0241
0
0.0473
0 .
0.0216
0.0108
0.0109
0.0186
0.0069
0.0644
0.0238
0.0057
0.0332
0
0.0357
0.0886
0.0848
0
0
0.0624
0.0064
0.0230
0.0434
0.0529
0.0337
0
0.0431
0.0060
0.0429
Cd
0.2791
0.2898
0.1478
0.2495
0.2526
0.3091
o.??:4
u.2282
0.1507
0.1569
0.1966
0.2996
0.1994
0.2525
0.2008
0.1382
0.3390
0.1842
0.1667
0.4818
0.1574
0.2069
0.3764
0.1299
0.1308
0.2296
0.1581
0.1488
0.2487
0.1313
0.1267
0.1590
0.1925
Cr
0.1234
0.9226
0.4368
1.0641
0.6015
0.8329
1.8923
0.2381
0.1762
0.2543
0.9614
0.5476
1.5573
0.7361
0.9413
0.7225
0.8753
0.6087
2.0752
0.8103
0.3088
2.1474
1.3573
0.2646
0.3701
0.7043
0.4092
0^8399
0.8345
0.2188
0.1534
0.2451
0.3219
Cu
33.3
19.6
11.4
35.2
10.0
24.9
30.2
11.9
12.3
24.4
06.7
22.2
20.9
35.3
15.7
16.0
61.9
16.0
47.0
120.5
38.2
39.3
40.3
34.3
17.2
16.8
41.2
53.2
55.8
58.2
28.2
23.5
23.5
Ni
1.3711
1.2084
0.7392
1.1375
1.0908
1.0734
0.6899
0.9593
0.6369
0.4871
0.8740
1.5188
0.7596
0.6809
0.4706
0.6910
0.5520
0.6007
0.9052
1.1497
0.5147
0.5579
0.9119
0.2963
0.4723
0.6890
0.5084
0.4827
0.6018
0.5002
0.6655
0.5167
0.3472
38
-------
APPENDIX B.
ANALYSIS OF BENZO [a] PYRENE
IN SOIL SAMPLES
Prepared by
Research Triangle Institute
Research Triangle Park, North Carolina
January 1973
39
-------
FOREWORD
This report, which is presented as Appendix B, was prepared by members of
the Engineering Division of the Research Triangle Institute, Research Triangle Park,
North Carolina, for the National Source Inventory Section, Applied Technology
Division, Environmental Protection Agency.
The work was performed by the Instrumentation, Measurements, and Device
Research Department under the general supervision of Mr. J.B. Tommerdahl, Mana-
ger. Mr. C.E. Decker was Project Leader. Mr. D.E. Wagoner provided the technical
support required to complete this task and is the author of this report. Mr. R.T.
Pickett assisted in the analysis of benzo[a]pyrene in soil samples.
Analysis of soil samples was conducted under two EPA purchase orders (Number
2PO-68-6424 and Number 3-02-00692) , which were completed separately but which
were combined for presentation in this document. The same sample handling and
analytical procedures were used in both projects.
INTRODUCTION
The objective of this work was to analyze soil samples for their benzo[a]pyrene
(BaP) content using EPA's current method for analysis of BaP in extracts of airborne
particulates. Nine samples were initially submitted by Mr. Carl V. Spangler, Project
Monitor, under Purchase Order No. 2PO-68-02-6424; subsequently, several samples
were extracted and analyzed for a second determination. Additional soil samples
were analyzed under Purchase Order No. 3-02-00692. Modifications were applied to
the analytical procedure as required to improve separation of BaP from other organic
materials by thin-layer chromatography. Alternative procedures other than Soxhlet
extraction with benzene for removal of BaP from the soil samples were investigated
at the request of the project monitor.
ANALYTICAL PROCEDURE
The procedure employed for the analysis of BaP in soils was adapted from EPA's
current procedure for analysis of BaP in extracts of airborne particulates. This
procedure involves Soxhlet extraction of BaP with benzene, thin-layer chromato-
graphic separation of BaP from other organics, and quantitation of BaP by measure-
ment of fluorescence emission at 540 nanometers (nm) .
40
-------
Alternate Procedure for Shaker Extraction:
A. Place 50 g of soil in a 250 ml extraction flask with 60 ml benzene.
B. Shake for 1 hour and filter through glass wool into a 250-ml beaker.
C. Add 60 ml of benzene, shake for 30 minutes and filter as in "B."
D. Add 60 ml of benzene, shake for 15 minutes and filter as in "B ."
E. Add 25 ml of benzene; hand shake for 2 minutes and filter; continue with
the following Step 5.
5. Sample volumes are reduced in 250-ml beakers on a steam bath (to approxi-
mately 25 ml) . The steam bath must be located in a hood.
6. Samples are filtered through sintered glass funnels into previously weighed
test tubes.
7. Samples are dried in a vacuum oven at 60°C. Place each test tube in a beaker
o± water.
8. Any remaining moisture is removed by placing test tubes in a desiccator.
Weigh test tubes to obtain extracted sample weight.
9. Prepare the A12O3 thin layer plate as follows:
A. Remove 1/16 in. of alumina from each edge of the plate.
B . Scribe a small dot 1/4 in. from each edge and 7/8 in. from bottom.
C. Scribe a line across the plate 15 cm above these two dots.
D. Activate plate at 110°C for 1 hour.
10. Samples are dissolved in methylene chloride. 1.0 ml/25 mg of residue. Let
the samples sit for a minimum of 15 minutes. Roll the tubes to get all of the
sample from the sides .
11. One hundred ml of hexane is poured into tanks. Alternate tank solution:
20 ml benzene, 80 ml pentane.
12. The template is positioned over the glass plate so that the bottom edge casts
a shadow connecting two previously scribed dots on plate. Pour sample into
5-ml beaker. Use thin-layer chromatography spotting capillaries, 5 pi, and
condition before use by drawing 2 to 5 ml portions and emptying into a
laboratory tissue. Spot 80 ul (or 40 yl) of methylene chloride solution of
the extracts by placing a row of eight slightly overlapping spots on the
43
-------
Initial Procedure for Analysis of Benzo[a]pyrene
in Soil Samples as Supplied by EPA
1. Soil samples will be thoroughly mixed, freed from foreign material, and
dried at room temperature and normal pressure. One-hundred-fifty grams
(g) of the soil will be crushed thoroughly and divided into two 50-g portions
and five 10-g portions. Samples will be placed in filter cups that have been
previously extracted with diethyl ether under reflux.
2. An appropriate sample (10- or 50-g sample) will be refluxed for a period of
8 hours with benzene in a Soxhlet apparatus.
3. The resulting benzene solution will be reduced in volume on a steam bath
located in a hood.
4. The sample will be filtered through a sintered glass funnel into a previously
weighed test tube, dried under vacuum at 60°C, and weighed.
5. The benzene soluble fraction will be dissolved in methylene chloride (1.0
milliliter/25 milligrams of residue) . The test tube will be capped and sample
allowed to dissolve.
6. Eighty microliters of the methylene chloride solution will be spotted on an
aluminum oxide (A^C^) thin layer plate that has been activated at 110°C and
stored in a dessicator over phosphorous pentoxide. In addition to the sample,
a standard of benzo[a]pyrene will be run simultaneously.
7. The plate will be placed in a developing tank with the appropriate solvent
to effect a complete separation of BaP from the other constituents.
8. The plate, while still moist, will be removed and placed in an ultraviolet
viewing chamber. The sample spot corresponding to BaP plus the standard
will be scribed.
9. The spots will be scraped onto glassine paper and transferred to a dry,
screwcap vial tube containing 8 ml of pentane/acetone (95/5) . The vials will
be shaken (> 10 minutes) and allowed to settle. The supernatent liquid will
be transferred through a coarse porosity sintered glass funnel into a dry,
screwcap test tube. The adsorbent will be washed with additional portions
of pentane/acetone and all portions combined. Test tubes will be evaporated
(in water bath at 35 to 40°C) in a vacuum oven at approximately 50 torr, until
the solvent is evaporated. The sample will be dried in a vacuum oven with-
out water bath for at least 10 additional minutes.
41
-------
10. The fluorimeter will be blanked with concentrated sulfuric acid. Five ml of
concentrated sulfuric acid will be added to each sample tube and 10.0 ml to
the standard test tube, The test tubes will be stoppered and shaken to allow
bubbles to escape (approximately 5 minutes) and transferred to a cuvette.
The fluorescence is read in a filter fluorimeter. Fluorimeter setting will be as
follows: excitation wavelength = 470 nm; emission wavelength = 540 nm .
Identification will be by comparison of relative Rf values of pure BaP versus Rf
values of the experimental samples. The concentration of BaP in soil will be deter-
mined by comparing the fluorescence of the sample solution with the fluorescence of
a sulfuric acid solution of a standard amount of BaP that has been carried through the
thin-layer chromatograph and elution steps.
Adapted Procedure
The following sample handling and analytical procedures were initiated after dis-
cussion with the project monitor. (Preliminary precautions included storing samples
in boxes protected from light after delivery to RTI by EPA and storing samples, when
not under investigation, in a refrigerator - this includes interim steps in the analysis.)
1. Soil samples are dried at room temperature in a dry box containing silica gel
with steady flow of purified dry nitrogen.
2. Rocks and other foreign material are removed from soil and crushed thoroughly
by mortar and pestle.
3. Soil is weighed in extraction thimbles that have been previously extracted
with diethyl ether under reflux. Soil is weighed on glassine paper and
poured directly into extraction flasks if the shaker extraction technique is
used.
4. Samples are extracted on Soxhlet apparatus for 8 hours, and the procedure
is continued to Step 5 unless alternate shaker procedure is used.
In an attempt to shorten the time of extraction, a mechanical shaker was employed
in the initial separation. In all cases the amount of organic residue extracted was
greater by the Soxhlet method. This increased organic loading also puts a greater
burden on the thin layer separation. Because of the many factors in the analysis, the
Soxhlet method of extraction should be used for completeness of recovery when many
varied soil sample types are being extracted. The mechanical shaker extraction may
be employed when soil samples from the same area are being evaluated on a relative
basis. The shaker extraction procedure utilized in this project is described in the
following paragraphs.
42
-------
Table B-l (continued). RESULTS OF SAMPLE ANALYSIS FOR SOILS TAKEN IN WEST VIRGINIA
(COAL REFUSE BANKS) AND IN CHARLESTON, S.C.
EPA sample
Charleston, S.C.
C-38A-B
C-39A-1-B
C-39A-2-B
C-39A-3-B
C-398-B
C-45-B
C-46D-B
C-47-D-B
C-52-B
C-52C-B
C-53B-B
C-62A-B
NSC-1773
26C-B
27C(CHK-39A2)-1B
-2B
-3B
-4B
-5B
-6B
28C-37A'
RTI
sample no.
(1374-124 )
-18
-18(2)
-19
-19(2)
-20
-20Ae
-21
-22
-23
-24
-25
-26
-27
-28
-29
-30
-31
-32
-33
-34
-35
-36
-37
-38
Peak position
Sample
524
523
539
539
543
539
530
522
521
541
521
542
522
543
519
523
523
536
531
531
539
530
525
524
Standard
542
540
539
540
540
542
542
542
542
541
541
541
542
542
542
544
544
544
542
542
542
544
544
544
AA
ia»
17
Oa
la
3a
39
12
20
21
0*
20
la
20
la
23
21
21
18
11
11
3a
14
19
20
Concentration,
ug/kg
823
2,502
953
829
97,612
64,600
1,875
150
1 ,095
33,528
489
27,709
231
3,520
255
178
206
827
1,518
809
2,354
7,087
1 ,920
3,099
Eluting solvent
Benzene/pentane
n-Hexane
Benzene/pentane
n-Hexane
Benzene/pentane
Benzene/pentane
. n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
Development Method
Single Shaker
Multiple Soxhlet
Single Shaker
Multiple Soxhlet
Single Shaker
Single Shaker
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
Multiple Soxhlet
-------
Table B-l (continued). RESULTS OF SAMPLE ANALYSIS FOR SOILS TAKEN IN WEST VIRGINIA
(COAL REFUSE BANKS) AND IN CHARLESTON, S.C.
EPA sample
Charleston, S.C.
30C-43B
32-C(48-0)
33-C(49-B)
C-36 39B
C-38 25A
C-39 110-A
C-39 110-B
C-40-46A-B
C-43.2A
C-49,67A
C-47J6A
RTI
sample no.
(1374-124- )
-39
-40
-41
-42
-43
-44
-45
-46
-47
-48
-49
Peak position
Sample
522
523
523
525
523
525
524
523
522
521
521
Standard
537
537
537
537
537
537
540
540
540
544
544
AX
15
14
14
12
14
12
16
17
18
23
25
Concentration,
yg/kg
500
912
982
910
944
1,408
1,807
2,595
5,489
201
378
Eluting solvent
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hexane
n-Hj/ane
Development
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Multiple
Method
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
Soxhlet
CO
Peak comparison z 4 nm.
Second spotting, both from methylene chloride sample of Soxhlet extract and spotted on same plate with two standards.
Second spotting, both from methylene chloride sample of Soxhlet .extract.
Third spotting, both from methylene chloride sample of Soxhlet extract spotted on same plate with two standards.
eSecond spotting from methylene chloride sample of mechanical shaker.
f1.04 yg added.
92.0 pg added, read at 538.
-------
Results indicate l.hal Ihc opliinuin TLC solvrnl syslrm ol n hi-x.mr i:; |i rclcr rci I
over the 20/80 benzene-pentane mixture. Multi-development was required over 95
percent of the samples analyzed. The use of a scanning versus a filter fluorimeter
is required because of the large number of interfering species that must be overcome
in the separation. The scanning fluorimeter allows one to compare the fluorescence
emission curves of the BaP standard and sample, which were separated vn situ from
the Al2O^ plate. The excitation wavelength can also be varied in an attempt to
resolve the fluorescence emission spectra. In the cases in which the fluorescence
emission of the sample differed more than 4 nanometers, the concentration of the
sample was calculated on the basis of the relative fluorescence emission at the wave-
length of the standard separated from the aluminum oxide plate; consequently, the
results reported in these cases can be compared to other results in which only a
filter fluorimeter is used.
The fluorescence method for B in soil requires an involved separation by thin
layer chromatography. A poor separation or inaccurate marking of the BaP from the
organic constituents on the thin layer plate will usually produce very high results.
RECOMMENDATIONS
1. The analysis of soil samples for BaP should be done with a scanning spectro-
fluorimeter. A TLC scanner affixed to the spectrofluorimeter would permit
location of the desired spot on the thin layer plate and quantitation without
removal of the substrate from the plate.
2. Following extraction and evaporation of solvent, the residue should be
dissolved in an organic solvent such as pentane and extracted with a dilute
acidic solvent to remove basic organic interferences.
3. An extensive evaluation of the TLC system should be undertaken to determine
the optimum solvent system and adsorbent. The matrices of various soil
samples are quite different.
4. Further investigation should be devoted to development of a high-speed
liquid chromatographic procedure for separation and detection of BaP.
49
-------
TECHNICAL REPORT DATA
(Please read /Hstructions on the reverse before completing)
1. REPORT NO.
EPA-450/2-75-004
4. TITLE AND SUBTITLE
Benzo[a]pyrene and Trace Metals in Charleston, S. C.
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
June 1975
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Carl Spangler and Noel de Nevers
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park. NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Charleston, S. C. , along with some other cities in the Southeastern Coast
and Gulf Coast area, has an anomalously high incidence of deaths resulting from lung
cancer - about 50 percent higher than the national average. Benzo[a]pyrene (BaP) and
trace metals are widely suspected of being causative agents in lung cancer. A survey
of BaP and trace metals in the ambient air in Charleston reveals, however, that the
air concentrations are lower than the national averages, generally falling in the
0.01 to 29 percentile among American cities.
To test the view that atmospheric concentrations of BaP can readily be inferred
from soil concentrations, soil samples were taken in Charleston at sites roughly corre-
sponding to the area in which air was subject to testing in the air sampling program.
There are few values from the United States with which to compare the Charleston soil
values, though the soil concentrations of BaP there are somewhat higher than observed
urban values in the Soviet Union, for example. Accordingly, from the limited data
available, the Charleston soil values of BaP do not appear extraordinarily high.
Thus, it seems safe to infer that the abnormally high death rate resulting from
lung cancer is not due to higher-than-normal exposure to the agents addressed.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Benzo[ajpyrene
Trace metals
Health effects
Pollutant measurements
13. DISTRIBUTION STATEMENT
Jnlimited
19. SECURITY CLASS {This Report)
Unclassified
21. NO. OF PAGES
58
20. SECURITY CLASS (This pane)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
50
------- |