-------
Figure 3
2378-TCDD in Surface Soils
Four Midwestern Sites
2.5
2.0
Q
O
O
I 1.0
CO
CM
0.5
N
Geometric Mean
Arithmetic Mean
//
<t-
&
#/'
/
CDC LEVEL OF CONCERN
FOR RESIDENTIAL SOILS
-------
Figure 4
2378-TCDD in Surface Soils
Dow Chemical-Midland Plant
1 2
1
0L
2378-TCDD (ppb) in
SURFACE SOILS
1.0 3.0
L
3.5 4.0
Concentration (Parts Per Billion)
4.5 35.5
36.0
-------
TCDD, while the random site had a concentration of 36 ppb. All other samples
had less than 0.5 ppb 2378-TCDD.1 These results are consistent with the
results of more comprehensive independent sampling conducted by Dow Chemical
that confirmed a few areas with relatively high concentrations above 1 ppb near
chlorinated phenols production areas and widespread site contamination by
2378-TCDD in the range of a few parts per trillion up to 1.0 ppb. 4/
One value above 1.0 ppb (2.03 ppb) was found on the east perimeter of the
Dow plant adjacent to the Union Carbide Corporation - Linde Division plant
(Figure 5). Supplemental sampling by Dow Chemical indicates only a limited
area of high concentration. 5/ As a result of discussion with USEPA Region V,
Dow Chemical has enclosed a Targe area around the site with fencing to prevent
inadvertent public access.
As shown in Figure 6, the residential and public use area samples with the
highest concentrations of 2378-TCDD were found to the north-northeast of the
Dow plant. That area is downwind of generally prevailing winds with respect to
the plant. 2378-TCDD data from public use and residential open areas with
winds generally prevailing from the Dow plant (downwind, 0-90ฐ) were compared
with all other Midland open area data (90-360ฐ). This comparison shows a
statistically significant difference at the 80* confidence level (t-test),
indicating higher deposition in the downwind area. The data for all residantial
sites in Midland show that downspout and roof dripline samples had an average
concentration higher than the average concentration for samples obtained from
open yards by a factor of about 3.4. The distribution of these data are such
that there is a statistically significant difference in the open area and
downspout data at the 95% confidence level (t-test). These data are consistent
with the hypothesis that air emissions from Dow Chemical are the likely source
of dioxin found in Midland outside the plant perimeter.
Estimates were made of upper bound (99th percentile) concentrations of
2378-TCDD that might be expected in Midland outside the Dow Chemical plant. The
distribution of 2378-TCDD in Midland soils is displayed in Figure 7. The data
appear to be distributed in a log-normal fashion, as are data for Dow plant
ilJSEPA, the state of Michigan, and Dow Chemical have entered into a
consent order pursuant to Section 106 of the Comprehensive Environmental
Response, Compensation, and Liability Act (Superfund) which requires Dow
Chemical to cover two of the three high concentration areas with asphalt. The
purposes of the covering are to prevent worker exposure to high levels of
2378-TCDD and to prevent possible migration of 2378-TCDD off-site by windblown
dusts. The third area of high concentration (near the hazardous waste inciner-
ator) is not generally accessible and will be considered as part of a long-term
remedial investigation of the Dow plant which will likely include supplemental
soil sampling in and around the perimeter of the plant and ground water quality
evaluations.
14
-------
Figure 5
2378-TCDD in Surface Soils
Perimeter-Dow Chemical-Midland Plant
I 3
I 2
2378-TCDD (ppb) in
SURFACE SOILS
f
0.5 1.
Concvntr
1.5
ition (Parts Per Billion)
2.0
-------
Figure 6
2378-TCDD
SURFACE SOIL CONCENTRATIONS
MIDLAND. Ml
2378 TCDD
(PARTS PER BILLION)
1+
0.5-1.0
0.1 - 0.5
0.05 0.10
0.001 - 0.05
OPEN AREA SAMPLE
DOWNSPOUT SAMPLE
25.0
36.0)
DOW
CORNING
DOW
CHEMICAL
COMPANY
CONSUMERS POWER CO.
COOLING PONDS
-------
o
w
0
1
9
8
(0
I 6
(0
co 5
4
3
2
1
0
Figure 7
2378-tCDD in Surface Soils
:ity of Midland
2378-TCDD (ppb) in
SURFACE SOILS
.05 .10 .15 .20
Concentration (Parts Per Billion)
.25
-------
perimeter (Figure 5). Although the quantity of data is not sufficient to
statistically determine the distribution of these data (Chi-square test),
log-normal statistics were used to estimate 99th percentile surface soil
concentrations that may be expected in public use and residential areas, at
residential downspouts, and near the perimeter of the Midland plant. The use
of lo^-normal statistics provides higher estimates of expected 99th percentile
concentrations than would normal statistics. Accordingly, 99% of the open area
surface soil concentrations of 2378-TCDD in Midland away from Dow Chemical plant
perimeter are expected to be less than 0.21 ppb; 99% of the residential downspout
area surface soil concentrations are expected to be less than 0.43 ppb; and 99%
of the surface soil concentrations encountered near the perimeter of the plant
are expected to be less than 3 ppb. Based upon these values and the public
health reviews of the data (Appendix E), further efforts to more completely
characterize 2378-TCDD surface soil concentrations in the city of Midland do
not appear warranted. The probability of finding significant levels of
2378-TCDD (e.g., greather than 1 ppW away from the perimeter of the Dow
Chemical plant appear to be quite low. Supplemental sampling near the perimeter
of the Dow Chemical Midland Plant to detect additional areas that may have high
concentrations will be considered as part of the remedial investigation of the
facility.
The Dow Chemical hazardous waste incinerator is reported to be the most
significant current source of 2378-TCDD air emissions from the Midland plant. 4/
Estimated current annual emissions based upon emission sampling programs
conducted by Dow Chemical are about 0.33 gram. 4/ Estimated annual emissions
from several other documented sources (Midland plant powerhouse; 564, 1009 and
1058 Building burners; 2,4-D chlorinolysis vent; 2,4-D process scrubber vent;
and dichlorophenol process scrubber vent) total less than 0.02 gram. ฃ/ Dow
Chemical has also "fingerprinted" the dioxin found in Midland soils with various
in-plant sources through isomer-specific analyses of tetrachlorodioxin Isomers
(TCDDs). The conclusion made by the company from this work is that TCDDs found
in Midland area soils are similar, though not identical, to TCDDs found in
incinerator ash and incinerator stack particulates. 4/
The computed geometric mean concentration of 2378-TCDD from Midland public
use and residential areas is 0.03 ppb (Table 1). Thirty-seven percent of the
27.9 square mile surface area of the city of Midland is estimated to be
impervious (e.g., paved areas, building roofs). 6.7/ Assuming the depth of
dioxin contamination is about six inches (see sample D-2-6, Table 2, Appendix D),
and there is no dioxin on impervious surfaces, the mass of 2378-TCDD in Midland
away from the perimeter of the Dow Chemical plant is estimated at about 500 gm
(1.10 Ibs). From these data, it is obvious that estimated current annual
emissions from Dow Chemical cannot acccount for the mass of dioxin found in
Midland soils. Accordingly, past emissions from Dow Chemical waste incineration
operations and other processes must account for the vast majority of the
2378-TCDD currently found in Midland soils. Based upon personal observation,
it appears possible that windblown dusts from contaminated surface soils in the
plant may have also contributed to contamination outside the plant.
18
-------
Dow Chemical began Incinerating chemical process wastes at Midland during
the 1930s. 4/ Incineration operations were upgraded from time to time in
response to changing production at the plant and more restrictive air pollution
control regulations. The combustion conditions and air pollution control
systems for the hazardous waste I incinerator were last substantially upgraded
during the 1978 to 1981 period. 4/ With changing production operations at the
plant, changes 1n incinerator operations, and without actual emission data, it
is not possible to estimate annual emission rates before the 1978-1981 period
with a high degree of confidence However, based upon incinerator practice at
Dow Chemical, It is likely that higher emissions occurred prior to 1978. Average
annual emissions from 1930 to 1
account for the current levels (
was computed assuming steady stat
in the environment of 12 years.
978 of about 40 grams would be necessary to
f 2378-TCDD in Midland soils. This estimate
B conditions and with a half-life for 2378-TCDD
8/ A wide range of estimated average annual
emission rates could be made. TlRe magnitude of the emission rate estimate is
heavily Influenced by estimates of the mass of dioxin in Midland soils, and
assumptions regarding the half-life of 2378-TCDD in soils and other factors.
Depending upon assumptions made, one could conclude that the estimated current
mass of 2378-TCDD in Midland soi
(longer half-life and correspon
estimated mass of 2378-TCDD is at
half-life and corresponding higher emission rates).
s is in the upper range of historical values
ling lower emission rates), or the current
the lower range of historical values (shorter
Given the order-of-magnitude nature of
historical data, it is not possible to estimate
these estimates and lack of
historical exposures of 2378-TCDD
of certainty. Nonetheless, these estimates
n in Midland from soils were somewhat higher
Public exposure to dioxin from Midland
future. Atmospheric emissions from Dow
ignificantly reduced, and the current soil
burden should continue to decrease slowly over time.
to the public with any degree
suggest public exposures to diox
in the past than current exposures
soils will likely decrease in ;he
Chemical have apparently been ;
lard
B. Other PCDDs and PCDFs
Tables 7, 8, and 9 of Appendi
selected samples from the Midi
natural areas, respectively.
certain qualifications and limitations
other PCDDs and PCDFs. Samples
relatively low levels (<0.5 ppb)
These data appear to be consistent
the Minnesota natural areas (Tabl
in Middletown, Ohio (Table 8). Li
'ligh ppb levels of several PCDDs
nore extensive data developed bj
lost samples obtained from Midli
n'gher levels of PCDDs and PCDFs
the comparison sites. The dist
sublic use area samples is generlally
x D present data for other PCDDs and PCDFs for
area; Middletown, Ohio; and the Minnesota
Reference is made to Appendix C, page 3, for
regarding analyses of soil samples for
obtained upwind from Midland (Table 7) had
of HyCDDs and OCDD and no detectable PCDFs.
with "natural background" levels found in
9) and levels found in most samples obtained
mited data for the Dow Chemical plant indicate
and PCDFs. These results are in the range of
' Dow Chemical for the Midland plant. 4,9/
nd public use areas contained significantly
than found at the Midland upwind sites and at
ibution of PCDD homologues in the Midland
consistent with that found in hazardous
19
-------
waste and chemical waste incinerator emission samples. 9.10/ Relatively low
levels of 2378-TCDD were found compared to relatively high levels of HyCDDs and
OCDD. These data further support the hypothesis that air emissions from Dow
Chemical have been the primary source of dioxins found In Midland.
The data for the Middletown, Ohio, upwind samples appear to be somewhat
anomalous (Table 8). Samples 13377 and 13400 were duplicate field samples
obtained at the same location. The data for 2378-TCDD show excellent agreement
(0.003 ppb and 0.004 ppb). However data for HgCDDs, HyCDDs, and OCDD do not
agree. Also, as discussed below, data for other toxic organic pollutants for
these samples are not in complete agreement.
Bumb, et.al.. suggest that chlorinated dioxins are by-products of combustion
of most types of organic material and are widespread in the environment, ll/
Czuczwa, et.al., suggest the most significant source of PCDDs and PCDFs to tFe
atmosphereTTsthe combustion of domestic and chemical wastes that contain
chlorinated compounds. 12/ USEPA has reported findings of 2378-TCDD and other
TCDDs in municipal waste combustor emissions, and there have been several
studies completed in Europe with similar findings. 13,14/
Data from this study demonstrate the highest levels of 2378-TCDD and other
PCDDs were found at and near the Dow Chemical - Midland Plant where 2,4,5-tri-
chlorophenol and derivatives were manufactured and where significant amounts of
chlorinated compounds and other chemical process wastes have been incinerated
over a long period of time. Trace levels of 2378-TCDD and other PCDDs found in
a few of the samples obtained near other industrial sites with processes and
combustion operations different than those at Dow Chemical may be related to
fallout from combustion. The distribution of the data at the other industrial
sites do not suggest a dominant local source. Four natural areas had no
detectable 2378-TCDD and only trace levels of HyCDDs and OCDD. These low
levels may also be related to combustion.
The results of this study do not suggest widespread contamination by 2378-
TCDD, other PCDDs and PCDFs at levels that may be significant with respect to
public health or adverse environmental impacts. EPA's National Dioxin Study
should provide sufficient data to characterize the extent of 2378-TCDD contami-
nation in the United States and more information about dioxins and combustion
(Tier 4).
C. Other Toxic Chemicals
Selected soil samples from the Midland area; Dow Chemical; Middletown, Ohio;
and the Minnesota natural areas were subjected to broad scan organic chemical
analyses for acid compounds, base/neutral compounds, and pesticides, including
PCBs at low ppm and high ppb detection levels. These data are presented in
Appendix D, Tables 11, 12, and 13. The results are somewhat unexpected in
that relatively few compounds were quantitatively detected in most samples,
particularly those from inside the Dow Chemical plant. The major distinguishing
20
-------
feature of these data 1s the hit
carbons (PAHs) found In Middlet
Middle town Works steel plant (Tab
by-product coke plant operated
for duplicate field samples pbt
Middletown appear somewhat diffe
relatively low In concentration
had 13 PAHs detected In the 1-2
six detected. Not all of the det
Phthalate compounds were de
levels) in many samples at and si
Including some of those from Minne
potential low-level widespread
environment or potential low-lev
addition to the data reported
organic compounds were tentativel
0. 2378-TCDD in Municipal Se
USEPA did not achieve the qua
for 2378-TCDD analyses of the Mid
sewage sludge samples. The com
(Appendix C, Appendix D, Tables
reported 0.021 ppb (49% recovery)
(69% recovery). Dow Chemical d
data Indicate low-level contamina
surface runoff entering the sewe
these data about comparable leve
municipalities. Lamparski, et.a
(0.002, 0.011, 0.016 ppb) from
sewage sludge sold as a fertili
TCDDs found In those samples is su
resulting from the chlorination
potable waters and wastes. 15/
plant include intermittent chTori
chlorination of the trickling f
practices could result in chlor
reacting with the sludge which is
system. Sufficient data are no
findings of 2378-TCDD in municipa
The amount of 2378-TCDD in Midi an
of the sludge as a soil conditio
conditioner for tree plantings b
used in the past by some Midland
to 2378-TCDD from sewage sludges
population in Midland.
ler incidence of polynuclear aromatic hydro-
wn samples taken near the ARMCO, Inc. -
e 12). The likely source of the PAHs is the
y ARMCO, Inc. As noted above, the data
ined from the upwind sampling station near
ent. Although the analytical results are
or both samples, the duplicate field sample
Dm range, while the original sample had only
cted compounds were the same in both samples.
ected (below contract laboratory detection
htly above levels found in laboratory blanks,
ota natural areas. These data indicate either
istribution of phthalate compounds in the
1 phthalate contamination of samples. In
Tables 11-14 for toxic pollutants, other
identified in certain samples (Tables 15-18).
age Sludges
ity assurance objective for surrogate recovery
and, Michigan and Middletown, Ohio, municipal
lex sample matrices caused poor recoveries
4). For the Midland sludge sample, USEPA
and Dow Chemical reported 0.019 ppb 2378-TCDD
not analyze the Middletown sample. These
ion of the Midland sludge that may be due to
age system. No statements can be made from
s in sewage sludge from Middletown or other
report 2378-TCDD at similar low levels
samples of Milorganiteฎ, a dried municipal
er. 15/ The source of 2378-TCDD and other
gestecTto be the condensation of chlorophenols
f naturally occurring phenolic compounds in
Operations at the Midland sewage treatment
ation of final effluent filters and occasional
Hers to control excess growth. 16/ These
lation products and free available chlorine
Itimately processed in a thermal wet oxidizing
currently available to fully resolve the
sewage.
surface soils cannot be accounted for by use
er. The sludge is currently used as a soil
the Midland parks department and had been
esidents for similar purposes. 17/ Exposure
not likely to be significant for the general
21
-------
VII. CONSIDERATION OF PUBLIC HEALTH RISKS
All of the data from this study and supplemental 2378-TCDD data for
Tlttabawassee River fish and 2378-TCDD ambient air data in and around the Dow
Chemical plant were forwarded for review to the U. S. Department of Health and
Human Services, Public Health Service, Centers for Disease Control (CDC), and
USEPA's Chlorinated Dioxins Work Group (CDWG). Table 2 presents a summary of
the fish, ambient air, and soil data provided to CDC. Appendix E (Attachment 1)
presents the CDC review of health risks associated with exposure to 2378-TCDD,
other PCDDs and PCDFs and other toxic chemicals found in Midland soils,
Tittabawassee River fish and ambient air. The CDWG applied its recommended
"2378-TCDD equivalents" approach to public use area soil samples with the
highest levels of PCDDs and PCDFs to determine if those samples would indicate
potential public health risks due to higher chlorinated PCDDs and PCDFs (see
Appendix E, Attachment 2). The major conclusions from these reviews are
presented below:
1. Current levels of 2378-TCDD in Midland residential soils, including
downspout areas, are well below one part per billion and do not represent
an unacceptable public health risk.
2. Ambient air levels of 2378-TCDD, as characterized by limited available
data, are unlikely to be a risk from bioaccumulation through the
respiratory route.
3. The warning on consuming fish from the Tittabawassee River should be
continued.
4. The levels in soils of other PCDDs, PCDFs, and other chemicals all less
toxic than 2378-TCDD, do not represent an unacceptable public health
risk.
5. The available data represent current soil levels and the CDC can make
no comment about past environmental levels and potential risks that
may have existed in the past.
Potential public health risks will be reviewed again when complete data
from the multi-media investigations are available.
22
-------
TABLE 2
2378-TCDD
Summary of Recent Environmental Measurements
Midland, Michigan Area
Number
of
Measurements Range
Ambient Air* (March 1983-February 1984)
Total - all measurements
Dow Plant
Dow Plant Fence Line
City of Midland
*Data provided by Dow Chemical Company
CO
Tittabawassee River Fish (August- September 1983)
Carp - whole fish composite (5 fish)
Carp - individual skinless fillets
Catfish - skinless fillet composite (5 fish)
Smallmouth Bass - skin-on fillet composite (5 fish)
Walleye - individual skin-on fillets
Surface Soils (October-December 1983)
Dow Plant
Dow Plant Perimeter
Public Use and Residential Areas
Residential Areas*
Residential Downspouts*
10
2
6
2
1
25
1
1
5
15
9
17
8
8
0.010-0.21
0.019-0.022
0.010-0.21
0.019-0.16
190
12-530
75
5.1
2.8-5.1
0.01-25.0
0.01- 2.03
0.003-0.17
0.009-0.076
0.013-0.27
pg/m3
pg/m3
pg/m3
pg/m3
ppt
ppt
ppt
ppt
ppt
ppb
ppb
ppb
ppb
ppb
Arithmetic
Mean
0.061
0.021
0.064
0.090
50
3.9
2.38
0.74
0.037
0.025
0.104
Arithmetic
Standard
Deviation
0.069
0.075
102
0.8
6.41
1.43
0.042
0.022
0.103
Geometric
Mean
0.036
0.038
28.6
3.8
0.24
0.17
0.026
0.020
0.062
*Data for four additional sites available 12/15/84.
-------
REFERENCES
1. Dioxin Strategy, Office of Water .Regulations and Standards, Office of Solid
waste and Emergency Response, Dioxin Strategy Task Force, U.S. Environmental
Protection Agency, Washington, D. C., October 20, 1983.
2. State of Michigan and USEPA Region V Study of Dioxins and Other Toxic
Pollutants; City of Midland, Dow Chemical , and Michigan Rivers; quality
Assurance Project Plan, USEPA Region 71 Central Regional Laboratory,
April 1, 1984.
3. Jones, D. R., Craig, J., Blending and Aliquoting of Soil Samples, IIT
Research Institute, Chicago, Illinois, December 19134.
4. Point Sources and Environmental Levels of 2378-TCDD (2,3,7,8-tetrachlorodi-
benzp-p-dioxln) on the Midland Plant Site of the Dow Chemical Company and
in the City of Midland, Michigan, Dow Chemical Company, Midland, Michigan,
November 5, 1984.
5. Rio, J. M., Chairman Dioxin Initiatives Team, Michigan Division, Dow Chemical
U.S.A., Midland, Michigan to (Gary Amendola, USEPA Region V, Eastern
District Office, Westlake, Ohio) November 15, 1984, ALS, 3 pp.
6. Personal communication from James Schrader, Planning Department, City of
Midland, May 1985.
7. Combined Sewer Overflow Analysis Handbook for use in 201 Facility Planning;
Volume 1 - Procedures and Example Analyses, E. D. Driscol and Associates,
Inc., EPA Contract No. 68-01-6148, prepared for USEPA, Facility Requirements
Division, Policy Guidance Branch, February 1983.
8. Kimbrough, R. D., Falk, H., Stehr, P., Fries, G., "Health Implications of
2,3,7,8-Tetrachlorodibenzo dioxin (TCDD) Contamination of Residential Soil,"
Journal of Toxicology and Environmental Health, 14:47-93, 1984.
9. Point Sources and Environmental Levels of 2378-TCDD (2,3,7,8-tetrachlorodi-
TSenzo-p-dToxin) on the Midland Plant Site of the Dow Chemical Company and
in the city of Midland, Michigan (database), Dow Chemical company, Midland,
Michigan, November 5, 1985.
10. Review and Development of Chlorinated Dioxins and Emissions Data (draft),
Radian Corporation, Durham, North Carolina, EPA Contract No. 68-02-3513,
March 1983.
24
-------
REFERENCES (continued)
11. Burnt), R. R., Crunmet, W. B., Cutie, S. S., Gledhill, J. R., Hummel, R. H.,
Kagel, R. 0., Lamparski, L. L., Luoma, E. V., Miller, D. L., Nestrlck, T. J.,
Schadoff, L. A., Stehl, R. H., Woods, J. S., "Trace Chemistries of F1re:
A Source of Chlorinated Dioxins," Science, 210:385-390, 1980.
12. Czuczwa, J. M., McVetty, B. D., Kites, R. A., "Polychlorlnated D1benzo-p-
Dioxins and Dibenzofurans in Sediments from Si ski wit Lake, Isle Royale,"
Science, 210:568-569, 1984.
13. "Interim Evaluation of Health Risks Associated with Emissions of Tetra-
chloroinated Dioxins from Municipal Waste Resource Recovery Facilities,"
Office of the Deputy Administrator, USEPA, November 1981.
14. "TCDD Emissions for Municipal Waste Combustors," Office of Solid Waste and
Emergency Response, USEPA (memorandum), December 16, 1983.
15. Lamparski, L. L. Nestrick, T. J., Stenger, V. A., "Presence of Chiorodibenzo
Dioxins in a Sealed 1933 Sample of Dried Municipal Sewage Sludge,
Chemosphere, 13:361-365, 1984.
16. Personal communication with Merlin Lavkack, City of Midland, Wastewater
Treatment Plant, June 1985.
17. Personal communication with George Young, City of Midland, Wastewater
Treatment, June 1985.
25
-------
APPENDIX A
Soil Screening Survey at Four Midwestern Sites;
Draft Study Plan, September 22, 1983
-------
Draft
9/22/S3
SOIL SCREENING SURVEY
AT
FOUR MIDWESTERN SITES
INTRODUCTION
Limited data published by Dow Chemical suggest that dioxins are present in soils in the
city of Midland in the parts per trillion range and somewhat higher within the Dow
Chemical - Midland Plant boundaries. Dow Chemical has also reported limited data in the
same range for other cities. There are currently not enough data available to adequately
assess whether the levels of dioxins reported by Dow Chemical represent typical levels in
Midland or whether higher levels that may be of concern from a public health standpoint
may be found at certain locations. There are also not sufficient data available to
determine whether dioxin levels reported by Dow Chemical for Midland and other
locations are representative of "background" levels that might be found in almost any
location. Finally, there are no substantial soil data available for Midland and other
locations for several other toxic pollutants that are emitted or discharged from Dow
Chemical operations and other chemical and industrial operations. The design of this
study is based upon the premise that the primary method by which surface soils become
contaminated is from deposition of air emissions from point sources of the pollutants.
While other methods of soil contamination are possible (e.g., application of pesticides or
contaminated sewage sludges, land disposal of contaminated solid and liquid wastes)) air
deposition is the most likely method of contamination of surface soils in the study areas.
This study is planned to address some of the above questions. Although the study is
principally ah environmental study, it is being conducted to provide sufficient data to
state and federal health agencies such that they can determine whether there is a health
concern, and whether comprehensive health-related studies or remedial action are
warranted. For each city, several samples will be taken in areas with the greatest
likelihood of showing contamination. Thus, positive results should represent worst-case
conditions in each case and the need for any health-effects studies can be determined
accordingly. The results from this study will be evaluated by EPA, the Centers for
Disease Control, and the respective state agencies. After the results have been
evaluated, the results and all conclusions regarding possible health effects will be
presented to federal, state, and local elected officials prior to publication of any
information.
OBJECTIVES
The objectives of this screening study are as follows:
1. To provide sufficient data such that state and federal health agencies can determine
whether more comprehensive health-related environmental soil studies or remedial
clean-up actions are warranted. Should more detailed health-related studies be
required, supplemental sampling, analytical, and quality assurance protocols will be
developed and implemented. The results from this study should be compatible with
those that may be obtained in future studies.
2. To determine the levels of PCDDs and PCDFs at the parts per trillion level and other
toxic organic pollutants at the parts per billion level in soils from four midwestern
sites with contrasting levels and types of industrial development. These data will
-------
-2-
indicate whether levels of dioxins reported by Dow Chemical for several locations
represent "background" conditions; and allow for comparison of levels of dioxins and
several other toxic pollutants at different sites. Pollutants to be studied in addition
to PCDDs and PCDFs include PCBs, pesticides, acid pollutants (principally chlori-
nated phenols), and base/neutral pollutants (principally chlorinated benzenes and
polynuclear aromatic hydrocarbons). Data for these pollutants are necessary to assist
in the interpretation of the data for PCDDs and PCDFs and to focus ambient air
sampling, point source emission sampling, and in-plant soil sampling at Dow Chemical
which will follow this study. Volatile pollutants are not being studied since they are
not likely to be found on soils and because their physical/chemical properties
generally make them less persistent in the environment than other types of
pollutants.
3. To distinguish any patterns in soil contamination that may indicate whether and to
what extent the Dow Chemical - Midland Plant is a source of pollutant contamination
in Midland, Michigan.
4. The analytical objectives. for the compounds to be studied including limits of
detection, levels of quantitation, precision, accuracy, completeness and comparability
are specified in the analytical protocols and quality assurance plan for this study
presented in Attachment A.
THE FOUR MIDWESTERN SITES
1. Midland, Michigan
The,-city-of Midland has a population of about 35,000 people. The major industries are
the Midland Plant of the Michigan Division of Dow Chemical Company and the Dow-
Corning Silicone Products Plant. The Dow Chemical - Midland Plant falls within
Tiers 1, 2, 3, and * of the U.S. EPA National Dioxin Strategy in that 2,*,5-
trichlorophenol (2,4,5-TCP) was produced (Tier 1); 2,4,5-TCP was used to make
pesticide products (Tier 2); 2,4,5-TCP and derivatives were formulated into pesticidal
products (Tier 3)j and, the plant is also a combustion source (Tier 4). Based upon
limited data published by Dow Chemical and Dow Chemical's status as a former
producer and user of 2,^,5-TCP, dioxin contamination in the Midland area at some
level is expected. Based upon other production operations at Dow Chemical, other
toxic pollutants are likely to be present as well. Sewage treatment plant sludge from
the Midland sewage works has reportedly been used as a soil conditioner in Midland.
2. Site 2
The second midwestern site will be a municipality with significant chemical
production facilities and other combustion sources. The site will not be a Tier 1 or
Tier 2 site as defined by EPA's draft National Dioxin Strategy.
3. Site 3
Site 3 will be a municipality with significant industrial production facilities largely
unrelated to chemical manufacturing. These facilities are likely to include an
integrated steel plant. This site will also not be a Tier 1 or Tier 2 site.
-------
-3-
* Siteป
Site 4 will be a rural site that should be largely unaffected by any nearby industrial or
municipal point sources or nearby major combustion sources.
SAMPLING DESIGN
For the three municipalities with significant industrial point sources, sampling will be
conducted around the perimeters of the industrial plants as well as in public use areas such
as parks, playgrounds, and schoolyards. The public use areas will be selected to show
possible pollutant gradients with distance from the sources. For the rural site, soil
sampling will be conducted in open areas and at residences. Background samples well
away from each city will also be collected. For Midland, Michigan, and Site 2, sampling
will also be attempted at private homes located on concentric rings around the industrial
facilities. The results of the residential sampling will be used to determine if there is a
pattern or trend of pollutant concentrations with distance from the industrial facilities.
1. Sampling Sites
a. Midland, Michigan
Air quality modeling conducted by the Michigan Department of Natural
Resources, Air Quality Division, suggests that ground level concentrations of
contaminants released from the Midland Plant will be highest near the north
fence line of the plant. Thus, samples taken around the perimeter of the Dow
plant will be concentrated in that area as opposed to equally spaced samples
arosnd the entire perimeter. About 11 plant perimeter samples wUl be taken.
The plant perimeter samples are likely to have the highest contaminant levels.
Wind patterns at Midland indicate that winds are from the northwest to the
south-southwest about 55 percent of the time, Figure 1. Thus, heavier deposition
of pollutants emitted by Dow Chemical operations may be expected in an arc
from the north-northeast to the southeast. Sample sites at public use areas and
residences will be concentrated in this arc rather than randomly spaced
throughout the city.
To determine possible levels of exposure in public use areas, about 11 samples
from schoolyards, parks, and playgrounds will be obtained. Core samples will be
obtained at selected locations for possible future analyses.
Sampling at private residences will be conducted for two purposes: (1) to
supplement the public use area sampling described above, and (2) to determine if
there is a trend or pattern of pollutant concentrations with distance from the
Dow plant. Airborne particles tend to collect on exposed surfaces such as
building roofs. These particles then can be washed from the roofs by subsequent
rains, caught by the gutters, and deposited in the soils surrounding the
downspouts. Dioxins and related pollutants are known to adhere to soil and dust
particles. Since downspouts are good concentrators of airborne dusts and soil,
soil samples obtained in the vicinity of downspouts can serve as indicators of
environmental contamination. Additional soil samples from yard areas
unaffected by roof drainage will provide information on ambient levels of
-------
pollutants and will indicate possible contamination by routes other than air
deposition, such as direct application of pesticides or contaminated sewage
sludges. Efforts will be made to assure that the samples collected at the various
sites are comparable. Sampling conditions will be made as similar as possible for
each sample type and location.
The 20 residential sampling locations are arranged in four concentric rings at
varying distances from the center of the Dow Chemical - Midland Plant. The
residential samples taken from the first (innermost) and third concentric rings
will be analyzed first. The remaining residential samples will be collected, but
analyses will be contingent upon findings from analyses of the first set. At each
residence three samples will be taken: (1) a composite of individual samples
taken within one foot of the downspout, (2) a composite of individual samples
taken about three feet from the downspout, and (3) an open yard composite
sample taken away from the downspout. If higher concentrations of pollutants
are found at the outer (three foot) downspout ring than at the inner (one fool)
ring, this may indicate that emission rates from potential point sources are lower
now than in the past, or that pollutants are migrating away from the downspout.
Figure 2 presents tiie approximate sample site locations for the plant perimeter,
public use area, and residential sampling. (Note that a few of the residential
sampling sites are located within the boundaries of the Dow Corning plant. If
possible, downspout sampling from buildings will be conducted. If not, open area
samples will be obtained.) In addition, about 10 samples of dust or soil will be
collected from areas near the Dow Chemical plant gates likely to reflect track-
out of pollutants by foot or vehicular traffic. Four samples will be collected
along the railroad tracks near the plant boundary. Local background composite
soil samples will be collected at four locations upwind of the Dow plant site,
about 10 miles south and west of the facility. At two of the background
locations, downspout samples will be obtained; and, at the other two locations,
public use areas will be sampled. Core samples will be obtained at selected
locations for possible future analysis. The results from these samples would be
used to determine whether higher concentrations of pollutants are found below
the surface of the ground. The field sampling team will also be. instructed to
obtain a limited number of samples from home roof gutters and other areas
where airborne contaminants may collect. Finally, a sewage sludge sample will
be obtained from the Midland STP to indicate the possible impact of sludge
application on soil contaminant levels. Sample types and numbers are
summarized in Table 1.
b. Site 2
About ten composite soil samples will be obtained at Site 2. These samples will
be from the perimeter of the industrial complex and public use areas that may be
impacted by air emissions from local industries.
c. Site 3
Sampling at Site 3 will be similar to that for Midland, Michigan. Approximately
20 to 30 samples will be obtained. These will include perimeter samples around
the industrial facilities, public use area samples, and residential samples.
-------
-5-
d. Siteป
About four open area samples and one residential sample will be obtained at
Site 4.
SAMPLING METHODS
1. Residential Samples
Two samples will be collected adjacent to each residential downspout. These samples
will be composites of several small cores or plugs arranged in two concentric
semicircles extending out about one foot and three feet from the bottom of the
downspout. Drip lines may be sampled in a composite fashion at homes without
gutters as an alternative if a sufficient number of homes with gutters cannot be
found. At sites where splash pads are used to divert rainwater away from the
foundation of the building, a similar procedure will be used. The composite samples
will be taken at one foot and three foot rings away from the end of the splash pad.. A
third composite sample, comprising a similar number of cores selected from a grid
will be collected from an area at a distance from each downspout and isolated from
the effects of its discharge. The size of the grid will be dependent upon the size of
the yard. However, grid sizes between 3 meters by 3 meters and 10 meters by
10 meters will be sought. The composite sample from the open yard will be collected
first, followed by the three foot and one foot ring composite samples adjacent to the
downspout. Photographs will be taken of each sampling location. Information about
each sampling site will be obtained from the home owner.
Areas with bare soil will be sampled preferentially to grassed areas. Before sampling,
any grass will be trimmed to just above the. soil surface at the points to be cored.
Next, the cores, about two inches in diameter and several inches deep, will be
extracted. The top one-inch layer of each core will be shaved off and retained as the
sample. The remaining soil will be returned to the core holes, and repairs to the
lawns will be made by refilling the holes with topsoil and plugging with purchased sod.
As each soil grab sample is taken, the sample will be placed on a cleaned, disposable
aluminum pan and miscellaneous debris (e.g., twigs, roots, wood chips, stones,
pebbles, and other nonsoil material that can be distinguished) will be removed with
cleaned tweezers. Since the objective is to determine the concentration of pollutants
in soils which can be transferred to humans through ingestion, inhalation or dermal
contact, the miscellaneous debris must be removed to prevent biasing the analytical
results in either direction. A new aluminum pan and new tweezers will be used for
each composite sample. The sampling team will also use new disposable gloves at
each composite sample site. The grab sample will then be deposited in a new, cleaned
gallon paint can. This can will become the sample repository for each composite
sample. Each composite sample will be thoroughly blended in the laboratory to make
the composite sample as homogenous as possible. Aliquots of the composite samples
will be shipped blind to the analytical laboratories for analyses. Portions of the
composite sample will be retained for possible future analyses.
2. Public Use Area Samples
Composite soil samples will be collected from a 10 meter by 10 meter grid, at
appropriate intervals. Collection methods will be similar to those described above.
Areas with bare soil will be sampled preferentially to grassed areas.
-------
-6-
3. Dow Plant Perimeter Samples
Plant perimeter samples also will be collected as composites from a grid. However,
grid size and shape may vary depending upon the particular characteristics of the
sample location. The surface soil layer will be sampled at each site.
* Track-out Samples
Eleven track-out samples will be collected from the perimeter gate areas of the plant
with the greatest vehicular and pedestrian traffic. Emphasis will be placed on the
northwest corner of the plant, where production of pesticides and other chemicals
with the potential for dioxin contamination is believed to have occurred. These soil
or sediment samples will be collected from road gutters of low areas likely to receive
runoff that may have been contaminated with pollutants carried out of the plant.
Five of the 11 samples will be analyzed first, with analyses of the remaining six
samples contingent upon results from the first five. Four samples along the main
railroad lines entering and leaving the plant will be collected and held for possible
future analyses.
5. Core Samples
Core samples will be collected at selected locations for possible future analyses. The
core sample will be made up of three separate grab samples, the first beginning at the
ground surface to 6 inches in depth, the second beginning at 6 inches to 12 inches, the
third beginning at 12 inches to 18 inches.
SAMPLE-HANDLING AND EQUIPMENT
The soil sampling equipment will be a coring device such as a metal tulip bulb planter. A
new, cleaned coring device will be used for each composite sample. A new, cleaned
disposable aluminum pan will be used as a receptacle to remove miscellaneous debris from
each set of individual grab samples used to form a composite sample.
1. Details on specific sampling locations with rationale for the selection of each
sampling point will be carefully documented prior to the time of sample collection.
The documentation of each sampling location will be maintained in the team leader's
site log along with information about the site obtained from the home owner.
2. Log book entries, sample tags, and field record sheets with identification of sampling
locations will be completed for each sample and will include date, time, and name or
initial of the team leader.
3. Each sample container will be identified by a sample identification tag and unique
sample number.
*. The top one inch of the grab samples for each sampling site will be shaved off with a
cleaned, disposable knife. This sample will be used to form the composite sample.
The samples will be kept cool during shipment and when held in the laboratory prior
to analyses. The sample will be thoroughly blended in the laboratory prior to
analyses.
-------
-7-
5. All of the disposable sampling equipment (coring devices, gloves, knives, aluminum
pans, and tweezers) will be collected and disposed of under RCRA.
CLEANING OF SAMPLING EQUIPMENT
Each piece of new, disposable sampling equipment will be cleaned in the laboratory prior
to use in the field. Cleaning will consist of a soap and water wash, a water rinse, an
alcohol rinse, and a trichloroethylene (TCE) rinse.
Equipment cleaning in the field will not be required. However, the field team will have on
hand appropriate solvents to clean equipment in the field should the need arise.
FIELD QUALITY ASSURANCE
The field quality assurance plan will consist of the following:
1. In order to verify that cleaning of new, disposable equipment and sample containers
has been effective, a second TCE rinse for each type of equipment will be submitted
in a standard sample container for analysis (the knives and tweezers will be cleaned
together).
2. One sample container blank will be carried into the field each day and handled in the
same fashion as the other sample containers. Upon reaching the laboratory, a soil
sample without 2,3,7,8-TCDD (analytical result N.D. at 1 ppt) will be placed in the
sample container blank. This sample will then be process and analyzed with actual
samples.
3. A duplicate field sample will be obtained at a rate of one duplicate sample per ten
samples. The duplicate sample will consist of a second composite sample container
filled with one-half of each grab sample obtained from the aluminum pan after debris
has been removed.
LABORATORY QUALITY ASSURANCE
A comprehensive laboratory quality assurance plan and analytical protocols are presented
in Attachment A.
CHAIN OF CUSTODY
Standard EPA chain of custody procedures will be followed throughout the survey.
-------
TABLE I
SAMPLE SUMMARY
Type
Residential
Public Use Area
Plant Perimeter
Track-out
Municipal Sewage Sludge 1
Background
TOTAL
Number
of
Locations
20
5
11
5
11
15
Ige 1
_J_
72
Samples
per
Location
3 Composite
3 Cores
1 Composite
2 Cores
1 Composite
1 Composite
1 Composite
1 to 3 Composites
Total
Samples
Collected
60
15
11
10
11
15
1
8
131
Samples for
Immediate
Analysis
30
11
~
11
5
1
8
66
The above sample summary does not include field and laboratory quality
assurance samples. In addition, the field sampling team will be instructed to
collect samples from a limited number of catchment areas (e.g., low spots in
roof gutters, catch basins, etc.).
Note: A sample summary for Sites 2, 3, -and 4> will be included in the.final
study plan.
-8-
-------
NNW
NE
N
WNW>
ENE
WSW
sw
SE
SSW
FIGURE I
Surface Wind Rose at the Midland' Nuclear Plant (adjacent to Dow
Chemical)* On.-site data at 10 meters aboveground, March 1, 1975,
through February 28, 1977, Bars show the direction from which
the wind blows. Calms are hourly average windspeeds less than
0.3.m/s.
-------
FIGURE 2
PROPOSED SOIL SAMPLING LOCATIONS
MIDLAND, MICHIGAN
DOW CHEMICAL COMPANY
DOW CORNING
SAMPLE TYPES
Residential
A Public Use Area
i II Plant Perimeter
-------
APPENDIX B
In-Pi ant Soil Screening Survey, Dow Chemical - Midland Plant;
Draft Study Plan, November 28, 1983
-------
Draft
11/28/S3
IN-PLANT SOIL SCREENING SURVEY
DOW CHEMICAL - MIDLAND PLANT
INTRODUCTION
This study is planned in conjunction with the city of Midland soil screening survey being
performed at locations outside the Dow Chemical - Midland Plant. The data to be
gathered in this study will supplement data reported by Dow Chemical that indicate the
presence of dioxins in soils within the Midland Plant. These data will indicate whether
contaminated soils and dusts within the plant may be a source of contamination outside
the plant from transport by winds or trackout, or by transfer through the wastewater
treatment facilities.
OBJECTIVES
The objectives of this study are as follows:
1. To determine concentrations of PCDDs, PCDFs, and other toxic pollutants in surface
soils across the Dow Chemical - Midand Plant and specifically in the areas of former
trichlorophenol production and the waste incinerators.
2. To compare soil concentrations of PCDDs and PCDFs within the Midland Plant with
concentrations outside the plant.
SAMPLING DESIGN
The sampling program will be conducted in two phases. Phase 1 includes limited surface
soil sampling at selected locations as outlined in this plan. Phase 2, to be conducted at a
later time, may include additional surface soil sampling and core sampling at certain
locations. The Phase 2 sampling will be conducted after the results of Phase 1 sampling
and additional data from Dow Chemical are available.
Sampling will be concentrated in the vicinity of former production facilities involving
2,4,5-trichlorophenol and existing facilities in which phenoxy chemicals and agricultural
chemicals are manufactured, and the waste incinerators and wastewater treatment
facilities. Other samples will be gathered from points approximately on the centers of
grid squares measuring approximately 3/8 mile on a side. The selection of actual sample
sites will be determined on the sampling date by EPA, thus Dow Chemical will not have
prior notification of the specific sampling locations.
The attached map shows the approximate locations of the Phase 1 sampling points. Four
samples are to be taken in and near the 703 and 830 Building incinerators, and four
samples in the phenoxy (agricultural chemical) area. Seven additional samples will be
obtained from the grid, as shown. Phase 2 sampling sites will be selected at a later time.
SAMPLING METHODS
Composite soil samples will be collected at each sampling site from a 10 meter by 10
meter grid, at appropriate intervals. Areas with bare soil will be sampled preferentially
to grassed areas. Areas that have recently been disturbed will be avoided. Before
-------
any grass will be trimmed to just above the soil surface at the points to be
cored. Next, the cores, about two inches in diameter and several inches deep, will be
extracted. The top one-inch layer of each core will be shaved off and retained as the
sample. The remaining soil will be returned to the core holes. The sample site locations
will be staked or otherwise permanently marked at the time of sampling to allow for
positive identification and resampling, as necessary.
As each soil grab sample is taken, the sample will be placed on a cleaned, disposable
aluminum pan and miscellaneous debris (e.g., twigs, roots, wood chips, stones, pebbles, and
other nonsoil material that can be distinguished) will be removed with cleaned tweezers.
Since one of the objectives is to determine the concentration of pollutants in surface soils
which can be transferred to humans through ingestion, inhalation or dermal contact, or
transferred outside the plant by winds or through the wastewater treatment facility, the
miscellaneous debris must be removed to prevent biasing the analytical results in either
direction. A new aluminum pan and new tweezers will be used for each composite sample.
The sampling team will also use new disposable gloves at each composite sample site. The
grab sample will then be deposited in a new, cleaned one-gallon paint can. This can will
become the sample repository for each composite sample. Each composite sample will be
thoroughly blended in the laboratory to make the composite sample as homogenous as
possible. Aliquots of the composite samples will be shipped blind to the analytical
laboratories for analyses. Portions of the composite sample will be retained for possible
future analyses.
SAMPLE HANDLING AND EQUIPMENT
The soil sampling equipment will be a coring device such as a metal tulip bulb planter. A
new, cleaned coring device will be used for each composite sample. A new, cleaned
disposable aluminum pan will be used as a receptacle to remove miscellaneous debris from
each set of individual grab samples used to form a composite sample.
1. Details on specific sampling locations with rationale for the selection of each
sampling point will be carefully documented at the time of sample collection. The
documentation of each sampling location will be maintained in the team leader's site
log along with historical information about the site obtained from Dow Chemical.
Photographs will be taken of each sampling site.
2ป Log book entries, sample tags, and field record sheets with identification of sampling
locations will be completed for each sample and will include date, time, and name or
initials of the team leader.
3. Each sample container will be identified by a sample identification tag and unique
sample number.
*. The top one inch of the grab samples for each sampling site will be shaved off with a
cleaned, disposable knife. This sample will be used to form the composite sample.
The samples will be kept cool during shipment and when held in the laboratory prior
to analyses. The sample will be thoroughly blended in the laboratory prior to analysis.
5. AH of the disposable sampling equipment (coring devices, gloves, knives, aluminum
pans, and tweezers) will be collected and disposed of under RCRA.
-2-
-------
CLEANING OF SAMPLING EQUIPMENT
Each piece of new, disposable sampling equipment will be cleaned in the laboratory prior
to use in the field. Cleaning will consist of a soap and water wash, a water rinse, an
alcohol rinse, and a trichloroethylene (TCE) rinse.
Equipment cleaning in the field will not be required. However, the field team will have on
hand appropriate solvents to clean equipment in the field should the need arise.
FIELD QUALITY ASSURANCE
The field quality assurance plan will consist of the following:
1. In order to verify that cleaning of new, disposable equipment and sample containers
has been effective, a second TCE rinse for each type of equipment will be submitted
in a standard sample container for analysis (the knives and tweezers will be cleaned
together).
2. One sample container blank will be carried into the field each day and handled in the
same fashion as the other sample containers. Upon reaching the laboratory, a soil
sample without 2,3,7,8-TCDD (analytical result N.D. at 1 ppt) will be placed in the
sample container blank. This sample will then be processed and analyzed with actual
samples.
3. A duplicate field sample will be obtained at a rate of one duplicate sample per ten
samples. The duplicate sample will consist of a second composite sample container
filled with one-half of each grab sample obtained from the aluminum pan after debris
has been removed.
LABORATORY QUALITY ASSURANCE
A comprehensive laboratory quality assurance plan and analytical protocols are presented
in Attachment A.
CHAIN OF CUSTODY
Standard EPA chain-of-custody procedures will be followed throughout the survey.
-3-
-------
-------
APPENDIX C
Quality Control Summary
-------
SOIL SCREENING SURVEY
AT
FOUR MIDWESTERN SITES
QUALITY CONTROL SUMMARY
USEPA REGION V CENTRAL REGIONAL LABORATORY
PARTICIPATING ANALYTICAL LABORATORIES
U.S. Environmental Protection Agency
Toxicant Analysis Center
Bay St. Louis, Mississippi
Environmental Monitoring System Laboratory
Research Triangle Park, North Carolina
Dow Chemical Company
Analytical Laboratory
Midland, Michigan
May 1985
-------
INTRODUCTION
The quality of the polychlorfnated dlbenzodloxln (PCDD) and polychlorlnated
dlbenzofuran (PCDF) data generated for the EPA Region V Midland Michigan
Soil Survey was determined based on Internal and external audits. Each
sample batch sent to the participating labs Included a blind blank soil
and control soil. In addition, batches contained blind field duplicates,
and homogenlzatlon check samples. All samples were coded and not traceable
to the sample location. Labs did not communicate while the analyses were
being conducted. Data were submitted to Harda A. Kuehl, CRL QC Coordinator
for QC assessment and sample decoding. Ms. Kuehl was the sole possessor
of the sample codes which were kept under custody along with the data.
This summary presents the results of Internal lab prepared QC audits and
Internal blind QC audits. In addition, the comparabllHty between lab
results 1s presented. Assessment of the data was done 1n accordance with
the "State of Michigan and U.S. EPA Region V Dloxlns and Other Toxic
Pollutants, City of Midland, Dow Chemical, Michigan Rivers, Quality
Assurance Project Plan for Laboratory Analysis 4/1/84".
I. Methodology
Before any study samples were analyzed, EPA and Dow conducted a method
comparability study. The extraction method decided upon was Soxhlet
extraction as outlined In Analyt1cal_Chem1stry, Vol. 52, No. 13, pp. 2045-
2053, November, 1980. Analytical standards for 2,3,7,8-TCDD and other
PCDD/PCDFs were exchanged and found to vary less than 20% between labs.
The clean up methodology used by Dow consisted of HPLC/RPHPLC, as
described In Analytical Chemistry, Vol. 52, No. 8, pp. 1239-1245, July,
1980 with an additional carbon column reverse eluted with toluene as
described by D.C. Stalling ASTM STP 686, pp. 302-323, 1979.
Quantltatlon and calculation procedures for each lab have been described
by the labs and are set out below. Percent solids determinations were
done by the Illinois Institute of Technology Research Institute with a
precision of < 5% as determined by In-lab duplicates.
A. EPA - Research Triangle Park, NC.
"The HRGC-HRMS method of analysis Is briefly described. 60mm SP-2340
fused silica capillary columns were used for analysis of TCDD and TCDFs.
Retention times of 2378-TCDD were 30 to 35 minutes and 2378-TCDF were 36
and 41 minutes respectively. Retention times did change after a number
of highly contaminated samples were analyzed, but this did not cause any
problems. 20m SP-2330 fused silica and 15m SE-54 UCOT glass capillary
columns were used In the analysis for penta through octa CDDs and CDFs.
The HRGC-HRMS multiple Ion selection (MIS) analysis was performed using
an Integration rate of 30 milliseconds per mass and a jump time of 12
milliseconds. This speed Is sufficient to produce responses slmlllar to
those of an EC or FID detector for components elutlng from a capillary
column. The Individual HRGC-HRMS-MIS analysis sequence and exact masses
of reference and CDD and CDF compounds are described:
-------
2.
0 PFK, 318.9793; JCDDs, 319.8965./321.8936; 13C12-TCDD, 333.9338.
0 PFK, 292.9825; TCDFs, 303.9016/305.8987; TCDDs, 319.8965/321.8936;
13C12-TCDD, 333.9338.
0 PKF, 330.9793; 13C12-TCDD, 333.9338; PCDDs; 355.8546/357.8517; HxCDDs,
389.8156/391.8127.
0 PKF, 330.9793; 13C12-TCDD, 33.9338; PCDFs, 339.8597/341.8567; HxCDFs,
373.8207/375.8178.
0 PFK, 418.9729; Hepta CDDs, 423.776/425.7737; OCDD, 457.7377/459.7348;
13C12-OCDD, 471.7750.
0 PFK, 404.9761; Hepta CDFs, 407.7817/409.7788; OCDF, 441.7428/443.7398;
13C12-OCDD, 471.7750.
Analytical standards that were used In analysis for the Identification and
quantification of tetra through octa CDDs and CDFs are listed:
13C12-2378-TCDD
2378-TCDD
1368/1379-TCDD
2:2 type TCDDs
3:1 type TCDDs and 1234-TCDD
other pairs of TCDD Isomers
2378-TCDF
12378-penta-CDD
13478-penta-CDF
124679-hexa-CDF
123478-hexa-CDD
1234689-hepta-CDF
1234678-hepta-CDD
OCDD
OCDF
13C12-OCDD
Extracts of municipal Incinerator fly ash that contained all tetra
through octa CDDs and CDFs were analyzed previously on slmlllar capillary
columns.
This limited supply of CDD and CDF reference standards was sufficient to:
determine method efficiency for tetra through octa CDDs and CDFs;
Identification of 2378-TCDD and the other 21 TCDD Isomers; Identification
of hepta CDDs and CDFs, OCDD and OCDF. Conclusive Identification of
TCDFs, penta and hexa CDDs and CDFs can not be made with this limited
supply of reference Isomers. However, retention time windows were
determined relative to the Internal standard and recent literature.
Labeled and native CDD and CDF Isomers were colnjected prior to and during
the analysis of test samples to provide the data and labeled/native ratio
response required to determine method efficiency, concentrations of
respective compounds and the minimum limit of detection for each sample.
Typical amounts of analytical standards Injected for these purposes were:
-------
3.
250pg 13C12-TCDD, lOpg 2378-TCDD.
250pg 13C12-TCDD, lOpg 2378-TCDD, 16pg 2378-TCDF.
250pg 13C12-TCDD, 50pg each of penta and hexa CDF.
250pg 13C12-TCDD, 50pg of hexa CDD.
lOOOpg 13C12-OCDD, lOOpg each of hepta CDD and OCDD.
lOOOpg 13C12-OCDD, lOOpg each of hepta CDF and OCDF.
NOTE: nanogram amounts used for determination of high concentrations.
Specific qualifications and limitations should accompany the results:
0 Quantification of TCDD Isomers Is based on concentration of 2378-TCDD
analytical standard.
0 Quantification of penta through octa CDDs and CDFs 1s based on specific
standards previously mentioned.
0 The analysis reported for 2378-TCDF 1s for a very narrow time window.
One or more of these TCDF Isomers In test samples has exact retention
time of the 2378-TCDF analytical standard. NOTE: However, conclusive
assignment of 2378-TCDF can not be made because sufficient reference
standards are not available.
0 Several TCDF Isomers are present In much higher concentration
that reported for 2378-TCDF. However, these Isomers could not be
quantified because they elute 1n the early time frame of TCDD Isomers
and the Instrumentation has certain limitations. These analyses for
the other TCDF Isomers can be performed on the remainder of each
extract If necessary.
0 Conclusive Isomer assignments for penta and hexa CDDs and CDFs can not
be made because sufficient reference standards are not available.
0 HRGC-HRMS peak matching analysis In real time were performed on a
number on Ions to determine and confirm exact masses that correspond
to PCDDs and PCDFs. COCL loss analysis Indicative of TCDDs and
analyses to determine the absence of hexa, hepta, and octa chlorinated
dlphenylethers were performed on 13412. The distribution of CDD and
CDF Isomers In this sample Is slmlHar to those detected 1n most of
the other samples In this study.
The analytical results Indicate:
0 Method efficiency of tetra through octa CDDs and CDFs was adequate for
almost all samples.
0 Certain samples caused peak broadening and loss In MS sensitivity due to
enormous contamination present.
0 Results generated for "blind" QA samples were acceptable 1n most cases.
However, large differences are noted In some. For example, 13399 and
13412 were derived form one actual test sample but the concentrations
detected are quite different. ECL Indicates the original sample
consists of about 25% small pebbles. This may explain the difference
In results reported for the two samples.
0 10% to 30% of each extract 1s stored for reference 1n case there are
any questions and additional analyses need to be performed.
-------
4.
B. Dow Chemical Company
"The reporting format Is: a(b), where a Is the determined concentration,
and (b) Is the limit of detection (LOO). The LOO Is defined as 2.5X peak
to valley noise In an adjacent region of the HRGC-LRMS (mass resolution
1000, El Mode) mass chromatogram. Whenever the determined concentration
of a given analyte exceeds 10X LOO (analyte signal Is > 25X noise) we do
not report the LOO. Species not detected are listed as "nd" along with a
specific LOD.
Regarding TCDD Isomers, the data presented are Isomer-specific and have
been corrected for the observed recovery of 3.0 ng of [13CJ-2378-TCDD
Internal standard mechanically administered onto each sample prior to
extraction. All TCDD Isomers are determined via HRGC-LRMS at m/z 320 and
322, and quantftatlons are based upon a primary reference standard of
native 2378-TCDD. Regarding 2378-TCDF, the data presented are believed
to be Isomer specific, but until additional TCDF Isomers become available
we cannot prove Isomer specificity. These results have been corrected
for the observed recovery of 200 pg of [13C3-2378-TCDF Internal standard
added to each sample prior to extraction, and are determined via HRGC-
LRMS at m/z 304 and 306 (also monitor M+ for a C16-d1phenyl ether) using
a primary reference standard of native 2378-TCDF. Regarding HCDDs, the
data are Isomer specific with regard-to the groupings presented 1n the
tables and have been corrected for the observed recovery of 1.0 ng [13C]-
123478-HCDD Internal standard added to each sample prior to extraction.
All HCDD Isomers are determined via HRGC-LRMS at m/z 388 and 390 using a
primary reference standard of a mixture of two native HCDD Isomers prepared
at Dow. Whenever native CDDs concentrations are sufficiently high so as
to preclude monitoring of the Internal standard, we report our findings
as absolute, which means that the values are those observed and are not
corrected for the Internal standard recovery. Regarding H7CDDs, the data
are Isomer specific and have been corrected for recovery as Indicated on
the tables. The H7CDD Isomers are determined via HRGC-LRMS at m/z 424
and 426 using a primary reference standard of native 1234678-H7CDD.
Regarding OCDD, the data have been corrected for the observed recovery of
5.0 ng of [13C1-OCDD added to each sample prior to extraction. OCDD Is
determined via HRGC-LRMS at m/z 458 and 460 using a primary reference
standard of native OCDD.
The analytical procedure employed for these samples Is summarized as
follows: Each soil sample was subjected to exhaustive benzene Soxhlet
extraction for a period of 20 hours. The crude extracts were concentrated
to 20 mL volume via atmospheric pressure distillation and subsequently
diluted with 80 mL of hexane. This solution was passed through a multi-
layered chromatographlc column containing silica, sulfurlc acid on silica,
and aqueous sodium hydroxide on silica. The total column effluent was
concentrated to dryness and redlssolved in 5 mL of hexane. This solution
was passed through a 10% AgN03 on silica column and a high-aspect ratio
flash chromatography column containing basic alumina. The resulting
crude CDDs/CDFs fraction was concentrated to dryness and subsequently
quantitatively Injected Into a semi-preparative reverse phase HPLC system
for component fractlonatlon. The RP-HPLC fractions containing HCDDs,
H7CDDs, and OCDD were examined directly by HRGC-LRMS on a Kratos MS-80
operating at a mass resolution of 1000 In the El Mode employing single
Ion monitoring conditions as previously described. The RP-HPLC TCDD
fractions and 2378-TCDF fraction were subjected quantitatively to further
normal phase adsorption (silica)-HPLC fractionation and the resulting
residues subjected to HRGC-LRMS examination as described above.
-------
5.
BLANK RESULTS
For each batch of samples, both labs prepared an Internal method blank.
This blank consisted of the glassware reagents and clean up columns
used for the samples. An empty soxhlet thimble was extracted and the
extract carried through the rest of the procedure. Neither lab detected
any 2,3,7,8-TCDD In any of the blanks (Table 1). However traces of
hepta CDD (H7CDD) and OCDD were found. This may be due to the ubiquitous
nature of these pollutants as evidenced by the field blank results.
Field blanks were collected for every sampling location. These blanks
consisted of a sample paint can container, strawberry huller, tulip
bulb planter, spoon, and knife used In the collection of real samples.
These Items were rinsed with trlchloroethylene and placed Into the can.
The EPA lab rinsed the can and contents with the extracting solvent and
analyzed the extract. The results (Table 2) Indicate the presence of
400-900 plcograms of OCDD. Based on a typical can filled with 500
grams of real sample, this translates to approximately 1-2 ppt.
In each batch, a blind soil blank was Included. This sample originated
from nursery top soil.Three positives were found (Table 3). Two positive:
(2ppt, 0.1 ppt) were near the laboratory detection limit. One positive
of 7ppt did not Invalidate the entire batch of results as between lab
comparability was reasonable. Two Independent contract laboratories
analyzed this sample (n-20). After a review of the data 3 points were
discarded due to lab Induced contamination, (Table 4).
Althrough the low detection limits requested of the contractors (l-5ppt)
were not reached, no Indication of 2378-TCDD was found. All results In
Tables 3 & 4 are on different allquots of the blended blank sample.
These allquots were contained In separate jars with unique untraceable
codes. Details of the preparation of this sample are available on videotape
and will be described In the final report from the Illinois Institute of
Technology Research Institute.
-------
NOTE:
6.
There was a change 1n the labeling system for the last three batches.
They are prefixed with 2476, then numbered sequentially starting again
with one.
Table 1. Lab Method Blank Results
BATCH
1
2
3
4
5
6
7
8
9
10
11
12
-1
-2
-3
2,3,7,
EPA
nd (10)
nd (10)
nd (10)
nd (20)
nd (10)
nd (10)
nd (10)
nd (10)
nd (50)
nd (30)
nd (10)
nd (60)
nd (2)
nd (1)
nd (3)
8-TCDD (pg)
DOW
nd (22)
nd(22), nd(31), nd(45)
nd (10)
nd(20), nd(22)
nd(15), nd(15)
nd(9), nd(8), nd(6), nd(13), nd(8)
nd(2), nd(4)
nd (10)
nd (10)
nd ( 6)
nd ( 6)
nd ( 6)
PCDD/PCDF (pg)
170 pg OCDD (EPA)
46 pg H7CDD (DOW)
90 pg OCDD (DOW)
nd ( 6) (Batch 2 & 3 run together)
Table 2. Field Blank Results - EPA
Field Blank
A
B
C
D
2378-TCDD (pg)
nd (27)
nd (24)
nd (25)
nd (30)
OCDD (pg)
632 (216)
735 (191)
890 (225)
437 (244)
-------
7.
Table 3. Blind Soil Blank Results
BATCH #
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
2378 TCDD (ppt)
DOW
7 (1)
nd (2.5)
nd (0.6)
nd (0.8)
nd (0.6)
0.1 (0.1)
nd (0.3)
320
0.3 (0.2)
EPA
nd (1)
2 (1)
nd (2)
nd (2)
nd (1)
nd (2)
nd (1)
nd (1)
nd (1)
nd (3)
nd (2)
nd (3)
nd (1)
nd (1)
nd (2)
OTHER PCDD/PCDF (ppt)*
H6CDD:3.7/nd(15); H7CDD:14.9/31; OCDD;504/271
%
*61ven as Dow/EPA values
-------
8.
LAB
CODE
2,3,7,8-TCDD
(ppt)
Referee
A
34*
11*
<4
<6
<5
<6
<6
<3
<18*
<4
Referee
B
<8
<7
<3
<9
<7
<5
<9
< - Not detected at matrix dependent level cited
*data point Invalid due to Internal lab contamination
-------
9.
III. DUPLICATES
In batches designated by Region V, the EPA lab chose a sample at random
and analyzed two allquots. Dow selected batches at random to perform In
1ab_dupl1cates on. These samples consist of 2 separate 10-20 gram allquots
taken out of the same sample jar. EPA analyzed the duplicates In the
same analytical run whereas Dow analyzed the allquots 1n two separate
analytical runs. Results (Table 5) Indicate that the QA Plan objective
of RPD < 30% or + detection limit for TCDD and TCDF were not met. The
precision of OCDD was not within the objective of 60% RPD or + detection
limit, either. However, the data are valid for purposes of project
objectives considering the relatively low levels encountered.
In order to determine the Inherent Imprecision In the sample preparation
and splitting process, horoogenlzatlon checks were Inserted Into sample
batches. These were coded as a regular sample and were analyzed blind.
Two separate subsample jars of the sample were analyzed In the same analytical
run and 1n different analytical runs. Results (Table 6) Indicate a
coefficient of variation of between 10-27%.
F1eld_dupl1cates consisted of two separate 500 gram samples allquoted out
of the soil composite contents 1n the field and placed In separate
sample cans. Results (Table 7) Indicate rather wide and Inconsistent
variation In field sampling. The RPD ranges for Dow analyses were 8-23%
and EPA analyses 0-110% RPD.
In summary, the greatest variation occured In field duplicates which Is
due to field subsampllng error plus sample homogenlzatlon and analytical
error. The smallest degree of variation Is associated with the
homogenlzatlon/spllttlng procedure. The analytical precision Is better
within an analytical run than between analytical runs.
-------
10.
Table 5. In - Lab Duplicate Precision
LAB
EPA
DOW
SAMPLE
13319
13343
13366
13412*
14176
DE013004
13304
13317*
13341
13315
13331
DE013103
2378 TC
sample
28
6
nd(2)
271
74
nd(l)
33
160
220
52
67
32
;DD (p
dup.
29
5
nd(l)
103
72
nd(l)
28
82
240
23
24
34
)t)
RPD.
4
18
36
31
16
64
9
77
94
2
2378 TC
sample
27
;DF i
dup
34
PPt)
RPD
23
OCDD
sample
375000
(ppt)
dup
120000
RPD
103
Note: Only batch 8 has other PCDD/PCDF data due to request for full scan analysis
*sample Inhomogenelty noted by lab personnel.
Table 6. Homogenlzatlon Check Precision
2378 TCDD (ppt)
Type
between run
between run
within run
EPA
12,17
169,213
DOW
25,23
220,240,250,
220,210
CV
27%
5%
10%
Table 7. Field Duplicate Precision
BATCHES
1.2
2,3
3,3
6,7
7,9
11,12
1,1
2378
San
DOW
51
82
23
2.6
TC
iple
EPA
57
26
17
3
3
34
nd(l)
:DD i
Dup'
DOW
47
96
29
ppt)
1cate
EPA
26
67
21
3
4
20
nd(l)
RF
DOW
8
16
23
'D
EPA
74
110
21
0
28
52
-------
11.
IV. SPIKE RESULTS
Each sample was spiked with a surrogate for 2378-TCDD. Other PCDDs and
PCDFs were also added If a full scan analysis was requested. Table 8
summarizes the TCDD surrogate recovery. Actual recoveries for each sample
are found in Table 12. Overall batches, EPA mean recovery was 83.9%,
Sd=16.1%, m ป 135. Ten EPA recoveries did not meet the QA objective of
50-115%. Two if these samples were mlnicipal sewage sludges and seven
were control samples not typical of soils found at the study sites
(See Section V). Dow mean recovery was 72.9%, Sd=9.21, m=88. A Dow
recovery (116%) did not meet the QA objective. A test for significant
difference between the two lab means was significant at p=0.05. Both
labs added 13C12-OCDD to samples requiring full scan analysis. Actual
OCDD recoveries for each sample are found In Table 13. Overall EPA mean
recovery was 75.9%, Sd=9.63%, n = 22. Overall Dow mean recovery was
87.1%, Sd=5.19, n = 11. A test for significant differences between the
two labs means was significant at p=0.05. No recoveries were outside the
QA objective of 40-110%.
The EPA lab prepared in-lab spikes of a real soil samples in batches
designated by Region T.Results (Table 9) met the QA objectives except for
the 145% recovery. OCDD outliers were due to contribution from the native
example. The QA objective for OCDD was 70-130%. Hexa CDF recovery (0%)
was also outside the QA objective of 40-160%. Associated sample batch
data were not rejected.
-------
12.
V. CONTROL SAMPLE RESULTS
In each batch, a blind control son was Included. This sample originated
from an Industrial site near a city of Chicago Incinerator. Initial
results of this sample prior to homogenlzatlon Indicated 9-25 ppt 2,3,7,8-
TCDD and a host of other PCDD/PCDFs. After homogenlzatlon 1t appeared
that <5 ppt was present. Results (Table 10) Indicate that the nature of
the sample which was found to contain high ppb levels of PAH's, methyl
naphthalenes and benzene with long hydrocarbon chains prevented the EPA
from consistently analyzing 1t with acceptable TCDD recoveries. Two
Independent contract laboratories analyzed this sample (Table 11). After
a review of the data by Jay Smith, CDC Statistician, It was decided that
analysis of this sample was not a valid means of accepting or rejecting
associated sample batches since the extrapolation of this diffident
sample matrix behavior to the study samples was not technically correct.
All results in Tables 10 & 11 are for different allquots of the blind
control sample. These allquots were contained 1n separate jars with
unique untraceable codes. Details of the preparation of this sample are
available on videotape and will be described In the final report from
Illinois Instituted of Technology Research Institute.
-------
13.
table 8. TCDD Surrogate Recovery Results
BATCH *
1
2
3
4
5
6
7
8
9
10
11
12
-1
-2
-3
DOW
X
73.1
76.3
74.3
71.9
74.3
70.4
78.8
67.2
85
65
63
67.5
72.8
73.9
Sd
6.09
19.9
8.46
7.14
10.1
3.93
7.57
6.12
4.00
6.36
1.71
5.44
n
11
10
10
11
10
5
10
6
1
2
1
2
4
9
EF
Y
74.0
87.0
87.3
87.8
79.1
74.2
88.4
90.8
79.6
91.4
74.6
78.1
90.8
82.8
88.8
>A
Sd
14.3
15.6
10.1
9.45
15.3
12.4
18.9
9.21
14.6
13.8
20.8
18.7
18.1
15.5
16.3
n
10
10
9
9
9
9
10
9
9
9
9
8
10
8
8
SURROGATE CONCENTRATION INFORMATION
Dow - Batches 1-12, Episode 2476, Batch 1
5.0 ng 13C 2378 TCDD added
Dow Batch 7
1.5 ng 13C 2378-TCDD added
Dow Episode 2476, Batch 1-2
3.0 ng 13C 2378 TCDD added
EPA Batches 1-3
5.0 ng Cl 37 2378-TCDD added
EPA Batches 4-8, Episode 2476, Batch 1-3
5.0 ng 13C 2378-TCDD added
EPA Batches 9-12
10.0 ng 13C 2378-TCDD added
-------
14.
Table 9. EPA In-lab Matrix Spike Results
EPA
BATCH #
1
3
5
7
10
11
12
12
-1
2378 TCDD
ppt added
10
8
9
10
10
8
11
20
12
ppt recovered
13
7
9
11
11
7
16
18
16
%
Recovery
130
87
100
91
110
87
145
90
133
EPA
BATCH
7
12
2378 TCDF
added
16
32
ppt
recovered
16
29
%
recovery
100
91
OCE
added
100
100
)D ppt
recovered
244*
391**
%
Recovery
ปป
* native cone 290 ppt
** native cone 271 ppt
EPA OTHER PCDD/PCDF RECOVERY
1,2,4,7,8 P5CDF
1,2,4,6,7,8 H6CDF
1,2,4,7,8 H6CDD
(25 ppt): 48%
(50 ppt): 0%
(50 ppt): 64%
-------
15.
Table 10. BLIND CONTROL SAMPLE
BATCH
1
2
3
4
5
6
7
8
9
10
11
12
DOW
4 (1)
3.7 (3.6)
3.4
4.7 (0.8)
3.4 (0.7)
4.3 (0.6)
3.4
2378
EPA
32
12
nd
nd
nd
6
5
nd
nd
nd
13
nd
TCDD
(3)
(3)
(10)
(7)/15 (5)
(11)
(2)
(3)
(19)/nd (7)
(54)
(42)
(4)
(6)
Units
COMMENTS
EPA 50% Recovery
EPA 40% Recovery
= ppt
EPA 34%, 22% Recovery; 2 injections
of same extract
EPA 23% Recovery
2378 TCDF: H6CDD: H7CDD: OCDD
37/17 477/640 1680/NA
EPA 10%, 19% Recovery
EPA 22% Recovery
EPA 13% Recovery
EPA 46% Recovery
6300/NA
Recovery refers to 13C 2378-TCDD recovery.
Dow/EPA.
NA = Not Analyzed
Other isomer values given as
For the last three batches, Episode 2476-Batch 1,2 & 3, a previously analyzed
positive sample was used for the blind control. Since no true values exist,
the previously reported values are used for comparison.
2,3,7,8 - TCDD
Batch
2476-1
-2*
-3
Dow
250
88
60
EPA
95
not reported
40
RPD
90%
40%
Previous Value
169 (EPA)
220 (DOW)
240 (DOW
250 (EPA)
290 (DOW)
49 (EPA)
90 (DOW)
* The QC for Batch 2476-2 is not usable. EPA did not submit a result for the
control. Wide deviations between the three sequential samples listed below
indicate a possible sample mix-up.
-------
16.
Sample
ID
DOW
EPA
DE013008
DE013009
DE013010
Field Sample
Blind Control
Blind Blank
0.2 (0.2)
88
320
90
not reported
ND
Table 11. Control Sample Referee Lab Results
LAB
CODE
2,3,7,8-TCDD
(ppt)
Referee
A
<30
< 7
< 3
< 3
< 7
< 3
<30
Referee
B
< 3
< 9
<30
<20
<37
<24
-------
17.
VI. COMPARABILITY
In Tables 12 & 13, the comparfslon between EPA and Dow data Is made by a
RPD calculation. The QA objective for TCDD precision was < 50% RPD. It
appears that 32% of the data did not meet this goal.
X TCDD concentration
(ppt)
< detection limit
1 - 10
11 - 20
21 - 35
36 - 60
61 - 80
81 - 100
101 - 500
501 +
_% RPD
X
0
46
63
41
53
46
36
55
76
sd
0
38
47
36
31
14
17
43
-
n
16
12
8
12
7
3
4
11
1
Tests for significance (p =0.05) between the concentration levels Indicate
that the generally smaller RPD's observed at TCDD levels >61 ppt are not
statistically different from those seen at lower levels.
Other PCDD/PCDF comparability In Table 13 Indicates discrepancies In
H7CDD concentrations between labs. The QA objectives for these
homologues was < 50% RPD for 2378-TCDF and < 100% RPD for all others.
10% of the data points fell outside these goals. No significant
correlation (p =0.05) between concentration level and RPD was found.
In September, partial comparability data were sent to Jay Smith of Center
for Disease Control for statistical verification of a suspected bias.
CDC concluded that Dow consistently had results that are aproxlmately 40%
higher than EPA. This bias Is statistically significant. Possible
sources of this bias were discussed.
Dow described their subsampling procedure. They stored samples 1n the
freezer. An Ice collar near the cap formed during storage. The samples
were brought to room temperature, and shook to Incorporate the condensed
water. Rocks, roots found In the samples were excluded by Dow when a
subsample was taken. The material was broken up with a spatula to attempt
to get finer particles. A 20 gram aliquot was then taken.
-------
18.
EPA broke up large balls of soil, but Included roots, and rocks 1n their
subsample. It appears that Dow analyzed more of the finer particle size
material In each sample. It has been shown that dloxln 1s preferentially
bound to smaller particle sizes. This could explain the consistent high
bias 1n Dow results. Subsequent batches (6,8,10,11,12) analyzed after this
discovery Indicated a less consistent bias when Dow subsampled like EPA.
Both EPA and Dow chemists recommend a consistent field procedure to define
the original soil sample as everything that will pass through a 1mm (18
mesh) polypropylene sieve. These sieves could then be disposed of to
avoid cross contamination. It was decided not to do any further
Investigation of analysis of split samples, since given the differing
homogeneity of this study's samples even after blending, and extended
holding times differences between labs, data comparability would probably
not get any better than 40%. Nonetheless, caparablllty of the data 1s
adequate to satisfy project objectives.
-------
TABLE 12. BETWEEN LAB COMPARABILITY
scc#
13300
13301
13302
13303
13304
13305
13306
13307
13308
13309
13312
13313
13314
13315
13315
13317
13318
13319
13320
13321
13324
13325
13326
13327
13328
13329
13330
13331
13332
13333
13336
13337
13338
B
A
T
C
H
1
2
3
4
2378 TCDD (ppt)
DOW
51
28
23
93
33 [28]D
200
30
370
4 (1)
7 (1)
47
17 (2)
25 (2.7)
52 [23]R
33
160 [82]R
ND (2.4)
45
3.7 (3.6)
ND (2.5)
96
415
29
0.96 (0.7)
92
38
90
67 [24]R
ND (0.6)
3.4
5.3 (0.6)
ND 0.8)
ND (0.9)
EPA
57 (5)
9 (1)
13(1)
54 (3)
20 (1)
157 (6)
22 (1)
235 (15)
32 (3)
NO (1)
26 (3)
13(1)
12 (1)
17 2)
18 (1)
26 (1)
ND 1)
28 (1)
12 (3)
2 (1)
67 (1)
273 (13)
21 (1)
4 (1)
76 (7)
31 (4)
49 (1)
24 (1)
ND (2)
ND (10)
4 (1)
ND (1)
ND (1)
RPD
11
103
56
53
49/33
24
31
45
156
57
27
70
101/30
59
144/104
0
47
106
35
41
32
122
19
20
59
94/0
0
28
0
0
% Recovery Surrogate TCDD
DOW
72%
79%
67%
67%
65% [71%]D
71%
70%
86%
78%
78%
71%
68%
61%
63%
66%
65%
53%
98%
116%
102%
75%
72%
75%
69%
71%
64%
76%
79%
66%
96%
66%
78%
61%
EPA
112%
75%
73%
64%
70%
63%
71%
71%
58%
83%
96%
92%
108%
72%
100%
86%
94%
92%
50%
80%
97%
74%
94%
82%
100%
86%
80%
100%
73%
40%
92%
88%
96%
-------
TABLE 12. BETWEEN LAB COMPARABILITY
Mil
scc#
13339
13340
13341
13342
13343
13344
13345
13348
13349
13350
13351
13352
13353
13354
13355
13356
13357
13360
13361
13362
13363
13364
13355
13366
13367
13368
13369
B
A
T
*
*
H
4
5
6
2378 TCDD (ppt)
DOW
ND (0.7)
220 [240]D
210
9.2
5.4 (0.9)
4.7 (0.8)
ND (0.8)
6.7
ND (0.6)
0.8 (0.6)
ND (0.8)
1.4
11
1.8 (0.6)
ND (0.7)
ND (0.6)
3.4 (0.7)
290
22
38
ND (0.7)
4.3 (0.6)
EPA
ND (1)
169 (6)
213 (6)
9 (1)
6 (2)
ND (7) [15(5)]R
ND (2)
5 (2)
ND (1)
ND (2)
ND (3)
ND (2)
11 (3)
ND (2)
ND (2)
ND (1)
ND (11)
250
ND (2)
19 (2)
ND (2)
28 (2)
3(2)
ND (2)
14 (2)
6 (2)
ND (2)
RPD
0
26
1
2
11
-/104
0
29
0
-
0
-
0
-
0
0
~
15
15
30
0
33
% Recovery Surrogate TCDD
DOW
65%
65% [67%]D
78%
72%
82%
75%
82%
73%
75%
82%
65%
68%
62%
68%
93%
67%
90%
76%
69%
64%
71%
72%
EPA
96%
94%
68%
84%
76%
34% [22%]R
96%
94%
94%
86%
54%
82%
52%
82%
78%
100%
23%
21%
86%
64%
76%
64%
104%
70%
68%
66%
70%
-------
TABLE 12. BETWEEN LAB COMPARABILITY
see*
13372
13373
13374
13375
13376
13377
13378
13379
13380
13381
13387
13388
13390
13391
13392
13393
13394
13395
13396
13397
13386
13400
13401
13403
13404
13405
13406
13407
13408
13409
13402
13412
13413
13414
13415
13416
13417
13384
13385
13389
B
A
T
C
H
7
B
9
10
2378 TCDD (ppt)
DOW
0.3 (0.2)
NO (0.1)
2.4
99
0.7 (0.2)
2.6
ND (0.2)
ND (0.1)
3.4
0.1 (0.1)
ND (0.5;
ND (0.6)
4.4 (0.5)
8.4
33
55
910
EPA
ND (2)
ND (3)
3 (1)
76 (
ND I
3)
2)
3 (1)
ND (1)
ND (2)
5
ND
4
3)
1)
ii
ND (2)
ND (2)
12 (3)
15 I
78 <
[3)
[3)
108 (2)
ND (4)
ND(19),ND(7)
ND
ND
4
1
*)
(2)
ND (4)
ND
18
2)
3)
45 (5)
3500 (39)
223
ND
ND
10
(7)
:D
54)
t>)
271 (7)
36000 (290)
25000 (500)
ND (27)
ND (24)
ND (25)
ND (30)
ND
ND
(3)
(42)
2030 (42)
RPD
0
22
26
-
14
0
0
38
ป
-
0
93
56
81
65
76
% Recovery Sur
DOW
70%
76%
64%
91%
82%
77%
85%
84%
84%
75%
60%
62%
66%
73%
74%
56%
85%
EPA
106%
64%
104%
58%
76%
106%
98%
106%
60%
100%
94%
80%
106%
92%
86%
88%
81%
84%
101,19%
106%
72%
98%
90%
70%
71%
80%
100%
51%
84%
22%
104%
100%
66%
102%
100%
96%
80%
103%
13%
72%
-------
TABLE 12. BETWEEN LAB COMPARABILITY
scc#
14175
14176
14177
14179
14180
14181
1418Z
14183
14184
14187
14188
14189
14190
14191
14192
14193
14194
14195
1419
B
A
T
C
H
11
12
2378 TCDO (ppt)
DOW
W (3)
90
470
EPA
21 (8)
74 (5)
69 (3)
4600 (83)
130 (13)
40 (2)
463 (12)
NO (2)
13 (4)
440 (16)
314 (14)
ND (31),NO(24)
420 (13)
212 (8)
20 (3)
146 (10)
10 (3)
ND (3)
ND (6)
RPD
10
26
40
% Recovery Surrogate TCDO
DOW
69%
61%
63%
EPA
49%
52%
55%
101%
103%
73%
94%
97%
59%
119%
84%
6%, 6%
90%
62%
63%
65%
61%
81%
46%
2378 TCDD (ppt)
scc#
DE012901
-02
-03
-04
-05
-06
-07
-08
-09
-10
B
A
T
C
H
2476
-1
DOW
250
ND (0.3)
EPA
ND (1)
ND 3)
2 (1)
ND (1)
ND (1)
MD (2)
NO (1)
ND (1
95 (3)
ND (1)
RPD
99%
% Recovery Surrogate TCDD
DOW
63%
72%
EPA
100%
58
112
100
104
62
98
94
80
100
-------
2378 TCDD (ppt)
scc#
OE013001
-02
-03
-04
-05
-07
-08
-09
-10
DE013101
-02
-03
-04
-05
-06
-07
-08
B
A
T
C
H
-2
-3
DOW
150
0.2 (0.2)
88
320
50
350
32,34
34
35
183
60
0.3 (.2)
EPA
ND (1)
ND (1)
ND (2)
ND (1)
ND (1)
75
90
-
ND (1)
9 (2)
112 (8)
19 (2)
28 (4)
24 (2)
32 5)
40 (4)
ND (2)
RPD
67%
199%
_
-
139%
103%
51,57%
19%
37%
140%
40%
-
% Re<
DOM
75%
73%
71%
72%
80%
63
71,73
81
73
78
74
72
:overy Surrogate TCDD
EPA
92%
90
70
88
80
102
52
ซ.
88
102%
92
88
54
86
88
110
90
-------
BETWEEN LAB COMPARABILITY PCOD/PCDF
Table 13.
SCCI
1 4479
1 4 Jit
1 4471
1 JJ/'i
13375
1447C
1 4477
1 JJff
1447ft.
1 44DA
1 JJnU
1 4401
1 JJnl
1 4407
1 JOfW
13391
13392
13393
13394
B
A
T
C
H
7
7
7
8
8
8
8
1368
Dow
260
""*"
2JC
.0
14
24
54
220
TCDD
EPA
135
n/4
30
30
100
120
RPD
63
73
22
60
59
1379
Dow
84
] c
5.7
10
22
80
TCDD
EPA
65
fuf
10
10
50
56
RPD
25
55
0
78
35
1378 1
Dow
27
itrff 5
1.9
3.2
7.6
26
rCDD
EPA
19
nd
nd
17
23
RPD
35
0
0
76
12
2378 T(
Dow
1 2
0.2
1 6
31
3 2
1 6
nd (3)
nd 121
47
nd (?\
roi \c.)
4 5
3.7
3.8
7.8
NR
:DF
EPA
nd (1)
nd (2)
nrf f2l
13
nd (5)
2
nd (2)
nd H)
17
nd M)
IIU \ 1 /
ฃ
nd (5)
nd (5)
13
15
RPD
o
o
82**
n
99
o
47
90
0
0
50
HTcno
Dow
15
2.9
56. 1
583
31.8
8.8
nd (11
1.9
477
3 7
46.2
75.8
190
169
EPA
nd
nd I
67
243
nd |
nd {
nd {
nd (
640
nd 1
mi 1
nrf
86
63
335
413
19)
5
39)
10)
12)
11)
!15)
20)
cu/
RPD
18
82
o
o
o
o
o
59
18
55
84
HiCDD
Dow
14.7
310
3900
147
42
8.7
6.2
1680
U.o
30
175
330
930
570
EPA
73
47
350
417
23
91
25
31
145
346
376
2280
2350
RPD
1
104**
12
161**
18
58
164**
122**
69
131**
66
13
84
122**
OCDD
Dow
350
200
2200
18000
670
340
260
170
6300
540
1 jo
960
1900
5700
3800
EPA
172
130
3100
12000
840
195
196
92
MA
271
277
680
860
7000
7000
RPD
34
40
22
28
66
72
34
75
20
59
13 c-ocn
Recovery
Dow
97
absolute
on
89
91
absolute
79
84
87
82
absolute
D
EPA
ฃ7
78
7C
M'
M.
O7
78
76
75
95
** Outside QA Plan goal of SOX RPD TCDF, 100X RPD others
* All below Detection
NR ซ Not Recovered
Absolute ซ Not determined due to native concentration
NA Not analyzed for
-------
25.
VII. SUMMARY_OF_EPA_DATA_QUALITY
A. Precision
1. Field precision as measured by absolute difference (Ri)
between aTl field duplicates was:
"R"ป = 15 ppt
Sd = 15 ppt
n = 6
The 95% confidence limit for the precision of each field
duplicate Is calculated as 2.46 R*:
TCCD
Results (ppt) Ri (ppt)
57,26 ~ 11
26,67 41
17,21 4
3,3 0
3,4 1
34,20 14
2. Lab precision as measured by absolute difference (Ri)
between all lab duplicates was:
"R~t = 1 ppt
Sd = 1 ppt
n = 3
TCDD
Resul ts__( ppt) RMppt)
28,29 1
6,5 1
nd (l),nd (2) 0
74, 72 2
nd (1), nd (1) 0
271,103 168*
* Analyst noted extreme Inhomogenelty.
3. Sample_homogen1zat1on_prec1s1on as measured by the coefficient
of variation between TCDD measurements of different allquots of
three samples was:
x CV = 14%
n = 12
-------
26.
B. Accuracy
1. TCDD
2,3,7,8-TCDD accuracy expressed as the percent recovery for the mean
surrogate (13C or C137) recovery was:
x = 83.9%
Sd = 16.1%
n = 135
95% CI = (52-116%)
2. OCDD
OGDD accuracy expressed as the percent recovery for the mean 13C-OCDD
surrogate recovery was:
x" = 75.9%
Sd = 9.63%
n = 22
95% CI = (56-96%)
C. Completeness
A total of 10 EPA sample TCDD surrogate recoveries were judged as
outliers. Two of these samples were municipal sewage sludges and
seven were control samples not typical of soils found at the study
sites (See Section V). Completeness as expressed by the percent of
valid TCDD data obtained for the study was:
(124 x 100) = 99%
(125
D. Comparability
Comparability, expressed as mean RPD between EPA and Dow results was
as follows:
2,3,7,8-TCDD OCDD
x = 41% CV = 105% 7 = 48% CV = 38%
Sd = 43% Sd = 18%
n = 73 n = 14
-------
APPENDIX D
Study Results
-------
Figures
1
2
APPENDIX D
Study Results
Sample Site Location Map, Henry, Illinois.
Sample Site Location Map, Mlddletown, Ohio
Tables
1 2378-TCDD, Dow Chemical - Midland Plant
In-Pi ant Surface Soil Samples
2 2378-TCDD, Site #1 - Midland, Michigan Area
Surface Soil Samples
3 2378-TCDD, Site #2 - Henry, Illinois Area
Surface Soil Samples
4 2378-TCDD, Site #3 - Mlddletown, Ohio Area
Surface Soil Samples
5 2378-TCDD, Site #4 - Minnesota Natural Areas
Surface Soil Samples
6 2378-TCDD, Control and Blank Samples
7 PCDDs and PCDFs, Site #1 - Midland, Michigan Area
Surface Soil Samples
8 PCDDs and PCDFs, Site #3 - Mlddletown, Ohio Area
Surface Soil Samples
9 PCDDs and PCDFs, Site #4 - Minnesota Natural Areas
Surface Soil Samples
10 PCDDs and PCDFs, Control and Blank Samples
11 Toxic Organic Pollutants, Site #1 - Midland, Michigan Area
Surface Soil Samples
12 Toxic Organic Pollutants, Site #3 - Mlddletown, Ohio Area
Surface Soil Samples
13 Toxic Organic Pollutants, Site #4 - Minnesota Natural Areas
Surface Soil Samples
14 Toxic Organic Pollutants, Blank and Control Samples
-------
APPENDIX D
(continued)
Study Results
Tables
15 Tentatively Identified Compounds, Site #1 - Midland, Michigan Area
Surface Soil Samples
16 Tentatively Identified Compounds, Site #3 - Mlddletown, Ohio Area
Surface Soil Samples
17 Tentatively Identified Compounds, Site 14 - Minnesota Natural Areas
Surface Soil Samples
18 Tentatively Identified Compounds - Blanks and Control Samples
Surface Soil Samples
-------
LEGEND
Q_ perimeter Sample (PER)
- Residential Area Sample
(M, N, & P)
Q. Upwind Sample (UPW)
OPER-18
PPENDIX D
FIGURE 1
SAMPLE SITE LOCATION
MAP} HENRY, ILLINOIS
-------
ARMCO
STEEL
CORP
LEGEND
Q- perimeter Sample (PER)
- Residential Sample (G, H, & K)
Q. public Use Area Sample (P)
SAMPLE SITE LOCATION
MAPjMIDCXJELTOWN, OHIO
-------
APPENDIX D
Table 1
2378-TCDD
Dow Chemical - Midland Plant
In-Plant Surface Soil Samples
Sampl e
Number
13404
14176
14190
14192
13406
14180
14178
14193
14194
13405
14187
14182
13407
13412
13413
Field
Identification
Station 1
Station 2
Station 3
Station 4
Station 5
Station 6
Station 7
Station 8
Station 9
Station 10
Station 11
Station 12
Station 13
Station 14
Station 15
Location
South of 492 Building
South of 1005 Building; Southwest of 703 Building
South of 703 Building
West of 703 Building
Southwest of 956 Building; East of 703 Building
Northwest of 1159 Building; North of Shot Pond
llth and J Streets - Northwest Corner
8th and G Streets - Northwest Corner at Steam Pipeline
Northwest of 1050 Building at F Street
South of 543 Building; West of 14th Street
16th and G Streets - Southwest Corner
16th and G Streets - Northwest Corner
17th and G Streets - Northwest Corner
West of 934 Building
South and East of 874 Building North of RR Tracks
2378-TCDD (DL)
(ppb)
0.018
0.074
0.42
0.020
3.50
0.13
4.60
0.15
0.010
0.045
0.44
0.46
0.22
0.27
25.0
[36.0]R
(0.003)
(0.005)
(0.013)
(0.003)
(0.039)
(0.013)
(0.083)
(0.010)
(0.003)
(0.005)
(0.016)
(0.012)
(0.007)
(0.007)
(0.50)
(0.28)
% Recovery % Solids
71
52
90
63
100
103
101
65
61
80
119
94
51
100
66
92.
96.
96.
95.
96.
98.
96.
99.
90,
99.
99.
96.
99.
100.
85.5
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported In parts per
billion (ppb).
(2) % Recovery - Recovery of Internal standard (Cl37 2378-TCDO or 13c 2378-TCDD)
expressed as percent.
(3) % Solids - Solids content of sample determined after sample homogenlzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) [ ]R = Repeat analysis of same sample.
-------
APPENDIX D
Table 2
2378-TCDD
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Sample Field 2378-TCDD (DL)
Number Identification (ppb) % Recovery % Solids
Upwind Areas
13354 UPW-l-L ND (0.002) 82% 94.9%
13343 UPW-1-1 0.006 (0.002) 76% 98.6%
13401 UPW-2-L ND (0.004) 90% 94.0%
13395 UPW-4-L ND (0.004) 84% 98.1%
13342 UPW-4-D 0.009 (0.001) 84% 84.2%
Track Out and Perimeter Samples - Dow Chemical-Midiand Plant
13353 TO-4-G 0.011 (0.003) 52% 99.3%
13360 TO-6-G 0.25 (0.018)* 21% 99.0%
13367 TO-9-6 0.014 (0.002) 68% 99.0%
14188 PER-2-L 0.31 (0.014) 84% 82,8%
14177 PER-2-1 0.069 (0.003) 55% 99.0%
14191 PER-5-L 0.21 (0.008) 62% 99.2%
13402 PER-8-L 0.010 (0.005) 104% 99.0%
13389 PER-9-G 2.03 (0.042) 72% 95.6%
14181 PER-10-L 0.040 (0.002) 73% 97.1%
Public Use Areas
13362 P-l-L 0.019 (0.002) 64% 99.5%
13364 P-2-L 0.028 (0.002) 64% 89.7%
13374 P-5-L 0.003 (0.001) 104% 99.7%
13392 P-6-L 0.015 (0.003) 86% 80.2%
13393 P-7-L 0.078 (0.003) 88% 98.8%
13340 P-8-L 0.17 (0.006) 94% 96.5%
13375 P-9-L 0.076 (0.003) 58% 89.2%
13391 P-10-L 0.012 (0.003) 92% 96.6%
13394 P-ll-L 0.108 (0.002) 81% 100.0%
-------
APPENDIX D
Table 2 (continued)
2378-TCDD
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Sample Field 2378-TCDD (DL)
Number Identification (ppb) % Recovery % Solids
Residential Areas
13007 A-l-L 0.075 (0.006) 102% 83.0%
13008 A-l-1 0.090 (0.008) 52% 89.7%
13101 A-3-L 0.009 (0.002) 102% 88.5%
13102 A-3-1 0.112 (0.008) 92% 76.5%
13328 B-l-L 0.076 (0.007) 100% 93.2%
13305 B-l-1 0.16 (0.006) 63% 97.0%
13306 B-3-L 0.020 (0.001) 71% 97.8%
13325 B-3-1 0.27 (0.013) 74% 87.1%
13103 B-4-L 0.019 (0.002) 88% 84.7%
13104 B-4-1 0.028 (0.004) 54% 88.1%
13317 C-l-L 0.026 (0.001) 86% 78.7%
13303 C-l-1 0.054 (0.003) 64% 84.4%
13314 C-3-L 0.012 (0.001) 100% 98.6%
13307 C-3-1 0.24 (0.015) 71% 98.1%
13105 C-4-L 0.024 (0.002) 86% 91.0%
13106 C-4-1 0.032 (0.005) 88% 88.9%
13318 D-l-L ND (0.001) 94% 84.0%
13331 D-l-D 0.024 (0.001) 100% 99.5%
13319 D-2-6 0.028 (0.001) 92% 94.9%
13316 D-3-L 0.018 (0.001) 100% 90.9%
13329 D-3-1 0.031 (0.004) 86% 99.8%
13312 E-l-L 0.026 (0.003) 96% 97.5%
13330 E-l-1 0.049 (0.001) 80% 97.0%
13301 E-3-L 0.009 (0.001) 75% 97.6%
13304 E-3-1 0.020 (0.001) 70% 99.1%
13302 F-l-L 0.013 (0.001) 73% 99.0%
13313 F-l-1 0.013 (0.001) 92% 96.8%
Miscellaneous
14175 Sludge 0.021 (0.008)* 49% 88.7%
-------
APPENDIX D
Table 2 (continued)
2378-TCDD
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported 1n parts per
billion (ppb).
(2) % Recovery - Recovery of Internal standard (Cl37 2378-TCDD or "C 2378-TCDD)
expressed as percent.
(3) % Solids - Solids content of sample determined after sample homogenlzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field Identification of samples:
Location Type
UPW - Upwind L - Yard, lawn, or open area composite
TO - Track Out 1 or D - Downspout or drlpllne composite
PER - Perimeter G - Open area grab sample
P - Public Use
A-K - Residential
* Data not valid. Quality assurance objective not achieved.
-------
SITE
Sample Field
Number Identification
APPENDIX D
Table 3
2378-TCDD
n - Henry, Illinois
Surface Soil Samples
2378-TCDD (DL)
(ppb)
Area
% Recovery
% Solids
Upwind Area - Henry, Illinois
12901 UPW-7-L
12902 UPW-7-1
Perimeter Samples
13002 PER-16
13003 PER-17
13004 PER-18
13012 PER-18CD]
Residential Areas
12903 M-l-DL
12904 M-l-L
12905 N-l-L
12907 N-l-LCD]
12906 N-l-1
13001 P-l-L
12908 P-l-l
NO (0.001)
NO (0.003)
NO (0.001)
NO (0.002)
ND (0.001)
NO (0.001)
0.002 (0.001)
ND (0.001)
ND (0.001)
NO (0.001)
NO (0.002)
ND (0.001)
ND (0.001)
100%
58%
90%
70%
88%
80%
112%
100%
104%
98%
62%
92%
94%
89.5%
88.1%
88.0%
98.7%
90.8%
90.8%
90.6%
92.7%
87.5%
87.3%
83.5%
90.1%
88.6%
-------
APPENDIX D
Table 4
2378-TCDD
SITE #3 - Middle town, Ohio Area
Surface Soil Samples
Sample Field
Number Identification
Upwind Area - Hamilton,
13400 UPW-5-L
13377 UPW-5-L
Perimeter - ARMCO, Inc.
13366 PER-12-L
13337 PER-12-1
13387 PER-13-L
13363 PER-13-D
13372 PER-14-L
13376 PER-15-L
Public Use Areas
13349 G-l-L
13386 P-ll-L
13327 P-12-L
13338 P-13-L
13339 P-14-L
13390 P-15-L
13352 P-15-1
Residential Areas
13336 G-2-L
13348 G-2-1
13355 H-l-L
13403 H-l-1
13361 K-l-L
13351 K-l-1
13350 K-2-L
Miscellaneous
2378-TCDD (DL)
(ppb)
Ohio
0.004 (0.002)
0.003 (0.001)
, Mlddletown Plant
ND (0.002)
ND (0.001)
0.004 (0.003)
ND (0.002)
ND (0.002)
ND (0.002)
ND (0.001)
ND (0.002)
0.004 (0.001)
ND (0.001)
ND (0.001)
ND (0.002)
ND (0.002)
0.004 (0.001)
0.005 (0.002)
ND (0.002)
ND (0.002)
ND (0.002)
ND (0.003)
ND (0.002)
% Recovery % Solids
98% 79.3%
106% 79.9%
70% 87.4%
88% 88.4%
94% 86.4%
76% 87.4%
106% 84.9%
76% 88.4%
94% 77.5%
72% 87.5%
82% 87.5%
96% 83.2%
96% 82.0%
106% 78.8%
82% 67.3%
92% 87.7%
84% 86.7%
78% 95.9%
70% 94.7%
86% 81.3%
54% 88.3%
86% 86.4%
14189 Sludge ND (0.031)* 6% 92.3%
-------
APPENDIX D
Table 4 (continued)
2378-TCDD
SITE #3 - Middletown, Ohio Area
Surface Soil Samples
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported in parts per
billion (ppb).
(2) % Recovery - Recovery of internal standard (Cl37 2378-TCDD or 13c 2378-TCDD)
expressed as percent.
(3) % Solids - Solids content of sample determined after sample homogenization,
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field identification of samples:
Location Type
UPW - Upwind L - Yard, lawn, or open area composite
TO - Track Out 1 or D - Downspout or dripline composite
PER - Perimeter 6 - Open area grab sample
P - Public Use
A-K - Residential
* Data not valid. Quality assurance objective not achieved.
-------
APPENDIX D
Table 5
2378-TCDD
SITE 14 - Minnesota Natural Areas
Surface Soil Samples
Sample Field
Number Identification
13373
13378
13379
13388
IWD-l-L
BSP-4-L
KR-l-L
PT-l-L
location
Itasca Wilderness
Bluestem Prairie
Kettle River
Pemblna Trail Preserve
2378-TCDD (DL)
(ppb)
ND (0.003)
ND (0.001)
ND (0.002)
ND (0.002)
% Recovery % Sol Ids
64% 94.7%
98% 98.0%
106% 79.5%
80% 53.5%
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported 1n parts per
billion (ppb).
(2) % Recovery - Recovery of Internal standard (Cl37 2378-TCDD or 13C 2378-TCDD)
expressed as percent.
(3) % Sol Ids - Solids content of sample determined after sample homogenlzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field Identification of samples:
Location
UPW - Upwind
TO - Track Out
PER - Perimeter
P - Pu&Hc Use
A-K - Residential
Type
L - Yard, lawn, or open area composite
1 or D - Downspout or dripllne composite
G - Open area grab sample
-------
APPENDIX D
Table 6
2378-TCDD
Control and Blank Samples
Sample
Number
14196
14184
13368
13308
13320
13333
13344
13344
13357
13380
13396
13396
13409
13385
14183
14195
13309
13321
13332
13345
13356
13369
13381
13397
13408
13414
13415
13416
13417
13384
Field
Identification
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Control Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Blank Soil
Can Blank (PER-5)
Can Blank (In-Plant)
Can Blank
(Middle town)
Can Blank #6
Blank Soil
2378-TCDD (DL)
ND (0.006)*
0.013 (0.004)
0.006 (0.002)
0.032 (0.003)
0.012 (0.003)
ND (0.010)*
ND (0.007)*
0.015 (0.005)*
ND (0.001)*
0.005 (0.003)
ND (0.019)*
ND (0.007)*
ND (0.054)*
ND (0.042)*
ND (0.002)
ND (0.003)
ND (0.001)
0.002 (0.001)
ND (0.002)
ND (0.002)
ND (0.001)
ND (0.002)
ND (0.001)
ND (0.001)
ND (0.001)
ND (0.027)
ND (0.024)
ND (0.025)
ND (0.030)
ND (0.003)
% Recovery
46%
59%
66%
58%
50%
40%
34%
22%
23%
66%
10%
19%
22%
13%
97%
81%
83%
80%
73%
96%
100%
70%
100%
106%
84%
102%
100%
96%
80%
103%
% Solids
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
98.4%
53.7%
53.7%
53.7%
53.7%
53.7%
53.7%
53.7%
53.7%
NR
53.7%
53.7%
53.7%
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported In
parts per billion (ppb).
(2) % Recovery - Recovery of Internal standard (Cl37 2378-TCDD or
13c 2378-TCDD) expressed as percent.
(3) % Solids - Solids content of sample determined after sample
homogenlzatlon, expressed as percent. Analytical results not
adjusted for moisture content.
(4) NR - Not reported.
* Data not valid. Quality assurance objective not achieved.
See Quality Assurance Summary, Appendix C.
-------
Sample No.:
Field ID.:
Location:
PCDDs (PL)
2378-TCDD
Total Iso TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
PCDFs (DL)
2378-TCDF
Total TCDFs
Total
Total
penta CDFs
hexa CDFs
Total hepta CDFs
XDF
SITE
13401
UPW-2-L
PI easant
View School
ND (0.004)
ND (0.004)
ND (0.024)
ND (0.024)
0.15 (0.024)
0.34 (0.026)
ND (0.004)
ND (0.008)
ND (0.022)
ND (0.031)
ND (0.051)
APPENDIX D
Table 7
PCDDs and PCDFs
#1 - Midland, Michigan
Surface Soil Samples
Upwind
13395
UPW-4-L
4853 W. Kent
ND (0.004)
ND (0.023)
ND (0.023)
0.17 (0.034)
0.33 (0.034)
ND (0.004)
ND (0.008)
ND (0.023)
ND (0.028)
ND (0.045)
Area
Dow Chemical
In-Plant
13406 13412
Station 5 Station 14
Incinerator West of 934 Building
3.5 (0.039) 0.27 (0.007)
0.32
0.24 (0.067)
4.0 (0.067)
75.0 (0.9)
375.0 (1.3)
0.45 (0.06) 0.027 (0.007)
0.90 (0.14)
3.1 (0.13)
15.4 (0.38)
8.6 (0.48)
Notes: (1)
Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported in parts per billion (ppb).
-------
APPENDIX D
Table 7
PCDDs and PCDFs
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Public Use Areas
Sample No.:
Field ID.:
Location:
PCDDs (DL)
2378-TCDD
Total 1so TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
PCDFs (DL)
2378-TCDF
Total TCDFs
Total penta CDFs
Total hexa CDFs
Total hepta CDFs
OCDF
13374
P-5-L
County Line Rd.
0.003 (0.001)
ND (0.001)
ND (0.014)
0.067 (0.007)
0.35 (0.013)
3.1 (0.096)
ND (0.002)
ND (0.002)
ND (0.01)
ND (0.01)
0.065 (0.02)
0.044 (0.023)
13392
P-6-L
Mapleton School
0.015 (0.003)
0.040
ND (0.035)
0.063 (0.035)
0.38 (0.028)
0.86 (0.027)
ND (0.005)
ND (0.008)
ND (0.024)
0.14 (0.043)
0.10 (0.071)
13393
P-7-L
Longvlew School
0.078 (0.003)
0.17
Interference
0.34 (0.02)
2.3 (0.055)
7.0 (0.068)
0.013 (0.007)
ND (0.025)
0.26 (0.036)
0.72 (0.021)
0.64 (0.037)
13375
P-9-L
Virginia Park
0.076 (0.003)
0.29
0.10 (0.018)
0.24 (0.018)
0.41 (0.093)
12.0 (1.5)
0.013 (0.002)
0.040 (0.01)
0.064 (0.01)
0.50 (0.034)
0.37 (0.049)
13391
P-10-L
Central School
(ball diamond)
0.012 (0.003)
0.040
ND (0.034)
0.086 (0.034)
0.35 (0.031)
0.68 (0.031)
ND (0.005)
ND (0.007)
ND (0.029)
0.16 (0.062)
0.11 (0.070)
13394
P-ll-L
Bullock School
0.11 (0.002)
0.22
0.12 (0.022)
0.41 (0.022)
2.4 (0.042)
7.0 (0.052)
0.015 (0.003)
0.11 (0.017)
0.17 (0.037)
0.82 (0.045)
0.66 (0.045)
Notes: (1) Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported In parts per billion (ppb).
-------
Sample No.:
Field ID.:
Location:
PCDDs (PL)
2378-TCDD
Total Iso TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
APPENDIX D
Table 8
PCDDs and PCDFs
SITE #3 - Middle town, Ohio Area
Surface Soil Samples
Upwind
13377
UPW-5-L
5130 Princeton
0.003 (0.001)
ND (0.001)
ND (0.010)
ND (0.010)
0.023 (0.007)
0.20 (0.009)
13400 13387
UPW-5-L PER-13-L
5130 Princeton 2820 Packaging
0.004 (0.002) 0.004 (0.003)
ND (0.002) ND (0.003)
ND (0.044) ND (0.020)
0.072 (0.044) ND (0.020)
0.20 (0.17) 0.15 (0.014)
10.6 (0.23) 0.28 (0.014)
Perimeter
13372
PER-14-L
Perimeter 14 - West
ND (0.002)
ND (0.002)
ND (0.014)
ND (0.009)
0.073 (0.018)
0.17 (0.026)
13376
PER-15-L
Oxford Road
ND (0.002)
ND (0.002)
ND (0.039)
ND (0.039)
0.12 (0.009)
0.84 (0.071)
PCDFs (PL)
2378-TCDF
Total TCDFs
Total penta CDFs
Total hexa CDFs
Total hepta CPFs
OCOF
0.002 (0.001)
NP (0.004)
0.006 (0.004)
ND (0.001)
ND (0.008)
ND (0.008)
NP (0.018)
NP (0.026)
ND (0.016)
ND (0.036)
ND (0.023)
ND (0.056)
ND (0.010)
ND (0.028)
NP (0.026)
ND (0.046)
ND (0.012)
ND (0.012)
ND (0.017)
ND (0.024)
ND (0.005)
ND (0.015)
ND (0.015)
0.043 (0.023)
0.050 (0.026)
Notes: (1)
Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported In parts per billion (ppb).
-------
APPENDIX D
Table 9
PCDDs and PCDFs
SITE #4 - Minnesota Natural Areas
Surface Soil Samples
Sample No.:
Field ID.:
Location:
PCDDs (PL)
2378-TCDD
Total Iso TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
13373
IUD-1-L
Itasca Wilderness
ND
ND
ND
ND
0.047
0.13
(0.003)
(0.003)
(0.010)
(0.005)
(0.009)
(0.019)
13378
BSP-4-L
Bluestem Prairie
ND (0.001)
ND (0.001)
ND (0.012)
ND (0.012)
0.091 (0.010)
0.20 (0.012)
13379
KR-l-L
Kettle River
ND
ND
ND
ND
(0.002)
(0.002)
(0.004)
(0.011)
0.025 (0.009)
0.092 (0.011)
PCDFs (PL)
2378-TCDF
Total TCDFs
Total penta CDFs
Total hexa CPFs
Total hepta CPFs
OCPF
NP
ND
ND
NP
ND
ND
(0.002)
(0.002)
(0.010)
(0.010)
(0.017)
(0.024)
NO (0.002)
NP (0.009)
NP (0.009)
NP (0.009)
NP (0.010)
NP (0.001)
NP
NP
(0.013)
(0.013)
NP (0.009)
NP (0.014)
Notes: (1)
Concentrations of PCDPs, PCPFs, and detection
levels (PL) reported 1n parts per billion (ppb).
-------
APPENDIX D
Table 10
PCDDs and PCDFs
Control and Blank Samples
Blank and Control Samples
Sample No.:
Field ID.:
Location:
PCDDs (PL)
2378-TCDD
Total Iso TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
13380
Control Soil
13381
Blank Soil
13414
Can Blank
PER-5
13415
Can Blank
In-Plant Blank
13416
Can Blank
Blank Middle town
13417
Can Blank
Blank #6
0.005 (0.003)
ND (0.003)
0.081 (0.033)
0.64 (0.033)
NA
NA
ND (0.001)
ND (0.001)
ND (0.015)
ND (0.015)
0.031 (0.005)
0.27 (0.007)
ND (0.027)
ND (0.027)
ND (0.097)
ND (0.097)
ND (0.17)
0.63 (0.22)
ND (0.024)
ND (0.024)
ND (0.10)
ND (0.10)
ND (0.15)
0.74 (0.19)
ND (0.025)
ND (0.025)
ND (0.098)
ND (0.098)
ND (0.17)
0.89 (0.23)
ND
ND
ND
ND
ND
(0.03)
(0.03)
(0.11)
(0.11)
(0.19)
0.44 (0.24)
PCDFs (PL)
2378-TCDF
Total TCDFs
Total penta CDFs
Total hexa CDFs
Total hepta CDFs
OCDF
0.017 (0.005)
NA
NA
NA
NA
ND (0.001)
ND (0.011)
ND (0.011)
ND (0.012)
ND (0.013)
ND (0.039)
ND (0.039)
ND (0.021)
ND (0.35)
NA
NA
ND (0.022)
ND (0.022)
ND (0.032)
ND (0.14)
ND (0.15)
ND (0.15)
ND (0.023)
ND (0.023)
ND
ND
ND
ND
(0.14)
(0.13)
(0.19)
(0.19)
ND (0.027)
ND (0.027)
ND (0.074)
ND (0.12)
ND (0.16)
ND (0.16)
Notes: (1) Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported in parts per billion (ppb)
(2) NA - Not analyzed. Peak broadening in GC/MS
system prevented quantitation.
-------
APPENDIX D
Table 11
Toxic Organic Pollutants
SITE II - Midland, Michigan Area
Surface Soil Samples
(Results In parts per million (ppm))
Upwind
In-Plant
CAS
Number
Acid Compounds
Sample Number:
Field ID.:
13401
UPW-2-L
106-44-5 4-methylphenol
122-83-2 2,4-d1chlorophenol
876-36-5 pentachlorophenol
108-95-2 phenol
65-85-0 benzole acid
Base/Neutral Compounds
206-44-0 f1uoranthene
91-20-3 naphthalene
117-81-7 b1s(2-ethylhexyl) phthalate
84-74-2 d1-n-buty1 phtifelate
117-84-0 dl-n-octyl phthalate
84-66-2 dlethyl phthalate
56-55-3 benzo (a) anthracene
50-32-8 benzo (a) pyrene
205-99-2 benzo (b) fluoranthene
207-08-9 benzo (k) fluoranthene
218-01-9 chrysene
208-96-8 acenaphthylene
120-12-7 anthracene
191-24-2 benzo (ghi) anthracene
13. OT
1.3TC
1.3TC
1.3T
13395
UPW-4-L
13404
#1
13406
#5
13405
#10
13407
#13
13412
#14
13413
#15
12.3T
0.01T
0.01T
1.2TC
1.8C
1.2T
0.01T
5.0
11.0
8.1
0.01T
1.2
0.02T
0.01T
4.4
-------
APPENDIX D
Table 11 (continued)
Upwind In-Plant
CAS Sample Number:
Number Field ID.:
13401
UPW-2-L
13395
UPW-4-L
13404
#1
13406
#5
13405
#10
13407
#13
13412
#14
13413
#15
Base/Neutral Compounds (continued)
86-73-7 fluorene
85-01-8 phenanthrene 0.01
53-70-3 dlbenzo (a,h) anthracene
193-39-5 Indeno (1,2,3-cd) pyrene
129-00-0 pyrene 0.01
132-64-9 dlbenzofuran
91-57-6 2-methyl naphthalene
118-74-1 hexachlorobenzene 0.01
131-11-3 dimethyl phthalate 0.5
100-51-6 benzyl alcohol
87-86-5 pentachlorophenol
83-32-9 acenaphthene
Pesticides/PCBs
309-00-2 aldrin
57-74-9 chlorodane
50-29-3 4,4J-DDT
72-55-9 4,41-DDE 0.01
12672-29-6 PCB-1248 1.3
11097-69-1 PCB-1254
11096-82-5 PCB-1260
Notes and Symbols:
(1) { } - Coeluted Isomers, Indistinguishable concentration.
(2) C - Value corrected for blank concentration.
(3) T - Trace, less than detection limit.
(4) Not detected.
-------
APPENDIX D
Table 11
Toxic Organic Pollutants
SITE #1 - Midland, Michigan Area
Surface Soil Samples
(Results in parts per million (ppm))
Perimeter
Public Use
CAS
Number
Acid Compounds
106-44-5 4-methylphenol
122-83-2 2,4-dlchlorophenol
876-36-5 pentachlorophenol
108-95-2 phenol
65-85-0 benzole acid
Base/Neutral Compounds
206-44-0 f1uoranthene
91-20-3 naphthalene
117-81-7 b1s(2-ethylhexyl) phthalate
84-74-2 di-n-butyl phthalate
117-84-0 d1-n-octyl phthalate
84-66-2 diethyl phthalate
56-55-3 benzo (a) anthracene
50-32-8 benzo (a) pyrene
205-99-2 benzo (b) fluoranthene
207-08-9 benzo (k) fl uoranthene
218-01-9 chrysene
208-96-8 acenaphthylene
120-12-7 anthracene
191-24-2 benzo (ghl) anthracene
86-73-7 f1uorene
85-01-8 phenanthrene
Sample Number:
Field ID.:
13367
PER-9-6
13402
PER-8-L
13392
P-6-L
13393
P-7-L
13375
P-9-L
13391
P-10-L
13394
P-ll-L
13374
P-5-[Re]
3.3
0.2T
0.04T
0.04T
13.2T
1.3TC
1.3TC
1.3T
0.3T
O.IT
0.2T
0.04T
0.04T
,0.08T*,
I *i
0.08T
0.04T
0.08T
1.1TC
1.1TC
1.1T
0.04T
0.04T
0.04T
0.04T
0.04T
0.04T
0.2T
0.2T
-------
APPENDIX D
Table 11 (continued)
CAS
Number
Sample Number:
Field ID.:
Perimeter
13367 13402
PER-9-G PER-8-L
Public Use
13392
P-6-L
Base/Neutral Compounds (continued)
53-70-3 dibenzo (a,h) anthracene
193-39-5 indeno (1,2,3-cd) pyrene
129-00-0 pyrene
132-64-9 dibenzofuran
91-57-6 2-methyl naphthalene
118-74-1 hexachlorobenzene
131-11-3 dimethyl phthalate
100-51-6 benzyl alcohol
87-86-5 pentachlorophenol
83-32-9 accenaphthene
3.5
0.08T
0.04T
13393 13375 13391
P-7-L P-9-L P-10-L
0.04T
1.3T 0.08T
0.08T
13394
P-ll-L
13374
P-5-[Re]
0.3T
Pesticides/PCBs
309-00-2
57-74-9
50-29-3
72-55-9
12672-29-6
11097-69-1
11096-82-5
aldrin
chlorodan
4,41-DDT
4,4l-DDE
PCB-1248
PCB-1254
PCB-1260
0.01
0.02
0.03
0.05
Notes and Symbols:
(1) { } - Coeluted isomers, indistinguishable concentration.
(2) C - Value corrected for blank concentration.
(3) T - Trace, less than detection limit.
(4) Not detected.
-------
APPENDIX D
Table 12
Toxic Organic Pollutants
SITE #3 - Middletown, Ohio
Surface Soil Samples
(Results in parts per million (ppm))
Upwind
Perimeter
Public Use
Residential
CAS
Number
Sample Number:
Field ID.:
13400
UPW-5
13377
UPW-5 [D]
13387
PER-13-L
13372
PER-14-L
13376
PER-15-L
Acid Compounds
106-44-5
122-83-2
876-36-5
108-95-2
65-85-0
4-methyl phenol
2,4-dichlorophenol
pentachloro phenol
phenol
benzoic acid
___
___
_..
...
...
11. 6T
._
...
...
...
0.04T
---
---
___
...
...
---
0.08T
Base/Neutral Compounds
206-44-0
91-20-3
117-81-7
84-74-2
117-84-0
84-66-2
56-55-3
50-32-8
205-99-2
207-08-9
218-01-9
208-96-8
120-12-7
191-24-2
86-73-7
f 1 uoranthene
naphthalene
b1 s{ 2-ethyl hexyl ) phthal ate
di-n-butyl phthal ate
di-n-octyl phthal ate
di ethyl phthal ate
benzo (a) anthracene
benzo (a) pyrene
benzo (b) fl uoranthene
benzo (k) fl uoranthene
chrysene
acenaphthylene
anthracene
benzo (ghi) anthracene
f 1 uorene
1.4T
LOT
LOT
0.6T
___
0.6T
___
_
___
1.2T
___
___
_
...
1.4
...
1.2TC
2.3C
...
1.2T
1.2T
2.3T
2.3T
2.3T
2.3T
1.2T
1.2T
2.3T
1.2T
1.5T
1.5TC
1.5TC
1.5T
1.5T
2.9T
2.9T
2.9T
2.9T
_ซ_
1.5T
2.9T
0.2T
0.04T
-
...
-
0.08T
0.4
0.6*
i */
0.2T
0.04T
___
0.2T
2.3
...
0.4T
...
...
...
1.0
1.4
2.0*
i *>
1.2
0.3T
0.2T
0.8T
O.IT
LIT
LIT
1.1TC
LIT
2.2T
2.2T
2.2T
1.4T
2.3T
1.4T
1.4T
1.2T
1.2T
L2T
-------
APPENDIX D
Table 12 (continued)
Upwind
Perimeter
CAS
Number
Sample Number:
Field ID.:
Base/Neutral Compounds (continued)
85-01-8 phenanthrene
53-70-3 dibenzo (a,h) anthracene
193-39-5 Indeno (1,2,3-cd) pyrene
129-00-0 pyrene
132-64-9 dibenzofuran
91-57-6 2-methyl naphtha!ene
118-74-1 hexachlorobenzene
131-11-3 dimethyl phthalate
100-51-6 benzyl alcohol
87-86-5 pentachlorophenol
83-32-9 acenaphthene
Pestlcldes/PCBs
309-00-2
57-74-9
50-29-3
72-55-9
12672-29-6
11097-69-1
11096-82-5
aldrln
chlorodane
4,4'-DDT
4,4'-DDE
PCB-1248
PCB-1254
PCB-1260
13400 13377
UPW-5 UPW-5 [D]
2.2
1.4T
0.4T
2.0
ซ
M*ปW
1.2T
2.3T
1.2T
WMM
13387
PER-13-L
1.5T
2.9T
1.5T
ป
13372
PER-14-L
O.IT
0.04T
0.3T
0.2T
0.04T
ซ
13376
PER-15-L
1.0
0.7
1.2
0.04T
0.04T
0.04T
0.08T
0.05
0.03
0.06
Public Use
13386 13390
P-ll-L P-15-L
1.1T
1.1T
1.4T
Residential
13403
H-l-1
1.2T
1.2T
0.2
Notes and Symbols:
(1) { } - Coeluted Isomers, indistinguishable concentration.
(2) C - Value corrected for blank concentration.
(3) T - Trace, less than detection limit.
(4) Not detected.
-------
APPENDIX D
Table 13
Toxic Organic Pollutants
SITE #4 - Minnesota Natural Areas
Surface Soil Samples
(Results In parts per million (ppm))
Minnesota Natural Areas
CAS
Number
Add Compounds
Sample Number:
Field ID.:
13388
PT-l-L
106-44-5 4-methylphenol
122-83-2 2,4-dlchlorophenol
876-36-5 pentachlorophenol
108-95-2 phenol
65-85-0 benzole acid
Base/Neutral Compounds
206-44-0 f1uoranthene
91-20-3 naphthalene
117-81-7 b1s(2-ethylhexyl) phthalate
84-74-2 di-n-butyl phthalate
117-84-0 dl-n-octyl phthalate
84-66-2 diethyl phthalate
56-55-3 benzo (a) anthracene
50-32-8 benzo (a) pyrene
205-99-2 benzo (b) fluoranthene
207-08-9 benzo (k) fluoranthene
218-01-9 chrysene
208-96-8 acenaphthylene
120-12-7 anthracene
191-24-2 benzo (ghi) anthracene
86-73-7 f1uorene
24.5T
2.5TC
2.9C
2.5T
13379
KR-l-L
13378
BSP-4-L
0.04T
6.6
0.08T
0.04T
O.IT
0.04T
0.04T
13373
IWD-l-L
O.IT
0.2T
0.08T
0.2T
-------
APPENDIX D
Table 13 (continued)
Minnesota Natural Areas
CAS Sample Number:
Number Field ID.:
Base/Neutral Compounds (continued)
85-01-8 phenanthrene
53-70-3 dlbenzo (a,h) anthracene
193-39-5 Indeno (1,2,3-cd) pyrene
129-00-0 pyrene
132-64-9 dlbenzofuran
91 -57-6 2 -methyl naphthal ene
118-74-1 hexachlorobenzene
131-11-3 dimethyl phthalate
100-51-6 benzyl alcohol
87-86-5 pentachlorophenol
83-32-9 acenaphthene
Pest1c1des/PCBs
309-00-2 aldrin
57-74-9 chlorodane
50-29-3 4,4' -DDT
72-55-9 4, 4 '-DDE
12672-29-6 PCB-1248
11097-69-1 PCB-1254
11096-82-5 PCB-1260
Notes and Symbols:
13388
PT-l-L
2.5T
2.5T
...
...
...
...
...
...
...
...
...
...
...
...
...
13379 13378
KR-l-L BSP-4-L
0.04T
... ...
... ...
0.04T
... ...
0.04T
... ...
... ...
0.04T 0.08T
... ...
...
... ...
...
... ...
... ...
0.04
0.03
... ...
13373
IWD-l-L
...
...
...
...
...
0.04T
ป*ปM
ซปปป
...
...
...
...
...
...
...
ซ
...
...
(1) { } - Coeluted Isomers, Indistinguishable concentration.
(2) C - Value corrected for blank concentration.
(3) T - Trace, less than detection 1
(4) Not detected.
Imlt.
-------
APPENDIX D
Table 14
Toxic Organic Pollutants
Surface Soil Samples
(Results 1n parts per million (ppm))
Blanks and Controls
CAS
Number
Acid Compounds
Sample Number:
Field ID.:
E9199 E9200
Blank Control
Soil Soil
106-44-5 4-methylphenol
122-83-2 2,4-d1chlorophenol
876-36-5 pentachlorophenol
108-95-2 phenol
65-85-0 benzole add
Base/Neutral Compounds
206-44-0 f1uoranthene
91-20-3 naphthalene
117-81-7 bis(2-ethylhexyl) phthalate
84-74-2 dl-n-butyl phthalate
117-84-0 d1-n-octyl phthalate
84-66-2 dlethyl phthalate
56-55-3 benzo (a) anthracene
50-32-8 benzo (a) pyrene
205-99-2 benzo (b) fluoranthene
207-08-9 benzo (k) fluoranthene
218-01-9 chrysene
208-96-8 acenaphthylene
120-12-7 anthracene
191-24-2 benzo (ghi) anthracene
86-73-7 f1uorene
85-01-8 phenanthrene
0.01T
1.1
0.7
0.4
0.02T
1.0
0.6
0.01T
2.0
E8064
Blank
Soil
3.4TC
3.4TC
E8065 E8076
Control Can Lab
Soil Blank Blank
1.1
0.9TC
1.0
1.8T
2.1
1.8T
0.9T
1.8T
3.3
E8061 E8063 E8051
Blank Control Can
Soil Soil Blank
2.8
0.3T
O.IT
O.IT
O.IT
O.IT
0.06T
3.4
2.4
1.8T
1.4T
1.2T
LOT
2.2
1.8T
0.2T
0.8T
0.6T
4.0T
-------
APPENDIX D
TABLE 14 (continued)
Blanks and Controls
CAS
Number
Sample Number:
Field ID.:
E9199
Blank
Soil
E9200
Control
Soil
E8064
Blank
Soil
E8065
Control
Soil
E8076
Can
Blank
Lab
Blank
E8061
Blank
Soil
E8063
Control
Soil
E8051
Can
Blank
Base/Neutral Compounds (continued)
53-70-3 dibenzo (a,h) anthracene
193-39-5 Indeno (1,2,3-cd) pyrene
129-00-0 pyrene
132-64-9 dlbenzofuran
91-57-6 2-methylnaphthaiene
118-74-1 hexachlorobenzene
131-11-3 dimethyl phthalate
100-51-6 benzyl alcohol
87-86-5 pentachlorophenol
83-32-9 acenaphthene
0.7
0.4
1.3
Pestlcldes/PCBs
309-00-2
57-74-9
50-29-3
72-55-9
12672-29-6
11097-69-1
11096-82-5
aldrln
chlorodane
4,4'-DDT
4,4l-DDE
PCB-1248
PCB-1260
PCB-1260
* 0.087 ppb
1.8T
1.8T
2.0
0.9
1.9
0.2T
0.6T
2.4
LOT
4.0
Notes and Symbols:
(1) { } - Coeluted Isomers, Indistinguishable concentration.
(2) C - Value corrected for blank concentration.
(3) T - Trace, less than detection limit.
(4) Not detected.
-------
APPENDIX D
Table IS
Tentatively Identified Compounds
SITE fl - Midland, Michigan Areas
Surface Soil Samples
(results in parts per Billion (ppm))
Upwind
Inplant
Perimeter
Public Use
CAS
Number
17301-30-3
54410-98-9
127-18-4
79-34-5
33374-28-6
62016-37-9
630-07-9
638-66-4
74685-33-7
17301-27-8
52783-43-4
629-96-9
69576-82-5
29053-04-1
79-00-5
55499-02-0
103-23-1
Sample Number:
Field Identification:
Compound list
Undecane, 3,8-dimethyl
1-nonene, 4,6,8-trlmethyl
ethene, tetrachloro
ethane, 1,1, 2,2- tetrachloro
1,2-benzene dlcarbozylic acid
2, butoxyethyl butyl ester
octane, 2,4.6-trimetnyl
pentatriacontane
octadecanal
3-elcosene, (E)-
undecane. 2,10-dimethyl
nona decanal
1-elcosanol
l,lt-bicyclo[2.2.2]ocune
(alpha-methylpropyl )-
ethane. 1.1,2-trichloro-
hexanedloicalid, b1s(2-
ethylhexyl) ester
13401 13395 13404 13406 13405 13407 13412 13413 13367 13402 13374
UPV- UPU- PER- PER-
2-L 4-L fl 15 110 113 114 115 8-t 9-G P-5
39 tt fi ..........
6 1ft ...........
HfiltA ..........
12.8
1.99 0.66
1.99
2.99
*
13392 13393 13375
P-6-L P-7-l P-9-L
1 AQ
27
I W
1.6
13391
P-
10-BD
2.7
23.8
1.33
1.55
1.14
13394
P-
11-L
-------
Upwind
APPENDIX D
Table IS (continued)
(results In parts per million (ppn)}
Inplant
Perimeter
Public Use
CAS
Number
5675-51-4
74685-29-3
630-06-8
21048-65-9
44635-30-6
57-88-5
1002-84-2
57-10-3
111-04-9
4CCt.C1.Jt
tt.47.fi
471-68-1
1058-61-3
123-08-0
4429-77-0
18835-33-1
37571-80-5
559-74-0
Sample Number:
Field Identification:
Compound List
I.l2-dodecand1o1
9-elcosene, (E)-
hexatrlacontane
1-decanol, 2-ethyl-
S-e1cosene, (E)-
cholest-S-en-3-ol, (3-.beta.)-
pentadecanolc acid
hexadecanolc acid
7.hปxitarซn.l .nl 3 7 11 1C.
tetramethyl-[R-[R*. R*-(ฃ)]J
rnnrt. >.-ซ. l.nl 11. hซta 1.
ct1i*mact.th**fh_l.n1 11. h*ta '
olean-12-ene
st1gmast-4-en-3-one
benzal dehyde-4-hydroxy
cjrcloheptadecanol
1-hexacosene
stl9Basta-5>22-d1en-3r
ol, (3-.beta. 22E)-
0:A-Fr1edooleanan-3-one
unknown aliphatic acid
13401 13395 13404 13406 13405
ura- UPW-
2-L 4-L 11 15 110^
3.29
39.5
11.5
5.84
4.94
5.49
I 0.66 - -
13407 13412 13413 13367 13402 13374 13392
PER- PER-
113 114 115 8-L 9-6 P-5 P-6-L
0.49 2.8
0.64UK 0.68UK 4.8 4.6
20
So
- - - _ . .. 10
043
1.05
1.06
1.10
2.1
5.7
0.87
0.85
0.76
13393 13375 13391
P-
P-7-L P-9-l 10-80
1.9
2.3
2 5
1 i
3 4
1.9
0.47
13394
P-
_U-L
1.1
2.8
4 2
066
1.4
4.2
-------
Upwind
APPENDIX D
Table IS (Continued)
(results In parts per Million (ppซ0)
Inplant
Perimeter
Public Use
CAS
Number
134-96-3
9AQA.AA.A
tH3W"^O"*
6624-79-9
143-07-7
481-17-4
57-11-4
15968-05-5
77-90-7
54932-67-1
6330-09-2
31504-88-8
605-45-8
3790-71-4
288-13-1
60-12-8
Sample Number: 13401 13395 13404 13406 13405 13407
UPU- UPW-
Fleld Identification: 2-L 4-L fl IS 110 ซ13
Compound list
3,5-dlMethoxy
1-dotrfacontano) -
dodecanolc acid .....
st
-------
Upwind
APPENDIX 0
Table 15 (continued)
(results In parts per Million (ppm))
Inplant
CAS
Nimber
103-82-2
104-54-1
501-52-0
544-63-8
5009-33-6
1454-85-9
490-99-3
106-02-5
593-45-3
629-99-2
112-95-8
630-01-3
661-19-8
100-41-4
108-94-1
92-69-3
6738-04-1
3933-94-6
2432-11-3
141-04-8
6093-03-4
Sample dumber: 13401 13395 13404
UPU- UPH-
Fleld Identification: 2-1 4-1 11
Compound List
benzeneacetlc acid
2-propen-l-ol, 3-phenyl-
benzene propanolc acid
tetradecanolc acid - -
ll-dodecen-2-one
1-heptadecanol
cyclohexanol, S-nethyl-2-(l-
ethylethyl)-, (1-. alpha., 2)
oxacyclohexadecan-2-one
octadecane
pentacosane
elcosane
hexacosane
l-doco$anol
benzene, ethyl -
cyclohexanone
l.l'-btphenyl-4-ol
l.l'-biphenyl, 2-phenoxy-
l.l'-blpbenyl, 4-phenoxy-
l.r:3M"-terpheny1-2'-ol
hexanedlolc acid. b1s(2-ซ*thyl-
propyl) ester
l.r:3' ,r-terpbenyl-4'-ol
\ 13406 13405 13407 13412 1341.
15 110 113 114 115
* <
....
....
....
....
-
....
.....
.....
.....
.....
.....
.....
0.7 4.7
0.84
3.7
1.2
1.7
3.2
1.1UK
9.4
Perimeter
Public Use
PER-
8-L
PER-
9-6
P-
P-5 P-6-L P-7-l P-9-L 10-BD
P-
U-L
2.4
3.0
1.5
0.5
0.51UK
1.2
0.65UK
1.2UK
0.51
1.1
1.5
13.0
4.7
-------
Upwind
APPENDIX 0
Table 15 (continued)
(results in parts per million (ppm))
Inplant
CAS
Number
131-18-0
17851-53-5
28080-85-5
52161-54-3
29812-79-1
922-28-1
6570-92-9
55334-42-4
84-61-7
108-10-1
629-20-9
98-82-8
33021-02-2
105-05-5
98-86-2
768-49-0
21898-96-4
Sample Number: 13401 13395 13404
UPU- UPW-
Field Identification: 2-L 4-L 11
Compound List
1,2-benzenedicarboxylic acid.
dipentyl ester
1.2-benzenedicarboxylic acid,
butyl 2-methylpropylester
10-undecenoic acid, octyester
benzene, l,lMl.4-dimethyl-l-
butene-l,4-diyl) bis-
hydroxylMine, o-decyl-
heptane. 3,4-diwethyl
pentane, l-brซmo-3.4-dlMethy1-
dodecane, 1,2-dibromo-
l,2-benzenedicarboxyl1c acid.
dlcyclohexylester
2-pentanone, 4-methyl- ...
1,3,5,7-cyclo octatetraene
benzene. (1-methyl ethyl )-
propane, l-(l.l-dlmethylethoxy)-
2 -methyl -
benzene, l,4-diethy1-
ethanone, 1-pnenyl-
benzene (2-methyl-l-propenyl)-
tricyclo 4.3.1.13,8 undecane,
1-brono-
13406 13405
15 110
1.9UX
6.1UK
3.1UK
1.3UK
1S.OUK
2.2UK
0.84UK
7.8UK
4.0UK
m
m
13407 13412 1341.
113 114 115
1.1
8.4 0.4!
4.5
1.5
6.3
2.3
1.4
1.6
Perimeter
Public Use
PER-
8-L
PER-
9-G
P-
P-5 P-6-L P-7-l P-9-l 10-BD
P-
U-l
-------
APPENDIX 0
Table IS (continued)
Upxlnd
(results In parts per million (ppro))
Inplant
Perimeter
Public Use
Sample Number:
Field Identification:
CAS
Number Compound List
2049-95-8 benzene (1,1-dimothyl propyl)-
101-84-8 benzene, l.l'-oxybis-
100-42-5 benzene, ethenyl-
17104-67-5 2-naphthonitrlle, 5,6,7,8-
tetrahydro
780-25-6 benzene methanamine,
n-(pheny1methylene)-
56728-02-0 benzene l,r-2-methy1-2-
(phenylthio) cyclopropyliden
35825-28-6 stannane, cyclopropyltriethyl
1120-21-4 undecane
33046-84-3 2-pentenal, S-phenyl-
108-88-3 benzene, methyl
5131-66-8 2-propanol, 1-butoxy-
2091-29-4 9-hexadecenoic acid
13401 13395 13404 13406 13405
UPU- UPU-
2-L 4-L 11 15 110
.
13407 13412
*13 114
0.62
1.2UK
0.79
1.3UK
5.7UK
0.69UK
26.0
0.5RS
0.57UK
0.31
13413 13367 13402 13374 13392 13393 13375 13391
PER- PER- P-
115 8-L 9-6 P-5 P-6-L P-7-L P-9-L 10-BD
0.49
0.47UK
0.37
0.74UK
1339'
P-
11-1
Notes:
UK * Unknown, not in NBS library.
RS * Reasonable identification;
retention time compatibility.
-------
APPENDIX D
Table 16
Tentatively Identified Compounds
SITE #3 - Mlddletown, Ohio
Surface Soil Samples
(results in parts per million (ppm))
Upwind
Perimeter
Public Use
CAS
Number
4537-
4536-
2719-
2719-
54986-
4534-
4534-
4534-
10544-
15-9
86-1
62-2
64-4
44-6
49-0
50-3
54-7
50-0
33374-28-6
103-23-1
630-06-8
Sample Number:
Field Identification:
Compound List
benzene (1-butylheptyl)
benzene(1-propyloctyl)-
benzene!1-pentylheptyl)-
benzenet1-propylnonyl)-
benzene(l,33-trimethylnonyl)
benzened-pentyloctyl )-
benzene(l-butylnonyl)-
benzened-hexyloctyl )-
sulfur, Mol(58)
phenanthrene-methyl isomer
aliphatic hydrocarbon
polyunsaturate
1,2-benzenedicarboxylic
acid, 2-butoxyethyl
butyl ester
hexanedioicalid, bis(2-
ethylhexyl)ester
hexatriacontane
13400 13377 13387 13372 13376
UPW-5
UPW-5
6.0
2.7
10.0
3.1
2.3
4.4
2.0
4,3
3.1
2.0
5.5
3.30
[D] PER-13 PER-14 PER-15
47.6
2.78
13.9
13386
P-ll
13390
P-15-L
Residential
13403
H-l-1
24.9
-------
APPENDIX D
Table 16 (continued)
(results in parts per million (ppm))
Upwind
Perimeter
Public Use
Sample Number:
Field Identification:
CAS
Number Compound List
117-82-8 1,2-benzenedicarboxil1c
acid, b1s(2-methoxyethyl)
124-25-4 tetradecanal
506-52-5 1-hexacosanol
17301-30-3 undecane, 3,8-dimethy1-
127-18-4 ethene, tetrachloro
79-34-5 ethane, 1,1,2,2-tetrachloro-
638-66-4 octadecanal
74685-33-9 3-elcosene, (E)-
5675-57-4 1,12-dodecandlol
1002-84-2 pentadecanoic acid
100-83-4 benzaldehyde, 3-hydroxy
148-53-8 benzaldehyde, 2-hydroxy-
3-methoxy
57-10-3 hexadeconoic acid
150-86-7 2-hexadecen-l-ol, 3,7,11,15-
tetramethyl-[R-CR*,R*-(E)]]
111-03-5 9-octadecenoic acid
(Z)-2,3-dihydroxypropylester
25154-56-7 nonacosanol
4651-51-8 ergost-5-En-3-ol, (3-.beta.)-
83-47-6 stigmast-5-en-3-ol,
(3-.beta. 24S)
13400 13377 13387 13372 13376
UPW-5
UPW-5 [D] , PER-13 PER-14 PER-15
1.86
12.4
27.8
1.47
3.09
5.05
1.55
1.83
10.3
21.8
2.02
-
-
_
_
1.3
0.51
0.83
3.6
0.83
1.101
2.6
0.64
1.3
0.79
13386
P-ll
1.73
1.44
11.8
19.4
13390
P-15-L
Residential
13403
H-l-1
15.2
1.42'
1.57
0.930
-------
APPENDIX D
Table 16 (continued)
(results In parts per million (ppm))
Upwind
Perimeter
Sample Number:
Field Identification:
CAS
Number Compound List
471-68-1 olean-12-ene
1058-61-3 stigmast-4-en-3-one
2091-29-4 9-hexadecenolc acid
630-07-9 pentatrlacontane
62108-21-8 decane, 6-ethyl-2-methyl-
74685-29-3 9-e1cosene(E)-
10486-19-8 trldecanal
630-06-8 hexatrlacontane
74645-98-0 dodecane, 2,7,10-trlmethyl-
50871-03-9 1-decene, 3,4-dimethyl
62016-37-9 octane, 2,4,6-tr1methyl
13400
UPU-5
;
13377
UPU-5
[D]
:
13387 13372
PER-13 PER-14
0.72
0.91
13376
PER-15
0.62
0.95
0.89
Public
13386
P-ll
16.2
3.93
Use
13390
P-15-L
23.3
-23.0
2.62
13.1
Residential
13403
H-l-1
4.76
1.14
1.78
-------
APPENDIX D
Table 17
Tentatively Identified Compounds
SITE #4 - Minnesota Natural Areas
Surface Soil Samples
(results In parts per million (ppm))
Minnesota Natural Areas
Sample Number:
Field Identification:
CAS
Number Compound List
74645-98-0 dodecane, 2,7,10-trlmethyl-
33374-28-6 1,2-benzenedicarboxyllc acid
2-butoxyethyl butylester
50871-03-9 1-decane, 3,4-dlmethyl-
62016-37-9 octane, 2,4,6-trlmethyl-
13828-37-0 cyclohexanemethanol
4-(l-methylethyl)-c1s
148-53-8 benzaldehyde, 2-hydroxy-3-
methoxy
57-10-3 hexadecanolc acid
123-08-0 benzaldehyde, 4-hydroxy
143-07-7 dodecanolc acid
2004-39-9 1-heptacosanol
150-86-7 2-hexadecen-l-ol, 3,7,11,15-
tetramethyl-[R-[R*. R*-
83-47-6 stigmast-5-en-3-o1,
(3-.beta., 24S)-
1058-61-3 st1gmast-4-en-3-one-
83-48-7 st1gmasta-5,22-d1en-3-ol,
(3-.beta., 22E)-
13388
PT-l-L
3.92
35.2
2.45
3.92
5.23
13379
KR-l-L
0.59
1.1
0.44
0.48
0.43
13378
BSP-4-L
1.2
2.2
2.0
1.4
13373
IWD-l-L
-------
CAS
Number
13151-81-0
2490-48-4
14021-23-9
20475-86-9
Sample Number:
Field Identification:
Compound List
undecane, 6-cyclohexyl-,
6-cyclohexyl
1-hexadecanol, 2-methyl
D-Friedoolean-14-ene,
3-methoxy-(3-.beta.)-
urs-12-en-24-o1c acid,
3-oxo-, methyl ester, {+)
APPENDIX D
Table 17 (continued)
(results in parts per million (ppm))
Minnesota Natural Areas
13388
PT-l-L
13379 13378
KR-l-L BSP-4-L
0.78
4.4
3.3
13373
IWD-l-L
0.9
-------
APPENDIX D
Table 18
Tentatively Identified Compounds
Surface Soil Samples
(results 1n parts per million (ppm))
Sample Number:
Field Identification:
CAS
Number Compound List
123-08-0 benzaldehyde, 4-hydroxy
621-59-0 benzaldehyde, hydroxy,
methoxy Isomer
55638-41-0 5-octen-2-one-6-methyl-8-
(2,6,6-trlmethyl-l-
cyclohexen-1-yl)-
2091-29-4 9-hexadecenolc acid
62016-37-9 octane, 2,4,6-trlmethyl-
96-48-0 2(3H)-furanone, dlhydro
3648-21-3 l,2-benzened1carboxy11c acid
dlheptylester
629-99-2 pentacosane *
57-10-3 hexadecanolc acid
4536-86-1 benzene(1-propyloctyl)-
18435-22-6 tetradecane, 3-methyl-
2719-62-2 benzened-pentylheptyl )-
2719-64-4 benzene(1-propylnonyl)-
2400-00-2 benzene!1-ethyldecyl)-
4534-49-0 benzeneU-pentyloctyl)-
4534-50-3 benzene(1-butylnonyl)-
4534-54-7 benzene(l-hexyloctyl)-
4534-52-5 benzene(1-ethylundecyl)-
E8061
Blank
Soil
2.0
1.3
1.8
B 1 a
E8051 E8064
Can Blank
Blank Soil
13.5
n k s
E8076
Can
Blank
and
E9199
Blank
Soil
.0.6
4.0
Control s
E8063 E8065
Control Control
Soil Soil
E9200
Control
Soil
0.26
0.573
0.57
3.13
0.59 UK
4.0
16.0
4.9
3.2
6.7
3.2
5.2
2.0
3.11
2.61
10.2
4.34
3.6
5
2
09
97
2.75
2.4 UK
11.0
3.6
2.6
-------
APPENDIX D
Table 18 (continued)
(results In parts per million (ppm))
CAS
Number
4534-51-4
2719-61-1
10544-50-0
1002-17-1
17301-30-3
54004-41-0
25368-56-3
211662-16-8
62108-21-8
54410-98-9
74645-98-0
25368-56-3
100-83-4
21964-48-7
2719-63-3
2400-03-5
4534-54-7
2400-04-6
613-12-7
593-45-3
55162-61-3
Sample Number:
Field Identification:
Compound List
benzene(1-propyldecyl )-
benzened-methyl undecyl )-
aliphatic hydrocarbon
sulfur, Mo1(58)
methyl phenanthrene
dimethyl naphthalene Isomer
decane, 2,7-dlmethyl-
undecane, 3,8-dlmethyl-
l-pentanolt 4-methyl-2-propyl-
2-nonanone, 9-hydroxy-
2,4-dodecandlenal, (E,E)-
decane, 3-ethy1-2-methyl-
1-nonene, 4,6,8-trfmethyl-
dodecane, 2,7,10-trlmethyl-
2-nonanone, 7-hydroxy-
benzaldehyde, 3-hydroxy-
1,12-trldecadlene
benzenetbutyl octyl)-
benzene(1-propylheptadecyl)-
tetradecane, 7-phenyl-
benzene(1-butylhexadecyl)-
anthracene, 2-methyl
octadecane
tetracontane, 3,5,24-trlmethyl-
E8061
Blank
Soil
E8051
Can
Blank
B 1 a
E8064
Blank
Soil
n k s
E8076
Can
Blank
and C
E9199
Blank
Soil
o n t r o
E8063
Control
Soil
1 s
E8065
Control
Soil
E9200
Control
Soil
7.59
11.4
18.0
8.06
9.17
3.97
15.3
13.8
10.7
1.3
0.73 UK
2.4
3.4
8.4
5.6
3.3
2.3
2.95
1.9 UK
2.1 UK
3.2 UK
1.4 UK
0.45
1.0 UK
-------
APPENDIX D
Table 18 (continued)
(results in parts per million (ppm))
Sample Number:
Field Identification:
CAS
Number Compound List
630-06-8 hexatriacontane
2922-51-2 2-heptadecanone
17302-27-1 nonane, 2,5-dimethyl
294-62-2 cyclododecane
131-18-0 1,2-benzenedicarboxylic acid,
dipentylester phthalate
56554-96-2 2-octadecanal
1921-70-6 pentadecane, 2,6,10,14-tetramethyl
629-59-4 tetradecane
544-76-3 hexadecane
569-41-5 naphthalene, 1,8-dimethyl-
18335-17-6 benzened-propylpentyl )-
4170-84-7 benzene(l,l-diethylpropyl)-
17376-04-4 benzene(2-iodoethyl)-
E8061
Blank
Soil
E8051
Can
Blank
B 1 a
E8064
Blank
Soil
n k s
E8076
Can
Blank
and C
E9199
Blank
Soil
o n t r o
E8063
Control
Soil
1 s
E8065
Control
Soil
E9200
Control
Soil
1.5 UK
0.5 UK
0.92 UK
1.3 UK
0.74 UK
1.0 UK
4.4
1.3
1.4
2.1
2.8 UK
0.64 UK
1.3 UK
Notes:
UK = Unknown, not in NBS library.
RS = Reasonable identification;
retention time compatibility.
-------
APPENDIX E
Consideration of Public Health Risks
Attachment 1 Centers for Disease Control (January 22, 1985)
Attachment 2 USEPA, Chlorinated Dloxlns Work Group (April 4, 1985)
-------
DEPARTMENT OF HEALTH & HUMAN SERVICES
Public Haalth Sซrvicซ
Centers for Disซau Control
o,te January 22, 1985
From Chief, Superfund Implementation Group
Memorandum
Attachment 1
Centers for Disease Control
(January 22, 1985)
Subject Review: Preliminary Dioxin Study
Midland, Michigan
To Louise A. Vabinski
Public Health Advisor
6TA lesion V
The material you submitted on the subject site has been reviewed by staff of
both the Hational Institute for Occupational Safety and Health, and the
Center for Environmental Health, Centers for Disease Control. I hope that
you find their comments useful.
COMMOTS
The reviewers found that the precision, accuracy, and quality control ware
acceptable. They were aware that all field samples had not been analyzed.
They believe that the study represents a state-of-the-art attempt to measure
the polychloroinated dibenzo dioxlns (PCDD's) and furans (PCDF's) in soil.
The surface soil 2,3,7,8-TCDD content is well below the 1 ppb level at which
some action to limit human exposure to residential soils needs to be
considered (Kimbrough, H.D., Falk, H. Stehr, P. and Tries, G., "Health
Implications of 2,3,7,8-Tetrachlorodibencodioxin (TCDO) Contamination of
Residential Soil." 3. Toxicol. Environ. Health 14:47-93, 1984.) Thus, we
agree that levels found in residential soils, including downspout areas do
not represent a significant health risk to either the persons living there
or the public at large.
the results from public use areas are also well below the 1 ppb level and
represent an acceptable risk as defined in Kimbrough et. al., 1984.
The soil data in the chemical plant site are consistent with data from
similar chemical production facilities. Without seeing the actual work
site, it is difficult to assess the degree of potential worker exposures
from this soil contamination. However, based on experience at similar
facilities, it is doubtful whether any significant dioxin exposure would
occur unless the soil was disturbed due to construction or similar
activities. An asphalt or other cover material would be a prudent approach
to minimize this hazard.
The ambient air levels presented in Exhibit 5 at the maximum of 0.21 pg/m3
are unlikely to be a risk for bioaccumulation through the respiratory route.
-------
Pag* 2 - Louis* A. Pabinski
tto f**l that th* warning on sating flab from th* Tittabawaaa** tlv*r ahould
b* continued b*caua* of th* likelihood that fiซh. particularly bottom
f**d*rs. will eontinu* to bioaccumlat* th* anall awnmta pr*a*nt in
detritus and ป*diJMmt.
Th* i*vปle of othซr chemicals, all less hazardous than 2,3.7,8-TCDD. ar*
fairly low and ar* within th* rang* of lev*la found g*n*rally in th*
enviroRMnt and do not r*pr***nt an unaec*ptabl* rl*k.
Th* l*velg ef oth*r dioxine (PCDD's) and furan* (PCDP'ป> that w*r* found in
varioua ar*aa Clxhiblt 7) ar* also I*M than 1 ppb. Th* ralatlv* proportion
of PGDD** and PCDP'a in th* various aamplaa appaar aittllar although th* high
proportion of "BD'v" m*k*a thia difficult to *ป. Th*a* eh*mlcalป ar*
Ittas toxie than th* 2r3,7,8-TCOD, Son* in-plant SMasur*in*nta ar* hlgh*r
(incinarator, w*8t.of building 934>, but ar* unllk*ly to pos* hasard,
sine* dir*et contact of th* g*n*ral public and of work*ra would b* at a
niniatn.
W* auซt atat* that th*ป* data r*pr*a*nt current l*v*la and w* can nak* no
cotcnant about past *nvironiD*ntal l*v*la, and pot*nti*l rlika that nay hav*
*xiat*d in th* paat. /
Gซorgi
-------
Attachment 2
.. USEPA, Chlorinated Dioxins Work Group
(April 4> 1985)
| UNITED STATES ENVIRONMENTAL PR
' WASHINGTON, D.C. 20460
April 4, 1985
OFFICE OF
PESTICIDES AND TOXIC SUBSTANCES
MEMORANDUM
SUBJECT: Assessment of CDD/CDF Levels Found in
Midland, Michigan
FROM: Donald G. Barnes ,jn
-------
-2-
Further, CDC has indicated that in many exposure situations
other than residential soils, lซvels higher than 1 ppb could be
tolerated. An exception discussed by CDC is contaminated soil
in pastureland. Such cases would require closer analysis;
however, the CDC understands that none of the Region V samples
came from areas in which livestock graze.
For areas within the perimeter of the Dow plant, some of
the soil samples did contain residues levels above 1 ppb. It
is the understanding of the CDWG that Dow and Region V have
taken action at these locations to reduce the levels and/or the
associated exposure.
Fish
The data supplied by Region V indicate that some of the
specimens of certain species of fish taken from the Tittawabassee
River exceed the 25 and 50 ppt "levels" of concern of 2,3,7,8-
TCDD identified by the Food and Drug Administration (FDA).
The data from this set of samples appears to be consistent
with data from earlier studies on fish taken from the same river.
It is our understanding that other CDDs/CDFs were not
found in these samples; therefore, the 2,3,7,8-TCDD equivalents
approach was not used in this case.
Further, it is our understanding that on the basis of the
earlier studies, governmental authorities in Michigan have
issued a recommendation against consumption of fish from the
Tittawabassee River. Based upon the similarity between the
earlier and the recent data, the CDWG sees no reason to relax
whatever restrictions have been deemed appropriate to address
health concerns associated with consumption of fish from this
river.
Attachments
-------
TABLE 6
PCDDs and PCDFs
SITE II - MIDLAND, MICHIGAN AREA
SURFACE SOIL SAMPLES
Public Use Areas
Sample No.: 13374 13392 13393 13375 13391 13394
Field ID.: P-5-L P-6-L P-7-L P-9-L P-10-L P-ll-L
Central School
,<, Location: County Line Rd. Mapleton School. Longvlew School Virginia Park (ball diamond) Bullock School
\\A. "
^ PCDDs -tDL)
l 2378-TCDD 0.003 0.001'
.01 Total Iso TCDDs ND 0.001
.xTotal penta CDDs ND 0.014
.evTotal hexa CDDs 0.067 (0.007
oOTotal hepta CDDs 0.35 (0.013
'oOCDD 3.1 0.096
PCDFs (DLJ,
> .<ปป 0.015 (0.003)
I - 0.040
ND (0.035
,"" 0.063 i 0.035
- 0.38 (0.028
_^ 0.86 (0.027;
00 t
,12378-TCDF NO 0.002)- ND (0.005)
ooiTotal TCDFs ND 0.002) -
.'Total penta CDFs ND 0.01) - ND (0.008
,ปiTotal hexa CDFs ND 0.01) - ND 0.024
<06iTotal hepta CDFs 0.065 0.02) " 0.14 0.043
oOCDF 0.044 0.023Jb_ 0.10 0.071
.ซซ 0.078 (0.003) .air 0.076 (0.003]
- 0.17 ../JL 0.29
- Interference 0.10 (0.018
.*ป 0.34 (0.02) .ซ** 0.24 (0.018
- 2.3 (0.055) .ป* 0.41 (0.093
^ 7.0 (0.068) ^ 12.0 (1.5)
1- 0.013 (0.007) .0-* 0.013 (0.002,
ND (0.025) ~ 0.040 (0.01)
0.26 (0.036). ."v 0.064 (0.01)
0.72 (0.021) .ปซ' 0.50 (0.034
0.64 (0.037) - 0.37 (0.0491
I ,^ 0.012 (0.003)
* 0.040
ป ND (0.034
ป' 0.086 (0.034
- 0.35 (0.031
~1 0.68 (0.031!
I ,<>< ND (0.005]
."ป ND (0.007
.ป NO (0.029
)',-' 0.16 (0.062,
1 - 0.11 (0.070,
*tg ซi*
<**^^ sฃ
.oป 0.11 (0.002)."v
- 0.22 .'*
- 0.12 (0.022).<
."ซ> 0.41 0.022)/
- 2.4 0.042)/
- 7.0 0.052)-
,t>if .|!H
1- 0.015 (0.003K
> - 0.11 (0.017)-
i - 0.17 (0.037)/
- 0.82 (0.045).-'
- 0.66 (0.045)'
Notes:
(1) Concentrations of >CU0s, PCDFs, andSfet^ctlon
levels (DL) reported In parts per billion (ppb),
-------
Attachment 1
TABLE 6
PCDDs and PCDFs
SITE fl - MIDLAND, MICHIGAN AREA
SURFACE SOIL SAMPLES
Upwind
Dow Chemical
In-Pi ant
13401 13395
UPU-2-L UPH-4-L
Pleasant
View School 4853 U. Kent
ND
ND
ND
ND
0.15
0.34
0.004
0.004
0.024
0.024
0.024
0.026
ND {
ND
ND
0.17
0.33
[0.004)
0.023)
0.023
0.034
0.034
ND (0.004) - ND (0.004)
ND
ND
ND
ND
0.008]
0.022
0.031
0.051,
ND <
ND
ND
^_ ND
0.008
0.023
0.028
0.045
PCDDS (DL)
1 2378-TCDD
tH Total Iso TCDDs
.a Total penta CDDs
!oV Total hexa CDDs
00i Total hepta.CDDs
' cOCDD
PCDFs (DL)
.' 2378-TCDF
.001 Total TCDFs
.1 Total penta CDFs
.iiTotal hexa CDFs
<*>iTotal hepta CDFs
oOCDF
Notes: (1) Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported In parts per billion (ppb).
3.5 (0.039) **
0.45 (0.06)
0.27 (0.007
0.32
0.24
4.0
75.0
375.0
0.067
0.067
0.9)
1.3)
0.027 (0.007
0.90
3.1,
15.4
8.6
0.14
0.13
0.38;
0.48
.0*
"*
.131
-------
Attachment 2
DRAFT
CHLORINATED DICKINS WORKGROUP POSITION DOCUMENT
March 16, 1985
RISK ASSESSMENT PROCEDURES FOR MIXTURES OF CHLORINATED DIOXINS
AND -DIBENZOFURANS (CDDs and CDFs)*
I. Summary
EPA increasingly Is confronted with the need to determine the
risks inherent in exposure to materials such as soot, incinerator
flyash, industrial wastes, soils, etc. These materials often involve
the potential for exposure to a mixture of chlorinated dibenzo-p-dioxins
(CDDs) and chlorinated dihenzofurans (CDFs). In response, the Chlorinated
Dioxins Work Group (CCWG) is proposing a method for assessing the
human health risks posed by mixtures of these chemicals.
The CDWG has discussed several approaches for making such assessments
and has concluded that a direct biological assessment of the toxicity
of complex mixtures of CDDs/CDFs is preferred. Therefore, research
to develop appropriate methods of this type should be supported. In
the interim, however, the CDWG believes that a reasonable estimate
of the toxic risks can be made by taking into account the distribution
of CDD/DCF congeners or homologues that are estimated to have the
greatest toxic potential. This document describes the recommended
procedure for generating the "2,3,7,8-TCDD equivalents" of complex
mixtures of CDDs/CDFs, based upon congener- or homologue-specific
data and for using such information in assessing risk.
* Refer to Appendix for precise nomenclature used in this paper.
-------
-2-
The recommendations are summarized in the Table III.
The CDWG acknowledges that this procedure is not based on a
thorougly established scientific foundation. It represents a consensus
recommendation on science policy. Consequently, assessors and risk
managers are urged to use informed discretion when deciding to what
situations the procedure can be appropriately applied.
II. The Need for a Procedure for Assessing the Risk Associated with
Exposure to Complex Mixtures of CDDs/CDFs
During the late 1970s, the Agency was faced with assessing the
human health significance of exposure to 2,3,7,8-tetrachlorodibenzo-
p-dioxin (2,3,7,8-TCDD). In preparation for the cancellation hearings
for the herbicide 2,4,5-trichlorophenol and silvex, the Agency's
Cancer Assessment Group generated a quantitative cancer risk assessment
for 2,3,7,8-TCDD. This assessment was later adapted for use in the
Water Quality Criteria (WQC) Document for 2,3,7,8-TCDD. In addition
to carcinogenicity concernes, the \JQC contains an assessment of
systemic toxicitv, based on reproductive effects produced by 2,3,7,8-TCDD.
The Agency's attention for CDDs and CDFs has expanded more recently.
For example, the current draft of the Health Assessment Document
prepared for the Air Program contains a quantitative risk assessment
for a mixture of hexachlorodibenzo-p-dioxins (HxCDDs), based upon
carcinogenicity studies conducted by the National Cancer Institute.
However, as early as the late 1970s, it became clear that exposure
situations existed in the country which involved more than simply
-------
-3-
2,3,7,8-TCDD and/or the mixture of two HxCDDs, the only CDDs/CDFs
for which toxicological data existed that lent themselves to Quantitative
risk assessments. Specifically, data on emissions from combustion
sources (e.g., hazardous waste and municipal waste incinerators) and
contents of waste from certain industrial production processes indicated
that the majority of the 75 CDDs and 135 CDFs could be detected in
the environment.
Given the high potency and strong structure-activity relationship
exhibited in quantitative studies of CDDs and CDFs, the CDWG recognizes
that the potential risks-posed by the congeners other than 2,3,7,8-TCDD
and the mixture of two isomers of HxCDD also need to be addressed.* Detailed
consideration of the toxicity the vast majority of the CDDs/CDFS is limited
by the lack of long-term toxicology studies. Further, it is
unlikely that many expensive long-term test results will be available
soon. For example, research on 2,3,7,8-TCDD has been underway for
more than two decades at an estimated cost in the hundreds of millions
of dollars. Although this chemical has been investigated to a much
greater extent than any of the other CDDs/CDFs, unanswered questions
remain. As noted below, the CDWS believes that it would be unwise,
uneconomical and unnecessary to conduct such extensive testing on
each of the CDD/CDF congeners prior to conducting an assessment of
their risks.*
* In the early 1980s, the Agency developed an approximate method for
assessing the risks of the emission of 2,3,7,8-TCDD-like compounds
associated with the high temperature incinceration of PCBs and combustion
of municipal waste (USEPA, T980; USEPA, 1981); cf. Table III. The
procedure presented in this document is a refinement of that approach.
-------
-4-
III. Approaches to Hazard Assessment for CDD/CDF Mixtures
A. Overview.
1. Preferred Practical Approach Toxicity Assay of Mixtures
In the first instance, an assessment of the toxicity of a mixture
of chemicals is best accomplished by direct evaluation of its toxic
effects, e.g., by determining the effects of chronic exposure in an
experimental animal. Such an assessment is time consuming and costly
and would theoretically have to be performed for each of the many
mixtures of environmental importance. Therefore, this idealized
approach is not likely to be achieved.
An alternative, practical approach to hazard assessment of a
mixture is to have an assay that indirectly provides a measure of
the mixture's potential chronic toxicity. In the case of mixtures
containing CDDs and CDFs, short term assays are under development
that directly determine the 2,3,7,8-TCDb-like response which can be
used as a measure of the toxicity of the mixture as a whole. Such
assays, which take advantage of the similar toxic manifestations
induced by CDDs and CDFs, have been used to assess the potential
health hazards of exposure to CDD/CDF-contaninated soot from PCB
fires (Eadon, 1982; Gierthy, 1984; Gravitz, 1983), and predicting
the potential toxicity of incinerator flyash (Sawyer, 1983). The
development of such "mixture assays" assays is progressing rapidly.
While additional work is required to more fully validate the assay
findings for specific toxic endpoints, especially chronic effects,
-------
-5-
data have been presented that indicate correlations with subchronic
effects of CDDs/CDFs (Safe, 1984). The CDWG recognizes the importance
of this approach in implementing its regulatory strategy for
2,3,7,8-TCDD-like chemicals and encourages research in this area.
2. Alternative Approach Additivity of Toxicity of Components
In the absence of more fully developed "mixture assays", however,
the CDWG recognizes the viability of a second approach to assessing
the risk posed by a mixture of CDDs/CDFs. First, components in a
mixture of CDDs and CDFs are identified and quantified. Then, the
toxicity of the mixture is estimated by adding the toxicity of each
of its components.
In the case of most environmental mixtures, however, this method
cannot be directly applied, since congener-specific analyses for
the 75 CDDs and 135 CDFs potentially present in the mixture are
seldom available. In addition, there is generally little information
available on the in vivo toxic potency of most of these congeners.
3. An Interim Approach 2,3,7,8-TCDD Toxicity Bquivlance
Factors (TEFs)
The CDWG recognizes a third alternative for estimating the risks
associated with exposure to complex mixtures of CDDs/CDFs. First,
as in approach #2, information is obtained on the concentrations of
homologues and/or congeners present in the mixture. Then, reasoning
on the basis of structure-activity relations and results of short
term tests, the toxicity of each of the components is estimated and
expressed as an "equivalent amount of 2,3,7,8-TCDD". Combined with
-------
-6-
esttmates of exposure and known toxlcity information on 2,3,7,8-TCDD,
the risks associated with the mixture of CDDs/CDFs can be assessed.
Key to the approach is the 2,3,7,8-TCDD Toxicity Equivalence Factors
(TEFs) are are derived below.
The general approach of TEFs outlined here is not unique:
several organizations have used similar approaches; cf. Table III.
The CDWG recommends that the TEF procedure be adopted as a
matter of science policy on an interim basis. The approach will
enable the Agency to deal with many, but not all, of its problems;
e.g., which Superfund sites should be given administrative priority,
to what extent a hazardous waste site should be cleaned up, which
manufacturing wastes can be delisted as EPA hazardous wastes, and how
to esimate the risks associated with the emission of CDDs/CDFs
from combustion sources.
The remainder of this document discusses the TEF approach in greater
detail, illustrates its use in risk assessment, and identifies additional
research, the results of which would strengthen the basis of this interim
approach.
-------
-7-
IV. The 2,3,7,8-TCDD Toxicity Equivalence Factors (TEFs) Approach
to Assessing the Toxicity of Complex Mixtures of CDDs/CDFs
2,3,7,8-TCDD is one of 75 CDDs. Exceptionally low doses of
this compound elicit a wide range of toxic responses in many animals;
e.g., adverse reproductive effects, thymic atrophy, and a "wasting
syndrome" leading to death. EPA's Cancer Assessment Group (GAG) has
determined that there is sufficient evidence to treat qxialitatively
2,3,7,8-TCDD as a potential human carcinogen and to estimate quantitatively
that it is the most potent animal carcinogen evaluated by the Agency
to date. Limited data suggest that some of the 74 other CDDs may
have similar toxic effects, again at very doses.
Moreover, these toxicity concerns are not restricted to CDDs.
Limited experimental data, supplemented by strong structure/activity
relationships in j.n vitro tests that are correlated wi.th in vivo
toxic effects of these compounds, indicate that some CDFs exhibit
"2,3,7,8-TCDD-like" toxicity (Bandiera,' 1984; Safe, 1984).
Thft cellular biochemical mechanisms leading to the toxic response
resulting from exposure to CDDs and CDFs are not known in complete detail.
However, over the last few years experimental data have accumulated which
suggest that an important role is played by an intracellular protein,
the Ah receptor. This receptor binds polycyclic aromatic molecules,
including CDDs and CDFs. In animals, the binding of 2,3,7,8-TCDD
related compounds to this receptor has been correlated with the
expression of several systemic toxic effects including sensitivity
to acute toxic effects (LD5Q values), thymic involution, chloracnegenic
-------
-8-
response, and the induction of several enzyme systems, some of which
have been linked to carcinogenic pathways (Poland.and Knutson, 1982;
Bandiera et al., 1984).
Table I contains information on a variety of endpoints: acute
toxicity, carcinogenic!ty, reproductive effects, receptor binding,
enzyme induction, and in vitro cell transformations. The data are
normalized to unity for 2,3,7,8-TCDD. For example, 2,3,7,8-substituted
HxCDDs have about 5% the Ah receptor binding strength of 2,3,7,8-TCDD,
they are about 70% as potent in the ability to induce the enzyme
AHH; and their carcinogenic potency is about 4% that of 2,3,7,8-TCDD.
Note that for these effects the LOELs or NOELs for 2378-congeners
are a thousand-fold lower than those for the non-2378 congeners.
Kociba (1984) has recently presented similar data.
The structure/activity generalizations based on the data in
Table I bear ait the generalizations in the literature concerning
the congeners that are most likely to be of toxic concern (Poland,
1982; Gasiewicz, 1982; Bandiera et al., 1984). That is, congeners
which are substituted in the lateral 2, 3, 7, and 8 positions are
likely to exhibit toxic effects at lower doses than other congeners.
This includes the fifteen tetra-, penta-, hexa- and heptachlorinated
CDDs and CDFs listed in Table II.*
* The CDW3 is aware that certain investigators (Grant, 1977;
Olie, 1982; Commoner, 1983; and Ontario, 1983) have broadly defined
the congeners of conern to include all those tri- to hepta- congeners
-------
-9-
which are substituted with at least three chlorines in the four
lateral (2,3,7 and 8) positions. The CDWG has reviewed the toxicity
data and does not find it to argue strongly for this extended range
of concern. Further, the increased level of complexity invoked by
including these additional congeners is to suggest a greater level
of accuracy and resolution than the CDWG believes is warranted.
The CDWG is also aware that receptor binding data suggest a
relatively high toxicity for 1,2,4,6,7-PeCDF. Examination
of stereochemial models point out that the 4/6 positions on CDFs are
arguably "more lateral" than the 2/8 positions (Bandiera et al, 1984),
However, this increased receptor binding acitivity is not reflected
in an increased ability of 1,2,4,6,7-PeCDF to act as an enzyme
inducer (cf. Table I), an endpoint which has been shown to correlate
with subchronic toxicity (Safe, 1984). Therefore, the CDWG is not
treating 1,2,4,6,7-PeCDF as a "2378-congener" at this time; however,
additional data could lead to a change in this position.
-------
-10-
The associated "2,3,7,8-TCDD equivalent factors" were assigned
as follows. The relative carcinogenicity responses (Table I) for
2,3,7,8-TCDD and the mixture of two 2378-HxCDDs* provide the TEF
for 2378-HxCDD. The relative toxicity of 2378-PeCDD was taken to be
the root mean square of the 2378-TCDD and 2378-HxCDD values. The
remaining assignments in Table II and Table III (righthand column)
are based on a rough assessment of the data in Table I, subject to
constraints:
1. The CDFs are likely to be less toxic than their corresponding
CDDs, based on compartive toxicity data of 2,3,7,8-TCDD and
2,3,7,8-TCDF in various species (Moore, et al, 1979).
2. As a matter of judgment, the CDWG believes that the uncertainties
in the procedure limit discrimination of relative toxicity to
order of magnitude estimates.
In the sane vein, TEFs for the non-2378 isomers are assigned values
which are 1% of the TEFs of the 2378-isomers in the same homologous
group.
While it could be argued that the hepta- congeners are of minimal
concern, the CDWG recognizes that in some mixtures the hepta congeners
are predominate; therefore, we have chosen not to exclude them entirely.
The general approach of estimating relative toxicities discussed
here has been taken by other groups in reaching decisions regarding risk.
Table III lists the TEFs used by these other workers.
* Refer to Appendix for nomenclature
-------
-11-
The TEFs assigned (except for the HxCDDs) do not rest on
the results of long term animal studies. Generally, they are based
on estimation of relative toxicity in in vitro tests whose relationship
to the chronic effects of concern is largely presumptive. However,
experimental results continue to supplement the view that the short
term assays are providing important fundamental information on the
toxicity of the CDDs/CDFs. For example, for the higher chlorinated
CDFs, inducation of certain enzymes correlates rather well with
thymic atrophy and body weight reduction noted in subchronic rat
studies (Bandiera et al., 1984; Safe, 1984).
It should also be noted that the structure/activity relationships
are not universal and cannot be said to have established, at this
point, a causal relationship between the Ah receptor and all forms
of toxicity of CDDs and CDFs. For instance, it has recently been
noted that die development of porphyria in mice does not correlate
with Ah phenotype and that genes other than Ah influence the development
of 2,3,7,8-TCDD-induced hepatotoxicity in mice (Greig, 1984).
In summary, in the view of the CDWG, there is a sufficient scientific
support for the TEF approach to estimating risks associated with
CDDs/CDFs that the Agency should adopt the approach, on an interim
basis, as a matter of science policy.
-------
-12-
V. Applications to Risk Assessment
In general, an assessment of the risk to human health of a
mixture of CDDs and CDFs involves the following steps:
1. Analytical determination of CDDs. and CDFs in the sample.
2. Multiplication of congener concentrations by the TEFs in Table III.
3. Stagnation of the products in step 2 to obtain the "2,3,7,8-TCDD
equivalence" of the sample.
4. Determination of human exposure to the mixture in question,
expressed in terms of equivalents of 2,3,7,8-TCDD.
5. Combination of exposure from step 4 with toxicity information
on 2,3,7,8-TCDD (usually carcinogenicity and/or reproductive
effects) to estimate risks associated with the mixture.
In cases in riiich the concentrations of the fifteen congeners
of concern are known:
2,3,7,8-TCDD Equivalents - 3) (TEF of each 2378-CDD/CDF congener
x the concentration of the respective congener) +
"2 (TEF of each non-2378 CDD/CDF congener
x the concentration of the respective congener)
Examples of this calculation for several environmental mixtures
are provided in Table IV.
In cases where only the concentration of homologous groups is
known; i.e., no isomer-specific data available, different approaches
are possible. For example, the assumption that the 2378-congeners
of concern constitute all of the CDDs and CDFs present in the mixture is
likely to provide an upper bound estimate of the toxicity. Alternatively,
one could assume that the occurrence of each of the congeners in the
mixture has equal probability (01 le, 1982; Qommoner, 1982). For
-------
-13-
instance, 2,3,7,8-TCDD is one of 22 possible TCDDs and would
constitute about 4 % of a mixture of equally probable isomers.
In other situations, particular knowledge of chemical reaction parameters,
process conditions, and results fron related studies, (e.g., congener
distributions in emissions from combustion sources) might enable one to
estimate the relative occurrence of 2378-congeners. However, one
must be careful to explicitly explain and justify whatever assumptions
are made.
The calculated "2,3,7,8-TCDD equivalents" can then be used to
assess the health risk of a mixture. As an explicit example, consider
a municipal solid waste (MSW) combustor whose particulate emissions,
the CDD/CDF mixture in question, were exactly like the electrostatic
precipitator (ESP) catch cited in columns 5 and 6 of Table IV. The
sample is estimated to contain 70 ppb 2378-TCDD equivalents; i.e.,
70 picograms of 2378-TCDD equivalents per milligran of mixture.
Suppose that an exposure analysis indicates that a person living
downwind frcm the incinerator receives an average daily dose of 1 ng
of the mixture/kg body weight. This exposure estimate is combined
with the carcinogenic potency of 2,3,7,8-TCDD (1.6 x H)5 per rog/kg-day
(U.S. EPA 1984)) to generate the upper bound of the excess
risk of developing cancer for a person living downwind frcm the
facility emitting the mixture under consideration, assuming lifetime
exposure:
Upper limit of
excess cancer risk = [potency] x [exposure]
- [1.6 x 105 per mg TCDD/kg-day] x [69.6 pg TCDD/
mg mixture x 10*9 mg TCDD/pg TCDD x 10-3 ng
mixture/kg-day x 1(H> mg mixture/ng mixture
- 10-8
-------
-14-
Use of the different assumptions regarding relative toxicities
(see Table III) influence the calculation of 2,3,7,8-TCDD equivalents
only slightly. For example, using analytical data from an Agency
study on emissions from a particular municipal waste combustor (EPA
1984), the 2,3,7,8-TCDD equivalents calculated using the assumptions
listed in Table III are generally within an order of magnitude.
VI. Comparisom with Other Approaches to Determining 2.3.7,8-TCDD
Equivalents
When the TEF approach is compared with several in vitro and in
vivo bioassays for directly determining 2,3,7,8-TCDD equivalents in
complex mixtures of CDDs/CDFs (the prefered approached discussed in
Section II above), the results are found to be in general agreement.
Eadon (1982) noted that CDD/CDF-contaminated soot from a PCB transformer
fire contained 58 ug 2,3,7,8-TCDD equivalents/gm soot (estimated by
an in vivo assay of the acute toxicity of a soot extract). This is
to be compared to an estimate of its toxicity using the TEFs shown
in Table III of 37 ug 2,3,7,8-TCDD equivalents/g.
In another study, Sawyer et al. (1983) and Hutzinger et al.
(1981) developed three estimates for the 2,3,7,8-TCDD equivalents in
a MSW flyash sample: 3.9, 4.6, and 32 ppb, using two different enzyme
induction assays, and the Ah receptor binding assay, respectively.
The use of TEF approach gives a 2,3,7,8-TCDD equivalence of 1.6 ppb.
Recently, Safe has noted a stronger correlation of toxicity with the
enzyme inducition assays than with the receptor assay. Therefore,
the TEF result is within a factor of 3 best short term assay estimates.
-------
-15-
VII. Research Needs
As noted above, the CDWG recommends that research be conducted
to develop bioassays that will directly assess the toxicity of complex
mixtures of CDDs/CDFs. In addition, research should be conducted
which will provide a firmer basis for the TEF approach and guide
appropriate modifications thereof. This research should be aimed at
1. Validating and completing the entries in Table I.
2. Investigating additional short term assays which can test the
mechanistic hypothesis which underlies the TEF approach.
3. Investigating correlations between the short term assays and
results of longer term assays.
-------
-/fc-
VI. REFERENCES.
Bandiera, S. et al., 1983. Competetive binding of the cytosolic
tetrachlorodibenzo-p-dioxin receptor. Biochem. Pharmacol.
32: 3803-3813.
Bandiera, S. et al., 1984. Polychlorinated dibenzofurans (PCDFs):
Effects of structure on binding to the 2,3,7,8-TCDD cytosolic
receptor protein, AHH induction and toxicity. Toxicol. 32:131-144.
Bradlaw, J. et al., 1979. Induction of enzyme activity in cell
culture: A rapid screeen for detection of planar polychlorinated
organic compounds. J. Assoc. Off. Anal. Chem. 62: 904-916.
Bradlaw, J. et al., 1980. Comparative induction of aryl hydro-
carbon hydroxylase activity in vitro by analogues of dibenzo-ฃ-
dioxin. Cosmet. Toxicol. 18: 627-635.
Commoner, B. et al., 1984. Environmental and economic analysis
of alternative municipal solid waste disposal technologies. I. An
assessment of the risks due to emissions of chlorinated dioxins
and dibenzofurans from proposed New York City incinerators. Center
for the Biology of Natural Systems, Queens College, CUNY, Flushing,
New York. May 1 .
Cooper Engineers, 1984. Air emissions and performance testing of
a dry scrubber (quench reactor) dry venturi and fabric filter
system operating on flue gas from combusion of municipal solid waste
at Tsushima, Japan.
Czuwa, J.M., and R. Hites, 1984. Environmental fate of combustion-
generated polychlorinated dioxins and furans. Environ. Sci.
Technol. 16: 444-450.
Des Rosiers, P., 1984. PCBs, PCDFs, and PCDDs resulting from
transformer/capacitor fires: An overview. Proc. 1983 PCS
seminar. Research project 2028, Electric Power Research Institute.
Palo Alto, California.
Eadon.G. et al., 1982. Comparisons of chemical and biological
data on soot samples from the Binghamton State Office Building.
New York State Department of Health (Unpublished report).
EPA, 1984. Assessment of emissions of specific compounds from
a resource recovery municipal refuse incinerator. Office of
Toxic Substances. EPA 560/5-84-002. June.
Gierthy, J. F. and D. Crane. 1984. Reversible inhibition of in
vitro epithelial cell proliferation by 2,3,7,8-tetrachlorodibenzo-
ฃ-dioxin. Toxicol. Appl. Pharmacol. 74: 91-98.
Gierthy, J.F., and D. Crane. 1985. In vitro bioassay for dioxin-
like compounds. A.C.S. meeting, Miami, FL. April (preprint
extended abstract).
-------
-17-
Grant, D.L., 1977. Proc. 12th annual workshop on pesticide residues
analysis. Winnipeg, Canada, p. 251.
Gravitz, N. et al., 1983. Interim guidelines for acceptable
exposure levels in office settings contaminated with PCB and
PCB combustion products. Epidemiological Studies Section, California
Department of Health Services. Sacramento. November 1.
Greig, J.B. et al., 1984. Incomplete correlation of 2,3,7,8- tetra-
chlorodibenzo-ฃ-dioxin hepatotoxicity with Ah phenotype in nice.
Tox. Appl. Pharmacol. 74:17-25.
Harding, D.H. 1982. Chlorinated dioxins and chlorinated dibenzo-
furans ambient air guideline. Health Studies Service, Ministry
of Labor. December.
Hassoun, E., et al. 1984. Teratogenicity of 2,3,7,8-tetrachloro-
dibenzofuran in the mouse. J. Tox. Envl. Health 14:337-351.
Hutzinger, 1981. Polychlorinated dibenzo-p-dioxins and dibenzo-
furans: A bioanalytical approach. Chemosphere 10: 19-25/
Knutson, J., and A. Poland, 1980. Keratinization of mouse
teratoma cell line XB produced by 2,3,7,8-tetrachlorodibenzo-ฃ-
dioxin: An in vitro model of toxicity. Cell 22:27-36.
Kociba, R.J. and 0. Cabey. 1984. Comparative toxicity and biologic
activity of chlorinated dibenzo-p-dioxins and furans relative to
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Presented at the
Fourth International Symposium on Chlorinated Dioxins and
Related Compounds, Ottawa, Canada. October.
Lamparski, L.L. et al. 1984. Presence of chlorodibenzodioxins
in a sealed 1933 sample of dried municipal sewage sludge,
Chemosphere 13: 361-365.
McKinney, J., and E. McConnell, 1982. Structural specificity and
the dioxin receptor. Perg. Ser. Env. Sci. 5:367-381.
Murray, F.J. et al., 1979. Three-generation reproduction study of
of rats given 2,3,7,8-tetrachlorodibenzo-ฃ-dioxin in the diet.
Toxicol. Appl. Pharmacol. 50:241-252.
National Cancer Institute (NCI). 1980a. Bioassay of 2,3,7,8-Tetra-
chlorodibenzo-ฃ-dioxin for possible carcinogenicity (gavage study).
Carcinogenesis Testing Program. DHHS pub. no. (NIH) 80-1765.
National Cancer Institute (NCI). 1980b. Bioassay of a mixture
of 1,2,3,6,7,8- and 1 ,2 ,3,7 ,8,9-hexachlorodibenzo-j>-dioxlns
for possible carcinogenicity (gavage study). DHHS pub. no.
(NIH) 80-1754.
Neal, R.A., et al., 1982. The toxicokinetics of 2,3,7,8-tetra-
chlorodibenzo-p-dioxin in mammalian systems. Drug Metab. .Rev.
13: 355-385.
-------
Olle, K., et al., 1983. Formation and fate of PCDD and PCDF
froo combustion processes. Chemosphere 12: 627-636.
Poland, A., et al., 1976. 3,4,3*,4'-tetrachloro azoxybenzene and
azobenzene: Potent indueera of aryl hydrocarbon hydroxylase.
Science 194: 627-630.
Poland,A. and J. C. Knutson, 1982. 2,3,7,8-Tetrachlorodebenzo-
2-dioxin and related halogenated aromatic hydrocarbons: An
examination of the mechansim of toxicity. Ann. Rev. Pharmacol.
Toxlcol. 22: 517-554.
Rappe, C., 1984. Analysis of polychlorinated dioxins and furans.
Env. Science and Techno1. 18: 78A-90A.
Rizzardini, M., et al., 1983. Toxicological evaluation of
urban waste incinerator emissions. Chemosphere 12: 559-564.
Safe, S., et al., 1984. Polychlorinated dibenzofurans:
chemistry, biochemistry, and toxicology. Presented at the
Fourth International Symposium on Chlorinated Dioxins and
Related Compounds. Ottawa, Canada. October 1984. (Abstract).
Sawyer,!., et al., 1983. Bioanalysis of polychlorinated dibenzo-
furan and dibenzo-p-dlioxin mixtures in fly ash. Chemosphere
12: 529-534.
Schwetz, B.A. et al., 1973. Toxicity of chlorinated dibenzo-p_-
dioxins. Env. Health Persp. 5:87-89.
Swiss Government (Bundesamt fur Umweltschutz, Bern), 1982. [Envi-
ronmental pollution due to dioxins and furans from chemical rubbish
incineration plants]. Schriftenreige Umweltschutz, No. 5.
Tong, H.Y. et al., 1984. Identification of organic compounds
obtained from incineration of municipal waste by HPLC and GC/MS
J. Chromatographv 285:423-441.
USDHHS. 1983. Levels of concern for hexa- (HCDD), hepta- (HpCDD)
and octachlorodibenzo-p-dioxins (OCDD) in chicken and eggs.
Memorandum from N. Bolger and 6.N. Biddle to J. Taylor. April 29.
USEPA, 1981. Interim evaluation of health risks associated with
emissions of tetrachlorinated dioxins from municipal waste resource
recovery facilities. Office* of the Deputy Administrator. November 19,
USEPA, 1984a. Health Assessment Document for Polychlorinated
Dibenzo-2-dioxins. External Review Draft. May. EPA 600/8-84-014A.
USEPA, 1984b. Thermal degradation products from dielectric fluids.
December. EPA 560/5-84-009.
-------
-14-
v.d. Berg, M., et al., 1983. Uptake arid selective retention In rats
of orally administered chlorinated dioxins and dibenzofurans
from fly ash and fly ash extract. Chemosphere 12:537-544.
Weber, H., et al., 1984. Teratogenicity of 2,3,7,8-tetrachloro-
dibenzofuran (TCDF) in mice. Toxicol. Letters 20: 183-188.
-------
POTENCIES Of PIPKINS RELATIVE TO 2,3,7,8-TCDD
CHEMICAL
CDOs:
Mono thru tri
2378-TCDD
TGDDs
2378-PeCDD
PeCDDs
2378-HxCDDs
HxCDDs
2378-HpCDDs
HpCDDs
OQDD
CDFs;
Mono thru tri
2378-TCDF
TCDFS
2378-PeCDF
12467P6CDF
PeCDFs
2378-HxCDFs
HxCDEs
2378HpCDFs
HpCDFs
GUINEA PIG
<10-*(e)
1 (a)
<.001(a)
.67 (a)
.002(a)
.03 (a)
.004(a)
.002(a)
.28;.5(a)
.017(a)
CARCINO- REPROD. TER. RECEPTOR ENZYME INDUCTION CELL FIAT (XB)
GENICm EFFECTS BINDING AHH EROD KERATIN. CELL ASSAY
(ref) (ref) (ref) (ref) (ref) (ref) (ref ) (ref)
.OOl-.Ol(e) <.001(f) .Ol(e)
Kb) l(c) l(i) l(e) l(e) Kg) l(e) l(j)
<.001(k) <.01-.16(e) <.001-.02(g) <.001-.01(e)
Ke) .02-.2(g) .5(e)
<.001(g)
,04(b) .01(c) .05(e) .001-. Kg) ,005(e)
<.001(g)
.002-.004(g,f)
<.001(f)
<.00001(k) <.001(f)
ฃ.001-.02(d,h)<.001(d) <.001(d) <.001(e)
.03-.13(i,k) .3/.25(e/h) .01-.04(frh) ,l(h) .05(e) ,l(j)
.001-.05(d,e) COOKd) _<.005(d)
.13/.7/.6(d/e/h) <.3(d) .l(d)
.15(h) ,002(h) <.001(h)
.001-.l(d,e) ฃ.001(d) <.001(h)
.04-.5(e,h) .05-.2(h) ,l-.5(h)
.001(e,h) .002(h) ,006(h)
,004(g)
<.001(h) <.001(f) _
a. McKinney and McConnell, 1982; b. USEPA 1984a; c. Murray et al., 1979; Schwetz et al., 1973;
Wfeber et al., 1984.; d. Bandiera et al., 1983; e. Knutson and Poland, 1980; f. Bradlaw et al., 1979;
g. Bradlaw et al., 1980; h. Bandiera et al., 1984; i. Hassoun et al., 1984; j. Gierthy and Crane, 1985;
k. Weber et al., 1984;
-------
TABLE II
CDD/CDF ISOMERS OF MOST TOXIC CONCERN3/
DIOXIN
Isomer
TEFb/
DIBENZOFURAN
Isomer
TEF
2,3,7,8-TCDD 1
1,2,3,7,8-PeCDD 0.2
1 ,2,3,6,7,8-HxCDD 0.04
1 ,2,3,7,8,9-HxCDD 0.04
1 ,2,3,4,7,8-HxCDD 0.04
1 ,2,3,4,6,7,8-HpCDD 0.001
2,3,7,8-TCDF 0.1
1 ,2,3,7,8-PeCDF 0.1
2,3,4,7,8-PeCDF 0.1
1,2,3,6,7,8-HxCDF 0.01
1 ,2,3,7,8,9-HxCDF 0.01
1,2,3,4,7,8-HxCDF 0.01
2,3,4,6,7,8-HxCDF 0.01
1,2,3,4,6,7,8-HpCDF 0.001
1 ,2,3,4,7,8,9-HpCDF 0.001
a/ In each homologous group the relative toxicity factor for the
isomers not listed above is 1/100 of the value listed above.
b/ TEF ป toxic equivalency factor = relative toxicity assigned.
-------
TABLE III
SOME APPROACHES TO ESTIMATING RELATIVE TOXICITIES OF PCDDS and
PCDFS
BASIS/ SWISS3 GRANT5 NEW YORK ONTARIO^
COMPOUND/ OLIEC STATEฎ
COMMONER!
(Basis) enzyme
Mono and di
237-TrCDD
other TrCDDs
2378-TCDD
other TCDDs
2378-PeCDDS
otherPeCDDs
2378-HxCDDs
other HxCDDs
2378-HpCDDs
other HpCDDs
OCDD
2378-TCDFs
other TCDFs
2378-PeCDFs
other PeCDFs
2378-HxCDFs
other HxCDFs
2378HpCDFs
other HpCDFs
OCDF
0
0
0
1
0.
0.
0.
0.
0.
0.
0.
0
0.
0.
0.
0.
0.
0.
0.
0
0
*: congeners with
01
1
1
1
1
01
01
1
1
1
1
1
1
1
0
0.1*
0
1*
0
0.1*
0
0.1*
0
0.1*
0
0
0.1*
0
0.1*
0
0.1*
0
0.1*
0
0
chlorine
**: these isomers were
a = Swiss government,
b ป Grant, 1977
c - 01 ie et al., 1982;
d - Commoner, 1984
e = Eadon, 1982
not
1982
1983
LD50
0
0
0
1
0
1
0
0.03
0
0
0
0
0.33
0
0.33
0
0.01
0
0
0
0
atoms at
considered
f =
g
h =
i -
FDA9
CAh EPA1 EPA
1982 Current
Recommend .
various various
effects effects
0
0
0
0
0
0
0
0
0
0
0
0
3 of
.
0
1*
0
1*
.01
1*
.01
1*
.01
1*
.01
0 <0.
.02*
.0002
.02*
.0002
.02*
.0002
.02*
.0002
0
0
0
0
1
0
_**
_**
0.02
0.02
0.005
0.005
00001
0
0
0
0
0
0
0
0
0
the 4 lateral
0
0
0
1
0
1
0
1
0
1
0
0
1
0
1
0
1
0
1
0
0
Care in.
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
.01
.2
.002
.04
.0004
.001
.00001
0
.1
.001
.1
.001
.01
.0001
.001
.00001
0
positions.
Ontario, 1982
USDHHS, 1983
California, 1982
USEPA, 1981
-------
ISOMER
TCDDS
PeCDDs
HxCDDs
HpCDDS
OCDD
TCDFs
PeCDFs
HxCDES
qpCDFS
OCDF
Total TCDD
AIR PARTICS.
St. Louis 5
TEF Gone. TCDD
Eqts.
(ppb)
1 0.2 0.2
0.2 1 0.2
0.04 1.2 0.048
0.001 25 0.05
0 170
0.1
0.1
A -
0.001
0
Eqts. 0.5
1. USEPA 1984b.
2. Cooper Engineers, 1984.
3. Rappe, 1984.
TABLE IV
PCDDS/PCDFS IN SOME ENVIRONMENTAL SAMPLES
MSW LAKE 1982 MSW FLYASH 6
ESP DUST 5 SEDIMENT 5 MILORGANITE 4 ONTARIO OSLO
Cone. TCDD Cone. TCDD Cone. TCDD Cone. TCDD Cone. TCDD
Eqts. Eqts. Eqts. Eqts. Eqts.
(ppb) (ppb) (ppt) (ppt) (ppt)
550 0 206 206 541 541 ND -
10 2 0.1 0.02 - - 467 93 11 2.2
160 6.4 0.34 0.014 2768 110.7 591 24 51 2
120 0.24 0.5 0.001 7600 15.2 434 0.9 119 0.2
260 - 1.3 - 60000 - 467 - 186 -
40 4 0.13 0.013 -
80 8 0.14 0.014 -
280 28 0.38 0.038 -
160 0.16 1.13 0.001 -
40 - 0.14 -
70 0.10 332 &59 5
4. Lanparski et al., 1984 7. des Rosiers, 1984.
5. Czuwa and Hites, 1984.
6. Tong et al., 1984
-------
TABLE IV (Continued)
THERMAL DEGRADATION PRODS.
FROM DIELECTRIC FLUIDS *
run
8-13-40
ISOMER
TEF Cone.
(ng)
TCDDs
2378
other
PeCDDs
2378
other
HxCDDs
2378
other
HpCDDs
2378
other
OCDD
TCDFS
2378
other
PeCDFs
2378
other
HxCDFs
^^ 2378
other
HpCDFs
2378
other
OCDF
1
0.01
0.2
0.002
0.04
0.0004
0.01
0.0001
0
0.1
0.001
0.1
0.001
oS
o/boi
0.001
0.00001
0
0
0
0
0
0
690
43
7
0
0
Total TCDD Eqts.
TCDD
Eqts.
0
0
0
0
0
69
4.3
0.7
0
0
75
JAP. MSW^
Run Pt. A TEF Pt. B
8-30-61 ASKL
Cone.
(uq)
0
0
0
330
37
1400
6400
910
29
3.4
TCDD Cone. TCDD Cone.
Eqts. Eqts.
(lb/MMBTU(xlO:ฑ))
0
0
0
0.66
0
140
640
91
0.029
0
872
0.1 0
0.07 0
0.04 0
0.02 <
0.01
1.31 0
0.38 0
0.06 0
0.01 <
0.004
0
.1
.014
.002
.001
0
.131
.038
.006
.001
0
.3
0.58
0.47
0.36
0.08
0.04
1.25
0.46
0.06
0.02
0.01
TEF
TCDD
Eqts.
0.58
0.094
0.014
<.001
0
0.125
0.046
0.006
<.001
0
0.87
COMMERCIAL CPs SOOT FROM PCB
PCS FIRE!
246TCPJ PCP3
Cone. TCDD Cone. TCDD Gone
. TCDD
Eqts. Eqts. Eqts.
(ppn) (ppm) (ppn)
< .1 - < .1 -
1.6
1.6
<.l - <! -
2.5
2.5
<1 - 2.5 0.1
1.1
3.6
<1 - 175 0.35
7
7
<1 - 500 -
1.5 0.15 <.l -
16
12
17.5 1.75 <.l -
358
312
36 3.6 <.3 -
670
295
4.8 0.005 19 0.019
288
172
<1 - 25 - 40
5.5 0.5
1.6
.016
.5
.005
.044
.001
.007
0
1.6
.012
35.8
.312
67
.295
.288
.017
107
-------
APPENDIX
The following terminology and abbreviations are used in this
document:
1. The term "congener" refers to any one particular member of
the same chemical fanily; e.g., there are 75 congeners of
chlorinated dibenzo-ฃ-dioxins.
2. The term "homologue" refers to a group of structurally related
chemicals which have the sane degree of chlorination. For example,
there are eight homologues of CDDs, monochloroinated through
octochlor inated.
3. The term "isoroer" refers to substances which belong to
the same homologous class. For example, there are 22 isoners
that constitute the hcnologues of TCDDs.
4. A specific congener is denoted by unique chemical notation.
For example, 2,4,8,9-tetrachlorodibenzofuran is referred
to as 2,4,8,9-TCDF.
5. Notation for homologous classes is as follows:
Dibenzo-g-dioxin D
Dibenzofuran F
No. of Halogens Acronym Example
2 D 2,4-DCDD
3 Tr
4 T 1,4,7,8-TCDD
5 Pe
6 Hx
7 Hp
8 0
1 through 8 CDDs and CDFs
-------
-2-
6. Dibenzo-ฃ-dioxins and -dibenzofurans that are chlorinated
at the 2,3,7 and 8 positions are denoted as 2378 congeners;
e.g., 1,2,3,7,8-PeCDF and 2,3,4,7,8-PeCDF are both referred
to as "2378-PeCDFs".
-------
EXHIBIT I
-------
Study of
DIOXIN &other
TOXIC POLLUTANTS
Midland, Michigan
U.S. EPA Region 5
April 1985
-------
DIOXIN
A generic term for a group of
75 related compounds known
as polychlorinated dibenzo-
p-dioxins. The most toxic
compound of this group is
2,3,7,8-tetrachloro-dibenzo-p-
dioxin (2378-TCDD). A num-
ber of the dioxins, including
2378-TCDD, are formed as
inadvertent byproducts dur-
ing the manufacture of or-
ganic chemicals, particularly
chlorinated phenols. Certain
combustion sources such as
municipal and industrial
waste incinerators and acci-
dental electrical transformer
fires have been shown to be
sources of 2378-TCDD and
other dioxins. The levels of
2378-TCDD from these
sources have been relatively
low compared to levels from
chlorinated phenols produc-
tion. 2378-TCDD is very per-
sistent in the environment.
It does not readily degrade
into less toxic chemicals.
Tests on laboratory animals
indicate 2378-TCDD is one of
the most toxic chemicals
made by man. Based on its
estimated carcinogenic po-
tency, cancer risks of individ-
uals who consume fish con-
taminated with high levels of
2378-TCDD could be signifi-
cant. Exposure to Dioxin
contaminated soils could
also significantly increase
cancer risk.
As part of its National Dioxin
Strategy, the U.S. EPA has
begun a national dioxin study
to determine the extent of
dioxin contamination in the
environment. The results of
that study are expected by
the end of 1985.
'JSl frfe Saate of MWhfesfl's request, the U.S. Environ-
. ti#nfe& Pr(|teotl^;^erv5y,,ltias been conducting
^^refiet^ye ewwrtieh^rt stupes in the City of
Midland, inside the Dow Chemical Midland plant,
at companion sites across th$ Midwest. While
the primary focus of the studies was on possible
78-TCl30rtoontstmifiatten In the Midland
EPA also sampled for Other toxic chemicals.
F|isujts show that;
> Current dtoxin levels in surface soils away from
the Dow Chemical plant are welt tetow one part
p&r billion (ppb). According to the Centers for
Disease Control, one ppb is the level at which
some action must be considered to prevent
human exposure to the dfoxin-eontaminated soil
in residential areasii Current levefs of dioxln in the
Midland area, therefore, do not pose an unac-
ceptable public health risk.
* Inside the Dow Chemical plant, EPA found con-
centrations as high as 36 ppb in surface soils.
(P0W Ghe|rfc?al; rafted cohp^ntratJons as high
af ง1 p^yc^n60ifra^tl0ha insUe the plant are
generally in the lower range of levels found at
similar chemical manufacturing and disposal sites
in other parts of the country, EPA, the State of
Michigan, and Dow Chemical have entered into a
proposed consent order to provide interim cover-
ing of surface areas that have high concentra-
tions inside the plant. A publfe comment period
on the consent order began on March 20,1985
and will last until April 17, 1985.
Concentrations of dioxin found in the City of
Midland away from the Dow ptetnt are significantly
higher than the trace levels found In A few
samples taken in other cities. Most samples from
these comparison sites,and all samples from
natural areas,dld not contain dioxin at detectable
levels.
Air emissions from the Dow Chemical plant are
the likely source of dioxin contamination in the
Midland area.
* Current levels of toxic organic pollutants, other
than dioxins, in Midland soils are at levels found
elsewhere in the environment and do not pose an
unacceptable public health risk.
Because of high dioxin levels in fish taken from
the Tittabawassee River, a warning on eating fish
particularly bottom-feeding fish such as carp
and catfish should be continued.
-------
INTRODUCTION
This report presents a brief history of dioxin findings in the
Midland area, a description of EPA's soil study, the soil
study results, and conclusions from the Centers for
Disease Control (CDC) and EPA about health risks from
exposure to contaminated soils. In addition, data from
other locations, including cities that also have chemical
manufacturing plants and disposal sites, are compared to
data from Midland. Results and conclusions from fish
sampling in the Tittabawassee River are also presented.
The study in Midland has been incorporated into the
National Dioxin Strategy. The National Dioxin Strategy is
examining potential sources and risks of dioxin in the envi-
ronment. It is organized into seven tiers or categories
ranging from sites where dioxin was most likely to be found
(Tiers 1 and 2) to sites where contamination is not
suspected (Tier 7).
HISTORY OF DIOXIN FINDINGS
In June 1978, Dow Chemical advised the Michigan Depart-
ment of Public Health (MDPH) that it had found dioxin
(2378-TCDD) in fish caught from the Tittabawassee River.
The MDPH immediately issued an advisory against eating
fish from the river. That advisory is still in effect today.
Subsequently, the Michigan Department of Natural
Resources (MDNR), in cooperation with EPA, sampled fish
from the Tittabawassee River and confirmed Dow's
findings.
The agencies attempted to determine if Dow Chemical's
wastewater discharges, which flow into the Tittabawassee
River, contained the dioxin that was accumulating in the
fish.
However, at that time, regulatory agencies could not
detect dioxin at ultratrace levels in water because
analytical methods had not yet been developed that were
able to detect dioxin at very low levels.
In the fall of 1981, MDNR and EPA conducted a caged-fish
study in the Tittabawassee River to determine if Dow
Chemical was the source of the dioxin. Catfish were kept
in cages for 28 days both upstream and downstream of
Dow Chemical and directly in the plume of Dow's dis-
charge in the river. An experimental method of ultratrace
analysis for dioxin in water was also developed as part of
that study.
The preliminary results released in March 1983 demon-
strated that Dow Chemical's wastewater was the source
of dioxin. Subsequent analyses by Dow Chemical have
confirmed that finding.
In the spring of 1984, the State of Michigan, with EPA's
concurrence, issued a wastewater discharge permit and
an administrative order to Dow Chemical that limits the
discharge of 2378-TCDD. The order also requires the con-
struction of wastewater treatment systems as a first step
in reducing dioxin discharges.
In March 1983, the State of Michigan formally requested
EPA to conduct a comprehensive study of dioxins and
other toxic pollutants at Dow Chemical's Midland plant, in
the Tittabawassee River, and in the City of Midland. The
state and EPA planned the study in the spring of 1983. Soil
sampling was conducted in Midland, at Dow Chemical,
and at the comparison sites during 1983 and 1984. Dow
incinerator emissions, ambient air samples, Dow brine
system and wastewater samples, drinking water well
samples, and river sediment samples were also collected
in 1984.
SOIL STUDY
Study Objectives
The principal objective of the soil study was to determine
whether concentrations of dioxins and other toxic pollut-
ants in Midland surface soils could pose unacceptable
public health risks. The study was also conducted to deter-
mine the likely source or sources of ambient levels of diox-
ins in Midland; if levels found in Midland are comparable to
levels found in other industrial areas with combustion
sources and process operations different from those at
Dow Chemical; and whether levels found in Midland are
significantly different from background levels generally
found in the environment.
Conduct of the Study
The investigation of dioxin and other toxic pollutants in the
Midland area is one of the first and most comprehensive
studies conducted by EPA in which laboratory methods
capable of detecting 2378-TCDD at the low parts per
trillion (ppt) range were used.
(One ppt is 1,000 times less than one part per billion (ppb)
which is the level in residential soils at which CDC believes
some action to limit human exposure needs to be con-
sidered.)
Throughout this study, elaborate quality assurance controls
were used to ensure that accurate data were produced.
Field work was thoroughly documented. Duplicate soil
samples were obtained at certain locations. Clean, dis-
posable sampling equipment was used for each sample.
Each sample was coded with a unique identification
number. The samples were thoroughly blended in a labora-
tory before extraction and analysis.
All samples were then analyzed on a blind basis, that is,
EPA laboratories at Bay St. Louis, Mississippi, and
Research Triangle Park, North Carolina, did not know
where the samples had been taken or the identity of the
samples they were analyzing. Control and blank samples
were also submitted to the laboratories on a random, blind
basis. In addition, EPA provided split samples to Dow
Chemical on a blind basis for comparative analyses.
Dow's analyses confirmed EPA's results.
EPA believes that the data obtained accurately reflect sur-
face soil conditions in Midland, at Dow Chemical, and at
the comparison sites. This set of high quality data is suffi-
cient to evaluate health risks to Midland residents from
exposure to current levels of 2378-TCDD and other pollut-
ants in Midland soils.
-------
acres) of, native pr-aMe, to
Itasca Wilderness
pine-dominated fo
logged. Bluestem
ftath$aซjtiปetn
FIGURE 1
MAP OF EPA REGION 5
SHOWING LOCATIONS
OF SOIL STUDY SITES
.Th$ principal
cess vents and some combustion sources.
There .are no' 'tier 1 -or Tier ,8-, sffiss its'
cJeftfterfbyfiPA's, National Dfoxlr* Strateg
fn the Heftr^,.; yiteof^ area, Hemy>
rwpufatton Qf atooirt 2,?CXJ,
. .
, OWo, area.
-------
Soil Study Results
The study results for 2378-TCDD for each site are sum-
marized in Table 1. The Dow Chemical and City of Midland
data are highlighted in the centerfold of this summary
report.
Figure 2 compares the ranges of dioxin levels found in sur-
face soils at each of the comparison sites with levels found
in Midland. The data show that levels of dioxin in Midland
away from the Dow plant are well below the 1 ppb level of
concern established by CDC for residential soils. One sam-
ple obtained near the east fence line of the Dow plant had
a dioxin concentration of 2.03 ppb. Supplemental sampling
conducted by Dow Chemical around the site showed the
2.03 ppb concentration to be isolated to a small area. Dow
Chemical has extended its fencing around that site to pre-
vent public access.
The next highest sample obtained near the perimeter of
the Dow plant contained less than 0.55 ppb of dioxin. Fre-
quent and prolonged exposure of the public to soils around
the perimeter of the Dow plant is not likely.
Figure 2 also shows that, while levels of dioxin found in
Midland are well below CDC's level of concern for residen-
tial areas, those levels are higher than found around other
industrial sites and in natural areas.
Figures 3, 4 and 5 present the dioxin data for Midland
public use and residential areas, the perimeter of the Dow
plant, and inside the Dow plant. The data in Figure 3 show
that most values of 2378-TCDD in Midland public use and
residential areas are at or less than 0.10 ppb. All values
were less than 0.30 ppb.
Four out of the five highest values in the city were found at
residential downspouts, indicating that airborne emissions
from Dow Chemical are the likely sources of dioxin. Down-
spouts can be good indicators of atmospheric contamina-
XI
a
Q
Q
O
I-
I
00
N
CO
M
FIGURE 2
COMPARISON OF
MIDLAND SOIL DATA
TO OTHER SITES
Geometric Mean
Arithmetic Mean
CDC LEVEL OF CONCERN
FOR RESIDENTIAL SOILS
-------
2378-TCDD
SURFACE SOIL CONCENTRATIONS
MIDLAND, Ml
DOW *
CHEMICAL
COMPANY
-------
1+
0.5-1.0
0.1-0.6
0.05-0.10
0.001-0.05
OPEN AREA SAMPLE
DOWNSPOUT SAMPLE
CORNING
-------
s
3
7
3
2
1
0
nil
FIGURE 3
PUBLIC USE AND
RESIDENTIAL AREAS
MIDLAND, Ml
2378-TCDD (ppb) in
SURFACE SOILS
.05 .10 .15
.20 .25
Concentration (Parts Per Billion)
?$JSS$r ;"ฃ\
in
o
I
FIGURE 4
PERIMETER DOW CHEMICAL
MIDLAND PLANT
2378-TCDD (ppb) in
SURFACE SOILS
O.5 1.2 1.5
Concentration (Part* Per Billion)
5r
&
0
FIGURE 5
DOW CHEMICAL
MIDLAND PLANT
2378-TCDD (ppb) in
SURFACE SOILS
1.0 3.0 3.5 4.0
Concentration (Part* Per Billion)
4.5 35.5
36.0
-------
tion because airborne particles tend to collect on exposed
surfaces such as building roofs. These particles are
washed from the roofs by rain and deposited in soils
around the downspouts.
Figure 4 shows that although one value of 2.03 ppb was
found on the perimeter of the Dow plant, the next highest
value is less than 0.50 ppb. Data presented in Figure 5 indi-
cate that most values found inside the Dow Chemical plant
have less than 0.50 ppb of 2378-TCDD. A few sampling
locations inside the Dow plant had concentrations in the 3
to 5 ppb range and one location had a dioxin concentration
of 36 ppb. Supplemental sampling by Dow Chemical indi-
cates that most of the Dow plant had dioxin concentrations
less than 1 ppb, although one sample had a value of 51 ppb.
CDC has stated that, based on experience at similar
facilities, it is doubtful whether significant dioxin exposure
to workers would occur at the plant unless the soil was
disturbed because of construction or similar activities.
EPA, the State of Michigan, and Dow Chemical have
entered into a proposed consent order by which Dow
Chemical will provide interim covering of the two identified
high concentration areas with asphalt to minimize expo-
sure to plant workers and prevent migration of dioxin
offsite.
Consideration of Health Risks
The data from this study were reviewed by CDC to deter-
mine whether levels of dioxin and other pollutants in
Midland soils might pose unacceptable health risks to
Midland residents.
CDC concluded that levels of 2378-TCDD in Midland soils
are well below the 1 ppb level at which some action to limit
human exposure to residential soils needs to be consid-
ered; and, that levels found in residential soils, including
downspout areas, do not represent a significant health risk
to either the persons living there or the public at large.
The levels of other dioxins, furans, and other toxic
chemicals, which are all less toxic than 2378-TCDD, do not
represent an unacceptable public health risk. CDC also
concluded that levels of other toxic chemicals are within
the range of levels found generally in the environment.
CDC also states that its assessment is based upon current
environmental data, and no comment can be made about
past environmental levels and potential risks.
EPA's Chlorinated Dioxins Work Group at EPA Headquart-
ers in Washington, D.C., also evaluated the data and came
to the same conclusions as CDC.
Comparison with Other Dioxin Sites
Figure 6 compares data obtained at Dow Chemical and in
Midland with data from other chemical manufacturing
sites and disposal sites where dioxin has been found.
These are Tier 1 and Tier 2 sites for EPA's National Dioxin
Strategy. These data show that the levels of 2378-TCDD
found at Dow Chemical are, for the most part, in the range
of, or lower than, levels found at other Tier 1 and Tier 2
sites.
1000
120
g 100
Q 8O
O
00 <>
rป
w
C4 40
20
R
x
"**
M.
ซ*
*"
tf:
>
$-
'ปป
,
; -,.. . -r,/ ,^^f '^J^A^t^fi^Vfe'A^'. ;'-."., ^-^iS^1 l"i*.IM* ,s ปi*
FIGURE 6
MAXIMUM 2378-TCDD AT SELECTED
TIER 1 AND TIER 2 SITES
Onsite/Offsite Concentrations in
Surface Soils(in ppb)
,-S
/A
f-r
I
ui ui
^ ป
O) 0)
o ฐ
^O
/*
T O*
^
LEGEND
>1000= Greater than
71
/ NA
/ NO
r
"~^~Tl
/*A
-------
FISH STUDY
During the summer of 1983, MDNR collected fish from the
Tittabawassee River and several other Michigan rivers for
dioxin analyses.
For most rivers, samples of bottom-feeding fish such as
carp and catfish were collected along with samples of
game fish including walleye and bass. The fish were
analyzed by U.S. EPA's Environmental Research Labora-
tory in Duluth, Minnesota. The results for the Tittaba-
wassee River, which are summarized in Table 2, were
originally released to the public by MDNR in October 1984:
Whole fish composite samples of carp from 11 other
Michigan rivers had levels of 2378-TCDD ranging from
none detected to 8.6 ppt.
Figure 7 presents data obtained for the 25 carp fillets and
5 walleye fillets from the Tittabawassee River. The concen-
trations of 2378-TCDD in carp are much higher than con-
centrations in walleye. As shown in Table 2, the average
carp fillet concentration is about 50 ppt, while the average
for walleye is about 4 ppt.
The Food and Drug Administration recommends that fish
with more than 50 ppt of 2378-TCDD not be eaten and fish
in the range of 25 to 50 ppt be consumed only on a limited
basis. Based upon the data obtained from this study, both
CDC and EPA recommend that the MDPH warning on con-
sumption of fish from the Tittabawassee River be main-
tained.
As a result of a March 1984 consent decree with EPA, Dow
Chemical has agreed to analyze dioxin levels in fish from
the Tittabawassee River every 2 years beginning in 1985
and lasting until 1991. This program has been approved by
EPA and MDNR and will be monitored by the agencies.
The purpose of the program is to determine whether
wastewater treatment systems now being installed by Dow
Chemical to control dioxin are having a beneficial impact
on dioxin levels in Tittabawassee River fish.
FIGURE 7
1983 FISH COLLECTION
IN TITTABAWASSE RIVER
NEAR MIDLAND. Ml
2378-TCDD (ppt)
Carp
(skinless fillet)
Walleye
(skin on fillet)
10 20 30 4O 50 60 70 80 90
Concentration (Part* Per Trillion)
530
-------
TABLE 2
FISH STUDY RESULTS
2378-TCDD
(Results in Parts Per Trillion)
Tittabawassee River Fish
(August-September 1983)
Carp whole fish composite (5 fish)
Carp individual skinless fillets
Catfish skinless fillet composite (5 fish)
Smallmouth Bass skin-on fillet
composite (5 fish)
Walleye individual skin-on fillets
Number of
Measurements
1
25
1
1
5
Range
190 ppt
12-530 ppt
75 ppt
5.1 ppt
2.8-5.1 ppt
Mean
50 ppt
3.9 ppt
Note: A composite fish sample is made up of a number of individual fish blended together prior to analyses.
Bottom-feeding fish were sampled because they are more likely to pick up contamination In river sediment.
EPA'.,i U.S, Environmental Protection Agency -
GDC Centers for Disease Control
iylDNft Michigan Department of Natural
M0PH Michigan Department of Public Health
PPB Parts Per Stilton - -. ,'
PPT Parts Per Trillion '
2378-TCDD dioxin or 2,3,7,8 teirachlorodibehzo-p^lioxln
GLOSSARY
Ambient Air Air outside the property limits "of industrial or
other types of facilities.
Administrative Order An order Issued by an EP"A adminis-
trator in accordance with congressional authorization trjirt r&-
quires a party to correct environmental problems!,
Background Levels Concentrations of chemicals that are
found in the environment away from manmade pollution
sources.
Carcinogen
cancer.
A substance or agent that produces or incites
National Dioxin Strategy In December 1083, EPA
announced a National Dioxin Strategy to determine the extehl
of dioxin contamination throughout the country. The Strategy
provides a framework under which the Agency wflK
ซ Study the nature of dioxin contamination throughout the
United States and the risks to people and the environment
^Ili^rfi^ Biat thf eaten public health
co^wtfnaHon:
existing dioxin.
EPA established seven cate-
, Igatton and study. These sites
are most probably contaminated (Tiers 1
~"*-~-"tB no expectation of contamination
than 80_percent of the dioxin in the
ijtjrf^fter 1 and 2 sites. The other
I EPA s National Dioxin Study, which
"'"" .and is scheduled for com-
O/ie part per billion, the levtl at
considered to prevent human
, __ jntial; soils, is equivalent to one
:. A part per trillion is one thousand
ซausf,
periods
To-enswre ihe accuracy of laboratory
re often divided and sent to different labo-
ative analyses.
Any of a number of chemicals considered
regulatory attention because of the adverse
$ ean nave on man, animal life, or the
Marty are lethal in single high doses or
if administered at tow doses over longer
AVAILABLE INFORMATION
For more information about EPA's dioxin studies in the Midland area, please review the information repositories at
the following locations:
Grace A. Dow Memorial Public Library
1710 W. St. Andrews Drive
Midland, Ml 48640
(517)835-7151
Dr. Winifred Oyen, Director
Midland Health Department
125 W. Main Street
Midland, Ml 48640
(517) 832-6655
Ingersoll Township Hall
c/o Kurt Shaffner, Supervisor
4400 Brooks Road
Midland, Ml 48640
(517)835-5289
For additional information, please contact:
Vanessa Musgrave
U.S. EPA Community Relations Coordinator
230 South Dearborn Street
Chicago, IL 60604
(312) 886-4359
1-800-621-8431 (toll free, 8:30 a.m.-4:30 p.m., Central Time)
Gary A. Amendola
U.S. EPA Project Manager
Michigan Dioxin Studies
U.S. EPA Eastern District Office
25089 Center Ridge Road
Westlake, OH 44145
(216) 835-5200
-------
-------