A STUDY OF THE DISTRIBUTION AND FATE OF
POLYCHLORINATED BIPHENYLS
AND BENZENES AFTER SPILL OF
TRANSFORMER FLUID
\
(S
CONTRACT NO! 68-01-3232
DIVISION OF OIL AND SPECIAL MATERIALS CONTROL
OFFICE OF WATER PROGRAM OPERATIONS
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
U.S. ENVIRONMENTAL PROTECTION AGENCY
REGION IV- 1421 PEACHTREE ST., N.E.
ATLANTA,GEORGIA 30309
JANUARY 1976
-------
FOLLOW-UP STUDY OF THE
DISTRIBUTION AND FATE OF
POLYCHLORINATED BIPHENYLS
AND BENZENES IN SOIL AND
GROUND WATER SAMPLES AFTER AN .
ACCIDENTAL SPILL OF TRANSFORMER FLUID
CONTRACT NO.: 68-01-3232
DIVISION OF OIL AND SPECIAL MATERIALS CONTROL
OFFICE OF WATER PROGRAM OPERATIONS
U. S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D. C. .20460
and
THE ENVIRONMENTAL PROTECTION AGENCY
REGION IV - 1421 PEACHTREE STREET, NE
ATLANTA, GEORGIA 30309
PROJECT OFFICER,
GEORGE J. MOEIN
ENVIRONMENTAL PROTECTION AGENCY
ATLANTA, GEORGIA
JANUARY, 1976
-------
FOREWORD
The Environmental Protection Agency's interest-in the spills of
Polychlorinated Biphenyls (PCB) and related substances has been
demonstrated through a series of response oriented actions, and
joint EPA industry efforts to mitigate the damages caused by such
spills. In recent years some significant quantities of PCB have
spilled, mostly from transportation related media in remote and
populated areas of the United States. In the Southeastern United
States several such spills have occurred during the past three years
causing alarm to the public and large clean up expenses to the
industry.
This technically oriented study was designed to derive a PCB con-
centration profile in a spill area two years after the occurrence
of the spill. This study is somewhat unique, in that many months of
field work and laboratory analysis were spent to examine numerous
environmental factors and parameters to determine the fate of PCB
and Benzenes in the "natural environment". The findings and con-
clusions of this study should have significant value to EPA response
personnel who often have to determine and recommend a "safe level"
of clean up and removal operation; to the industry who frequently
pays for such operations; and to the general public who is ultimately
affected by the menace of spills. The tabular data contained here-
in are arranged as such that little interpretation is needed for under-
standing by laymen. Many diagrams, maps, and chromatograms have been
-------
included as supporting documentation and as reference for any
future work.
Many individuals have contributed to the success of this project
and should be acknowledged. 'Notably among them are Mr. Al J. Smith,
Chief, Environmental Emergency Branch of EPA, Region IV, who directed
the clean up and removal operation during the initial phases of the
spill and coordinated the Federal response activities; Mr. Kenneth
E. Biglane, Director, Oil and Special Materials-Control Division, EPA,
Washington, who conceived the idea for the follow-up study and pro-
1 . . v ^
vided the necessary funding; Messrs. Bill Loy and Tom Bennett of
Surveillance and Analysis Division, EPA,.Region IV, who assisted in
the sample analysis and the quality control phase.
It is reasonable to expect that subsequent studies of a more detailed
nature will be made in this area to answer many remaining questions.
George J. Mbein
Chief, Hazardous Materials Section
Environmental Protection Agency
Region IV, Atlanta, Georgia'
Hans Crump-Weisner
Co. Project Officer
Environmental Protection Agency
Washington, D. C.
ii
-------
TABLE OF CONTENTS
Page
No.
I. SUMMARY • . . • 1
II. CONCLUSIONS 2
III. RECOMMENDATIONS .'...' , 3
IV. INTRODUCTION . . . . . . 4
A. History of Spill 4
B. Brief Review of PCB Literature 5
1. Distribution of PCBs in the Environment 5
.2. Fate of PCBs in the Environment 6
3. Literature References Cited • 8
*•
C. Purpose of This Study *
V. TECHNICAL APPROACH . . U
A. Introduction *•*• •
B. Sampling Protocol and Collections . . ^
1. Criteria for Selection of Soil Sampling Sites ..... 1*
2. Initial Field Preparations ..... **
3. Drilling and Soil Sampling 15
4. Field Procedures Relating to Soil Sampling \ ^ 20
5. Site Selections for Sediment and Water 22
6. Sampling Procedures for Sediments, Water,
and Controls 23
iii
-------
TABLE OF CONTENTS
(Continued)
Page
No.
C. Preliminary Laboratory Preparations and Splits , ' '
of Soil Samples 24
D. Microbiological Studies . 25
1. Procedures for Microorganism Investigations 25
2. Determinations of Soil pH and Moisture 26
E. Analytical Methodology 27
1.. Analysis of Water, Sediment, arid Soil
for Polychlorinated Biphenyls (PCBs) ......... 28
2. Analysis of Water, Sediment, and Soil '
for Polychlorobenzenes 35
F. Isomer Verification 36
VI. EXPERIMENTAL RESULTS . . . . . . . . . . . . . . 41
A. Field Data 41
1. Cores . . 41
2. Geology . . 41
3. Environmental Samples . 46
4. Wells . 46
5. Climatology ........... 49
6. Soil Temperature ......... . 52
7. Drilling Equipment and Procedures . . . . . . . ... . 56
B. Microbiological Data \ . . . 56
C. Analytical Data from Electron-Capture Gas
Chromatography (EC/GC) • • • 78
D. .Special Soil Extraction Experiment .... 78
iv
-------
TABLE OF CONTENTS
(Continued)
Page
No.
E. Quality Assurance Data
F. Analytical Data for Gas Chromatography/
Mass Spectrometry (GC/MS) . . . . ............ 91
VII. FATE OF POLYCHLORINATED BIPHENYLS (PCBs) AND
POLYCHLOROBENZENES AFTER A TWO-YEAR EXPOSURE
IN A NATURAL ENVIRONMENT ......... ........ . . 92
A. Fate of Aroclor 1254 .................. , 92
1. Distribution of Aroclor 1254 In and
Around the Spill Site in 1975 . ........... 92
2. Degradation of Aroclor 1254 ..... ..... '. . . 108
B. Fate of Askarel Solvent ................. 113
C. Over-all Assessment of the Environmental
Impact of the Spill ...... ........ .....
1. Condition of the Area Immediately
After the Spill .............. ..... 117
2. Condition of Spill Area—August 1975 ....... . . U9
APPENDICES ....... ......... ............ I21
-------
LIST OF TABLES
Table No. l
1 Gas ChrOmatographic Operational Parameters—
PCBs ;.,.... . 31
2 Method Sensitivity—PCBs . . 32
3 Gas Chromatographic Operational /
Parameters—Askarel Solvent ............. 37
4 Method Sensitivity—Askarel Solvent 38
5 Description of Soil Types Characterizing
- Core Samples Taken in August 1975 45.
6 Climatological Data Showing Monthly
Averages for the Study Area 51
7 Rainfall Data Prior to and During
Well Sampling 53
8 Collection and Microbiological Data from
Soil Samples Taken in August 1975 58
9 Microbial Populations Relative to
Soil Types 65
10 Microbial Populations Relative to
pH of Soil Samples ' 66
11 Microbial Populations Relative to
Moisture Content of Soils 68
12 Microbial Populations Relative to
Core Sample Depths . ,70
13 Microbial Populations Relative to
PCB Concentration in Core Samples ......... 77
14 Collection and Analytical Data from
Three Test Core Sites Sampled Over
Four-Inch Intervals in August 197,5 ........ 79
vi
-------
LIST OF TABLES
(Continued)
Table No. Page
15 Collection and Analytical Data from
Core Sites Sampled Over Four-Inch
Intervals in August 1975 82
16 Collection and Analytical Data from
Core Sites Sampled Over Sixteen-Inch
Intervals in August 1975 83
17 Aroclor 1254 and Askarel Solvent
Analyses for Water, Sediment, and
Control Soil Samples 86
18 Results from Special Soil Extraction
Experiment—Aroclor 1254 . 88
19 In-House Quality Control Data—
Aroclor 1254 '...... 89
20 Percent Coefficient of Variation
Evaluation for In-House Quality Control
Data—Aroclor 1254 . .. . . 90
21 Comparison of Analytical Data from
Core Samples Collected in 1973 and 1975 106
22 Typical Homolog Composition of
Aroclor 1254 . Ill
vii
-------
LIST OF EXHIBITS
Exhibit No.
I Location of Spill Site in Relation
to General Topography . , 17
II Excavation Area and Location of
Core Sites with Contour Intervals ........ 42
»
III A Cross Section Profile of a
Northeast Intersect through
the Study Area 43
IV A Cross Section Profile of a
North-Northwest Intersect
through a Steep Surface Gradient
Section of the Study Area . 44
viii
-------
LIST OF FIGURES
Figure No. Page
1 Representative Chromatogram for ,
Aroclor ,1254 33
2 Representative Chromatogram for
Askarel Involved in 1973 Spill 34
3 Representative Chromatogram for
Solvent Portion of Askarel . 39
4 Representative Chromatogram for
Standard Chlorobenzene Mix . . .1 40
5 Location of Environmental ',
Sampling Sites ...'... 47
6 Locations of Well Stations Relative
to the Spill Site 48
7 Location of the Control Well ., 50
8 Temperatures of Various Soil Types
Over Four-Inch Intervals as Compared
with Ambient Conditions 55
9 Total Ion Current Chromatograms ... 93
10 Specific Ion Search for the 292* Ion 94
11 Specific Ion Search for the 326* Ion 95
12 Specific Ion Search for the 360+ Ion 96
13 Specific Ion Search for the 394+ Ion 97
14 Spectrum Typical of Four Chlorine
Biphenyl ......... 98
15 Spectrum Typical of Six Chlorine
Biphenyl 99
16 Spectrum Typical of Seven
Chlorine Biphenyl . . . 100
ix
-------
LIST OF FIGURES
(Continued)
Figure No. Page
17 Concentration Distribution of Aroclor
1254 in Core Samples, 1975 ............ 102
18 Concentration-Depth Profiles of
Core Site UC, 1973-1975 . . . . . 103
19 Concentration-Depth Profiles of
Core Site RL, 1973-1975 104
20 Concentration-Depth Profiles of
Core Site JF, 1973-1975 . . . 105
21 Comparative Chromatograms for
Aroclor 1254 in a Typical "Aged
Environmental Sample" (1975) ,
and Aroclor 1254 in the Askarel
Spilled in 1973 .... ....'......... 112
22 Solvent Concentration Profile—
Well Located Closest to Spill
Site (Station 7) . . 116
-------
I. SUMMARY
The area of an askarel spill, which had been cleaned up two years
prior to this study, was investigated for migration and/or degrada-
tion of residual PCB and lingering intrusion of the solvent into
ground waters. The'PCB was found unchanged while the solvent had
continued to leach into the underground water.
-------
II. CONCLUSIONS
No significant reduction in the concentration of Aroclor 1254 in the
'soil has occurred as the result of migration or degradation. There
is no way to clearly assess the effect of the original insult on the
soil microorganism population of the spill site since no microbiological
studies were conducted in 1973. There was, however, no evidence of
« X
permanent environmental damage detected in the spill area in 1975.
The more water-soluble components of the askarel solvent invaded the
ground water supply almost immediately after the occurrence of the
spill. Leaching was the migration mechanism responsible for the
intrusion of the lower chlorinated benzenes into the ground water.
-------
III. RECOMMENDATIONS
Because of the many diverse and interrelated effects imposed by a
particular spill environment, no threshold concentration of PCBs
\ ,
can be recommended which would be equally applicable to all land
spill occurrences. Although the data from this study can serve as
a guideline, each PCB spill will require an individual evaluation
and assessment since no appreciable migration or degradation was
detected in the specific environment Investigated. It is recommended
that more research be conducted with varying soil types and a more
favorable medium for sustaining microorganisms in greater abundance.
More knowledge of the toxicity of askarel on various microbla would
be required as an integral part of these degradation studies.
-------
IV. INTRODUCTION
A. History £f Spill.
On March 5, 1973, an accidental spill of approximately 1500
gallons of askarel occurred in a rural area near Kingston,
Tennessee. The spill resulted in the.environmental.contamina-
' 'I
tion of two watersheds because of its location on the crest of
a hill. Through the influence of rainfall, geology, and
characteristics of the overlying stratum of soil, the chemical
was subsequently dispersed through the soil both horizontally
and vertically.
The particular askarel spilled was composed of a commercial ,
mixture of polychlorinated biphenyls (Aroclor 1254) and a pro-
prietary solvent mixture of polychlorinated benzenes. Both
components of the askarel involved were chlorinated hydrocarbons
\
—a class of compounds noted for their persistence in the environ-
ment. Consequently, an extensive "clean-up" operation in the
I
affected area was started March 14, 1973, by the responsible
parties. To assure the effective removal and proper disposal of
the hazardous substances, the contaminated soil was packed and
sealed in metal drums before it was taken from the spill site.
The removal procedure resulted in extensive excavations in three
general areas. After ll,531 drums of contaminated soil were re-
moved, 'the excavated areas were sealed, backfilled, and packed
-------
with uncontaminated soil. The entire affected area was subse-
quently covered with top soil, seeded with grass, and landscaped.
On March 8, 1973, the Regional Office (Region IV) of the
Environmental Protection Agency (EPA) in Atlanta commenced a
sampling program of the affected areas to determine the concen-
tration level of contaminants in the soil and water table.
Stewart Laboratories, Inc. (SLI), a private laboratory located
in Knoxville, TN, joined EPA in the sampling program on March 21,
1973. This sampling program, for monitoring and detection pur-
poses, continued jointly by EPA and SLI for several additional
weeks. In addition to the EPA sampling program, Stewart
Laboratories, provided sampling and analytical assistance during
the excavation.
A monitoring program, approved jointly by EPA and TWQC (Tenn.
Water Quality .Control Board) was conducted for a 12-month period
following the cleanup operation.
B. Brief Review of_ PCS Literature.
1. Distribution £f_ PCBs jln the Environment. According to the
Interdepartmental Task Force on PCBs (1), the history of
PCBs started in 1929 when industry introduced them for use
as non-flammable oils in electrical transformers, condensers,
and in paint. During the next forty years,- industrial uses
of PCBs grew steadily. They have been widely employed as
plasticizers, as sealers in waterproofing compounds and
putty, in printing inks, in waxes, in synthetic adhesives,
-------
in cutting oils, as dielectrics, as hydraulic fluids, as
high-pressure lubricants, and as a heat-transfer medium (2).
Sales of PCBs in the United States came to about 34,000
tons in 1970, with cumulative production over the years
amounting to an estimated 4 x 10^ tons (3).
Although PCBs were never intended for direct release into
the environment, they were first identified by Jensen as
a potential food contaminant in 1966 (4). Since that time,
it has been demonstrated that .they are ubiquitous environ-
mental pollutants. Numerous studies (5-18) have confirmed
their presence in animals and the aquatic environments as
well as in humans.
2. Fate of PCBs jin the Environment. The sparsity of knowledge
about the fate of PCBs in the environment is Illustrated by
the recommendation of. the Interdepartmental Task Force on
PCBs (1) that "more scientific information about PCBs is
/
needed" relative to their occurrence, transfer, and cycling
in the environment. Only general statements can be made
about how PCBs reach the environment and how they reach
target organisms.
The biologically important characteristics of PCBs are their
insolubility in water, high solubility in fats, toxicity to
metabolic processes, and extreme stability. The combination
of persistence and accumulation in fat (during transition
-------
through a food chain) can result in considerable increases in
concentration of the compounds at higher levels in the food
chain. Like DDT, PCBs are.reported to degrade very slowly
under natural conditions (19). The highly chlorinated PCBs
(Aroclor 1254 and 1260) seem to persist in the environment
longer and are less toxic than the more rapidly degradable
low chlorine forms (20).
The metabolism of several PCB isomers in fish, rats, and
pigeons has been studied by Hutzinger and co-workers (21).
Mono-, di-, and tetrachlorobiphenyl isomers were converted
into their corresponding mono-hydroxy derivatives in rats
and pigeons. However, no hydroxylation of any of the isomers
was observed with fish. It was also observed that 2,2',4,4',
5,5'-hexachlorobiphenyl was not oxidized by any of the
animals in the study.
Laboratory photolysis studies on pure chlorobiphenyl isomers,
as well as Aroclor 1254, have been reported by a number of
investigators (22-26). Using hexane as the solvent, the de-
chlorination reaction was predominant. However, in the
presence of air and water, a number of polar products were
observed. Examination of the polar products.from the Aroclor
1254 irradiation indicated formation of both hydroxylated
and hydrated products. Although these results suggested
pathways in which degradation of environmental PCB might
-------
8
occur, the actual observation of degradation products in
"natural environmentally aged samples" has not been reported
to date.
3. Literature References Cited.
1. .Interdepartmental Task "Force\on.PCBs, U. S. Department
of Commerce Pub. COM-72-10419, 1972.
2. Bonelli, E. J., Am. Laboratory, February 1971.
. 3. Hammond, A. L., Science, 175. 155 (1972).
4. Jensen, Soren, PCB Conference, National Swedish
Environment Protection Board, Stockholm, 7 (Sept., 1970).
5. Risebough, R. W., Rieche, P., Peakall, D. B., Herman,
S. G., and Kirven, M. N., Nature, 220. 1098 (1968).
6. Veith, G. D., and Lee, G. F., Water Research, 4., 265
(1970).
7. Martell, J. M., Rickert, D. A., Siegel, F. R., Environ.
Sci. Technol., 2» 872 (1975).
8. Maugh, T. H., II, Science, 178, 388 (1972).
9. Veith, G. D., and Lee, G. F., Water Research, _5,
1107 (1971).
10. Bailey, S.,,and Bunyan, P. J., Nature, 236, 34 (1972).
11. Edwards, R., Chem. and Ind., 1340 (1971)
12. Price, H. A., and Welch, R. L., Env. Health Perspectives,
it 73 (1972). ,
13. Burns, J. E., Pesticides Mon. J., _7, 122 (1974).
14. Gruger, E. H., Jr., Karrick, N. L., Davidson, A. I.,
and Hruby, T., Envion. Sci. Technol., JJ» 121\(1975).
15. Koeman, J. H., et al., Nature, 221, 1126 (1969).
16. Bache, C. A., et al., Science, 177, 1191 (1972).
-------
17. Holden, A. V., Nature, 228. 1221 (1970).
18. Crump-Wiesner, H. J., Feltz, H. R., Yates, M. L., Jour.
Research U. S. Geol. Survey, _!, 603 (1973).
19. Waldbott, G. L., Health Effects of Environmental
Pollutants (C. V. Mosby Co., St. Louis, 1973), p. 225-9.
20. Monsanto Company, Presentation to the Interdepartmental
Task Force on PCBs, Washington, D. C., May 15, 1972.
21. Hutzinger, 0., et al., Science, 178, 312 (1972).
22. Stalling, D. L., 163rd National Meeting, ACS, New York,
N. Y., Symposium on PCB, No. 22, August 1972.
23. Hutzinger, 0., Jamieson, W. D., and Zitko, V.,
Nature, 225. 664 (1970).
24. Safe, S., and Hutzinger, 0., Nature, 232. 641 (1971).
25. Ruzo, L. R., Zabik, M. J., and Schuetz, R. D., Bull.
Environ. Contam. Toxicol., j}, 217 (1972).
26. Hustert, K., and Korte, F., Chemosphere, 1, 7 (1972).
C. Purpose of This Study.
«
The primary purpose of this project is to study the biodegradation
effects of a natural environment on the chemical components of an
askarel spill after a two-year time lapse. Extensive experimental
background data were obtained during and after the 1973 cleanup
procedures employed at the study site. Correlation of 'the 1973
data with the results of this follow-up study will provide a means
of assessing the distribution and fate of PCBs and polychlorinated
benzenes in natural environmentally aged samples.
The findings of this project may form the basis for deriving a safe
concentration level for PCBs in soil, The practical application
-------
10
of such a threshold level will assist EPA spill response personnel
to determine the degree of soil removal necessary in a land spill
situation. Sampling of water supplies adjacent to the spill area
will provide information on the rate of intrusion of lower chlori-
nated benzenes into ground water. Such information will be of
paramount importance in increasing the efficiency and effective-
ness of EPA's spill cleanup efforts.
-------
11
V. TECHNICAL APPROACH
A. Introduction.
Potential mechanisms for the loss of askarel from the spill
site include volatilization, leaching, and metabolic and/or
nonmetabolic degradation. The experimental approach to be
employed in this study has been designed in a manner which will
detect and assess the magnitude of each potential route for.the
removal of residual askarel from the spill site.
Data obtained by Stewart Laboratories, Inc., during the 1973
site cleanup included: (1) core samples in the bottom of
excavation areas immediately prior to fill; (2) core samples
at elevations geologically below the spill; (3) water and sedi-
ment samples from a spring well in a watershed below the site;
(4) water samples from domestic wells adjacent and peripheral
to the spill area; (5) water samples from sources in the general
vicinity of the spill area; and (6) water samples from a domestic
control well. The initial phase of this project involved a •
re-sampling of the area in order to determine migration and/or
degradation of the PCBs in soil and askarel concentration changes
in water.
The second phase of the study provided the analytical data
necessary to evaluate any changes which might have occurred over
the past two years. Identical analysis procedures, concentration
units, and detection limits were employed for the determination
-------
12
of PCB content in both the present study and the 1973 project
so that direct comparison of the two sets of data would be
facilitated. Careful choices were made as to which samples
should be confirmed and identified by GC/MS analysis. The.
requested isomer identifications were done by the Surveillance .
and Analysis Division, Region IV, EPA, in Athens, Georgia. The
techniques employed in the field and laboratory positively identi-
fied soil profile correlations which allowed for the interpre-
tation of chemical and biological data.
Core samples of soil were examined by established procedures
to determine the relative numbers and major types of microbial
flora. Among these were the true bacteria, actinomycetes, and
a wide variety of microscopic fungi, all of which were enumerated
and identified.
Algae and protozooans are also found in soil. However, they
are far less numerous than the other microorganisms (27) and
are more difficult to evaluate because of somewhat unique
cultural requirements and were, therefore, not investigated.
Since available literature indicates that very little is
(27) Microbiology. 3rd Ed., 1972. M. J. Pelczar and R. D. Reid. McGraw
Hill Pub., New York, New York.
-------
13
known concerning the degradation of PCBs in the natural environ-
ment, potential biodegradative microorganisms are not identifi-
able. The two-year lapse since the spill incident, however,
should have allowed for some valid degradation. The over-all
approach in determining microorganism abundance was to isolate
certain groups which were probable major participants in
chemical alterations. It was realized, however, that other
environmental conditions such as pH, moisture, type of soil,
and sample depth influence the relative numbers and varieties
of soil flora.
B. Sampling Protocol and Collections.
During the field sampling phase of the project, the contractor
collected 120 soil samples and 40 water samples in and around
the spill area. These samples were to constitute the basic
materials for the study of the distribution and fate of
residual PCBs in the natural environment. Probable causes
of measured and documented changes in the chemical status
of the spill components could then be addressed.
An initial over-all assessment was made of the requirements
for proper choice of primary sampling sites, procedures for
collection of samples and attendant field data, as well as
essential equipment and supplies. A final protocol was estab-
lished which required only minor modifications once the field
work began.
-------
14
1. Criteria for Selection of Soil Sampling gites. All analytical
data obtained should have reasonable reference to analyses
performed in 1973. The core sites chosen for comparison
were to represent a broad concentration range. The areas
from which these cores were taken needed to exhibit a full
spectrum of topographic variables which would characterize
the study area as to soil type, probable hydrology, microbial
environments, and potential migration of PCBs. Attempts
needed to be made to sample areas of undisturbed terrain
as well as those which had been excavated or top dressed
in 1973. Distribution patterns of the residual contamina-
tion in the 1973 cores were studied in order to replicate,
as closely as possible, the same contamination variability
for the 1975 sampling. The depth of sample takes would
be dependent on the assessment of all other criteria thus
mentioned. It would, therefore, be essential that an
appropriate number of available sites be chosen so that
.t
the interdependency of all these requirements could be up-
held; and unforeseen field situations would not preclude
an adequate sampling.
t
2. Initial Field Preparations. The spill area had received
'.
little maintenance since it had been backfilled and seeded
in 1973. Successive cuttings with rotary mowers and hand
raking were required to remove grass, weeds, and limbs from
the study area. Photographs, drawings, and field data were
-------
15
used to approximate the location of the excavation perimeter
and the 1973 core sites. Surveyors' flags and stakes were
used to mark these assigned areas. Photographs and rough
sketches were recorded as guides for final and more exact
designation.
3. Drilling and Soil Sampling. The extreme variability in depth
of required cores and the soil matrix (clay with fractured
.quartz) indigenous to the site area required the use of a
truck-mounted drill rig. The drill unit was thoroughly
washed and cleaned prior to arrival on site. It remained
in the study area throughout the sampling period.
A methodical drilling program was begun to determine the
excavation limits in all areas peripheral to probable
sampling sites. Continuous cores were pulled using a split
spoon sampler for the determination of soil profiles. Since
the floor of all excavated areas had been sealed with "Vis-
queen" and a layer of sand (2-4 inches) prior to backfilling,
the position of the perimeter of the disturbed area was
readily discernible. Appropriate adjustments on all drawings
and records were made as documentation. All field markers
were securely installed at their verified locations. During
this preliminary drilling period, extreme precautions were
taken against contamination of the surface of the study
areas. The positioning of the drill truck and the disposition
-------
16
of the test cores were under constant surveillance. Once
logging of core data was completed, the cores were tamped
back in their original holes.
All original 1973 core sites,external to the excavation had
been backfilled, packed, and covered with top soil. In
order to identify the core holes and differentiate fill
top soil from natural overburden, each core hole was found ' -
and drilled once with the split spoon. By visual observa-
tion and actual measurement of the soil strata, the depth
of unnatural cover material was ascertained. The sampler
was pressed gently and rotated in order to avoid compact-
ing. Impact techniques were restricted to instances where
i
chert refused penetration.
Available core sampling equipment included a split spoon
sampler, a Shelby Tube Sampler, and a California Sampler.
The order of desirability of the three core samplers were:
a. California sampler—consists of four sequential brass
units two inches in diameter and four inches long allow-
ing for ease of sample removal and differential choice
of sample take. The cost per tube is moderate and
they are readily available. Two limitations were the
questionability of penetration of quartz layers and
the sample size.
-------
ROAD CLASSIFICATION
Heavy-duty ______ Poor motor road ... -,- = = ,-
Medium-duty ^-_=__^ Wagon and jeep track
Light-duly == Foot trail
-EXHIBIT I Location of Spill Site in
Relation to General Topography
In developed areas, on
BACON GAP, TENN
N3545-W8430/7.5
t> ^ if,-2/A;sl .V'c-fn^.^^-r^--
SCALE 1:24000
o
1000
1000
2000
3000
5000
l fECI
utM GRID AND 1968 MAGNETIC NORTH
DECLINATION AT CENTER OF SHEET
. CONTOUR INTERVAL 20 FEET
.DASHED LINES REPRESENT HALF-INTERVAL CONTOURS
DATUM IS MEAN SEA LEVEL
THIS MAP COMPLIES WITH NATIONAL MAP ACCURACY STANDARDS
FOR SALE BY U.S. GEOLOGICAL SURVEY. WASHINGTON. D.C. 20242.
TENNESSEE DIVISION OF GEOLOGY, NASHVILLE. TENN. 372.19.
U.S. TENNESSEE VALLEY AUTHORITY. CHATTANOOGA, TENN. 37401 OR KNOXVILLE. TENN. 3/90?
A FOLDER DESCRIBING TOPOGRAPHIC MAPS AND SYMBOLS IS AVAILABLE ON REQUEST
-------
18
b. Split spoon sampler—an open-chambered tube two inches
in diameter and ^24 inches long which provides a
relatively rapid sample take. Two disadvantages of
this sampler are possibility of disturbed core and poten-
tial cross contamination of samples.
c. Shelby tube—a sturdy impact corer of three inches in
diameter and thirty-six inches in length and capable
of delivering substantially undisturbed samples. Three
disadvantages are that each sample has to be extruded
with a press, the tubes are very expensive, 'and the
units are marketed with a baked-on plastic liner to
prevent rust.
After careful consideration of the sampler types and the
inherent geology involved, the California Sampler was
f
initially chosen for trial. Because of extreme difficulty
encountered in removing clay and chert fragments from the
tubes in the field, the split spoon sampler was used for
all consequent core ^samplings.
Since the topography of the study area rolls at varying
angles around the spill site, a fairly comprehensive choice
of sampling spots was made. The Knox Dolomite (undivided)
sub-surface versus the overburden influx on some of the sites
external to the excavation broadened the coverage of the
-------
19
study region from the standpoint of degradation and/or move-
ment of PCBs. A topographic representation of the general
spill site is shown in Exhibit I.
Sample core drilling began on August 4, 1975. The first
three core sites chosen for replicate testing were drilled
in a four-cornered quadrant pattern. Each of the four cores
was taken approximately one foot from the center of the 1973
core location. All samples were taken over four-inch inter-
vals to allow for differentiation of gradient effects and
reproducibility studies.
In order to representatively sample the site variations as
specified in the protocol, all subsequent collections were
designed to take soil from one core hole adjacent to an
identified 1973 core hole. This phase of collections began
with core taken over four-inch intervals; however, sampler
*
refusal and tendency toward loss of sample integrity be-
cause of inherent quartz in the clay matrix precluded
sampling over such a short interval. A decision was made
to complete the sampling with cores composited over sixteen-
inch intervals.
^- . •
Upon completion of all collections peripheral to the
excavation area, core samples were taken at a level approxi-
mating some of the original sample points at the bottom of
-------
20
the backfilled areas. The location of these samples was
confined to an area within three feet of the 1973 sampling
point. Few sites were available for this phase of the study
since nearly all excavation areas received additional clean-
ups after final data had been obtained in 1973. Additionally,
the floor and walls of one deeply excavated area had been
impregnated with a material to inhibit movement of any
residual askarel. There was no ready access to the basin
sampling areas without breaking the seal to the environment.
4. Field Procedures Relating £o Soil Sampling. The techniques
employed in the actual sampling and handling of core samples
\ • '
and the collection of supplemental field data followed the
protocol with no modifications.
a. . Core samples—Each split spoon sample was carefully re-
moved onto a piece of aluminum foil supported on a ply-
wood base. It was measured, cut into four sections of
four-inch length, logged, characterized as to general
soil type, and sealed in labeled foil for transport to
the laboratory. \
\
b. Cleaning and contamination precautions—Prior to each
\ ' '
use, all drilling equipment, core cutting blades, etc.,
-------
21
were washed with water, and a brush, scrubbed in a bath
of technical grade hexane, and finally rinsed in
pesticide grade hexane. This operation was not performed
near sampling sites. Disposable gloves were used while
handling each sample take, and all throw-away materials
were kept carefully sealed. Foil strips were removed
from the roll and appropriately labeled in an area remote
to the core operations.
c. Temperature measurements—A general profile was made
of the temperature of the soil.at several points within
the study area. A four-inch temperature-sensing probe
with a strip chart recorder readout was inserted in the
test cores. Cross checks on the accuracy of these read-
ings were made by inserting the probe in soil at very-
ing depths in advance of core takes. General confirma-"
tion between these two procedures was obtained. The
probe was cleaned in the manner described in "b" above.
»
d. Mapping—A grid survey .on two-foot intervals was made
of the entire study area. The data from these measure-
ments were translated into a contour map depicting the
exact location of sampling sites and excavation limits
on a grid basis. Substantiating photographs were shot
over the entire region with identifying landmarks in
plain view.
-------
22
e. Restoration—A concerted effort was made to return the
1 / . -
area back to its original state. All core holes were
plugged; stakes, flags, and twine were removed; and all
equipment, tools, and supplies were returned to, the
laboratory facility.
5. Site Selections for Sediment and Water. A protected spring
well located in the watershed below the spill site had been
monitored during the 1973 cleanup and subsequently for about
a year. Since some positive data had been obtained from
-------
23
6. Sampling Procedures for Sediments, Water, and Controls.
Once the sites external to the spill were chosen and collec-
tion permission obtained from the well owners, grab samples
were taken after relatively dry weather (August 29, 1975)
and again after heavy rains (September 26, 1975).
a. Techniques—^Sediments were taken with a metal, roll-
sided scoop since the depth of water was quite shallow.
Each sample was placed in a pre-labeled glass jar whose
cap was lined with aluminum foil.
All water samples were taken directly into a pre-
labeled gallon glass jug. Well samples were obtained
at the tap from which they had previously been sampled
in 1973. Each tap line was allowed to flush for 2-3
minutes so, that samples could be drawn directly from
\
the well system itself.
Two control soil samples were taken from nearby areas
not influenced by the spill. The collections were made
at 3-6 inch depths below the surface. One of these was
from an exposed clay and chert matrix while the other
was primarily loam.
b. Cleaning and contamination precautions—The sediment
scoop was cleaned according to the procedure previously
given for field equipment. All glass containers were
-------
24
washed with distilled water, rinsed with technical grade
hexane, rerinsed with pesticide grade hexane, capped with
foil liners, and labeled at the laboratory. Special care
was taken to insure that caps and glass container mouths
did not come into contact with potential contaminants
in the field.
C. Preliminary Laboratory Preparations and Splits of Soil Samples.
Sediments were passed through a quarter-inch screen in order to
remove inorganic and organic debris. These samples were allowed
to air dry in a metal tray. The dried material was then passed
through a U. S. Standard No. 18 sieve to insure uniformity of
particle size. The sediments were then submitted to the
laboratory for PCB analysis.
Core samples were mixed and spread for air drying on their
individual aluminum foil sheets. As the samples were being
spread, an approximate 40-gram sample for moisture, pH, and
microbiological determinations was sealed in a pre-labeled
container and submitted for analysis. Forty representative
soil samples of approximately 100 grams each were sealed in
containers for potential subsequent anaerobic study.
v '. • i
The dried core aliquots for PCB determinations were passed through
a U. S. Standard No. 18 sieve after obvious rocks and debris had
been removed. The fines were then passed through a splitter in
-------
25
order to obtain an aliquot of an appropriate size for analysis.
This aliquot was submitted to the laboratory for analysis of
PCBs.
All sample preparations were performed under clean room condi-
i
1 tions. Disposable gloves were worn and all equipment and con-
tainers were pre-cleaned by washing and subsequent hexane rinses.
. • . i
Samples were kept distinct from each other, and each step of -
the preparations was done in a separately ventilated zone in
. the laboratory. .
D. Microbiological Studies.
One hundred forty-two soil samples were examined to determine the
populations of heterotrophic aerobic bacteria, fungi, and actino-
mycetes, which constitute the majority of microorganisms in soil.
Laboratory data concerned with microbial flora enumeration and
identification were researched in correlation with soil type, pH,
moisture, and sample depth.
1. Procedures for Microorganism Investigations. The various
techniques and culture media described by Parkinson, et al.
(28) were used for the isolation of the soil microorganisms.
Ten grams of each soil sample were suspended in 95 ml of
sterile distilled water to give a 1:10 dilution. From this,
(28) Methods for Study the Ecology of Soil Microorganisms. 1971.
D. Parkinson, T. R. Gray, and S. T. Williams. International Biological
Programme Bladswell Scientific Publications, Oxford, England.
-------
26
serial 10-fold dilutions were prepared. Dilutions of 1:100
through 1:1,000,000 were employed for determining plate
counts to enumerate the three microbial groups. The numbers
of heterotrophic aerobic bacteria were obtained by plating
the various dilutions of the soil samples in Plate Count
Agar (Difco Laboratories, Detroit, Mich.) and incubating
at 25°C for 72 hours. Colony counts were then determined
with the use of a Quebec Colony Counter (American Optical
Company, Buffalo, N. Y.). Determination of fungal counts
were conducted by plating dilutions of the samples in Sabour-
aud's Dextrose Agar (Baltimore Biological Laboratories,
Cockeysville, Md.) with incubation at 25°C for 96-120 hours.
Enumeration of the organisms was performed as described for
the heterotrophic bacteria. Actinomycete counts were made
by plating soil dilutions in Starch Casein Agar, containing
0.0002 percent actidione for inhibition of fungi (Kuster and
Williams, 1964X, followed by incubation at 25°C for 96-120
hours. Again, colony counts were made after incubation as
described above. If no growth was observed on the fungal
or actinomycete plates after 96 hours, incubation was con-
\
tinued and the plates read at 120 hours in order to detect
any slower growing organisms. The heterotrophic aerobic
bacteria grew readily and never required more than 72 hours
for maximum growth.
2. Determinations of Soil pH and Moisture.
The pH of each sample was obtained by preparing a slurry
with three grams of soil and three ml of .distilled water
-------
27
in a 10 ml beaker and making the determination with a
Labomatic Model 165 pH meter.
Moisture content of a representative five-gram aliquot of
I
each sample was obtained by drying to constant weight at
E. Analytical Methodology. j
The method of choice for the analysis of PCBs in all known
»
monitoring and regulatory applications is GC/EC (gas-liquid
\
chromatography utilizing electron capture detection) . This
analysis mode is, likewise, most frequently employed in the
assessment of the environmental impact and health effects of
PCBs. This overwhelming utilization is by no means intended
to imply that the analytical method is the ideal mode of analy-
sis for PCBs. Gas chromatography is not ah inherently definitive
analytical technique. It is subject to serious complications
when other electron-capturing components are present in the
samples in addition to the PCBs. The shortcomings of the method
can be successfully overcome when the analyst involved is fully
cognizant of the ramifications of the situation. Most analytical
techniques incorporate a liquid chromatography cleanup of liquid-
liquid extracts prior to GC/MS (gas chromatography/mass spectro-
metry) analysis. This procedure is % 90 percent effective in
separating PCBs from organochlorine pesticides.
-------
28
The utilization of two or more unlike columns in the gas
chromatpgraphie analysis is another means for establishing the
identity of gas chromatographic patterns. The application of
GC/MS is considered by most experts to be the desired technique
for confirming qualitative identification of gas chromatographic
patterns. Microcoulometry and thin layer chromatography are
also useful tools for positive identifications. It is appro-
priate to conclude, however, that GC/EC is a most effective
*
analysis mode for the detection and measurement of PCBs in
environmental samples when interfering substances are either
totally absent from the samples under study or when they have
been effectively removed prior to analysis.
1. Analysis of Water, Sediment, and Soil for Polychlorinated
Biphenyls (PCBs).
a. Background information—Since this project is a follow-
up study and many conclusions will be based on correla-
tion of current analytical data with that obtained at
the time of the spill (1973), a consentaneous decision
was made by all responsible parties that the GC/EC
analytical methodology should not change from that em-
ployed in 1973.
The method employed in 1973 for the analysis of water,
sediments, and soils for PCBs is basically the method
from the R & D Laboratories of Monsanto Company (Method
69-13) which is contained in the publication Manual
-------
29
of Analytical Methods prepared by the Perrine Primate
Research Laboratories of EPA (current designation:
Pesticides and Toxic Substances Effects Laboratory, NERC,
Research Triangle Park, N. C.). The contractor modi-
fied the Monsanto method to incorporate the Perrine
column selection and instrumental recommendations.
In principle, the contractor's method is unchanged
from the Monsanto method. The PCBs in water, sediment,
and soil are extracted into hexane. Interfering com-
ponents, if present, are then removed from the extracts
by chemical treatment and column adsorption chromato-
graphy. The amount of PCB present is determined by
electron capture gas chromatography.
• • ' \ '
b. Analytical method—The complete Monsanto method consti-
tutes the Appendix of this report, and only those pertinent
modifications employed by the contractor will be discussed
at this time. The chromatographic column used is the most
significant change. The column packing is 1.5% SP
2250/1.95% SP 2401 on 80/100 mesh Supelcon AW-DMCS
(Cat. #01-1947, Supelco, Inc.). This is a custom packing
prepared to the Perrine Research Laboratories' specifica-
tions and is especially suitable for the separation of
chlorinated pesticides and other chlorinated hydrocarbons.
The gas chromatographic conditions employed for the
-------
30
t
analysis of Aroclor 1254 and method sensitivity data are
contained in Tables 1 and 2, respectively. Typical
chrbmatograms of Aroclor 1254 and the askarel involved
in the 1973 spill are shown in Figures 1 and 2.
i
c. .Operational narrative—
Concentration and chemical clean-up. After the initial
liquid-liquid or liquid-solid extractions, exploratory
chromatograms were run to determine whether the extracts
would require further adjustment by dilution or concen-
tration to bring the peaks into a quantifiable range
consistent with the linear range of the detector. When
these initial chromatograms indicated the presence of
interfering materials, extensive clean-up procedures
were employed. In most instances, the hexane extracts
required no additional clean-up.
Contamination. To insure against the possibility of
undetected contamination, blanks were routinely carried
through all steps of the procedure.
Measurement. Initially, both the individual and total
peak height methods were employed to determine the
amount of Aroclor 1254 present. It was soon found,
however, that a calibration plot of the major isomer
peak height could be used for the quantitation of the Aroclor.
-------
31
Table 1. Gas Chromatographic Operational Parameters - PCB's
\
Instrument: Beckman GC-45
DETECTOR: Electron Capture (polarized helium plasma)
Source Current: 7ma Polarizing Voltage: 610 volts
Scavenger: He, C02, Rate 80, 1.2 ml/min.
GAS: Helium Carrier Flow: 60 ml/min.
COLUMN: Glass Length: 6 feet Diameter: 1/4"
Coating: SP 2250/SP 2401 Cone.: 1.5Z/1.95Z
Support: Supelcon AW-DMCS Mesh: 80/100
TEMPERATURE:
Column: 195°C Injection Port: 220°C
Detector: 250°C Detector Line: 240°C
SENSITIVITY: x 8 K Recorder Range: 1 mv
CHART SPEED: 1/2 inch/minute
-------
32
Table 2. Method Sensitivity - PCBs
Water: 0.5 parts per billion sensitivity
Absolute sensitivity = 0.1 x 10"' grams
Volume injected = 6 pi
Final volume extract = 10 ml
Sample size = 1000 ml
• (
Sediment: 0.05 parts per million sensitivity
Absolute sensitivity = 0.1 x 10"' grams
Volume injected = 6 pi
Volume extract =» 200 ml
Sample size = 100 grams
-------
Figure 1. Representative Chromatogram for Aroclor 1254
-------
Figure 2. Representative Chromatogram for Askarel Involved in 1973 Spill
4--80-U-I-
Aroclor 1254
-------
35
Chlorinated Pesticide Interferences. Since surface
water drainage of the area of interest was known to
involve some agricultural land, a chromatogram showing
the elution pattern of 13 of the more common chlorinated
pesticides was run using the instrumental conditions of
the Aroclor method. Pesticides in the mixture included
a - BHC, 6 - BHC, Lindane, Heptachlor, Aldrin, Hepta-
chlor Epoxide, p,p'-DDE, Dieldrin, Endrin, o,p'-DDD,
p,p'-DDD, o-p'-DDT, and p,p'-DDT. The major isomer
peak for Aroclor 1254 was free from interference from
any of these pesticides under the analysis conditions
employed.
\
2. Analysis of_ Water, Sediment, and Soil for Polychlorobenzenes.
a. Background information—Since the solvent for Aroclor 1254
in askarel is a mixture of chlorobenzenes, the fate of
the solvent in the environment of the spill area needed
to be determined. Because of chemical similarities
between PCBs and the solvent components, the analytical
method of choice was again electron capture gas chroma-
tography.
b. Analytical method—The solubility of the solvent compo-
nents in hexane made it possible for a solvent analysis
to be performed on the same extract prepared for the
-------
36
PCB1 analysis. Likewise, the versatility of the chroma-
tographic column selected for the PCB analysis allowed
for its use in the solvent analysis. Instrumental
parameters for the solvent analysis are found in Table 3.
Method sensitivity data are contained in Table 4. A
• / •
chromatogram of the solvent portion of the askarel is
shown in Figure 3. For reference, a chromatogram of
a mixture of various chlorobenzenes is also included .
as Figure 4.
F. Isomer Verification.
The Surveillance and Analysis Division, Region IV, Environmental
Protection Agency, Athens, Georgia, provided the identification
and verification of PCB isomers in the natural environmentally
aged samples using GC/MS. ^ •
-------
37
Table 3. Gas Chromatographic Operational Parameters - Askarel Solvent
•
Instrument: Beckman GC-45
•
DETECTOR: Electron Capture (polarized helium plasma)
Source Current: 7 ma Polarizing Voltage: 610 volts
Scavenger: He, C02, Rate 80, 1.2 ml/min.
GAS: Helium Carrier Flow: 60 ml/min.
COLUMN: Glass Length: 6 feet Diameter: 1/4"
Coating: SP 2250/SP 2401 Cone.: 1.5Z/1.95Z
Support: Supelcon AW-DMCS Mesh: 80/100
TEMPERATURE:
Column: 125°C Injection Port: 220°C
Detector: 250°C Detector Line: 240°C
SENSITIVITY: x 8 K Recorder Range: 1 mv
CHART SPEED: 1/2 inch/minute
-------
38
Table 4. Method Sensitivity - Askarel Solvent
Water: 0.006 parts per billion sensitivity
Absolute sensitivity « 0.003 x 10"9 grams
Volume injected - 6 yl
Final volume extract » 10 ml
Sample .size - 1000 ml
Sediment: 0.010 parts per million sensitivity
Absolute sensitivity - 0.003 x 10"' grams
Volume injected - 6 yl
Volume extract - 200 ml
Sample size - 100 grams
-------
Figure 3. Representative Chromatogram
Solvent Portion of Askarel
H-H-i -;••:•;-
I'!: H.J i!;'
-------
riTTr•' 1111:
hi! ! ! M
Figure 4. Representative Chromatogram for Standard
-------
41
VI. EXPERIMENTAL RESULTS
.JL Field Data.
The sampling phase of this project involved only minor adjustments in
the initial protocol. A total of 145 core, 2 soil control, 19 water,
and 3 sediment samples were ultimately submitted to the laboratory for
analysis. Field log information and analysis data for these samples
are presented in a later section of this report.
1. Cores. Forty-five core sites were sampled over a 10-day period.
Exhibit II shows the relative position of each of these sites with
reference to the 1973 excavation area, the local topography over a
2-foot interval, and all other core locations. Cross section profiles
at two intersects are shown in Exhibits III and IV as an illustration
of the gradient in the immediate area of the spill site.
2. Geology. The geological formations of the site area strike 40° north-
easterly on the average, and normal dips are 25-35° to the southeast.
The rocks underlying the total spill area are part of the Knox dolomite
group. A shaly limestone series known as the Chickamauga formation lies
to the southeast and upon the Knox group. The Knox of this area is
the usual sequence of thin to massive-bedded dolomite (high MgC03 as
compared to limestone with its high CaCC^) and is well fractured.
The overburden above the Knox group is thick—in the 50 to 150 'foot
range. The primary overburden is made up of clays mixed heavily with
chert fragments (see Table 5). On top of the clay is a zone of top
soil ranging from 0 to 4 feet in thickness. The top soil allows for
ease of perculation while the clays are penetrable primarily through
the fracture crevices of the chert.
-------
Exhibit II _
Excavation Area and Location of Core Sites
with Contour Intervals
Sccle: 1 inch "* 20 :"-ec
Datun is Mean Sea Level
N2
-------
Exhibit III. A Cross Section Profile of a Northeast Intersect
through the Study Area
Tcpsoil
Sc.ile: 1 inch'20 feet
U>
-------
44
Exhibit IV. A Cross Section Profile of a North Northwest
Intersect through a Steep 'Surface Gradient
Section of the Study Area,
Topsoil
Cross Section Profile
Intersect B
N
Scale: 1 inch^O feet
-------
45
Table 5. Description of Soil Types Characterizing
Core Samples Taken in August 1975
Code No. General Soil Characterizations
1 Reddish-brown clay with chert fragments
2 Yellowish-brown clay with chert fragments
3 Greyish-brown shale to reddish-brown clay
4 Reddish-brown clay
5 Greyish-brown silty clay with chert fragments
6 Greyish-brown to reddish-brown silty clay
with numerous chert fragments
7 Greyish-brown to reddish-brown clay with
chert fragments
8 Reddish-brown silty clay with chert fragments
9 Yellowish-brown silty clay with chert
fragments
10 Greyish-brown clay with chert fragments
11 Loam (Control)
-------
46
3. Environmental Samples. The spill occurred at a,point where
most of the drainage is geologically controlled by a north-
east trending hollow near the site. This valley is aligned
along the strike of the formations, and there are no visible
outcrops of bedrock to indicate that sub-surface liquids
would come to the surface after they entered the clay over-
burden.
A spring well located in the hollow seemingly provided an
excellent sampling site for detecting movement of suspected
spill materials in the ground water. In order to test this
thesis, the spring and the mouth of the creek draining the
hollow were sampled for water and sediment contamination.
Two soil control samples were taken for analytical and
microbiological analyses. One site was characterized by
clay and chert within the first three, inches of the over-
burden while the clay at the other site was covered with
two feet of loam. ' . x •
Locations of all of these environmental sites are given in
^ Figure 5. .
4. Wells. Six wells within close proximity of the spill were
\ ' i
chosen for analysis. These wells were included in the
study because they were all located along probable geological
- /
strike formations, and 1973 data on these wells were extensive.
The relative locations of the study wells are given in Figure 6.
-------
Heavy-duty ..
Medium-duly
Light-duly ...
ROAD CLASSIFICATION
.. _<•___> Poor motor road .... sa,,*,;
.. L xi. Wagon and jeep track
' Foot trail
In developed treat, only through road! are clasiilied
FIGURE 5. Location of Environmental
Sampling Sites
'SS^.Cj.'mTS;
KENTUCKY
MOYT- •"•""•"••" y^-
7/~^ -•-•'
7TENNESSE.E: •J^/
L '. ' _ .. r .. •^f
BACON GAP, TENN
N3545-W8430/7.5
1968
AMS 4055 I SE-SERIES VS4I
SCALE 1:24000
o
1000
1000
2000
3000
4000
5COC
70*,
JTH GRID *NQ 19M MACNCUC NORTH
OCCUNATION AT CENTER OF SHCCT
CONTOUR INTERVAL 20 FEET
DASHED LINES REPRESENT HALF-INTERVAL CONTOURS
DATUM IS MEAN SEA LEVEL
THIS MAP COMPLIES WITH NATIONAL MAP ACCURACY STANDARDS
FOR SALE BY U.S. GEOLOGICAL SURVEY. WASHINGTON. D.C. 20242.
TENNESSEE DIVISION OF GEOLOGY, NASHVILLE. TENN. 37219. ~ -.^
U.S. TENNESSEE VALLEY AUTHORITY. CHATTANOOGA. TENN. 37401 OR KNOXVILLE. TENN. 3/902
A FOLDER DESCRIBING TOPOGRAPHIC MAPS AND SYMBOLS IS AVAILABLE ON REQUEST "T-
-------
FIGURE 6. Locations of Well Stations Relative to the Spill Site
6 Well Location
and Depth
Scale:_ 1 inch s o.l mile
STA4O o
165'
-------
During the 1973 excavation and cleanup of the spill, a
control well was sampled on numerous occasions. This well
was included in the present study for comparative purposes
(see Figure 7).
The time of sampling of these water supplies was predetermined
in the protocol—after dry weather and after heavy rainfall.
This aspect of the study will be discussed under "climatology."
5. Climatology. Since the protocol for water sampling was
established on the basis of single grab samples to be taken
immediately after dry weather and again after heavy rains,
climatological data were obtained for a 2-1/2 year period
in order to substantiate monthly, as well as, seasonal patterns
of precipitation for the study area (see Table 6).
\
Eastern Tennessee receives its greatest rainfall during the
winter and early spring. This is due to the more frequent
passage of large-scale storms over and near the state during
these months. A secondary maximum of precipitation occurs
.in mid-summer in response to shower and thunderstorm activity.
This activity is especially pronounced in Eastern Tennessee
where July rainfall frequently exceeds the precipitation of
any other month. Normally, the lightest precipitation is
observed in the fall and is brought on by the maximum occur-
rence of slow-moving, rain suppressing high pressure areas.
Although all parts of the state are generally well supplied
-------
FIGURE 7. Location of the Control Well
'^-/fLGKTDO
UNITED STATES
DEPARTMENT OF THE INTERIOR
GEOLOGICAL SURVEY
STATE OF TENNESSEE
Scale 1:500,000
1 inch equals approximately 8 miles
O Well Location
and Depth
-------
51
Table 6. Climatological Data Showing Monthly Averages
for the Study Area (D (2)
Total Precipitation in Inches
Month
January
February
March
April
May
June
July
August
September
October
November
December
(1) Climatological Data, U. S. Dept. Commerce, Annual Summary 1973, Vol. 78,
No. 13, p. 2.
Ibid., Annual Summary 1974, Vol. 79, No. 13, p. 2.
(2) Climatological Data, U. S. Dept. Commerce, 1975, Vol. 80, Nos. 1-9.
1973
4.51
3.30
12.44
4.55
9.82
7.33
5.81
3.58
4.32
3.12
9.74
8.38
1974
10.00
5.41
6.97
3.54
7.36
2.72
1.64
5.43
3.10
1.59
4.26
7.04
1975
5.93
5.90
13.19
2.45
6.43
4.47
3.63
2.00
5.17
5.23
3.87
4.65
-------
52
with precipitation; there occurs, on the' average, one or more
prolonged dry spells each year during the summer and fall.
Based on the foregoing information, a decision was made to
obtain local daily rainfall information beginning with the
inception of the project. It was apparent that over-all dry
weather had prevailed from June 21 through August 27, 1975
(see Table 7). There had been minimal>measurable rain
\
(total =5.5 in.) with three days of trace precipitation
over the 68-day period; therefore, water samples were
collected on August 28 and 29. During the next 29-day period,
from August 29 through September 25, appreciable rain fell
at the study site (total = 6.42 in.) with traces on two
other days. In order to take advantage of the wet weather,
the water sites were resampled on September 26.
6. Soil Temperatures. Seasonal temperature variations in the
shallow layers of earth overburden are affected primarily
by incoming solar radiation and outgoing terrestrial radia-
tion. Normally, in the top two feet of surface materials
there is a definite diurnal gradient which varies among
various types of soils. However, daily fluctuations in
soil temperature during the summer months lags considerably
behind affective atmospheric temperature variations.
Differences in texture, structure, and organic matter tend
to determine the moisture capacity of soils and also in-
fluence their ability to absorb and transmit heat. Rain,
-------
53
Table 7. Rainfall Data Prior to and During Well Sampling
Date
(1975)
June 21
26
27
28
July 3
6
9
10
13
16
20
21
25
26
30
31
August 3
5
6
10
11
16
19
24
29
September 7
13
17
18
21
22
23
24
Precipitation
(inches)
0.02
0.19 I
0.45
0.01
0.05
0.24
0.41
Trace
0.05
Trace
0.04
0.78
0.02
0.10
0.78
0.78
0.10
0.04
0.56
Trace
0.37
0.20
0.12
0.20
0.46
0.30
0.09
Trace
1.93
0.28
Trace
0.98
2.38
-------
54
although usually at a lower temperature than the soil,
generally leads to an increase in the temperature of shallow
layers; however,, if it falls in sufficient quantity, it can
serve to increase the rate of conduction of heat from the .
deeper layers to the colder surface. This phenomenon can
more than compensate for the initial fall of temperature
due to rain.
Variations of soil types in the study area, as observed by
core drillings, were minimal (see Table 5). Secondly, the
over-all effect of rainfall on the area was judged to be
insignificant, since climatolbgical data indicated little
rainfall prior to soil temperature measurements. On
August 6, 1975, three soil types were measured for tempera-
ture gradients over four-inch intervals to a depth of fifty-
six inches (see Figure 8). The ambient temperature for a
24-hour period preceding the measurements ranged from a
high of 86.8°F to a low of 70°F. During the sampling period,
the ambient temperature moderated to a high of 82°F with a
low of 73°F.
All determinations indicated that the soil temperature was
highest within the first four inches and followed a general
downward trend over the remaining fifty-two inches. However,
slight gradients were observed at varying depths in the
three types of soil sampled. These deviations were probably
due to moisture differentials caused by the non-uniform
clay-chert matrix within a given sampling site.
-------
86
Figure 8. Temperatures of Various Soil Types
Over Four-Inch Intervals as Compared
with Ambient Conditions (see Table
5 for Soil Type Codes)
84
82
80
78
76
En
O
74
Soil Type 2
w
u w
e a,
E)
H o
w -;
W
72
Y
70
68
66
00
i
oo
O
N
(£>
0>J
O
00
N
I
CO
I
00
(O
n
i
N
n
DEPTH (inches)
o
(0
en
I
o
00
«f
I
10
I
00
to
CM
-------
56
7. Drilling Equipment and Procedures. The drilling equipment
used in the field was quite adequate and versatile in that
it provided for the use of an auger and three types of core
samplers, the weight of the truck became critical at one
stage after a rain when the sites which had not been cored
were on a steep slope. A smaller drill truck was brought in,
and provided relief from the problem.
The split spoon sampler was an excellent compromise in
that it was efficient, produced a core sample of sound in-
tegrity, worked well in the clay and chert matrix, and
i
cleaned with minimal difficulty.
B. Microbiological Data.
The soil samples, which were various types of clay and chert,
were found to have relatively low bacterial, fungal, and actino-
mycete populations. The data for each sample is shown in
Table 8. It may be noted that the bacterial population far
exceeded populations within the other two groups. Plate counts
for these organisms determined in fertile agricultural soil have
been shown to be much higher. Average bacterial counts may
exceed 15 million per gram of soil, while smaller fungal and
actinomycete populations may average 400 thousand and 700
thousand, respectively (29). Only one of the 142 samples was
shown to have a bacterial count as high as three million per
(29) Surges, A. 1958. Microorganisms in the Soil. Hutchinson and Co.
Publishers, London.
-------
57
gram, and only four others exceeded one million per gram of
sample. Fungal and actinomycete counts in the samples examined
were correspondingly lower than those usually found in fertile
soils. Additionally, the two control soil samples, taken from
i
areas remote to the spill, were surprisingly low for all three
types of organisms. v . .
Eleven different types of clay and chert were identified, with
the majority of microorganisms occurring in types designated
as 1 and 2 (see Table 9).
It may be seen that the average bacterial, fungal, and actino-
mycete counts for 81 samples in soil type 1 were lower than the
average for all 142 samples, while populations of the three
groups in soil type 2 were higher than the over-all average.
Bacterial and fungal counts for soil type 2 were approximately
two times as great, and actinomycete counts were greater than
four times the mean for all samples. The remaining 34 samples
were distributed among 8 soil types and constituted too small a
group in each instance to provide statistically valid data.
Analysis of microbial populations relative to pH of the soil
samples is summarized in Table 10. From these data it is
obvious that conventional distribution of the three groups of
organisms was observed. Bacterial and actinomycete counts were
highest in samples having a pH of 5:0 to 5.9 and lowest in a
pH range of 4.0-4.5. Conversely, the number of fungi were
highest in the latter range and lowest at a pH of 5.0 or above.
-------
TABLE a COLLECTIOB AND MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKES iN AUGUST, 1975.
Microorganism Counts/Gram of Soil
er . X Holatur*
Hucber (Inches) W of Soil
Field Sample Sazple Death Character
GF 7316
OF 7520
CF 7524
GF 7S28
CF 7517 23-27
CF 7521
CF 7525
CF 7529
C? 7518 27-31
CF 7522
CF 7526
CF 7530
CF /519 31-35
CF 7523
CF 7527
GF 7531
GF 7618. 17-33
CF 7532 21-25
CF 7536 25-29
CF 7540 29-33
CF 7544 33-37
JES-
4.6
4.4
4.7
4.4
4.5
4.5
4.4
4.5
4.0
4.4
4.4
4.6
4.1
4.8
4.7
4.6
4.7
4.6
Z Mol«tur«
by Weight
7.30
6.99.
4.65
5.47
4.28
5.65
2.54
3.35
10.0
4.38
10.3
5.58
15.6
10.8
10.6
8.65
9.07
9.00
Bacteria
2.310,000
34,500
< 10,000
- ' 25,000
< 10,000
•'^ < 10,000
< 10,000
25,000
< 10,000
< 10.000
50.000
< 10,000
< 10.000
78.000
570.000
306.000
191.000
354.000
Actlnomycetes
2.500
1.500
< 1.000
< 1.000
500 .,
< 1.000
3.000
5,000
< 1.000
< 1,000
1,000
< 1,000
< 1.000
500
700
600
350
< 100
Fungi
25,000
3,500
50,000
400
< 1.000
5,000
7.200
32.GCO.OOO
< 1.000
5,000
< 1,000
< 1,000
< l.COO
li9CO
6.500
1,550
700
5.800
in
00
-------
TABLE 8. COLLECTION* AND MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKEN IN AUGUST, 1975 (con't)
Field Sai.?le
Number
CF 7i.»3
CF 7537
CF 7541
CF 7545
GF 7534
CF 7538
CF 7542
GF 7546
CF 7535
CF 7539
CF 7543
GP 7547
CF 7548
CF 7556
CF 7564
CF 7572
C? 7552
CF 7560
GF 7568
CF 7576
CF 7549
CF 7557
CF 7565
CF 7573
CF 7553
CF 7561
CF 7569
CF 7577
Sample Depth
(inches) U)
25-29
29-33
33-37
19-23
23-27
Character
of Soil <2>
31-JS
4.4
4.8
5.2
4.5
4.6
4.4
4.5
4.2
5.0
4.4
4.3
4.2
4.4
4.4
4.2
4.3
4.3
4.0
4.1
4.8
4.4
4.9
4.4
4.1
4.7
4.3
4.3
4.5
Z Moisture
by Weight
4.45
3^30
9.29
6.60
6.85
6.21
3.92
7.20
6.29
12.7
10.5
2.84
15.0
18.2
14.1
12.8
22.8
19.3
21.5
20.8
19.5
19.1
22.7
17.8
20.9
19.6
22.1
22.7
Microorganism Counts/Cram of Soil
Bacteria
20,500
2,000,000
1,270,000
800,000
6C.OCO
945,000
195,000
650,000
645,000
< 10.000
< 10,000
< 10,000
37,500
83,000
79,000
T., 000
8.000
3i.OOO
2.000
< 1.000
43.500
38.000
51.000
37.000
1,000
14.500
< 1,000
11.000
Actinomvcetes
6,000
43.800
230,000
5,500
< 1,000
7,500
2,500
3.000
l.OCO
1,000
< 1,000
< 1.000
600
300
. 850
1,400
100
650
300
< 100
< 1.000
450
450
300
350
< 100
< 100
400
Fungi
< l.COO
8,000
12,000
15,000
< l.OCO
3,500
< l.OCO
11.000
< 1,000
< 1,000
< 1.003
3,000
600
120
10. COO
3.500
250
200
1.000
1.050
1.000
50
100.000
/ 1,000
< 100
100
< 100
1.250
Ul
VO
-------
TABLE 8. COLLECTION AND MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKEN IN AUGUST, 1975 (con't)
Field Staple
Xuaber
CF 7550
•CF 7553
CF 7566
CF 7574
GF 7554
C- 7562
CF 7570
CF 7578
CF 7551
CF 7559
CF 7567
CF 7575
GF 7555
CF 7563
CF 7571
CF 7579
CF 7580
CF 7581
CF 7582
CF 7533
CF 75S8
CF 7589
CF 75SO
CF 7591
CF 7592
CF 7593
CF 7594
CF 7595
Saople Depth
(inches) U)
35-39
39-43
43-47
47-51
15-19
19-23
23-27
27-31
9-13
13-17
17_-i
23-kS
25-29
29-33
33-37
37-41
Character
of Soil »)
pH
4.2
4.1
4.4
4.3
4.3
4.6
4.6
4.8
4.3
4.3
4.0
4.6
4.7
4.6
4.6
4.7
4.5
4.6
4.3
4.7
4.4
4.8
5.1
4.8
4.7
4.3
4.2
4.3
Z Moisture
by Weight
20.4
19.3
20.9
18.2
20.4
21.7
22.8
20.5
19.7
19.7
20.8
21.0
20.8
21.7
24.2
22.8
7.14
14.3
16.0
15.0
17.2
17.2
18.7
19.3
13.3
15.0
18.3
19.2
Microorganism Counts/Cram of Soil
Bacteria
28,000
26,000
18,000
20,500
28,000
500
< 1.000
2,500
23,500
< 1.000
3.500
8.500
< 1.000
500
< 1,000
f 1,000
139,500
89.000
24.500
19,000
97.000
197,500
309,500
192.500
161,000
276,500
43.000
70.000
Actinoayceces
800
< 100
< 100
300
500
50 -_
200
< 100
900
< 100
< 100
< 100
150
< 130
< ioc
< 100
850
5.000
600
< 100
600
Ii550
450
250
ISO
200
< 100
100
Fung:
SCO
200
< 100
1.000
5, SCO
500
300
600
200
< 100
5, COO
650
< 100
< 100
< 100
< 100
2,5iO
180
1,900
105
2,300
2,550
3,150
2,600
10,000
. 10.000
1.200
1,200
ON
O
-------
TABU 8. COLLECTIOB AND MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKES IN AUGUST, 1975 (con't)
Field Saaple
Xuaber
CF 7596
CF 7597
GF 7593
GF 7599
CF 7600
CF 7601
GF 7602
GF 7603
CF 7604
CF 76C5
CF 7606
CF 7637
.Gr 7608
CF 7539
CF 7610
CF 7611
C? 7612
CF 7613
GF 7614
GF 7615
CF 7616
CF 7617
C? 7619
CF 7620
Simple Depth
(Inches) U>
9-13
33-17
x/-21
21-25
20-24
24-28
28-32
32-36
33-37
37-41
41-45
45-49 .
49-53
53-57
57-61
61-65
9-25
25-41
9-25
21-:,
12-28
9-25
9-25
9-25
Character
of Soil (2)
5
5
1
1
6
6
6
6
7
7
7
1
1
1
1
1
1
1
1
1
1
8
4 '
10
pH
4.6
4.6
4.7
4.8
4.5
4.4
4.8
4.6
4.7
4.9
4.9
4.8
4.8
4.8
5.0
4.9
4.9
4.6
4.7
4.2
4.9
4.6
5.2
5.2
Z Moisture
by Height
6.89
6.71
14.6
16.5
17.9'
19.4
19.7
21.9
13.6
14.5
13.7
14.7 .
17.1
18.4
17.2
17.9
19.0
19.3
18.8
22.8
14.1
11.2
14.6
13.9
Microorganism Counts/Cram of Soil
Bacteria
112,000
113,000
29,500
35,500
80,500
81,000
40,500
7,500
40,000
25,500
29,000
2,500
1,500
< 1,000
159.000
35,000
320,500
12,500
4,000
6,000
126,500
219,000
262,500
219,500
Actlnoaycetes
550
40C
2,500
< 100
250
150
< 100
< 100
100
100
200
< 100
< 100
50
900
400
200
100
< 100
< 100 .
250
< 100
3CO
50
Tumi
7,403
4,850
950
750
. 2.950
1,700
2,250
350
1,900
2,250
900
100
600
250
3,700
500
7,700
500
5,400
350
4.030
9,000
1,400
600
-------
TABLE 8. COLLECTION ASD MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKEN IN AUGUST. 1*75 (con't)
Microorganism Counts/Gran of Soil
Field Sample
Number
GF 7621
CF 7622
CF 7623
CF 7624
CF 7625
CF 7626
GF 7627
CF 7630
CF 7631
CF 7632
CF 7633 .
CF 7634
CF 7635
CF 7656
GF 7637
C? 76id
GF 7639
CF 7641
CF 7640
Sample Depth
(Inches) I"
9-25
'25
25-41
41-57
57-73
19-35
3-19
9-25
25-41
: 9-25
25-41
3-19
4-20
9-25
25-41
3-19
9-25
25-41
3-19
Character
of Soil <2)
5
5
5
5
8
5
1
5
i
i
i
i
i
i
i
i
i
i
_ES_
5.1
5.6
5.5
5.4
4.9
5.5
5.2
5.3
5.0
5.0
5.2
4.8
4.6
5.0
4.8
_ ; 5.2
5.2
5.3
5.2
Z Moisture
by Ueigbt
13.8
16.0
13.9
14.8
23.4
15.3
24.7
9.00
7.52
9:00.
8.40
25.0
23.5
12.6
22.1
21.2 -
22.1
23.6
22.8
Bacteria
215.000
334.000
136.000
46.000
2.500
367.000
/ 4.500
112.500
157.500
218.500
155.000
100.000
113.500
82.500
29.500
160,000
60,000
74.500
48.000
Actinomycete*
250
850
250
200
< 100
400
< 100
750
SO
650
500
< 100
2CO
450
250
100
250
200
< 100
Fungi
1.800
350
1.400
350
SO
13.000
100
2.650
600
1.900
1.400
200
1.05C
950
1.100
750
1.500
1,100
5.000
N>
-------
TABLE 8. COLLEGIUM AND H1CEOBIOLOCICAL DATA FROM SOIL SAMPLES TAKEN IS AUGUST. 1975 (eon't)
Microorganism Countg/Craa of Soil
Field Sample
Kuaber
CF 7642
CF 7643
CF 7644
. CF 7645
CF 7646
GF 7647
. CF 7648
CF 7649 '
CF 7650
CF 7651
GF 7652
CF 7653
CF 7654
CF 7655
CF 7656
CF 7657
CF 7658
CF 7659
GF 7660
Sample Depth
(Inches) W
9-25
25-41
2-9 .
9-25
9-25
25-41
9-25
25-41
. 9-25
0-16
29-45
21-37
25-37
33-45
9-2-
3-19
10-26
26-42
42-58
Character
of Soil <2)
1
1
1
• 1
1
1
1
2
1
2
2
1
1
1
1
1
1
1
-ESL
5.0
5.1
5.3
5.3
5.1
5.8*
5.7
5.8
5.9
4.6
5.5
5.8
4.9
5.0
4.6
4.7
4.8
4.5
4.2
X Moisture
by Weight
19.8
18.8
21.4
18.5
20.0
23.3
16.6
15. 5
13.3
23.1
7.95
12.9
10.6
11.3
21.0
12.2
11.5
13.1
20.4
Bacteria
315.000
96.500
204.000
56,500
57,500
11.000
1,480,000
170.000
665,000
503
. 65.000
4.503
17.500
138.500
4.500
291.003
3.030.000
165.500
1.400
Actinonycetes
400
1.050
200
250
700
100
500
350
20.000
< 100
150
< 100
200
400
< 100
250
5.500
100
< 100
Fungi
700
4.000
1.350
1.200
1.650
800
2.700
900
3,300
100
1.400
200
620
850
< 100
700
800
250
50
-------
TABU 8. COLLECTION AND MICROBIOLOGICAL DATA FROM SOIL SAMPLES TAKEN IN AUGUST, 1975 (con't)
Microorganism Counts/Crag of Soil
Field Sample
Number
CF 7661
CF 7662
CF 7663
CF 7664
GF 7665
CF 7666
GF 7683
(control A)
GF 7684
(control B)
Sample Depth
(inches) U)
9-25
25-41
9-25
25-41
3-19
9-25
3-6
3-6
Character
of Soil (2)
10
10
5
5
5
5
1
11
Overall
_Efi_
5.2
5.0
5.0
5.0
5.0
5.3
5.0
6.2
Averages * '
Z Moisture
by Weight
15.6
18.9
14.1
12.8
18.9
16.5
8.40
3.60
Hif;h
Low
Bacteria
325,000
26,500
620,000
' 59,500
164,000
1.540,000
186,000
340,000
181.654'
180,823
Actinonycetes
800
100
4,000
100
200
450
4,000
11,000
2.860.
2,760
Fungi
2,600
< 100
1,700
350
550
14.500
1,500
1,500
3,343
3,259
Pleasured from the original 1973 surface prior to backfill.
See Table 5 for soil code descriptions. _
The high averages are calculated with less than values taken to be positive (<.l taken as .1). The low averages are
calculated with the less than value taken to be zero. The actual mean of the sample oust lie between the high average
and the low average. . '
-------
65
Table 9. Microbial Populations Relative to Soil Types
Average Microbiological Counts
Soil Type
1
2
3
4
5
6
7
8
9
10
11
(1)
Criterion (2)
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
(Per
Bacteria
' 136628
136530
364796
360722
84333
84333.
138667
138667
207182
207182
52375
52375
31500
31500
110750
110750
78000
78000
190333
190333
340000
340000
Gram of Soil)
Actinomycete
469
427 .
12557
12183
2150
2150
158
125
723
723
. 150
100
133
133
100
0
500
500
317
317
11000
11000
Fungi
2707
2696
6093
5685
1543
1543
3984
3984
3127
3127
1813
1813
1683
1683
4525
4525
1900
1900
1100
1067
1500
1500
Number of
Samples/Sell
Type
81
27
3
6
11
4
3
2
1
3
1
(1) See Table 5 for soil codes.
(2) The high averages are calculated with less than values taken to be positive
(<.l taken as .1). The low averages are calculated with the less than
values taken to be zero. The actual mean of the samples must lie between
the high average and the low average.
-------
66
TABLE 10. Microbial Populations Relative
to pH of Soil Samples
Average Microbiological Counts
Criterion^1) (Per Gram of Soil) Number
pH for •. . of
Ranges Average Bacteria Actinomycete Fungi Samples
4.0-4.5 High . 90324 1043 , 4444 49
Low 88447 861 4275
4.6-5.0 High 215644 1299 2901 66
Low 215250 1225 2845
5.0 Up High 264315 . 9970 2426 27
Low 264315 9959 2426
(1) The high averages are calculated with less than values taken to be
positive (<.l taken as .1). The low averages are calculated with
the less than values taken to be zero. The actual mean of the samples
must lie between the high average and the low average.
-------
67
It is well recognized that fungi thrive best in a relatively
acid environment while bacteria and actinomycetes, in general,
prefer ~a more neutral pH range for survival and growth.
Microbial counts were compared in relation to moisture content
of the soil samples. Determinations were made in the ranges
of 0-5, 5-10, 10-15, 15-20, 20-25, and 25-30 percent moisture
by weight. From Table 11 it can be observed that the highest
bacterial and actinomycete concentrations were present in
samples having 5-10 percent moisture. Fungal counts were the
greatest in samples having the least moisture. A single sample
,\ |
having a 25-30 percent moisture was insufficient to provide
valid data at that level. In fertile soils, rich in nutrients,
it might be expected that extremely high numbers of organisms
would be detected when moisture levels were high since water
is required to solubilize the nutrients, making them more readily
available to the soil populations. However, in clay and chert
where nutrients are extremely deficient, and where the pH range
is predominately between 4.0 and 5.0, soluble acid minerals
/
contributing to the low pH may function bacteriostatically or
bactericidally to maintain relatively low levels of organisms.
Very significant decreases in bacterial counts were observed
when moisture content of the samples exceeded 20 percent, and
for actinomycetes when moisture was above 10 percent. Fungal
populations decreased steadily as moisture content increased
to 20 percent. Above this level a slight increase in fungi
was noted between 20-25 percent moisture.
-------
68
Table 11. Microbial Populations Relative
to Moisture Content of Soils
Moisture
(% by Weight)
0%- 5%
5%-10%
10%-15%
15%-20%
20%-25%
25%-30%
»
Criterion'1' Average Microbiological Counts
for (Per Gram of Soil)
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
Bacteria
289500
289500
385630
385630
236156
236156
135940
135940
27240
27240
100000
100000
Actinomycete
7756
7756
11476
11476
1503
1503
407
407
200
200
100
100
Fungi
8633
8633
4813
4813
2202
2202
1976
1976
3791
3791
200
200
Number
of
Samples
9
23
32
42
35
1
(1) The high averages are calculated with less than values taken to be positive
. (<.l taken as .1). The low averages are calculated with the less than
values taken to be zero. The actual mean of the samples must lie between
the high average and the low average. ' .'
-------
69
It has been reported (30, 31, 32) that microbial counts in
fertile agricultural soil are the highest in the first 6-8 inches
below the surface, decreasing rapidly below that point to only a
few hundred per gram at depths of 6 feet or more. Although
these data were obtained from samples deficient in nutrients at
any depth, a similar trend could be observed. The major differ-
ence was revealed in Table 12 by bacterial counts which averaged
157,330 per gram between 0 and 8 inches and increased to 271,700
per gram between 8 and 29 inches. Below 29 inches the bacterial
population was reduced progressively to a total of 2500 per gram
in a single sample taken at a depth below 66 Inches. Actino-
mycete counts were highest between 0 and 2 inches and at a depth
of 24-29 inches, subsequently decreasing to 100-700 per gram at
all depths below 29 inches. Fungal counts were highest between
8 and 27 inches, and 'these also decreased progressively below
that level except for a very slight increase in numbers in 13
samples taken at a depth between 56 and 61 inches.
The fact that bacterial counts near the surface of the soil were
slightly lower than at depths of 8-29 inches may possibly be
attributed to the seasonal effect'of high summer temperatures
which would tend to kill some of the organisms by baking of the
soil to shallow depths.
(30) Waksman, S. A., and Starkey, R. L. 1931. Soil and the Microbe.
Wiley and Sons, Inc., Publishers, New York.
(31) Kuster, E. and Williams S. T. 1964. Selections of Media for Isolation
of Streptomycetes. Nature, London, 202:928-929.
(32) Frobisher, M. 1962. Fundamentals of Microbiology (p. 652), W. B.
Saunders Co. Publishers, Philadelphia, Pennsylvania.
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths
70
Average Microbiological Counts
Depth below
Surface
(inches)
0
(
i
2
3
4
5
6
7
8
9
10
11
12
13
Criterion u;
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
(Per Gram of Soil)
Bacteria
175500
175500
175500
175500
182625
182625
149800
149800
146500
146500
146500
146500
146500
. 146500
146500
146500
146500
146500
211554
211554-
287932
287932
289357
289357
284833
284833
256058
256058
Actinomycete
5033
5000
5033
5000
3825
3800
1615
1575
1486
1450
1486
1450
1486
1450
1486
1450
1486
1450
1392
1373
1535
1516
1194
1174
1168
1149
1079
1040
Fungi
1033
1033
1033
1033
1113
1113
1175
1175
1164
1164
1164
1164
1164
1164
1164
1164
1164
1164
2538
2536
2524
2521
2582
2579
2621
2619 v
4973
4970
Number of
Samples/ Dept
Category ,
3
3
4 ;
10
11
11
11
11
11 .
37
37
35
36
43
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths (Continued)
71
Depth below
Surface
(inches)
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Criterion*1)
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
**V W*. b»^^_ **rib1
(Per
Bacteria
287944
287944
283932
283932
283932
283932
251307
251307
278405
278405
319549
319549
350714
350714
297307
297284
340730.
340703
309183
308671
264718
264179
266586
266207
285828
285172
176157
174700
191766
190797
Iv^VhS^V^VE^A^tl.*^. W
Gram of Soil)
Actinomycete
1165
1118
1157
1111
1157
1111
1081
1038
1168
1143
1283
1261
1461
1444
1216
1189
1328
1304
1274
1204
1113
1038
5743
5691
9380
9308
' 8564
8439
9273
9195
>«*• w^
Fungi
4997
4994
4931
4928
4931
4928
4349
4344
1943
1938
2770
2765
3048
3042
2800
2793
3137
3132
4131
4102
3651
3621
3301
3262
3985
3919
4001
3912
2780
2715
N'-r-.ber of
Samples/Depth
Category
36
37
37
44
37
41
35
44
37
41
39
58
32
35
32
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths (Continued)
72
Average Microbiological Counts
Depth below
Surface
(inches)
29
30
- 31
32
33
34
35
36
37 .
38
39
40
41
Criterion ^
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
(Per Gram of Soil)
Bacteria
201563
200763
116774
115774
105714
103971
114328
113359
116919
115477
87303
85455
87303
85455
86929
85821
76734
75672
69816
69658
69816
69658
69816
69658
66738
66595
Actinomycete
7795
7703
798
682
821
633
742 \
598
685
529
414
247
414
247
330
241
303
219
224
208
224
208
224
208
221
207
Fungi
2586
2478
2044
1938
2039
1859
1826
1660
1826
1633
1754
1569
1754
1569
1317
1206
1224
1115
916
889
916
889
916
889
888
864
Number of
Samples/Dept*
.Category
40
31
35
32
43
33
33
28
32
19
19
19
21
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths (Continued)
73
Average Microbiological Counts
Depth below
Surface
(inches)
.42
A3
44
45
46
47
48
49
50
51
52
53
54
55
Criterion^1'
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
(Pe
Bacteria
87914
87914
74983
74983
74983
74983
64629
64629
54975
54975
54975
" 54975
54975
54975
44280
44280
16300
16300
16300.
16300
16300
16300
12475
12225
16133
15800
16133
15800
r Gram of Soil)
Actinomycete
214
200
233
217
233
217
214
186
188
138
188
138
188
138
170
110
133
67
133
67
133
67
113
63
117
83
117
83
Fungi
671
671
742
742
742
742
650
650
350
350
350
350
350
350
400
400
333
333
333
333
333
333
313
313
217
217
• 217
217
Number of
Samples/Dept
Category
7 . .
6 j
6
7
4
4
4
5
3
3
3
4
3
3
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths (Continued)
74
Average Microbiological Counts
Depth below
Surface .
(inches)
56
57
58
59
60
61
62
63
64
65
66
67
68
Criterion^
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
High
Low
l) (Per Gram of Soil)
Bacteria
16133
15800
41980
41780
54300
54300
80750
80750
80750
80750
65500
65500
18750
18750
18750
18750
18750
18750
18750 .
18750
2500
2500
2500
2500
2500
2500
Actinomycete
117
83
270
230
367
. 300
500
450
500
450
467
433
250 .
200
250
200
250
200
250
200
100
0
100
0
100
0
Fungi
217
217
880
880
1267
1267
1875
1875
1875
1875
1417
1417
275
275
275
275
275
275
275
275
50
50
50
50
v 50
50
Number of
Samples/Deptl
Category
3
5
3
2
2
3
2
2
2
2
1
1
1
-------
TABLE 12. Microbial Populations Relative
to Core Sample Depths (Continued)
75
Depth below
Surface
(inches)
69
70
71
72
73
Criterion^1)
for
Average
High
Low
High
Low
High.
Low
High
Low
High
Low
Average Microbiological Counts
(Per Gram of Soil)
Number of
Samples/Depth
Bacteria
2500
2500
2500
2500
2500
, 2500
2500
2500
2500
2500
Actinomycete
100
0
100
0
100
0
100
0
100
0
Fungi
50
50
50
50
50
50
50
50
50
50
Category
1
1
1
1
1
(1) The high averages are calculated with less than values taken to be
positive (<.1 taken as .1). The low averages are calculated with the
less than values taken to be zero. The actual mean of the samples must
lie between the high average and the low average.
-------
76
Analysis of microbial populations exposed to polychlorinated
biphenyl (PCB) in the soil suggested a possible stimulating
effect of this compound in concentrations between 0^05 and 5.0
mg/kg on both the bacteria and actinomycetes, as shown'in Table
». • .
13. The lowest bacterial counts were obtained in samples where
PCB levels were less than 0.05 mg/kg. Although only three samples
were contaminated with PCB in excess of 10 mg/kg, the bacterial
counts were greater than in relatively uncontaminated soil having
less than 0.05 mg/kg. Fungal counts were essentially unaffected
in all except the one sample having the highest concentration of
PCB.
In summary, the relatively low microbial counts are compatible
with the clay and chert soils which are largely devoid of nu-
trients. The soil acidity is also reflected by the low bacterial
counts. Excessive moisture did not appear to influence an in-
crease in the microbial populations of the soil; and, in fact,
the lowest bacterial and actinomycete counts were observed in
samples having a moisture content of 20-25 percent while the
highest numbers were found in samples having 0.5-10.0 percent
moisture.
The most interesting observation was the correlation between
microorganism counts and the soil samples containing
\. .
0.05-5.0 mg/kg of the polychlorinated biphenyl compounds.
Bacterial counts in 55 samples were shown to be 50 percent
higher than the mean for all samples, and actinomycete counts
were similarly 100 percent greater.
-------
77
TABLE 13. Microbial Populations Relative to
PCB Concentration in Core Samples
PCB Concentration
(mg/kg soil)
<.05
.05-5.0
5.0-10.0
10.0-30.0
30.0-66.6
Criterion*1)
for
Average
High
Low
High
Low
High
Low
High
Low
High
Low
Average Microbiological Counts
(Per Grain of Soil)
Bacteria
82825
81475
273535
273353
920750
920750
140750
140750
160000
160000
Actinomycete
Fungi
1066
936
5708
5643
1500
1475
375
375
100
100
3378
3242
3269
3249
4638
4638
2700
2700
750
750
Number
of
Samples
80
55
4
2
1
(1). The high averages are calculated with less than values taken to be positive
(<.l taken as .1). The low averages are calculated with the less than
values taken to be zero. The actual mean of the samples must lie between
the high average and the low average.
-------
78
C. Analytical Data From Electron-Capture Gas Chromatography (EC/GC).
During Phase Two of this project, a total of 145 core, three sedi-
• / ,
ment, and two control soil samples were analyzed quantitatively
for Aroclor 1254. Analytical and collection data for soils are
contained in three tables. Data for the three test core sites
are in Table 14, data for the five core sites sampled at four-
,inch intervals are in Table 15, and the remainder of the core
samples are in Table 16. The nineteen water.samples collected
for this project were analyzed for both Aroclor 1254 and the
askarel solvent. Results for these analyses plus those for
sediment and control soils are given in Table 17. No analysis
difficulties were encountered during the course of this project.
D. Special Soil Extraction Experiment.
In 1973, data relating to the concentration of PCBs in soil and
sediment were obtained using hexane as the solvent for the ex-
traction. Two other extraction systems were recommended for
consideration by the project officer for Contract No. 68-01-3232.
The first employs a dual hexane/acetone system (33), and the
second extracts with 15% methylene chloride in hexane (V/V) (34).
Eight of the larger samples were selected for comparative analysis
using the three extraction systems. In two instances, there was
(33) Crump-Wiesner, H. J., Feltz, H. R., and Yates, M. D., 1973, A Study
of the distribution of polychlorinated biphenyls in the aquatic
environment: Jour. Research U. S. Geol. Survey, v. 1, no. 5, p. 603-607.
(34) National Pollutant Discharge Elimination System, Appendix A, Fed. Reg.,
38, No. 75j Pti II (11-28-73).
-------
79
TABLE 14. Collection and Analytical Data From Three Test Core
Sites Sampled Over Four-Inch Intervals in August 1975
Field Sample
Core Site Sample Depth2 Character3 PCB Concentration
Identification1 Number (Inches) of Soil (mg/kg)
RL-1 GF7516 19-23 2 1.51 .
RL-2 GF7520 0.95
RL-3 GF7524 ' 1.52
RL-4 GF7528 14.5
RL-1 GF7517 23-27 2 <0.05
RL-2 GF7521 <0.05
RL-3 GF7525 <0.05
RL-4 GF7529 10.9
RL-1 GF7518 27-31 2 <0.05
RL-2 GF7522 <0.05
RL-3 GF7526 <0.05
RL-4 GF7530 0.18
RL-1 GF7519 31-35 2 <0.05
RL-2 GF7523 <0.05
RL-3 GF7527 <0.05
RL-4 , GF7531 <0.05
RL-5 GF7618 17-33 2 0.16
1See Exhibit II for general location
2Measured from the original 1973 surface prior to backfill
3See Table 5 for soil code descriptions
-------
80
TABLE 14. Collection and Analytical Data From Three Test Core
Sites Sampled Over Four-Inch Intervals in August 1975
(Continued)
Core Site
Identification1
UC-1
UC-2
UC-3
UC-4
UC-1
UC-2
UC-3
UC-4
Field Sample
Sample Depth2
Number (Inches)
GF7532
GF7536
GF7540
GF7544
GF7533
GF7537
GF7541
GF7545
Character3
of Soil
21-25
25-29
PCB Concentration
(mg/kg)
0.09
<0.05
<0.05
0.16
<0.05
0.70
0.16
2.24
UC-1
UC-2
UC-3
UC-4
UC-1
UC-2
UC-3
UC-4
GF7534
GF7538
GF7542
GF7546
GF7535
GF7539
GF7543
GF7547
29-33
33-37
<0.05
0.52
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
1 See Exhibit n for general location
2 Measured from the original 1973 surface prior to backfill
See Table 5 for soil code descriptions
-------
81
TABLE 14. Collection and Analytical Data From Three Test Core
Sites Sampled Over Four-Inch Intervals in August 1975
(Continued)
Core Site
Identification1
JF-1
JF-2
JE-3
JF-4
JF-5 ,
JF-6
JF-7
JF-8
JF-1
JF-2
JF-3
JF-4
JF-5
JF-6
JF-7
JF-8
JF-1
JF-2
JF-3
JF-4
JF-5
JF-6
JF-7
JF-8
JF-1
JF-2
JF-3
JF-4
JF-5
JF-6
JF-7
JF-8
Field
Sample
Number
GF7548
GF7556
GF7564
GF7572
GF7552
GF7560
GF7568
GF7576
GF7549
GF7557
GF7565
GF7573
GF7553
GF7561
GF7569
GF7577
GF7550
GF7558
GF7566
GF7574
GF7554
GF7562
GF7570
GF7578
GF7551
GF7559
GF7567
GF7575
GF7555
GF7563
GF7571
GF7579
Sample
. Depth2 Character3
(Inches) of Soil
19-23 1
23-27 1
V
27-31 1
31-35 1
,
•
35-39 1
39-43 1
43-47 1
47-51 1
i
PCS Concentration
(mg/kg)
0.36
<0.05
0.08
0.17
' <0.05
<0.05
<0.05
<0.05
0.27
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
0.18
<0.05
<0.05
0.25
0.24
<0.05
<0.05
<0.05
<0.05
See Exhibit II for general location
2Measured from the original 1973 surface prior to backfill
3See Table 5 for soil code descriptions
-------
82
TABLE 15. Collection and Analytical Data From Core Sites
Sampled Over Four-Inch Intervals in August 1975
Core Site
Identification1
EK-1
EK-2
EK-3
EK-4
SR-1
SR-2
SR-3
SR-4
SR-5
SR-6
SR-7
SR-8
CC-1
CC-2
CC-3
CC-4
SM-1
SM-2
SM-3
SM-4
FH-1
FM-2
FM-3
FM-4
FM-5
FM-6
FM-7
FM-8
Field
Sample
Number
GW580
GF7581
GF7582
GF7583
GF7888
GF7589
GF7590
GF7591
GF7592
GF7593
GF7594
GF7595
GF7596
GF7597
GF7598
GF7599
GF7600
GF7601
GF7602
GF7603
GF7604
GF7605
GF7606
GF7607
GF7608
GF7609
GF7610
GF7611
Sample
Depth2
(Inches)
15-19
19-23
23-27
27-31
9-13
13-17
17-21
21-25
25-29
29-33
33-37
37-41
9-13
13-17
17-21
21-25
20-24
24-28
28-32
32-36
33-37
37-41
41-45
45-49
49-53
53-57
57-61
61-65
Character3
of Soil
3
3
3
4
1
1
1
1
4
4
4
4
5
5
1
1
6
6
6
6
7
7
7
1
1
1
1
1
PCS Concentration
(rag/kg)
0.13
<0.05
0.27
0.13
0.08
0.24
0.89
<0.05
<0.05
<0.05
<0.05
<0.05
0.17
0.40
<0.05
0.14
<0.05
0.08
1.10
0.11
0.22
*0.05
0.14
0.09
-------
TABLE 16. Collection and Analytical Data from Core Sites
Sampled Over Sixteen-Inch Intervals in August 1975
83
Field- Sample
Core Site Sample Depth2 Character3 PCB Concentration
Identification1 Number (Inches) of Soil
LJ-1-4
LJ-5-8
MB-1-4
KH-1-4
SW-1-4
JR-1-4
GW-1-4
TE-1-4
JD-1-4
MM-1-4
MM- 5-8
MM- 9-12
MM-13-16
GE-1-4
CT-1-4
MT-1-4
MT-5-8
BT-1-4
BT-5-8
LP-1-4
JS-1-4
FA-1-4
FA-5-8
GF7612
GF7613
GF7614
GF7615
GF7616
GF7617
GF7619
GF7620
GF7621
GF7622
GF7623
GF7624
GF7625
GF7626
GF7627
GF7630
GF7631
GF7632
GF7633
GF7634
GF7635
GF7636
GF7637
9-25
25-41
9-25
21-37
12-28
9-25
9-25
9-25
9-25
9-25
25-41
41-57
57-73
19-35
3-19
9-25
25-41
9-25
25-41
3-19
4-20
9-25
25-41
1
1
1
1
1
8
4
10
5
5
5
5
8
5
1
5
1
1
1
1
1
1
1
7.36
<0.05
3.69
<0.05
23.8
5.00
3.23
<0.05
X 0.17
0.27
<0.05
<0.05
<0.05
0.32
0.16
4.35
,0.89
1.34
10.1
3.41
5.62
<0.05
<0.05
Exhibit E for general location
2Measured from the original 1973 surface prior to backfill
3See Table 5 for soil code descriptions
-------
84
TABLE 16., Collection and Analytical Data from
Sampled Over Sixteen-Inch Intervals
(Continued)
Core Site
Identification1
Field Sample
Sample Depth2 Character3
Number (Inches) of Soil
Core Sites
in August 1975
PCB Concentration
(mg/kg)
PT-1-4
FO-1-4
FO-5-8
AA-1-4
xEL-l-4
EL-5-8
PS-1-4
W-l-4
SS-1-4
SS-5-8
HH-1-4
HH-5-8
AM-1-4
VT-1-4
JE-1-4
BA-1-4
TR-1-4
CV-1-4
PQ-1-4
MD-1-4
BS-1-4
BS-5-8
BS-9-12
GF7638
GF7639
GF7641
GF7640
GF7642
GF7643
GF7644
GF7645
GF7646
GF7647
GF7648
GF7649
GF7650
\
GF7651
GF7652
GF7653
GF7654
GF7655
GF7656
GF7657
GF7658
GF7659
GF7660
3-19
9-25
25-41
3-19
9-25
25-41
2-9
9-25
9-25
25-41
9-25
25-41
9-25
0-16
29-45
21-37
25-37
33-45
9-25
3-19
10-26
26-42
42-58
, 1
1
1
1
1
1
1
1
1
2
1
2
2
1
1
1
1
1
1
66.6
<0.05
<0.05
0.34
<0.05
<0.05
2.04
<0.05
<0.05
<0.05
0.67
0.12
<0.05
0.17
<0.05
<0.05
<0.05
<0.05
0.21
0.15
7.45
0.26
0.13
'See Exhibit H for general location
2Measured from the original 1973 surface prior to backfill
3See Table 5 for soil code descriptions
-------
85
TABLE 16. Collection and Analytical Data from Core Sites
Sampled Over Sixteen-Inch Intervals in August 1975
(Continued)
Core Site
Identification1
HT-1-4
HT-5-8
ZZ-1-4
ZZ-5-8
BR-1-4
SL-1-4
Field Sample
Sample Depth2 Character3
Number (Inches) of Soil
GF7661 9-25
GF7662 25-41
GF7663 9-25
GF7664 25-41
GF7665
GF7666
3-19
9-25
10
10
5
5
5
5
PCB Concentration
(mg/kg)
0.16
, <6.05
0.29
0.50
0.91
2.05
1See Exhibit H for general location
2Measured from the original 1973 surface prior to backfill
3See Table 5 for soil code descriptions
-------
Table 17. Aroclor 1254 and Askarel Solvent Analyses for
. Water, Sediment, and Control Soil Samples
WATER .
Concentrations are expressed as yg/liter (ppb)
Date
Collected
8-28-75
8-29-75
8-29-75
8-29-75
8-29-75
8-29-75
8-29-75
8-29-75
8-29-75
8-29-75
9-26-75
9-26-75
9-26-75
9-26-75 '
9-26-75
9-26-75
9-26-75
9-26-75
9-26-75
Field
Location
Sta. 7
Sta. IB
Sta. 22
Sta. 43
Sta. 23
Sta. 28
Sta. 42
Sta. 40
Sta. 70
Sta. 89
Sta. 43
Sta. 23
Sta. 22
Sta. 40
Sta. 42
Sta. 28
Sta. 70
Sta. 7
Sta. 89
Sta. IB
Sample
Number
PP 2257
PP 2258
PP 2260
PP 2261
PP 2263
PP 2264
PP 2265
PP 2266
PP 2267
PP 2268
PP 2270
PP 2271
PP 2272
PP 2273
PP 2274
~ PP 2275
PP 2276
PP 2277
PP 2278
1,2,4- 1,2,3- 1,2,4,5- 1,2,3,4- penta-
Aroclor Total trichloro- . trichloro- tetrachloro- tetrachloro- chloro-
1254
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
. <0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
Solvent benzene benzene benzene benzene benzene
1.23 - 0.18 0.47 0.33 0.25
<0.006 - - -
<0.006 - - - -
<0.006 - - -
<0.006 - - - - -
<0.006 - - - • . - - -
<0.006 - - - - -
<0.006 - -
<0.006 - - -.-.-. -
<0.006 - - - -
<0.006 - - - -
<0.006 - - - - -
<0.006 - - - - -
<0.006 - -
<0.006 - - - . -
.. <0.006 - - - -
<0.006, - ' • '- - -
0.92 0.006 0.097 0.417 0.152 0.245
<0.006
Collection Permission Refused
SEDIMENT AND CONTROL SOILS
Concentrations are expressed as mg/kg (dry basis)
8-29-75
8-29-75
9-26-75
8-14-75
8-14-75
Sta. IB
Sta. 43
Sta. 43
Control A
Control B
PP 2259
PP 2262
PP 2269
GF 7683
GF 7684
<0.05
<0.05
<0.05
<0.05
<0.05
0.016
<0.010
<0.010
0.016
00
-------
87
insufficient sample available for triplicate determinations.
Comparative data are found in Table 18. Based on this experi-
ment, all three solvent systems are equally effective in the
extraction of PCBs from the clay soil present at the spill site.
Round-robin studies in 1973, including the contractor and the control
agencies involved with the askarel spill, showed similar
correlation between the use of hexane only as the extractant
when compared to a dual hexane/acetone solvent system.
E. Quality Assurance Data.
The quality assurance for this project was provided by an in-
house quality control program. Duplicate blind sample splits
for use as quality control checks were prepared as soon as the
air-dried samples were screened; therefore, the quality control
data are representative of both method and sample reproduci-
blllty. Results of these analyses are tabulated in Table 19.
The eight samples involved in the special solvent extraction
experiment were also included in the in-house quality control
program. In order to more accurately assess the precision
of the data, the results from these two sources were combined;
and an average percent coefficient of variation was determined for
the samples involved. Tabulation of the data is found in Table
20. The average coefficient of variation based on these results
is ± 6.6%.
-------
88
Table 18. Results from Special Soil Extraction Experiment
—Aroclor 1254
*t
Concentrations are expressed as rag/kg (dry-weight basis)
Solvent System
Sample
Identification
GF 7616
GF 7617
GP 7635
GP 7640
GF 7644
GP 7648
. GF 7654
GF 7675
Hexane
20.7
7.24
5.36
0.31
2.08
0.82
<0.05
4.35
' . * •.*.,> -. .-•
Hexane /Ace tone
20.7
5.46
6.12
0.35
2.22
'F
0.74
<0.05
3.50
Hexane-15*
Methylene Chloride
22.0
5.94
5.19
0.43
*
0.73
<0.05
*
•Insufficient sample for experiment
-------
89
TABLE 19. In-house Quality Control Data—Aroclor 1254
Concentrations are expressed as mg/kg (dry basis)
Sample Code Original Analysis Duplicate Analysis
GF 7545
GF 7589
GF 7608
GF 7609
GF 7610
GF 7611
GF 7616
GF 7617
GF 7627
GF 7635
GF 7640
GF 7644
GF 7648
GF 7654
GF 7675
2.24
0.24
<0.05
<0.05
1.27
1.21
23.8
5.00
0.16
5.62
0,34
2.04
0.67
<0.05
4.35
. 1.64
0.34
* <0.05
<0.05
1.22
1.21
18.0
5-65
0.06
5.39
0.31
2.08
0.82
<0.05
5.44
-------
Table 20. Percent Coefficient of Variation Evaluation for In-house
Quality Control Data—Aroclor 1254
\
Concentrations are expressed as mg/kg dry basis
90
Hexane Average
Sample
Code
GF 7616
Z C.O.V.
GF 7617
Z C.O.V.
GF 7635
Z C.O.V.
GF 7640
Z C.O.V.
GF 7644
Z C.O.V.
GF 7648
Z C.O.V.
GF 7654
Z C.O.V.
GF 7675
Z C.O.V.
Average
Value
21.0
5.86
5.54
0.35
2.11
0.76
<0.05
4.41
Original
Analysis
23.8
±13.
5.00
±15.
5.62
±1.4
0.34
±2.9
2.04
±3.3
0.67
±12.
<0.05
±0.
4.35
±1.4
QC Blind
Split
18.0
±14.
5.65
±3.6
5.39
±2.7
0.31
±11.
2.08
±1.4
0.82
±7.9
<0.05
±0.
5.44
±23.
Hexane
Solvent
20.7
±1.4
7.24
±24.
5.36
±3.2
0.31
±11.
2.08
±1.4
0.82
±7.9
<0.05
±0.
4.35
±1.4
Hexane/
Acetone
20.7
±1.4
5.46
±6.8
6.12
±10.
0.35
±0.
2.22
±5.2
0.74
±2.6
<0.05
±0.
3.50
±21.
Methylehe
Chloride
22.0
±4.8
5.94
.±1.4
5.19
±6.3
0.43
±23.
-
0.73
±3.9
<0.05
±0...
-
C.O.V.
(Z)
±6.9
±10.2
±4.7
±9.6
±2.8
±6.9
±0.
±11.7
-------
91
F. Analytical Data from Gaa Chromatography/Mass Spectrometry (GC/MS)..
At the;conclusion of Phase II, Task A—Laboratory Analysis of
Soil and Water Samples by EC/GC—a meeting was held at the EPA
Southeastern Environmental Research Laboratory In Athens, Georgia,
which involved Mr. Don Brown, Mr. Bill Loy, and Dr. Anna M.
Yoakum. After a careful review of the EC/GC data, the two samples
containing the highest concentration of Aroclor 1254 were selected
for further study utilizing gas chromatography/mass spectrometry
(GC/MS). One sample selected was from the original excavation
area, and the other came from below the surface of an area un-
disturbed in the original excavation. The two samples selected
for evaluation were correlated to an actual sample of the
askarel involved in the 1973 spill.
The GC/MS data outputs consist of total ion current chromato-
grams and specific ion searches of the data (for ions charac-
teristic of the Indicated polychlorinated biphenyls). The
chromatograms obtained (Figures 9-13) show that the same isomers
/
are present in the same ratios in the environmentally aged
sample extracts and in the askarel. These data indicate that
no selective degradation of the Aroclor 1254 isomers has
occurred. Mass spectra typical of four, six, and seven chlorine
biphenyls are presented in Figures 14-16P
-------
92
VII. FATE OF POLYCHLORINATED BIPHENYLS (PCBs) AND POLYCHLOROBENZENES
.. AFTER A TWO-YEAR EXPOSURE IN A NATURAL ENVIRONMENT
The experimental evidence presented in the preceding section of this
report will now be evaluated in an effort to determine the fate of
askarel after a two-year exposure in a natural environment. This evalua-
tion will be conducted in two parts. The first part deals with PCBs and
the second with the polychlorobenzene solvent. Only one commercial mix-
ture of PCBs, Aroclor 1254, was found associated with the spill. Data
evaluation for PCBs will, therefore, be restricted to this material.
\ • ' . - . '
Potential mechanisms for the loss of the askarel remaining in the
spill area after the cleanup include volatilization, leaching,
metabolic and nonmetabolic degradation. Each of these mechanisms
and any other loss pathway which appears appropriate will be con-
sidered in conjunction with the experimental evidence collected.
A. Fate £f Aroclor 1254.
1. Distribution of Aroclor 1254 In and Around the Spill Site in
1975. Aroclor 1254 was detected in 68 of 145 core samples
collected in and around the 1973 excavation areas. Concentra-
tions ranged from 0.05 to 67 mg/kg in the positive samples. A
concentration distribution of Aroclor 1254 in the core samples
is shown in Figure 17.
^
In order to effectively evaluate the 1975 distribution of
Aroclor 1254, pertinent facts relative to the spill itself
need to be considered. The magnitude of the 1973 spill
and the elevation contours of the semi-mountainous terrain
of the spill site assured an initial mass flow transport
i
process which resulted in complete saturation of the
-------
Figure 9. Total Ion Current Chromatograms
93
GF7638
Til
GF7616
Til
PYRANOL
Til
60 80 100 120 140
-------
94
Figure 10. Specific Ion Search-for the 292 Ion
Indicative of the Four Chlorine Biphenyl Isomers
GF7616
292!
X'ed puk« are the Mjor four chlorine blphenyl Isooers
-------
Figure 11. Specific Ion Search for the 326 Ion
Indicative of the Five Chlorine Biphenyl Isomers
95
PYRANOL
X
326!
Z'cd peaks are~thc major five chlorine bipheuyl isoaera
-------
Figure 12. Specific Ion Search for the 360 Ion
Indicative of the Six Chlorine Biphenyl Isomers
96
GF7638
360
x x
ww.
•r—-*•
-r—••;—-r
GF7616
3601
PYRANOL
360:
X'ed peaks are the major six chlorine biphenyl isomers
-------
Figure 13. Specific^Ioti Search for the 394 Ion
Indicative of the Seven Chlorine Biphenyl Isomers
97
GF7638
GF7616
PYRANOL
ft 394'
X'ed peaks are the major seven chlorine biphenyl isomers
-------
Figure' 14. Spectrum typical of four chlorine Biphenyl
SPECTRUM
I
8.
~
*
i4
S3 .- SS
ecu
ll
X
&
0
170 183 1S3 283 213 223 223 213 2S3 253 270 2e3 223 333 318 323 323 3*3 333 2S3 373,353 323 <1G3
E
VO
oo
-------
Figure 15. Spectrua typical of nix chlorine Biphenyl
SPECTRUM MJK35R 103-107
8.
-v»J_
a.
o
M
T
T
J&
**
170 163 1S3 233 213 228 223 213 2S3 268 273 283 223 333 310 323 323 313 333 333 373 333 2£>3
e •
VO
VO
-------
Figure 16. Spectrum typical of seven chlorine Biphenyl
MtfBeR 131 - 132
F-313232. CONC,MIX COL, 12J3Y10.C3,10-6-S
8.
8.
r-
fc5>
.
go
lil
I|l11l|ll1l|lli<|;iltflll.«lll>pl»fllll|llf>llll<|l»>ill»|. I I H|HII|lll'j...¥,JlllJ.fT«II ll|f..'f..Tj.I.,,.tl.|l "('" I
250 270 283 233 333 313 323 333 3*3 323 332 373 £83
173 163 133
e
218
233
o
o
-------
101
top soil with varying degrees of penetration into the clay
overburden. This transport process covered a relatively
large area in both horizontal and vertical directions.
The spread of the askarel was also affected by the movement
of contaminated surface water resulting from massive rain-
fall in the weeks immediately following the spill. Exca-
vation operations revealed that the distribution in the
clay was non-uniform; and numerous so-called "hot spots"
resulted from movement along the root systems of plants
and trees, as well as from movement in the fractured chert
frequently found in the clay matrix.
Based on 1973 data, three test core sites—RL, UC, and JF—
were selected for the initial phase of the study. Concen-
tration-depth profiles relating the 1975 data to the
original 1973 data are shown in Figures 18-20. As was
typical of the 1973 distribution, one so-called "hot spot"
was detected; but the magnitude of the Aroclor 1254 was
not excessive (<25 mg/kg). A comparative tabulation of
1973 and 1975 core data are given in Table 21.
One stated purpose of the study was to determine if a
reduction in concentration of Aroclor 1254 in the soil had
occurred either as a result of migration from the spill
area or from degradation. By direct comparison, the 1973
data is higher in 61% of the locations; but the order of
-------
100-
90-
Figure 17_.
Concentration Distribution o£_ Aroclor 1254 in Core Samples, 1975
102
80_
70-
60-
01
0)
i-t
a
I 50-
30-
20-
10-
0-
(<0.05)
(0.05-0.99)
(1.0-9.9)
(10.-24.)
Concentration Ranges, Aroclor 1254 (mg/kg)
-------
Figure 18. Concentration-Depth Profiles of Core Site UC,
1973-1975
103
Core depth profiles given in inches
UC-/
25
29
33
37
UC-3
21
25
29
33
37
1973 Core
21
23
29
33
37 \mz
AROCLOR 12S4 CONCENTRATION mg/Kg
-------
Figure 19. Concentration-Depth Profiles .of Core Site RL,
1973-1975
104
Core depth profiles given in inches
RL-I
19
23
27
31
35
RL-3
19
23
27
31
33
1975
PL-Composite
1973 Core " M
21
21
25
25
29
29
33
33
AROCLOR 1254 CONCENTRATION mg/Kg
<0.05 O.OS-.99 1.0-9.9 1.0-24.
RL-2
19
23
27
31
35
RL-4
*f pfc.
19 c-?T
23
27
31
35
XX
XX
XX
XX
XX
XX
XX
xx
xx
XX
XX
'XX
XXx
XX
XX
XX
XX
XXx
-------
105
Figure 20. Concentration-Depth Profiles of Core Site JF,
1973-1975
Core depth profiles given in inches
JF-I JF-2 JF-3 JF-4 1973 Core
19
23
27
31
35
39
43
47
51
19
23
27
31
35
39
43
47
51
19 S:::
23
27
31
35
39
43
47
51
19
23
27
31
35
39
43
47
\
51
AROCUOR 1254 CONCENTRATION mg/Kg
. 19
23
27
31
35
39
43
47
-------
Table 21. Comparison of Analytical Data from
Core Samples Collected in 1973 and 1975
Core Site
Collection Depth
Ranges (in Inches)
Identification Year 1973 Year 1975
Part I.
RL
UC
JF
Part II.
EK
SR
CC
SM
FM
Part III.
IJ
MB
KH
SU
JR
GW
TE
JD
4-Inch Test Cores
9-35
9-60
9-45
4-Inch Core Samples
9-35
9-45
9-36
9-36
9-66
16-Inch Core Samples
9-43
9-37
9-37
9-30
9-20
9-40
9-62
9-35
19-35
21-37
19-51
15-31
9-41
9-25
20-36
33-65
9-41
9-25
21-37
12-28
9-25
9-25
9-25
9-25
Analytical Results—1973
PCB Concentration No. of
(mg/kg) Samples
1.25-5.87
0.06-0.27
2.04-7.67
0.86-2.76
0.27-0.43
0.88-2.20
0.44-0.85
1.64-3.44
0.55-1.45
1.60-5.86
4.10-6.56
0.19-1.42
0.52-2.40
0.09-0.29
0.07-0.21
0.27-0.82
3
4
3
4
3
3
3
3
3
3
3
3
2
3
4
3
Analytical Results—1975
PCB Concentration No. of
(mg/kg) . Samples
<0.05-14.5
<0.05-2.24
<0.05-0.36
<0.05-0.27
<0.05-0.89
<0.05-0.40
<0.05-1.10
<0.05-1.27
<0.05-7.36
3.69
<0.05
23.8
5.00
3.23 ~
<0.05
0.17
17
16
32
4
8
4
4
8
2
1
1
1
1
1
1
1
-------
Table 21. Comparison of Analytical Data from
Core Samples Collected in 1973 and 1975
(continued)
Core Site
Identification
MM
GE
CT
MT
BT
LP
JS
FA
PT
FO
AA
EL
PS
DV
SS
HH
AM
VT
JE
BA
TR
CV
PQ
MD
BS
HT
ZZ
BR
SL
Collection Depth
Ranges (in Inches)
Year 1973
9-66
9-35
3-6
9-40
9-36
3-6
3-6
9-42
. 3-6
9-36
6-9
9-50
0-18
9-34
9-36
9-66
9-38
3-6
39-45
34-37
33-41
33-52
9-22
0-18
9-48
9-55
9-40
0-18
9-61
Year 1975
9-73
19-35
3-19
9-41
9-41
3-19
4-20
9-41
3-19
9-41
3-19
9-41
2-9
9-25
9-41
9-41
9-25
0-16
29-45
21-37
25-37
33-45
9-25
3-19
10-58
9-41
9-41
3-19
9-25
Analytical Results—1973
PCB Concentration
(mg/kg)
0.23-6.72
0.68-2.32
0.12
1.37-3.00
3.44-311.
3.26
3.85
1.85-8.50
0.50
0.05-0.21
0.60
0.11-1.00
0.09-0.69
0.21-0.58
0.46-5.70
<0. 05-1. 35
0.09-0.15
0.72
0.16-0.20
0.12
<0.05
<0. 05-1. 39
0.40-0.44
0.76-7.50
0.27-15.8
0.13-4.80
0.47-7.50
0.63-6.00
<0. 05-0. 20
No. of
Samples
4
3
1
3
3
1
1
3
1
3
1
3
3
3
3
3
3
1
2
1
2
4
2
3
4
4
3
3
3
Analytical Results—1975
PCB Concentration
(mg/kg)
<0. 05-0. 27
0.32
0.16
0.89-4.35
1.34-10.1
3.41
5.62
<0.05
66.6
<0.05
0.34
<0.05
2.04
<0.05
<0.05
0.12-0.67
<0.05
0.17
<0.05
<0.05
<0.05
<0.05
0.21
0.15
0.13-7.45
<0. 05-0. 16
0.29-0.50
0.91
2.05
No. of
Samples
4
1
1,,
2
2
1
1
2
1
2
1
2
1
1
2
2
1
1
1
1
1
1
1
1
3
2
2
1
1
-------
108
magnitude for the Aroclor 1254 concentrations are comparable
for both years; and no significant migration patterns are
observed.
2. Degradation of Aroclor 1254. Two possible degradation modes
exist—metabolic and nonmetabolic. Each degradation mechanism
will now be considered from the standpoint of applicability to
the system under study.
a. Chemical Transformation—The non-biological alteration
of a chemical introduced into any part of the environ-
ment is dependent on the moisture, pH, and temperature
of that environment; on the nature of reactive groups
on the agent; and on the presence of catalytic sites.
In addition, the nature and intensity of available
illumination determines photochemical reactions.
Irradiation (photolysis) of Aroclor 1254 under laboratory
conditions has produced dechlorinated products; and, in
the presence of air and water, hydroxylated and hydrated
•\ ' . • • •
products have been identified in the polar products of
the irradiation. Although these theoretical routes for
chemical degradation exist, the extreme stability
(chemical inertness) of Aroclor 1254 coupled with the
absence of illumination at the potential reaction site
make chemical transformation a highly improbable degrada-
tion mechanism.
-------
109
b. Biological Alteration—By analogy to the dechlorination of
DDT to ODD by soil microorganisms, biodegradation of Aroclor
1254 is considered to be a possible degradation mechanism
in a natural environment. The unique biochemical asset of
certain aerobic microorganisms to catalyze early steps in
degradation allows for the formation of metabolites which
can enter the common pathways of metabolism. The establishment
of measureable biodegradation is dependent on the ease of
physical or chemical sequestration of the PCB components due
to the structure of the molecule as it relates to microbial
enzymatic action. The intrinsic toxicity of the askarel,
environmental factors affecting microbial populations and
their specificity, and available time for the maximum develop-
ment of the degradation process are also significant considera-
tions in the interpretation of biological alterations of PCBs.
c. Aroclor 1254 Degradation Assessment—Aroclor 1254 is a mix-
ture of chlorinated biphenyl homologs. If no degradation
in the environment occurred, or if all homologs degraded
at the same rate, the ratio of homologs in "aged environmental
samples" should be the same as that in the askarel introduced
into the environment at the time of the spill. On the other
hand, if the homolog ratio in the "aged environmental samples"
from the spill site differs from that of the askarel spilled,
some process(es) must be operating in the environment to
remove different homologs at different rates. According to
-------
110
data supplied by the Monsanto Company (35), the typical
percent (w/w) composition .of Aroclor 1254 is similar to
that given in Table 22.
All chromatograms from the EC/GC analyses of 1975 samples
were carefully examined for peak alterations and for the
appearance of new peaks with reference to the chromato-
gram of the askarel released into the environment at
the time of the spill. Comparative chromatograms for
Aroclor 1254 in a typical "aged environmental sample"
and Aroclor 1254 in the askarel spilled in 1973 are
shown in Figure 21. By comparing the two chromatograms,
it is readily apparent that the ratio of homologs in the
"aged environmental sample" is the same as that in the
askarel introduced into the environment at the time of
the spill. These results indicate that either there has
been no degradation in the environment or that all
homologs have degraded at the same rate.
Using -gas chromatography-mass spectrometry (GC/MS), the
homolog distribution of selected 1975 samples was com-
pared to the distribution exhibited in a sample of
askarel from the 1973 spill. These data also confirm
that the ratio of homologs in the "aged environmental
samples" is the same as that of the askarel spilled in
1973.
(35) Monsanto Company, "Presentation to the Interdepartmental Task Force
on PCBs," Washington, D. C., May 15, 1972.
-------
Ill
Table 22. Typical Homolog Composition of Aroclor 1254 (35)
Homolog Aroclor
(No. of Chlorine/biphenyl) 1254*
2 <0.5
3 1.
4 21.
5 48.
6 23.
7 6.
8 NDt
*Percent (w/w) by GC/Mass using area correlation factors by homolog
response.
tND * None Detected, <0.01%.
(35) Monsanto Company, "Presentation to the Interdepartmental Task Force
on PCBs," Washington, D. C., May 15, 1972.
-------
Tf.tF
1OO
±r±
Askarel Spilled in 1973
;:
l :
-Representative Aged Environmental Sample—1975 J-J--J -j --!-j-
f
Figure 21. Comparative Chromatograms for Aroclor 1254 in a Typical
"Aged Environmental Sample" (1975) and Aroclor 1254 in
the Askarel Spilled in 1973
t-rt
H-
i I '
1 -
11
ill!'
-------
113
No evidence was found by GC/MS and/or EC/GC to indicate
the presence of degradation products in any of the "aged
environmental samples" collected in or around the spill
site. Primary biodegradation can be defined as minimum
alteration of the chemical structure of the material in
\
question to an extent that characteristic properties of
the original material are no longer evident. Based on
the experimental evidence collected, and considering the
definition just presented, it may be concluded that no
detectable reduction in the concentration of Aroclor
1254 in the soil has occurred as the result of chemical
transformation or biodegradation. An evaluation of the
occurrence of major soil microorganisms is consistent with
this conclusion in that the three microbial groups studied
were probably not present in sufficient numbers to support
measureable degradation of the PCB.
B. Fate of Askarel Solvent.
The scope of this project did not include the analysis of soils
for the askarel solvent. Only water samples, collected adjacent
*
to the study area, were included to provide information on the
rate of intrusion of polychlorobenzenes into ground water.
As a consequence, since biodegradation analyses were limited to
soil samples, an assessment of solvent degradation is not
possible. This discussion of the so-called "fate" of askarel
-------
114
solvent in the spill environment will be limited to the distribu-
tion of polychlorobenzenes in ground water adjacent to the spill
site.
The 1973 data showed only minimal movement of Aroclor 1254 in
the ground water. Only two ground water sampling stations gave
positive data for Aroclor 1254. These locations were a.spring
just below the spill area and a well less than one hundred feet
from the spill excavation area. The highest Aroclor 1254 con-
centration detected in a water sample was 2.1 ppb.
Migration of the'askarel solvent, however, was completely differ-
ent from that observed for the PCS, Aroclor 1254. Differential
migration of the askarel solvent in the spill environment was
observed in a well water sample-collected March 22, 1973. A
subsequent screening of selected water sources—wells, creeks,
ponds, etc.—revealed an interesting phenomenon. There were
two distinct patterns for the solvent concentration in the
\ •-
water samples. One closely resembled the component distri-
bution of the askarel solvent spilled. The other was almost
95% trichlorobenzenes.
*
Because of this observed phenomenon, a selected number of repre-
sentative core drillings at the spill site were analyzed for
askarel solvent. These data indicated that tetra-and penta-
chlorobenzene were preferentially retained in the soil while
the other solvent components—especially the trichlorobenzenes—
/'
had selectively moved out of the area. This study also showed
-------
115
that Aroclor 1254 was preferentially retained in the soil. Data
i
from monitoring programs confirmed that water samples with sol-
vent patterns resembling the .askarel solvent component distribu-
tion indicated possible contamination by surface water run-off
V
and seepage of surface water into a supply. Water with the high
trichlorobenzene concentrations indicated intrusion of these sol-
vent components into the ground water supply. A graphic repre-
sentation of solvent concentration in the well located closest
to the spill site is shown in Figure 22. Samples from this
station collected during the follow-up study still show solvent
present in the water at the 1.0 ppb concentration level.
These data clearly indicate that the more water-soluble compo-
nents of the askarel solvent invaded the ground water supply
almost immediately after the occurrence of the spill. Leaching
was the migration mechanism responsible for the intrusion of
the lower chlorinated benzenes into the ground water supply.
Based on the 1973 and 1975 data, it can be concluded that in-
trusion of lower chlorinated benzenes into a ground water
supply used for drinking water purposes occurred rapidly after
the spill. This same water supply contains minimal, but de-
tectable, quantities of askarel solvent two years after the
occurrence of the spill.
C. Over-all Assessment of the Environmental Impact of the Sp-tl-1-.—
The contractor did not observe the spill area prior to
the actual spill; however, observations began on March 21, 1973,
-------
EB0
330
^^
100
a M
tn H0
§ 30
o= 20
10
a
5
E
H
3
FIGURE Z2.
SOLVENT CONCENTRRTIDN PRDFILE
WELL LDCHTED CLD5E5T TD 5P1LL H1TE
(5THTIDN 7)
*-^
~I' I T
ii i
1373
T i
OUL1BN DHTE
i r
1H7H
s
•1
-------
117
approximately two and one-half weeks after the spill occurred.
Invasion of the spill area by workers and heavy equipment involved
in the cleanup had already transpired by this time. " There had been
heavy rainfall during the two and one-half week period immediately
after the spill. The area received 12.44 inches of precipitation
during the month of March which constituted a 7.23 inch departure
from the norm. All assessments will utilize, as an environmental
basis, the condition of the area at the time the contractor arrived.
1. Condition of the Area Immediately After the Spill.
a. Primary Spill Area—The primary spill area was located
adjacent to the intersection of a main north-south
highway and a secondary access road utilized mainly by
residents living in sparsely developed areas along the
lake (see Exhibit I). The property on which the spill
occurred was a segment of land designated for pasture
farming.
The spill site is situated in a watershed approximately
0.6 mile long which drains into the lake in a north-
easterly direction and generally runs parallel to the
secondary access road. The distance from the spill site
to the lake is approximately 0.5 mile. The main water-
shed also received drainage from several secondary
watersheds adjacent to it. The spring well, Station IB,
-------
118
is located approximately 0.2 mile from the spill site
in this north-east trending hollow. Most of the chemical
spilled migrated down the slope toward the north-east,
influenced primarily by the surface topography.
The primary spill area, which was later excavated,
covered an area approximately 250 feet long and 125 feet
wide. This area was sparsely wooded with pines and hard-
woods including: Shortleaf Pine (Pinus echinata),
Eastern White Pine (Pinus strobus), Virginia Pine (Pinus
virginiana), White Oak (Quercus alba), Southern Red Oak
(Quercus falcata), Mockernut Hickory (Garya tomentosa),
and Flowering Dogwood (Cornus florida). Natural under-
/ '
story plant associations were observed in the area at
this time. Two Southern Red Oaks and a Shortleaf Pine
with base diameters of 12-18 inches were removed from
the excavation area in 1973.
Although the cleanup operation was not intended in any
N^
way to be a terrestrial ecology study, cursory observa-
tions, were made on the biota of the immediate spill area
and the surrounding area at various times during the
period of March through July, 1973. •> No mammals, reptiles
or amphibians and very few birds were observed in the
vicinity at this time. Crustacea (crayfish) were observed,
and watercress blanketed the area of the spring well
(Station IB). There was some evidence of damage to trees
and understory plants in the area of heaviest contamination.
-------
119
b. Area Peripheral ^o Spill Site—It was suspected that the
chemical may have also migrated to the south-east across
the road and into another watershed. South-east from the
crest of the hill where the spill occurred the topography
is gently rolling through 'fields and wooded areas down to
the lake. The distance from the spill site to the lake
in this direction is approximately 0.4 mile. Little wild-
life was observed in this area. Evidence of wildlife
within pastured segments including dry weather ponds
was absent.
2. Condition of Spill Area - August 1975.
In 1973, the cleanup procedure resulted in extensive exca-
vation in the spill area. Contaminated soil was removed;
and the excavated areas were sealed, backfilled, and packed
with, uncontaminated soil. The entire affected area was
covered with top soil, seeded with grass, and landscaped.
this area had received little maintenance since 1973; and,
as a result, grass and weeds were overgrown, hindering field
operations. A rotary mower was used to cut the .overgrowth,
and the area was raked.
According to the property owner, several damaged trees had
been removed in the period between the 1973 cleanup opera-
tion and the latest study.
Again, the 1975 follow-up study was not intended to examine
the terrestrial ecology of the area; however, over a period
of approximately one month (July and August, 1975),
-------
120
observations made in the spill area .indicated no obvious
detrimental effects to any biota of the ecosystem. In the
course of sampling, several specific observations were made.
Crayfish and minnows were found in abundance in the spring
l '
well waters, as were various aquatic insects and plants.
The immediate spill area was infested with insects including
ants, spiders, ticks, grasshoppers, crickets, flies, mos-
quitoes, bees, and yellow jackets. Several garter snakes .
were seen in this area also. Segmented worms were observed
in core samples taken within a few feet of the surface.
-------
APPENDIX
Method for Analysis of Water and
Sediment for Polychlorinated Biphenyls
Monsanto Company
Analytical Chemistry—Method 69-13
February 1970
-------
ANALYSIS OF WATER AND SEDIMENT FOR
POLYCHLORINATED BIPHENYLS
SCOPE
This methodology was developed for the determination of the amount and
type of polychlorinated biphenyls (PCB) in water and sediment samples.
Absolute confirmation of PCB structures is no.t obtained with this method.
Structure proof can be obtained using additional techniques such as mass
spectrometry to further identify the GC fractions.
PRINCIPLE
The PCB(s) in water and sediment samples are extracted into an organic
solvent. Interfering components are then removed from the extracts by
chemical treatment and column adsorption chromatography. The amount
and type of PCB present is determined by electron capture gas chroma-
tography (EC/GC). .
REAGENTS
Hexane
Acetonitrile
Sodium Sulfate
Alumina Adsorption
Nanograde, Mallinckrodt Chemical Works,
Catalog No. 4159.
Nanograde, Mallinckrodt Chemical Works,
Catalog No. 2442.
Anhydrous, granular: AR grade,
Mallinckrodt Chemical Works, Catalog No.
8042. Heat at 400°C for one hour prior
to use.
(for chromatographic analysis) 80/200
mesh, Fisher Scientific Co., Catalog No.
A540. Heat at 400°C for a minimum period
of 4 hrs. and deactivate with 5% (w/w)
distilled water.
Alumina column preparation: Fill a
chromatographic column with hexane up to
the point where the reservoir joins the
column and push a glass wool plug to the
bottom with a glass rod. In a 50 ml beaker
measure 35 ml of deactivated alumina (i«30g).,
and pour this slowly into the column. Tap
or vibrate the column to settle the alumina
and top the alumina with 2-3 cm of anhydrous
sodium sulfate. Wash the column with 50-
100 ml of hexane prior to the addition of
the sample.
-------
Distilled Water Extracted with hexane to remove hexane
soluble electron capturing impurities.
Sulfuric Acid Analytical Reagent Grade, SG = 1.84
Potassium Hydroxide Analytical Reagent Grade
Ethanol . . Formula 2B
2.5% (w/v) Alcoholic Dissolve ^12.5 grams of AR grade KOH in
Potassium Hydroxide 500 ml of ethanol
9/1 (v/v) Sulfuric Carefully add 270 ml of AR-grade sulfuric
Acid - Water acid to 30 ml of distilled water in a 500
ml iced beaker
.PCS Standards Aroclor 1242, 1248, 1254, and 1260
APPARATUS
1. Separatory funnels equipped with ground glass stoppers and Teflon
stopcocks: 125, 250, 500, 1000 and 2000 ml capacities.
2. Kunderna-Danish Evaporative Concentrators, 500 ml capacity equipped
with 3-ball Snyder columns and graduated 5 ml capacity vials: Ace
Glassware Company, Catalog No. 6707.
3. Chromatographic columns, glass, 10" x 20 mm (OD) with a 5" x 50 mm
(OD) reservoir at the top, equipped with Teflon stopcocks.
4. Sintered glass filter funnels, 600 ml capacity, 90M.
5. Flat bottomed boiling flasks, 125 ml capacity: Ace Glassware
Company, Catalog No. 6896, Code - 04.
6. Liebig Condenser, 200 mm in length: Ace Glassware Company,
Catalog No. 5915, Code - 12.
7. Hot plates, Corning PC-100: Fisher Scientific Company.
8. Water bath, Thelco, Precision Scientific, Model No. 84, Fisher
Scientific Company.
9. Reciprocating variable speed shaker, Eberback Corporation, Fisher
Scientific Company.
10. 10 yl Hamilton Syringes, Catalog No. 701N.
11. 32 oz. all glass mortars and pestles.
-------
12. 8 x 12 x 2" (2-1/2 qt.) Pyrex baking dishes.
13. U. S. Standard Sieve, No. 30, Fisher Scientific Company.
14. Usual laboratory glassware.
SAMPLING
It is to be assumed that a rather wide variety of sampling techniques
were employed in collecting the samples submitted for analysis. In
general the procedures used were probably selected for ease of adaption
to the local situation. For this reason water and sediment samples
were usually treated as follows:
Water ,
Where possible the entire water sample, including the container in which
it was collected, was extracted with hexane. With larger samples, where
this was not physically possible, the containers were simply agitated
and a 250 ml portion used for analysis.
Sediment
Any excess water was decanted and the entire sediment transferred to a
glass baking dish to air dry at room temperature. The dried material
was transferred from the dish into a mortar and pestle and ground.
The ground sediment was sieved, remixed, and a 250g portion taken for
analysis.
PROCEDURES :
Extraction of Water Samples
1. Extraction of water samples - after agitating, transfer the entire
aqueous sample or a 1000 ml aliquot into a graduated glass cylinder.
Record the volume of the sample and quantitatively transfer it to
a separatory funnel with distilled water.
2. Rins> the graduated cylinder with one 50 ml portion of hexane and
add each to the separatory funnel.
3. Stopper the separatory funnel and shake vigorously for at least 2
minutes. Allow the layers to separate and transfer the lower
aqueous phase to a second separatory funnel.
4. Extract the water sample a second' time with a 50 ml portion of
hexane. After the layers have separated, add the first hexane
extract to the second separatory funnel and transfer the aqueous
layer to the original separatory funnel.
-------
5. Repeat the extraction with a third 50 ml portion of hexane. Discard
the aqueous layer and combine the hexane extracts.
6. Filter the combined extracts through a 4" funnel plugged with glass
wool which is covered with sodium sulfate. Collect the filtrate
in a Kunderna-Danish evaporative concentrator^, add a small boiling
chip, put the Snyder column in place, and reduce the hexane volume
to less than 5 ml by heating the apparatus in a 80-90°C water bath.
(CAUTION: SOLVENT VAPORS MUST BE VENTED TO A HOOD.)
7. After cooling, remove the 5 ml graduated, tube and transfer hexane
extract to an alumina adsorption column washing it in with several
5 ml portions of hexane.
8. Carefully add 100 ml of hexane to the column reservoir and collect
the total eluent in either a 250 ml volumetric flask or a Kunderna-
Danish evaporative concentrator.
9. If the column eluent is collected in a volumetric flask, dilute to
volume with hexane and proceed with the gas chromatographic analysis.
10. If the column eluent is collected in a Kunderna-Danish evaporative
concentrator, reduce solvent volume, cool, dilute to volume and pro-
ceed with the gas chromatographic analysis.
Extraction of Sediment and Soil Samples
1. Decant off any excess water and transfer the entire sediment sample
to a glass baking dish. Air dry at ambient temperature (heat should
not be applied).
2. When dry, transfer the soil/sediment to a mortar and pestle and grind.
Sieve the ground material through a No. 30 mesh sieve and weigh 250g
(to the nearest O.Olg) into a 16 oz. narrow neck screw cap (aluminum
foil liner) glass bottle.
3. Moisten the soil with water C\/ 10 ml) and add 150 ml of acetonitrile.
. Cap the bottle tightly, and mechanically shake for a minimum period
of 1 hour.
4. Quantitatively transfer the acetonitrile extract into a 600 ml
sintered glass filter funnel containing a 1/4" layer of anhydrous
sodium sulfate. Collect the filtrate in a 600 ml beaker (vacuum
filtration may be necessary).
5. After the acetonitrile has completely drained into the beaker, wash
the bottle twice with 50 ml portions of acetonitrile, adding each
wash to the funnel after the previous has completely percolated
through the sediment. .
-------
I ' •
6. Quantitatively transfer the extract to a Kunderna-Danish
evaporative concentrator, add a small boiling chip, put the Snyder
column in p-lace, and reduce the solvent volume to less than 5 ml
by heating the apparatus in a 80-90°C water bath. (CAUTION:
SOLVENT VAPORS MUST BE VENTED INTO A HOOD.)
7. After cooling remove the 5 ml graduated tube and transfer the con-
centrate of extracts to a 125 ml extraction flask with the aid of
several small portions of solvent.
8. Evaporate the extract just to dryness with a gentle stream of dry
filtered nitrogen and add 25 ml of 2.5% alcoholic potassium
hydroxide.
9. Add a boiling chip, put a water condenser in place, and allow the
solution to reflux for 45 minutes.
10. After cooling, transfer the solution to a 250 ml separatory funnel
with the aid of 25 ml of distilled water.
11. Rinse the extraction flask with 25 ml of hexane and add it to the
separatory funnel.
12. Stopper the separatory funnel and shake vigorously for at least
1 minute. Allow the layers to separate and transfer the lower
aqueous phase to a second separatory funnel.
13. Extract the saponification solution with a second 25 ml portion of
hexane. After the layers have separated add the first hexane
extract to the second separatory funnel and transfer the aqueous
alcohol layer to the original separatory funnel.
14. Repeat the extraction with a third 25 ml portion of hexane. Discard
the saponification solution and combine the hexane extracts.
15. Carefully add.25 ml of the sulfuric acid solution (9:1 concentrated
sulfuric acid/water) to the hexane extracts.
1$. Stopper the separatory funnel and shake vigorously for at least one
minute. Allow the layers to separate and discard the lower aqueous-
acid layer. Repeat this step until the acid layer is colorless.
17. Wash the hexane with 25 ml portion of water. Discard the water
wash. .
18. Filter the hexane extract through a 4" funnel plugged with glass
wool which is covered with a layer of sodium sulfate into a
Kunderna-Danish evaporative concentrator. ' •
19. Add a small boiling chip, put the Snyder column in place and
reduce the hexane volume to less than 5 ml by heating the
apparatus in a 80-90°C water bath.
-------
20. After cooling, remove the 5 ml graduated tube and transfer the
hexane extract to an alumina adsorption column washing it in with
several 5 ml portions of hexane.
21. Carefully add 100 ml of hexane to the column reservoir and collect
the total eluent in either a 250 ml volumetric flask or a Kunderna-
Danish evaporative concentrator.
22. If the column eluent is collected in a volumetric flask, dilute to
volume with hexane and proceed with the gas chromatographic
analysis.
23. If the column eluent is collected in a Kunderna-Danish evaporative
concentrator, reduce solvent volume, cool, dilute to volume and.
proceed with gas chromatographic analysis.
Electron Capture Gas Chromatographic Procedure
Instrument; F&M 402 Biomedical Gas Chromatograph
Detector: High Temperature Ni63 Electron Capture Cell
6 mm x 61 Glass Column, 4% XE-60 on 80/100 mesh
Chromosorb W, HP, AW-DMCS
Column Temperature: 160°C
Detector Temperature: 300°C Flow Rates
Injection Port Temperature: 195°C Helium Carrier ^ 60 ml/rain
Pulse: 150 Argon-Methane Purge
^ 120 ml/min
Using EC/GC as the determinative step, inject in duplicate 1-10 yl of
each solution into the chromatograph. By comparison with standard
solutions injected, in duplicate, under the same operating conditions,
determine the amount and type of Aroclor using the individual or total
peak height method.
The electron capture detector should also be used to guide the isolation
procedures. Water and sediment extracts can be checked for the presence
of PCBs and/or interferences by injecting yl portions of the extracts
at various points in the extraction and concentration schemes. In this
manner, it can be determined if the sample needs to be concentrated or
diluted and if the clean up procedures should be employed.
DISCUSSION
Extraction
The extraction of PCB's from water employing hexane as the extractant
was found to be quantitative and to be sufficiently simple and rapid for
use as a routine procedure.
-------
The evaluation of this method was based on spiking water samples with
standard acetone solutions of PCB's. The spiking method consisted of
adding the PCB's in acetone (25-50 yl) to 500 ml of tap water in a 32
oz. narrow neck screw cap jar. After thoroughly mixing, duplicate
225-250 ml aliquots were taken and subjected to the proposed sample
preparation and work up as outlined. The results were quantified by
preparing a calibration curve using standard hexane solutions of the
PCB's used to spike the water samples. The major isomer peak height
was used to construct the calibration plot.
The average recovery and deviation achieved substantiated the appli-
cability of the method for the quantitative recovery and analysis
of PCB's from water at the ppb-ppm level.
No PCB recovery experiments from spiked sediment and soil samples
have been performed. Instead, several of the residual solids
representative of some of the types of sediment or soil analyzed were
re-extracted with hexane/acetone (40/60) in a soxhlet extractor to
test for the efficiency of the acetonitrile extraction step. The
hexane, after isolation by dilution with distilled water was then
carried through the purification steps. Recoveries by soxhlet extrac-
tion have indicated that the acetonitrile extraction of PCB's was
essentially quantitative in the cases checked.
Sample Concentration
Concentration of sample extracts is necessary, prior to clean up by
chromatographic or chemical means, to reduce sample size and increase
sensitivity. The preferred method of concentrating allows minimum loss
through volatilization or chemical decomposition and requires a minimum
time. The three methods of solvent volume reduction most commonly used
are evaporation by exposure to a stream of air, evaporation employing
a KunJerna.—Danish evaporative concentrator equipped with a Snyder
column, and evaporation under reduced pressure. We have used all
three techniques and have not encountered any significant losses from
volatilization or chemical alternation. However, the Kunderna-Danish
evaporative concentrator and the stream of air were employed because
of the ease of use.
Column Adsorption Chromatography and Chemical Clean Up
Silica gel, Florisil and Alumina deactivated with 0, 1.0, 1.5, 2.0 and
5% water were investigated as adsorbants for the elimination of inter-
ferences. Alumina (5% water) was found to be more effective and
reproducible than either silica gel or Florisil. The activity of
alumina varies with age and lot, therefore, 5% water was added to the
alumina, after heating for a minimum of 4 hours at 400°C to insure a
reproducible activity.
-------
Saponification and subsequent extraction of the sample with sulfuric
acid is an effective way to remove a number of chlorinated hydrocarbon
interferences as well as other matrix interferences. PCB's are not
affected.
Electron Capture Gas Chromatography
Columns;
Column performance is the key to effective gas chromatographic analysis
and as such the choice of column materials is particularly important.
Ideally, the support employed should be inert, mechanically strong, and
of high surface area. For these reasons, Chromesorb W, HP, AW-DMCS
was used in all of our work.
A variety of polar and non-polar liquid phases were investigated. The
following columns were found to provide adequate separation, etc., for use .
in PCB analysis by electron capture: 4% (w/w) DC-200, SF-96, OV-17, SE-30,
SE-54, XE-60, Apiezon L, and 6% QF-1. DC-200 and XE-60 or QF-1 have been
found to be the most suitable of these liquid phases.
i
Another important consideration when working with an extremely sensitive
detector and consequently low levels of materials is column conditioning.
With polar phases such as XE-60 and QF-1, we have found that operating
a new column overnight at a temperature 25-50°C higher than to be used
during analysis results in a more stable column. A no-flow conditioning
technique is employed to condition non-polar columns. The column is
purged with carrier gas, heated for 30 minutes at an elevated tempera-
ture without carrier flow and then cooled to room temperature. At the
end of this cycle the carrier flow is resumed and the conditioning is
completed as in the case of the polar liquid phase. Two precautions:
during conditioning, the column should not be connected to the detector
and one should not exceed the maximum safe temperature of the liquid
phase.
Since all liquid substrates bleed to one degree or another and columns
eventually degrade, we characterize all new columns with two column
performance indicators - the number of theoretical plates (N) and a tall-
ing factor (T). p,pf--DDT is employed to check these parameters because
it is known to degrade on "poor" columns. In this manner, we can deter-
mine if the performance of a new column is satisfactory and when the
column performance begins to fall off. We consider a column good if the
number of theoretical plates per foot is on the order of 400-500 with
tailing factors of 1.0-1.3. Calculation of these parameters is shown
in the Appendix. Additionally, there should be no significant extraneous
peaks upon injection of a pure p,p'-DDT standard.
Other chromatographic conditions that can be adjusted are column
temperature and flow rates. Although resolution of a mixture increases
with decreasing temperature, a temperature should be chosen that allows
-------
the elution of all components within a convenient time period. The
temperatures given are optimum for 42% chlorinated biphenyl, tempera-
tures are increased when specifically analyzing for the higher chlori-
nated biphenyls, i.e., 54%, 60%, etc. The flow rates are optimum for
our instrument, column and detector system and, of course, should be
adjusted if better results can be achieved.
Two gas chromatographic systems have been used for PCB analysis - F&M
Model 402 and 5750. We find that any system of instrument and column
suitable for chlorinated pesticides is satisfactory for PCB analysis.
The bulk of analyses in our laboratories was carried out using the
system outlined. The use of the high temperature Ni63 electron capture
cell is highly recommended. The ability to operate at higher tempera-
tures prevents maintenance problems due to contamination from high
boiling components. Glass columns should also be employed.
Detection and Measurement;
Quantitative determinations employing the electron capture detector
are non-stoichiometric measurements made by comparing peak heights
or areas for known concentrations with those for unknown compositions.
Except for sharp peaks, peak area measurements are usually more
reproducible than peak height measurements but are extremely time
, consuming unless.a recording integrator is employed. However, peak
/ height measurements are as accurate as disc integration of triangu-
lation and if the peak shape represents a gaussian curve, the height
may be considered independent of the base. Three variations of the
peak height quantification procedure were employed.
Case I i EC gas chromatogram of PCB unknown unchanged with respect
to standard PCB with no evidence of interferences.
Case II EC gas chromatogram of PCB unknown altered with respect
to standard PCB with no evidence of interferences.
Case III EC gas chromatogram of PCB unknown unchanged with respect
to standard PCB with evidence of interference.
The amount of PCB's in Case I samples were determined by preparing a
plot of the major peak height vs. concentration; for Case II, a plot
of the total sum of all major peaks vs. concentration. With Case III
samples, a peak free from interference was used. When dominant inter-
ferences were present, one or more of the chemical clean up procedures
was employed.
In all cases, the response of the electron capture detector must be
linear for quantitative analysis. With our instrument any response
less than 50 at an attenuation of 8 x 10 fell into the linear response
range at a pluse rate of 150. This corresponds to approximately 5 x
10~9g of Aroclor 1242.
-------
Contamination;
In determining PCB's in water, soil and sediment by electron capture
gas chromatography, laboratory sources of contamination can be a major
problem. The samples and extracts should never be allowed to come in
contact with materials other than glass, Teflon or metal. Laboratory
glassware should be thoroughly washed with hot, soapy water, rinsed with
distilled water, acetone, and then hexane. All equipment should also be
rinsed again with hexane just prior to use and blanks should be frequently
carried through all steps of the procedures to insure against the possi-
bility of contamination. -
SENSITIVITY Two parts per billion
Absolute sensitivity - 0.5 x 10~9 grams
Volume injected - 5 yl.
Final volume of extract - 5 ml.
Sample size - 250 ml.
Monsanto Company
R & D Laboratories
Applied Sciences Section
St. Louis, Missouri
•U.S. GOVERNMENT PRINTING OFFICE: 1976 - 648-803/10309
------- |