&EPA
United States
Environmental Protection
Agency
Reg:on V
Toxic Substances Office
230 South Dearborn Street
Chicago, Illinois 60604
September 1980
905R80101
Evaluation Of
Suspected Environmental
Contamination Of The
Hemlock, Michigan
Area
HEMLOCK
-------
EVALUATION OF SUSPECTED
ENVIRONMENTAL CONTAMINATION OF
THE HEMLOCK, MICHIGAN AREA
U. S. ENVIRONMENTAL PROTECTION AGENCY
REGION V
TOXIC SUBSTANCES OFFICE
230 SOUTH DEARBORN STREET
CHICAGO, ILLINOIS 60604
SEPTEMBER 1980
-------
TABLE OF CONTENTS
LIST uF FIGuRES ii
LIST uF TABLES ii
i. EXECUTIVE SUMMARY i
II. INTRODUCTION 5
III. PREVIOUS INVESTIGATIONS CONDuCTEu BY MICHIGAN STATE
AND LOCAL AGENCIES 7
IV. EPA INVESTIGATION 12
A. Introduction 12
B. Environmental Samples From Residential Sites 14
1. Water 14
2. Water distillation residue 22
3. Soil 24
4. Dust 34
5. Sediment and sludge 38
b. Tests for dioxin..... 40
C. Environmental Samples From Hemlock's Municipal
Water and Sewage Treatment Systems 41
1. Water 41
2. Sludge 41
D. Biological Samples 44
1. Fish 44
2. Animal Tissues 47
3. Plant 48
E. Evaluation of Goose Wing Deformity 49
F. Fish Bioconcentration Studies 50
G. Samples From Dow's Brine System and Areas
Located Near It 53
H. Evaluation of Dow Chemical Company's Injection Well
Design and Construction 58
V. CONCLUSION 59
VI. REFERENCES CITED 60
APPENDIXES o3
-------
LIST OF FIGURES
Figure 1 - Map of Hemlock Area and Sampling
Sites 15
LIST OF TABLES
Table 1 - Samples Collected From Residential Sites 16
Table 2 - Results of Analyses of Water Samples From
Res i dent i a 1 Si tes 17
Table 3 - Residue From Water Distillation Units Used at
Res i dent i a 1 Si tes 23
Table 4 - Downspout Soil Samples From Residential Sites 25
Table 5 - Garden, Uncultivated, and Sump Soil Samples From
Residential Sites 26
Table 6 - Soil, Dust, Sediment, and Sludge Samples From
Residential Sites (Chemical Compounds Tentatively
Identified bv a Gas Chromatography-Mass Spectrometry
Scan) ". 27,28
Table 7 - Dust Samples From Residential Sites 35
Table 8 - Sediment and Sludge Samples From Residential Sites 39
Table 9 - Water and Sludge Samples From Hemlock's Municipal
Systems 42
Table 10 - Plant and Animal Tissue Samples 45,46
Table 11(a) - Fish Tissue Samples From 30 Day Bioconcentration
Study Using Water From Residential Site #4 51
Table ll(b) - Fish Tissue Samples From 30 Day Bioconcentration
Study Using Water From Hemlock Municipal Water System....! 52
Table 12 - Description of Samples Collected From the Dow Brine
System and Areas Located Near It 54
Table 13 - Samples Collected From Dow's Brine System and Areas
Located Near It 55,56
n
-------
Executive Summary
Background
Numerous complaints of human and domestic animal health problems in the Hemlock,
Michigan area have been reported to various governmental agencies since 1977.
Some area residents have attributed these health problems to chemical contamina-
tion of drinking water supplies. One concern of the residents was that ground-
water used for drinking might be contaminated due to the presence of brine wells
and a reinjection system operated by Dow Chemical Company in the vicinity of Hemlock.
PREVIOUS INVESTIGATIONS
From 1977 to 1979, the following investigations were conducted by Saginaw
County and Michigan State agencies:
* Michigan Department of Agriculture and Michigan State University
analyzed environmental and animal tissue samples for the presence
of nine contaminants. No evidence of environmental contamination
was found.
x Michigan Department of Public Health and Saginaw County Health Department
analyzed well water samples for significant levels of 27 parameters
and several pesticides, and examined the construction of four private
wells for compliance with state regulations. No contaminants were present
at levels which could be associated with the alleged health'effects. The
well construction was determined to be adequate and in compliance with
state regulations for private wells.
* Michigan Department of Natural Resources conducted a comprehensive ground-
water quality investigation in Hemlock, which included laboratory testing
of water samples for additional parameters, static bioassay tests and field
-------
surveys. These analyses did not indicate any chemical contamination
of the water other than the presence of sodium chloride (salt) in high
enough concentrations to affect the taste of the water.
* Michigan Department of Public Health and Saginaw County Health Depart-
ment conducted an epidemiological study to compare the frequency of health
complaints in Hemlock with that of a neighboring locality. Although this
study showed a greater frequency of reported health complaints in the
Hemlock area than in the control area, the investigation was unable
to establish a link between the complaints and private well water supplies.
EPA INVESTIGATION
In July 1979, the United States Environmental Protection Agency (EPA) was asked to
conduct additional investigations into the possibility of environmental contamina-
tion in the Hemlock, Michigan area. This report describes the results of that
investigation. A concurrent investigation of the alleged Hemlock health problems,
conducted by the Center for Disease Control, is described in a separate report.
EPA initiated its investigation with a field reconnaisance and the collection
of environmental and animal tissue samples which included:
x 15 water samples from 12 residential sites
* 17 soil samples from 9 residential sites
* Dust samples from 7 vacuum cleaners and 2 furnace filters
* 2 well sediment and 1 septic tank sludge sample
* Samples from the municipal water supply
* Samples from the municipal sewage treatment facility
* Fish samples from 3 Hemlock area creeks and the Tittabawasee River
* Algae samples from one well
" 8 domestic and wild animal tissue samples
-------
The environmental samples were analyzed for numerous parameters including:
arsenic, cadmium, lead, calcium, sodium, bromide, chloride, fluoride, iodide,
pesticides, PCBs, PPBs, dioxin, and various other organic chemicals. A gas
chromatograph/mass spectrometer scan, which has the capability of identifying
virtually an unlimited number of organic compounds, was run on the samples to
detect the presence of organic compounds that could not be detected by any specific
test. Approximately 15UO analyses were conducted on these samples. All analyses
were conducted with current state-of-the-art techniques, using the best technology
available at the time of the study.
Evidence of chemical contamination of the environment was not found other than
elevated levels of sodium chloride (the principal component of table salt),
which can affect the taste of the water. It is believed that ingestion of high
levels of sodium may be correlated with high blood pressure in some individuals
susceptible to hypertension and may be a risk to persons with heart or kidney
disease. Hemlock area residents who had been drinking water with elevated
sodium concentrations and who might be on sodium restricted diets have been
encouraged to contact their physicians about the sodium level in their water.
The concentrations of heavy metals and other contaminants for which health-based
standards exist, were below U.S. EPA's maximum contaminant levels for drinking
water.
Methylene chloride, a solvent commonly used in laboratories, was found in several
water samples including the control sample. The quality control data indicate
that the presence of methylene chloride was due to sample contamination after the
sample was collected, rather than contamination of the water supplies. Subsequent
sampling indicated the absence of methylene chloride in the water.
-------
The soil, sediment, sludge, and dust samples were found to contain several
naturally occurring organic compounds. In addition, traces of several phthalates,
a common component of plastic was found in several samples. The presence of
phthalates was not considered to be indicative of an environmental contamination
problem because of the widespread use of this chemical in paints, varnishes,
rust preventatives, adhesives, cleaning agents, floorwaxes, and plastics.
Traces of other compounds, which have been widely used in consumer products
were found in some of the household dust samples.
Numerous samples were analyzed for dioxin to determine if the area might be
contaminated with that extremely toxic compound. Dioxin was not found in any
samples taken from the Hemlock area. The presence of dioxin could only be
confirmed in a caged fish taken from the Tittabawassee River downstream from
the Oow Chemical Company, a waterway known to be contaminated with dioxin.
Thirty-day fish bioconcentration studies were conducted in two water supplies
to test for the presence of chemicals that could bioaccumulate. The results
of these studies do not indicate that the water supplies were contaminated
with chemicals.
Data on Dow Chemical Company's brine system were evaluated to determine if its
brine reinjection system could contaminate aquifers used for drinking water.
This study did not find evidence that these wells could be responsible for
cross-contamination between geological formations.
In summary, the EPA has performed a comprehensive study to determine if chemical
contamination had occurred in the Hemlock, Michigan area. A wide variety of
environmental samples were analyzed for the presence of an extensive number of
possible contaminants. No evidence of a chemical contamination problem in
Hemlock, Michigan was found.
4
-------
II. INTRODUCTION
Beginning in the fall of 1977, public agencies were receiving reports of human
and animal health problems in the Hemlock, Michigan area. Some residents of
Hemlock believed their health problems were caused by drinking groundwater
contaminated with chemicals. Some claimed, specifically, that chemical contamination
had resulted from Dow Chemical Company's operation of brine wells and reinjection
lines located throughout the area.
For over twenty years, Dow has pumped naturally occurring brine, located approxi-
mately 3,UUO feet below ground level, up to the surface. The brine is then carried
through a network of buried pipes to the Dow plant site in Midland, fourteen miles
north of Hemlock. Uow reports that after salts and minerals are removed from the
brine, it is returned to the underground formations by injection wells. Residents'
claims of contamination were based upon the theory that chemical wastes could have
contaminated the brine during processing and that operation of the injection
wells or occasional leaks occurring in the buried lines may have caused chemical
contamination of upper-lying aquifers used for drinking water supplies.
Between 1977 and 1979, a series of investigations were conducted by the Saginaw
County Health Department and several Michigan State agencies. These agencies
were unable to either identify contaminants in the water supplies or correlate
the use of private well water with health complaints in the Hemlock area.
In 1979 the U.S. Environmental Protection Agency (EPA) was requested to further
investigate the possibility of water contamination and to determine'the possibility
of other chemical contamination in the Hemlock area. To accomplish this, samples
were collected from selected residential sites in the area. Water, residues from
water stills, household dusts, well sediments, garden soils, and uncultivated
soils were among the samples collected. Tissue samples from domestic and wild
-------
animals in the Hemlock area were also taken. To identify chemical contaminants
which may have been present in the spent brine, various samples were collected
from the brine wells and lines, and from soil which may have been previously
subject to brine spills.
Extensive field studies, laboratory analyses, including tests for dioxin, and
bioconcentration studies were conducted. The Dow brine wells and reinjection
system were evaluated for possible cross-contamination between geological form-
ations. The results of these studies are presented in the following pages.
In conjunction with EPA's study, the Center for Disease Control (CDC) conducted
an investigation of human health problems in the area. Representatives from CDC
evaluated the extent of the human health problems reported in the area to deter-
mine if the occurrence of health problems was higher than expected. Results
from the CDC investigation will be presented in a separate report.
-------
III. PREVIOUS INVESTIGATIONS CONDUCTED BY MICHIGAN STATE AND LOCAL AGENCIES
The initial investigation of the Hemlock situation began in October 1977 after
a resident of the Hemlock area reported a high incidence of health problems
in her livestock. At this time a variety of samples and whole specimens of
livestock were submitted to the Michigan Department of Agriculture (MDA) for
analysis and examination. The Michigan Department of Agriculture Laboratory
Division, with assistance from the Animal/Health Diagnostic Laboratory of Michigan
State University (MSU), conducted autopsies and field investigations to determine
the cause of the livestock's problems. Between November 1977 and April 1978,
these laboratories also analyzed samples of well water, raw milk, animal feed,
and various animal tissues and organs.
The samples were analyzed for the following parameters:
1. arsenic 4. copper 7. PCBs
2. lead 5. nitrates 8. PBBs
3. mercury 6. fluoride 9. pesticides
This initial investigation concluded that the animal health effects were not due
to exposure to excessive levels of these nine parameters. Rather, the health
problems were attributed to chronic mastitis of the herd. Evidence was not
found that indicated environmental contamination.
A second investigation was initiated in mid-1978, when residents of the Hemlock
area expressed the additional concern that adverse human health effects might
be associated with environmental contaminants in this area. The residents were
especially concerned about the Dow Chemical Company's operation of brine rein-
jection wells in the Hemlock area. Some residents of that area believed that
toxic chemicals were entering this system from the Dow plant and contaminating
the groundwater used for drinking water supplies. To investigate these reports,
-------
the Saginaw County Health Department and the Michigan Department of Public
Health, Division of Water Supply, collected water samples from the Hemlock area
and examined the construction of four private wells which were suspected of
being contaminated. The construction of these wells was determined to be adequate
and in compliance with the Michigan State regulations. Tests were performed for
twenty-seven parameters in the water samples. All parameters were at levels which
could not be associated with the alleged health effects. The parameters were:
1. hardness (CaCU ) 10. bicarbonate 19. zinc
2. fluoride 11. carbonate 20. cadmium
3. chloride 12. sulfate 21. lead
4. nitrate 13. total solids 22. silver
5. sodium 14. pH 23. arsenic
6. potassium 15. conductance 24. barium
7. calcium 16. iron 25. selenium
8. magnesium 17. manganese 26. mercury
9. zinc 18. copper 27. chromium
In addition, the private wells which were potentially subject to agricultural
runoff were tested for some common pesticides (p,p'DDT; p,p'DDE; okp'DDT).
The results of this investigation, completed in the early part of 1978, did not
indicate environmental contamination of the water which could be linked to the
reported human and animal health effects.
Because some of the area residents desired more extensive testing and research,
they obtained the services of two independent laboratories to analyze further
for contaminants. These residents and their representatives presented results
of a water sample analysis at a meeting in October 1978, in Lansing, Michigan,
with representatives of the Michigan Department of Natural Resources. (MDNR),
Water Quality Division, and representatives from other Michigan State agencies.
The analysis indicated the presence of four organic compounds in the sample:
diethyl ether, 1,1,2-trichlorotrifluoroethane, toluene, and trichloroethylene.
The presence of these chemicals was based on a gas chromatograph/mass spectrometer
(GC/MS) analysis for purgeable hydrocarbons in the sample.
8
-------
A review of this data indicates that there was no quality assurance program
associated with these analyses. That is, there were no quality control samples
shtpped or analyzed to insure that contaminants were not present in the sampling
bottles themselves, or that contamination did not occur during shipment and
analysis of the samples. Because of the lack of quality assurance data and
due to the fact that the contaminants detected are commonly used solvents in
the analytical laboratory, it is impossible to conclude that these results
demonstrate chemical contamination of the sampled groundwater.
To determine if, in fact, the groundwater was contaminated, the MDNR conducted a
water quality investigation in the Hemlock area (Investigation of Groundwater
Quality in the Hemlock Area of Saginaw County, MDNR, April, 1979). This investi-
gation involved water sampling and analysis, surveys to identify potential sources
of groundwater contamination, determination of hydrogeological factors which
might influence groundwater quality, and a static bioassay study for general
toxicity. MDNR tested for the following parameters in the water samples:
"1. Diethyl ether, toluene, freon, and trichloroethylene, the substances
identified. . . as being in one sample from" a private well.
"2. Thirty-nine chemicals listed on the Critical Materials Register, (the
1978 list or a previous register). Thirty-four of these thirty-nine
chemicals are reported to be used or produced at the Dow Chemical facil-
ity in Midland.
3. The presence of organic chemicals using gas chromatography with as
many as three different GC scans per sample.
4. Several inorganic chemicals including heavy metals and salts.
5. General water quality parameters.
6. Total and fecal coliform." (j_d.at, p. 9)
The results from the analyses did not indicate chemical contamination of the water.
However, some groundwater sources contained high salt (sodium chloride) concen-
trations. The concentrations were sufficiently high to affect the taste of the
water.
In addition, the MDNR used a static bioassay study to determine the general
toxicity of the water. The bioassay involved exposing Daphm'a magna, an aquatic
-------
invertebrate, to water from the test locations under specific conditions and
monitoring the Daphnia for toxicological effects. The bioassay results were
inconclusive since the results neither supported or eliminated the possibility
of water contamination.
The MDNR study also included a search for localities where groundwater contamina-
tion could potentially occur. Neither map and file inspections nor site visits
by MUNR staff indicated any potential contamination sites. The study indicated
that factors such as past coal and oil exploration practices or some natural
hydrogeological conditions may affect groundwater quality; however, it appeared
unlikely that any contamination had resulted from natural or artificial sources
located in the deep formations from which brine was extracted. Hydrogeologic
factors which were believed to influence the Hemlock area aquifiers included
naturally occurring salts located in the upper bedrock formations.
The Michigan Department of Public Health (MUPH) and the Saginaw County Health
Department conducted an epidemic!ogical study (MDPH, 1979) to determine the
significance of the alleged human health problems in the Hemlock area. This
study was based on information obtained through interviewing sample populations
in the northern half of Fremont Township and the southern half of Richland Township.
A total of 88 households representing 3U3 residents responded. A comparison popu-
lation was selected from the Blumfield Township, located in the northeast section
of Saginaw County. Seventy-four households, representing 2fa3 residents in this
area were interviewed.
/
The study found more health complaints in the Hemlock area than in the comparison
area. Health problems most frequently reported in the Hemlock area included:
1.) skin rashes 7.) visual problems
2.) numbness 8.) nausea
3.) arthritis 9.) urine sugar
4.) injuries 10.) thyroid problems
5.) dizziness 11.) strokes.
6.) arm, leg, and lower back pains
ID
-------
Differences in reported health complaints could not be attributed to differ-
ences in age or sex between the two areas. Occupational or agricultural exposure
to chemicals and the use of private well water supplies were also considered in
the epidemiological study, but these factors were not associated with differences
in the incidence of health complaints.
In summary, investigations by the Saginaw County Health Department, MDA, MDPH,
and MDNR did not produce evidence of groundwater contamination or indicate the
presence of toxic chemicals in the environment which would cause animal or human
health problems.
11
-------
IV. EPA INVESTIGATION
^A. Introduction
In July 1979, EPA was requested to assist the State of Michigan by conducting
a more extensive investigation to assess whether a contamination problem existed
in Hemlock and the surrounding area. In response to that request, Region V staff
members met with representatives of the MDA, MDPH, MDNR, and the center for Disease
Control to consider plans for further evaluation of the Hemlock problem.
In September 1979, EPA and MUNR representatives jointly conducted a reconnaissance
survey of various sampling sites and homes in the Hemlock area. During the period
from October 1979, to February 1980, representatives from these agencies collected
a variety of environmental samples including water, soil, dust, sediment, and
sludge samples. These samples were collected from 12 residential sites, the public
water and wastewater systems of Hemlock, the Dow brine production and injection
system, and other selected locations near the Dow brine system. Tissue samples
from domestic and wild animals were also collected during this time.
During October, a 30 day fish bioconcentration study was conducted using private
well water from one of the residential sites. A similar bioconcentration study
was conducted in May 1980 using water from the Hemlock public supply. Fish tissue
samples were collected from these studies and analyzed for the presence of low-
level contaminants which would accumulate in the fish but would not be detected
by analyses of the water.
All samples were analyzed using current state-of-the-art methodology. Docu-
mentation of methods used are available through the Toxic Substances Office,
Region V, U.S. EPA, Chicago, Illinois.
To insure the reliability of the data generated, a rigid quality assurance program
was conducted for all samples. Numerous duplicate samples were taken, and blanks
12
-------
(samples which contain no contaminants) were prepared and analyzed. Spiked water,
sediment and tissue samples from U.S. EPA, Region V's Central Regional Laboratory
were shipped to and analyzed by the analytical laboratory. Additionally, in
those cases where questions were raised, such as in some of the volatile organic
samples, sites were resampled.
The environmental and tissue samples were analyzed for a wide variety of
parameters including:
1) Arsenic 8) Fluoride
2) Cadmium 9) Iodide
3) Lead 10) Purgeable organic compounds
4) Calcium 11) Nonpurgeable organic compounds
5) Sodium 12) Pesticides
6) Bromide 13) Polychlorinated biphenyls
7) Chloride 14) Polybrominated biphenyls
These samples were also "scanned" for other organic chemical compounds using gas
chromatography coupled with mass spectrometry (GC/MS).
Tissue samples and selected environmental samples were also analyzed for 2, 3,
7, 8-tetrachlorodibenzo-p-dioxin, commonly known as dioxin, for a number of
reasons. First, dioxin is a highly toxic chemical. Secondly, Dow Chemical
produces certain pesticides, including 2,4,5-T, silvex, and trichlorophenol
at their facility in Midland, fourteen miles north of Hemlock. Dioxin may
be formed as a by-product during the production of these chemicals (Federal
Register, August 2, 1978). And, thirdly, dioxin has been detected in fish
collected from the Tittabawasee, Saginaw, and Grand Rivers in Michigan, in 1978
(EPA study, 1978).
In addition to collecting and analyzing samples from the Hemlock area, EPA staff
evaluated the design and construction methods used by Dow Chemical Company for
their brine production and injection wells. Documents and information supplied
by Dow were reviewed to assess whether the construction of these wells would allow
contamination of upper-lying aquifers.
13
-------
B. Environmental Samples from Residential Sites
Samples were collected from 12 residential sites in the Hemlock area. A map
which is located on page 15, designates the location of these residential sites
and identifies them by site number. Table 1, page 16, gives the number and
types of samples collected from each of these sites.
1. Water
A total of 15 water samples were collected from the 12 residential sites. All
samples were analyzed for nine inorganic chemicals. Results from the analysis
of these samples are given in Table 2, page 17. In general, these chemicals
were at levels which are not known to be harmful to health.
The quality of the water did appear'to be affected at some locations by sodium
chloride. This condition is probably due to natural causes since wells deeper
than the glacial drift layer in this area typically contain saline water
(Michigan Water Resources Commission, 1963). According to U.S. Geological Survey
geologist, Floyd Twenter (personal communication, July 9, 1980), salty water is
frequently found in aquifers of rock formations below the glacial drift layer
and in nearby aquifers. Since the depth of the glacial drift in this area ranges
from 115 to 250 feet (MDNR, 1979), wells near or below the depth of the glacial
drift for a specific area, would tend to be salty due to the natural conditions
of the aquifer. What effect, if any, other factors have had in contributing
to the saline condition, especially in wells of shallow depth, can not be deter-
/
mined with available information.
Dr. Renate Kimbrough, a toxicologist at the Center for Disease Control in Atlanta,
(personal communication July 5, 1980) recommended that well water at site number
7 should not be used for drinking because of the high sodium level. Consuming
two liters of water each day (a typical amount for adults) from this water
supply and ingesting salt from food would result in the ingestion of roughly
14
-------
FIGURE 1
MAP OF HEMLOCK AREA AND SAMPLING SITES
. t
oute 46
derer Rd
Swan Cry
HEMLOCK
Rd.
-o
a:
o
a
on
I
-o
o
U-
s-
o
11
**
A
. F
3
O
Roosevel
Rd.
Nelson Rt
t
N
. 1 mile i
Brine line
Brine injection well
Brine production well
Private Residences (9 is located within Hemlock)
l)-
A - Fish collection point
B - Fish collection point
C - Fish collection point
D - Brine injection well
E - Smith Drain (collection
sediment)
F - Brine injection well
G - Sample collection point
in Marsh Creek
in Williams Creek
in McClellan Run
point of creek bed
from brine line
H - Sandpit near brine line
I - Brine injection well
J - Brine injection well
K - Brine production well
R - Location from which the deer was
taken
S - Location from which the squirrel
was taken
15
-------
CO
UJ
CO
LU
O
14
co
UJ
a:
o
CU C£
IQ O
I UJ
O
UJ
_l
_l
o
CJ
CO
UJ
_J
a.
co
o
^"
en
co
cu
.a
CU
co
CO
CM
in
-
-
-
CM
f""
'-
CO
s_
OJ
fO
3
-
-
-
-
-
-
,-
-
f
^
0
co
^^
3
0
Q.
C
0
Q
-
r-
r-
-
-
f»
r
o
CO
c
0)
13
(T3
CO
-
'-
-
r~
r
O
co
-D
0)
4^
n3
>
*r-
-fr>
U
c
-
-
fm
1
O
CO
CL
3
CO
-
-
-
-
-
j-
-
_^J
Li
^
a
E
3
U
>
CM
~_
3
a
5_
c;
P
r
! *
a>
o
c
3
U_
CM
-
[ *
C
Oi
E
r~
"O
------- An error occurred while trying to OCR this image.
-------
twice the recommended daily salt intake. Based on personal communication on
July 23, 1980 with Dr. Stephen Hessl (Head, Section of Occupational Medicine
at Cook County Hospital), increased sodium intake may be detrimental to persons
suffering from various kidney and heart ailments and hypertension. But, accord-
ing to Dr. Hessl, there is no evidence that a normal, healthy person would
develop health problems from the sodium levels present in the water samples.
Analysis for organic compounds indicated the presence of methylene chloride,
toluene and two types of phthalates in some of the water samples. The presence
of these compounds appeared to be due to laboratory contamination, based on the
control samples and the fact that these compounds were not found when the water
supplies were resampled.
The GC/MS scan identified three organic compounds, which are sometimes associated
with fecal contamination, in five of the water samples. Fecal coliform tests,
which are more reliable indicators for this type of contamination, were con-
ducted by MDPH and MDNR on water from these locations. These tests did not
indicate fecal contamination.
The following material in this section gives a more detailed evaluation of these
water samples.
Arsenic levels
Arsenic levels in the water samples ranged from below detection [5 parts per
billion (ppb)] to 19 ppb. All levels were below the EPA maximum contaminant level
of 50 ppb.
Cadmium levels
Cadmium levels ranged from below detection (1 ppb) to 1.2 ppb and are below
the EPA maximum contaminant level of 10 ppb. During the original analysis of
these samples, the cadmium level for the water sample from site number 11 was
reported to be 70 ppb, and from site number 12 the cadmium level was reported to be
18
-------
17 ppb. Because both of these levels exceeded the EPA standard for cadmium in
drinking water, the residents were advised not to drink the water until the
wells were resampled. The results of the second analysis indicated that
the cadmium levels in the water were normal and below the standard for both
locations. Apparently a laboratory error was responsible for the higher levels
reported initially. MDPH and MDNR also collected and analyzed water from these
two locations to recheck the cadmium levels. Their results also indicated that
the cadmium levels in these two wells were not elevated.
j.ead levels
Lead levels in the water samples ranged from .95 to 18 ppb and are below the EPA
standard of 50 ppb.
Calcium levels
Calcium levels ranged from 29 to 160 parts per million (ppm). There is no
EPA health-related standard for calcium in drinking water. Calcium is an essen-
tial nutrient and, by being in the water, may beneficially contribute to the
body's daily nutritional needs (National Research Council, 1979). Although
calcium in drinking water is not harmful to health, high levels may affect water
by causing it to be hard. Water containing a level of calcium ranging from 75
to 150 ppm is generally considered to be moderately hard and would affect the
quality of the water by interfering with the cleaning ability of certain soaps.
jodiurn levels
Levels of sodium ranged from 49 to 1,175 ppm with an average level of 293 ppm.
The one extreme value of 1,175 ppm was detected in the sample from site number
7 as compared to the second-highest value of 500 ppm detected from site number 11.
There is no EPA standard for sodium in drinking water; however, the American Heart
Association recommends that a person on a sodium-free diet should not continually
drink water containing more than 20 ppm of sodium. It is also recommended that a
person on a moderately restricted sodium diet should not continually drink water
19
-------
containing a sodium level greater than 270 ppm. Doctors will sometimes place
people with heart, circulatory, or kidney illnesses on sodium restricted diets.
Persons whose water supplies had levels of sodium above 270 ppm and who had a
medical history of such illnesses were encouraged to check with their physicians
to determine whether they should limit the amount of this water they drink.
Bromide levels
The bromide levels in the water samples ranged from below detection (less
than 0.01 ppm) to 0.82 ppm. Bromide is seldom found in water at detectable
levels other than sea water, brines, and water affected by salt formations
(Sneed, et. al., 1954). The salt content in the upper bedrock formations
(MDNR, 1979) appears to be the source of the bromide levels in the upper-lying
aquifers. According to Dr. Renate Kimbrough (personal communication, July 7,
1980), the levels of bromide detected in the water samples are not known to
cause health problems.
Chloride levels
The chloride levels ranged from 19 to 1,550 ppm with an average of 314 ppm and
a median of 135 ppm. The one extreme value of 1,550 ppm was detected in the
sample from site number 7. One would expect to find high sodium and high
chloride levels in the same sample since high sodium and chloride levels are
often associated with one another. There is no health-related standard for
chloride in drinking water; however, there is an EPA secondary standard for
chloride of 250 ppm. Secondary standards exist for parameters which affect
the aesthetic quality of the water, such as its taste, odor, color, pr appear-
ance, rather than affect health. Many people object to a salty taste in water
when chloride is above 250 ppm. In addition to affecting its taste, high
chloride levels may cause plumbing to deteriorate more quickly.
Fluoride levels
The fluoride levels ranged from below detection (less than 0.1 ppm) to 2 ppm.
These levels do not exceed the EPA standard of 2 ppm for that geographic area.
20
-------
Iodide levels
Levels of iodide in the water samples ranged from 0.0007 to 0.01 ppm. In areas
with an average iodide content in the soil, the drinking water usually contains
approximately 0.001 ppm (Ermolenko, 1972). Iodide levels in water vary depending
upon the location, soil and geological formations. There is no EPA standard for
iodide in drinking water, and the levels measured in the water samples are not
known to be harmful to health. In fact, iodide levels in water may contribute to
the body's daily nutritional needs. Low-iodide waters (around .001 ppm) can pro-
vide around 1-2% of a person's daily requirements and high-iodide waters (0.018
ppm) can provide from 24 to 44% of a person's daily needs for iodide (National
Research Council, 1979).
Purgeable Organics
The analysis for volatile organic compounds (purgeable organics) indicated the
presence of methylene chloride and toluene, common laboratory chemicals, in several
of these samples. The quality control samples, however, indicated that these
results were due to contamination occurring either during the shipment of the
samples from the sampling location to the laboratory or due to laboratory contami-
nation of the samples prior to analysis. This was confirmed when new samples were
analyzed and yielded negative results.
Nonpurgeable Organics
Two nonpurgeable organics, bis (2-ethylhexyl) phthalate and diethyl phthalate,
were detected in one of the water samples from site number 8 at levels of 11 and
13 ppb. Phthalates are a group of compounds which are commonly found in the envi-
ronment because of their extensive use in the manufacturing of plastics. Conse-
quently, phthalates could originate from many products found in the home. If
the phthalates were actually in the water at the time of sampling, their pre-
sence could be due to the use of plastic being used in the plumbing, such as in a
water softening device or plastic pipes. It is also possible that contamination
21
-------
of the water sample with phthalates occurred after the sample v/as collected.
The possibility of sample contamination is supported by the analysis of a
duplicate sample from site number 8 in which no phthalates were detected.
Pesticides, PCBs, PBBs
The water samples were analyzed for pesticides, PCBs (polychlorinated bi-
phenyls ) and PBBs (polybrominated biphenyls). These chemical compounds were
not detected in any of the water samples at a detection limit of 1 ppb.
Tentatively Identified Compounds
In five water samples, three organic compounds (cholestanol, hexadecanol, and
9-octadecen-l-ol) were tentatively identified. These compounds are found in
fecal material, and, therefore, may be indicative of fecal contamination.
Based on these tentative findings, MUPH and MDNR performed fecal coliform tests
on water from these locations. The fecal coliform test is a more reliable test
for detecting fecal contamination. The results of the fecal coliform tests
indicate that the wells were not contaminated when the tests were made.
2. Water Distillation Residue
Some residents in the Hemlock area had been distilling their water and reported
that once they began this practice, their health seemed to improve. The residues
from the water which remained in the distilling devices were sampled from two
units used at residences numbered 4 and 6. The results from these samples are
given in Table 3, on page 23.
Polychlorinated biphenyls (Arochlor-1254) were tentatively indicated in the residue
from site number 4 at a level of 5.7 ppb. However, due to the low level detected,
it could not be confirmed whether PCB was actually present. Its presence was in-
dicated when analyzed by electron-capture gas chromatography (GC-EC) but it could
not be detected when the sample was reanalyzed by gas chromatography coupled with
mass spectrometry (GC/MS). (GC-EC can detect lower levels of chemicals than GC/MS,
22
-------
Parameters
Table 3
RESIDUE FROM WATER DISTILLATION
UNITS USED AT RESIDENTIAL SITES
Site Number
Arsenic (PPb)
Cadmium (ppb)
Lead (PPb)
Calcium (pom)
Sodium (ppm)
Bromide (ppm)
Chloride PPm)
Fluoride DDITI)
Iodide (ppm)
Purgeable Organics
Nonpurgeable Organics
Pesticides (ppb)
PBBs ippb)
PCBs ippb)
PCBs (Arochlor 1254)
Tentatively
Identified
Compounds
7
11
2
NA
NA
4.8
3820
1.5
0.014
NA
ND
ND
ND
5.7*
cholestanol
hexadecanol
8
13
33
NA
NA
4.5
843
1.7 .
.04
NA
ND
!
ND
ND
ND
cholestanol
hexadecanol
9-octadecen-l-ol
ppb - parts per billion.
ppm - parts per million.
NA - Not analyzed for in the sample.
ND - Not detected 1n the sample.
* - Was detected by GC/EC but could not
be confirmed by GC/MS.
23
-------
but is not as accurate at identifying specific chemicals.) Analysis of the water
sample from this location did not show FCBs to be present in the water.
In addition, a 3U-day, fish bioconcentration study using water from this location
was conducted. If PCBs were present in the water, they should have accumulated
in the fish tissue and would have been detected when analyzed. Since PCBs were
not detected in the fish tissue samples collected from this bioconcentration
study, it can be concluded that PCBs were not present in the water from this
location.
Cholestanol, hexadecanol, and 9-octadecen-l-ol were tentatively identified in
the residue samples. Since these compounds indicated the possibility of fecal
contamination, water samples were collected and tested for fecal coliform bacteria.
Results from these tests did not indicate the presence of fecal coliform.
3. Soi 1
Seventeen soil samples were collected from nine residential sites. These con-
sisted of scrapings up to 2-4 inches in depth from the soil surface. Soil samples
were collected below gutter downspouts to help identify atmospheric contaminants
which may have been present in the air. During periods of rainfall, atmospheric
contaminants would be washed out of the air with the rain and would be concentrated
in the soil at these locations. The single soil sample collected near a basement
sump drain was expected to contain contaminants similar to those found in soil
below gutter downspouts. Garden soil samples and uncultivated soil samples were
collected to determine if chemical contaminants were present in the soil itself or
as a result of long-term deposition from the air, water drainage, or other routes.
The laboratory results from these samples are summarized on Tables 4 and b on
pages 25 and 26, respectively. Because of the large number of additional chemical
compounds identified by the gas chromatography-mass spectrometry scan, these com-
pounds are summarized separately in Table 6 on pages 27 and 28.
24
-------
Table 4
DOWNSPOUT SOIL SAMPLES FROM
RESIDENTIAL SITES
Site Number
PARAMETERS
(ppm)
Arsenic L (Leachable)
T (Total)
Cadmium L
T
Lead L
T
Calcium L
T
Sodium L
T
Bromide L
T
Chloride L
T
Fluoride L
T
Iodide L
T
Crganics
Bis (2 ethyl hexyl)phthal ate
Fyrene
Pesticides
PCBs
PBBs
Tentatively Identified Com-
pounds (quantity detected)
1
3.9
9.0
0.64
1.7
120
120
6,700
7,100
70
4,300
3.2
< 10
800
4,460
12.8
480
0.8
< 1
0.24
ND
ND
ND
6
3
0.31
3.2
0.12
0.40
12
20
13,000
18,000
60
4,400
4
<10
1,800
1,800
20
72
0.2
<0.1
ND
ND
ND
ND
9
4 ! 5
0.33
5
0.12
0.38
14
60
1,200
5,200
50
5,400
4.8
310
-650
670
5.6
<5
0.1
0.3
0.21
ND
ND
ND
3
0.33
0.98
0.12
0.15
4
7.4
2,000
4,700
80
3,700
< 2
<10
4,860
5,000
20
440
0.17
0.25
0.22
ND
ND
ND
4
6
0.51
1.9
1.1
1.8
10
100
5,000
7,200
70
4,700
< 2
<40
500
2,000
3.6
< 5
0.1
1
ND
ND
ND
ND
4
7
0.64
5.4
0.35
1.3
180
170
8,000
9,700
70
4,900
< 2
<10
510
500
6.4
90
<(0.1
<0.1
ND
ND
ND
ND
1
9
0.91
3.2
0.2
0.35
19
30
30,000
29,000
60
4,800
11
180
400
400
8.8
9
0.2
0.2
2
ND
ND
' ND
0
n
0.44
6.8
0.27
0.66
52
65
9,800
11,500
50 '
4,800
9.6
< 10
940
1,000
3.2
< 5
0.02
0.2
ND
ND
ND
ND
4
12
0.24
1.4
0.06
0.17
8
18
470
3,600
50
4,700
6.4
< 10
180
165
26
15
0.8
< 1
ND
ND
ND
ND
1
ND - Not detected in the sample.
ppm , - All values are expressed in parts per million.
< - Actual value, if present, is less than stated value.
L.eachable - Amount extracted from the sample by dilute acid washing.
Total - Total amount in the sample (leachable plus remaining).
25
-------
Table 5
GARDEN, UNCULTIVATED, AND SUMP SOIL SAMPLES FROM RESIDENTIAL SITES
Type of soil sample and location of samples taken by site number
Parameters
(ppm)
Arsenic L (Teachable)
T (total)
Cadmium L
T
Lead L
T
Calcium L
T
Sodium L
T
Bromide L
T
Chloride L
T
Fluoride L
T
Iodide L
T
Organics
Bis(2-ethylhexyl )phthalate
Pesticides
PCBs
PBBs
Tentatively Identified
Compounds (quantity detected)
4
1.4
6.2
0.07
0.23
14
23
2,000
4,000
50
3,600
6.4
200
190
230
3.4
<5
0.1
<0.2
1.5
ND
ND
ND
6
GARDEi\
5
1.4
4
0.15
0.21
9
23
3,200
6,000
70
3,000
6.4
<10
500
500
24
25
0.01
<0.1
ND
ND
ND
ND
3
SOIL
6 ! 9
1.1
5.8
0.13
0.21
12
18
4.200
5,200
40
3,700
1.6
<10
180
200
9.6
10
<0.01
0.4
ND
ND
ND
ND
3
1.3
6.4
0.42
2.1
11
28
18,000
27,000
100
5,700
3.2
170
660
660
8
110
0.3
0.5
ND
ND
ND
ND
2
UNCULTIVATED 5
4 6
2.4
3.5
0.05
0.11
5
6.9
600
1,900
50
2,900
4.8
<10
260
270
10
35
<0.01
0.2
ND
ND
ND
ND
1
0.35
1.3
0.15
0.19
6.9
19
7,500
11,000
70
2,100
6.8
285
1,070
6,440
5.6
5
0.1
<0.5
ND
ND
ND
ND
7
>OIL
12
0.6
3.6
0.09
0.21
3
17
2,700
5,500
70
4,000
11
<10
1,260
1,480
10
10
0.2
0.1
1.1
ND
ND
NO
4
SUMP
SOIL
12
0.93
2.0
0.22
0.53
47
55
6,000
6,000
160
4,400
2.6
<10
270
275
<0.8
<5
0.05
<0.1
ND
ND
ND
ND
4
(ppm) - All values are expressed in parts per million.
< - Actual value, if present, is less than stated value.
Leachable - Amount extracted from the sample by dilute acid washing.
Total - Total amount in the sample (Teachable plus remaining).
ND - Not detected in the sample.
26
------- An error occurred while trying to OCR this image.
------- An error occurred while trying to OCR this image.
-------
Inorganic Elements
teachable and total levels of inorganics were determined during the analyses
of the samples. The Teachable value of the sample was obtained by mixing the
sample with a weak acid, in order to extract the soluble portion of the sample
for separate analysis. The total value given represents the Teachable portion
plus the amount which remained in the sample after the first extraction. In
a few instances, the total value was reported to be slightly less than the
Teachable value. In these instances, the Teachable value was actually the total,
with the slight differences caused by acceptable experimental variation during
the analyses. The evaluation of the levels of inorganics in the samples was
based on the total levels which were detected in the samples.
The exact composition and characteristics of soil in a given location depend on
many factors such as climate, landscape, and the rocks and minerals present in
the area from which the soil was derived (Townsend, 1973). For this reason,
the levels of the inorganic elements will vary for each area. When evaluating
the levels detected, broad ranges rather than averages must be used in assessing
whether the levels detected indicate an excessive amount.
For the inorganic metals (arsenic, cadmium, and lead), the levels detected
in the samples from the Hemlock area are comparable to levels which have been
detected in soil from other areas. Natural background levels of arsenic in
soil range from 1-70 ppm (Bear, 1964) with an average of 6 ppm (Lisk, 1972).
In other locations, higher concentrations up to 158 ppm have been defected
because of the addition of arsenic to the soil (Carey, et. al., 1980). The
arsenic levels in Hemlock area samples ranged from 0.98 to 9 ppm which are
typical of naturally occurring levels.
29
-------
The cadmium levels detected in the samples ranged from U.ll to Z.I ppm. Cadmium
generally ranges from U.U1 to 7 ppm in soil (Bonn, et. al., 1979). Background
cadmium levels in rural soils normally average around U.I ppm, while some agri-
cultural land and soils in urban areas have concentrations around l.U ppm or higher
(EPA Health Assessment Document for Cadmium, 1979). Except for one sample, the
cadmium levels of the garden and uncultivated soil samples were generally at
U.2 ppm, an average background level. The one higher value of 2.1 ppm was detected
in garden soil from a home (site number 9) located in the town of Hemlock. The
higher cadmium level could have resulted from several factors including the
use of phosphate fertilizers in this area, and from other anthropogenic sources
which involve the burning of coal, oil, or gasoline (Lisk, 1972). The downspout
soil samples contained slightly higher levels of cadmium. This would be expected
since airborne cadmium which was deposited on the roof would accumulate at these
locations.
Lead content in soil is often more variable than levels of arsenic and cadmium.
Lead occurs naturally in soils at concentrations from ID to 21) ppm (Swaine,
1955); however, because of the presence of lead in such items as gasoline and
paint, levels are detected in soil ranging from 2U to l.UbO ppm (Solomon, Hartford,
197b). Lead levels detected in the Hemlock samples ranged from b.9 to 17U ppm
and fall in the range of expected values. The higher levels were detected in
the downspout soil samples and are probably due to a lead-based paint which
may have been used on these houses. These findings are in agreement with the
results of Solomon and Hartford (197b) who reported that the highest residential
lead levels were in soil samples collected next to houses and other painted
structures.
Calcium levels in soil samples ranged from 1,9UU-29,UUU ppm which are common
levels. Calcium is a major component of soil. In areas, which have been
3D
-------
affected by limestone deposits, calcium may be at levels of 250,000 ppm,
which is equivalent to 2b% of the weight of the soil (Hausenbui Her, 1972).
More typical values characteristic of loam, silt, and clay soils in the humid
and temperate regions range from 10,000-20,000 ppm (Bear, 19bb).
The measured sodium levels (2,100-5,700 ppm) also appear to be typical.
Bear (1964) reports that the sodium content of many soils ranges from 1,UUU to
1U,UUU ppm, and Vinogradov (1972) reports the average sodiurn content to
be around 7,DUD ppm.
Levels of bromide which are found in soils vary, depending on the organic
content of the soil (Ermolenko, 1972). Most bromide in soil originates
from the soil-forming rocks. In the elemental state it is usually present
in only trace amounts since it is easily leached from the soil (Hesse, 1971).
However, bromide tends to accumulate in organic soils because of the presence
of hutnic substances which bind bromide. Therefore, bromide levels depend more
on the amount of humic substances in a particular area than the type of soil
present. Ermolenko (1972) reports that sandy soils may contain only 0.1-1 ppm
bromide as compared to humus-rich soils, such as peats, which may contain
bU-bUU ppm of bromide. The bromides detected in the soil samples ranged
from below detection (less than 1U ppm) to 310 ppm.
Chlorides were detected in the soil samples at levels from Ibb to 6,44U ppm.
The usual range of chloride levels is bO-500 ppm; however, higher levels are
found in soils with accumulated salts (Hausenbui11er, 1972). The soil samples
collected from the gardens and uncultivated areas have more typical levels of
cnloride, averaging around bbO ppm. One exception to this is the uncultivated
soil sample from site number 4 with a concentration of d,440 ppm. This sample
was collected from a field located near a dirt road. In the past, brine was
-------
sprayed on dirt roads in the county for dust control. This practice may be
a contributing factor in the high chloride levels in this and similar areas.
The higher chloride levels of the soils under the downspouts would most
likely be due to chlorides which have been deposited from the atmosphere.
A common source of atmospheric chlorides is from the combustion of fossil
fuels such as coal and oil which contain varying amounts of chlorine (Eliassen,
1959).
Fluoride, another variable component of the soil, was detected at levels
ranging from below detection (less than 5 ppm) to 480 ppm. These levels appear
normal since fluoride levels typically range from 10 to 1000 ppm (Bear, 1964).
Fluoride is a typical component of certain minerals common in soil.
Iodide levels detected in the samples (less than 0.1-1 ppm) were below levels
typically found in soil. Bear (1964) reports the normal range for iodide in
soils to be 0.6 to 8.0 ppm.
Organic Compounds
Bis (2-ethylhexyl) phthalate was detected in five of the soil samples. This
compound is widely distributed in the environment because of its widespread
use in products such as paints, varnishes, and rust preventatives (Initial
Report of the TSCA Interagency Testing Committtee, 1978). The widespread use
of this chemical compound may be the cause of its presence in the soil samples.
Fyrene, another organic compound, was detected in downspout soil fro/n site
number 1. Pyrene belongs to a class of compounds known as polycyclic aromatic
hydrocarbons and is most often found as a by-product of incomplete combustion
of organic matter (Hoffmann, Wynder, 1968). The most likely source of
pyrene at this location would be from the wood-burning stove used in the
-------
house. The smoke produced from the stove would contain pyrene and similar
compounds and would distribute these compounds in the vicinity of the house.
Tentatively Identified Compounds
Thirty additional compounds were tentatively identified in various soil
samples by the GC/MS scan and are listed in Table 6, on pages 21 and 28. The
types of compounds identified are those which are commonly found in soil or would
be expected to be present in soil samples. Chemical compounds which would in-
dicate a contamination problem were not detected in these samples.
Most of these compounds probably originated from living organisms, either
directly from the organisms or from the natural breakdown of their components
or by-products. For example, dead plant tissue will undergo a complex decompo-
sition process by which its components will be chemically changed or broken
down in the soil. Buckman and Brady (1960) outlined the stages of decomposition
during which the original compounds in plant tissue (eg., oils, fats, resins,
starches, sugars, proteins) will eventually become the complex organic material
in soil known as humus. During this decomposition, various compounds will be
formed and decomposed. These compounds could account for some of the alcohols,
aldehydes, carboxylic acids, and other types of compounds in the soil.
Small microorganisms themselves, such as bacteria, would also contribute to
the presence of these organic compounds. Average soils may contain from 1U -
1UO million bacteria per gram of soil (Allison, 1973). These bacteria contain
many biochemical compounds such as the lipids in which a variety of'carboxylic
acids have been identified (Shaw, 1974).
Several of these compounds may have originated from a single natural compound
present in the soil. It has been determined that microbial activity on a
33
-------
specific organic compound will yield a variety of degradation products. For
example, carboxylic acids may undergo the following reactions when being decom-
posed by microorganisms (Meikle, 1972):
1) shortening of the carbon chain by two carbons through beta-oxidation
which would produce a different carboxylic acid.
2) removal of the carboxyl group producing an alkane or alkene.
3) replacement of the carboxyl group by a hydroxy group resulting in an
alcohol.
Consequently, the breakdown of hexadecanoic acid, for instance, may produce tetra-
decanoic acid, pentadecane, hexadecanol, and other products. Similar general re-
actions are known to occur with alkanes, alkenes, ketones, esters, and other types
of organic compounds.
4. Dust
Seven vacuum cleaner dust samples and two furnace filter dust samples were col-
lected from nine of the residential sites. These samples were collected to
identify contaminants which may have been deposited within homes from outside
sources. The results of the analyses of these samples are contained in
Table 7, on page 35.
The arsenic content of these dusts do not appear high compared to the results
from one study in which 61 household dust samples were analyzed for arsenic (Klemmer,
et. al., 1975). Klemmer found levels of arsenic ranging from 1.1 to 1,080 ppm.
The higher values were found in homes which had been treated for termites or con-
tained lumber which had been treated with certain preservatives. The arsenic
levels in the Hemlock samples ranged from 2.3 to 21 ppm.
/
Cadmium levels in these samples ranged from 2 to 25 ppm. These cadmium concen-
trations appear to be typical based on a study by Solomon and Hartford (1976) in
which household dust was found to contain cadmium levels ranging from 7 to 48 ppm
with an average of 18 ppm.
34
-------
Table 7
DUST SAMPLES FROM RESIDENTIAL SITES
Type of Dust Sample and Location of Sample
Taken by Site NumbeF
Vacuum Cleaner
Furnace
Filter
Parameters
(ppm)
Arsenic L(Leachable)
T(Total)
Cadmium L
T
Lead L
T
Calcium L
T
Sodium L
T
Bromide L
T
Chloride. L
T
Fluoride L
T
Iodide L
T
Organic Compounds
Bis(2-ethylhexyl)phtha-
ate
Butyl benzyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Di-n-octyl phthalate
Phenol
Fluoranthene
Napthalene
Pyrene
Pentachlorophenol
PBBs
PCBs
Arochlor 1232
Arochlor 1260
Pesticides
Chlordane
Tentatively Identified
Compounds (quantity
detected)
1
6.2
21
5.1
9.4
77
110
11,000
25,000
6,800
15,200
36
4250
11,300
34,000
22
425
1.2
<2.4
160
50
6.6
33
26
3.3
4.1
0.6
ND
ND
ND
13
3
2.4
2.7
10
L 11
230
280
100,000
110,000
4,500
11,000
-CO. 8
460
1,500
23,300
14
5
0.1
/2.5
280
61
1
10
0.4
ND
0.45*
ND
10
4
2.5
5.9
4.3
4.0
60
62
30,000
30,000
4,400
7,200
50
4.250
3,060
146,000
30
<50
2
-^.2.5
800
35
4
19
ND
1*
0.3*
10
5
2.2
2.6
17
25
170
280
75,000
83,000
4,600
9,200
<4
4150
18,500
77,500
9.6
00
1.0
1.9
24
4.3
1.1
ND
2.4*
ND
9
7
5.4
2.9
20
21
230
220
48,000
66,000
3,500
5,700
32
<120
2,900
9,700
5.6
^5
1.1
4.1.2
100
31
1.8
14
.92
1
28
1.7
ND
8.5*
"6.6*
ND
11
10
4.6
4.9
6.9
20
56
61
20,000
25,500
10,000
12,000
64
4500
29,500
87,000
7.2
5
1.2
^5
240
12
ND
ND
ND
3
11
2.5
2.3
3.4
3.6
37
46
23,000
32,000
68,000
86,000
44
4200
13,200
52,000
16
^20
-40.04
-£2.1
49
5.4
0.42
6.6
0.42
0.3*
i
10
6
5
14
3.3
3.7
50
51
7,000
72,000
2,600
55,000
8
1,500
2,420
2,610
14
420
0.2
^1.5
1,630
91
20
610
5*
4*
7
9
1.2
4.7
2.2
2
140
170
7,500
53,000
2,200
78,000
24
25
1,970
1,970
5.6
6
0.24
4.8
42
3
0.35*
11
ppm - All values are expressed in parts per million.
ND - Not detected in the sample.
^ - Actual value, if present, is less than stated value.
* - Was detected by GC/EC but could not be confirmed by GC/MS.
Leachable - Amount extracted from the sample by dilute acid washing.
Total - Total amount in the sample (Teachable plus remaining).
35
-------
Lead concentrations in the Hemlock dust samples ranged from 4o to ZtiO ppm. These
levels are slightly lower than those reported by Solomon and Hartford (I97b) who
found average levels of bUU ppm with a range of 17U to 1.44U ppm.
Among the organic compounds detected, the phthalate compounds were the most com-
mon. Their presence is expected because of the number of household products which
would contain such compounds. Perhaps the largest use of these compounds is in
manufacturing plastics. Some plastics may contain up to t>U% phthalates
(EPA quality Criteria for water, 1976). Uther products containing phtnalates
which would make them common in the home include:
paints adhesives
varnishes cleaning agents and compounds
rust preventatives household aerosols
flame retardant chemicals floorwaxes
chemical deodorizers
(Initial Report of the FSCA Interagency Testing Committee, 1978)
Phenol was detected in two of the dust samples. It is an organic compound
which can originate from many sources including chemical oxidation processes,
human and otner organic wastes, microbial degradation of pesticides and other
naturally occurring sources (EPA quality Criteria for Water, 1976). The large level
of phenol detected in the furnace dust from site number 6 may have originated
from the filter itself. To collect the dust, the fiberglass material contained
in the filter had to be removed and included in the sample. Phenol, which is
also, used in many chemical processes, could have been used in the manufacturing
of the filter or have been in a coating which may have been present on the filter.
In two dust samples, three compounds normally associated with by-products of
incomplete combustion were detected in small amounts. These compounds are fluor-
anthene, napthalene and pyrene. Although these compounds are widely distributed
in the environment and can be found in the air, water, soil, sediments, and plant
-------
and animal tissues (EPA Ambient Water Quality Criteria, PAH), their presence at
these detectable levels suggest that they were formed during combustion processes
in these homes. At site number 1, a wood-burning stove was used occasionally.
This stove is a likely source for these compounds. Although a specific source
was not identified at site number 7, these compounds could have been produced
from a variety of sources such as the burning of methane gas (Hoffman, Wynder,
1968), coal, oil and wood. The compounds can also be found in cigarette smoke
(EPA Ambient Water Quality Criteria, Fluoranthene).
Pentachlorophenol, a chemical compound commonly used as a pesticide and as a
wood preservative, was found in trace amounts in three of the dust samples.
Additional sources of pentachlorophenol in the home environment may include such
products as:
adhesives textiles
oils carpet shampoos
paints fabrics
rubber (Conklin, Fox, 1978)*
These products are sometimes treated with pentachlorophenol because of its effec-
tiveness as a fungicide and bactericide, and in preventing mildew.
Two types of PCBs (polychlorinated biphenyls), designated as Arochlor 1232 and
Arochlor 1260, were identified by electron capture gas chromatography (GC/EC) in
7 of the dust samples at levels ranging from 0.3 to 8.5 ppm. However, these
compounds could not be detected by gas chromatography/mass spectrometry when
the samples were reanalyzed. These PCBs, if actually present in the samples,
are believed to be at background levels for dust. PCBs were once widely used
in many products and would still be found in some products within the home.
Arochlors 1232 and 1260 were used in adhesives, surface coatings, paints,
sealants, and textiles, and as a de-lustering agent for rayons (Interdepartmental
Task Force on PCBs, 1972). Other possible sources of PCBs in the home environment
37
-------
include older fluorescent lights, and appliances and electrical equipment which
nave components containing HCtfs (MacLeod, iy79).
Jr. Thomas 1'iurphy, of UePaul university, recently completed a study in which
dust samples collected from homes and buildings in the Lake Michigan area were
analyzed for HCds. Preliminary results (letter of April Zb, lyyu from or.
1'iurphy) snowed HCd levels in the dust samples to range from 1.2 to bd ppm with a
median level of 9 ppm. In an unpublished report, Dr. Murphy states that these
levels of HCds in dust appear to be representative of background levels of
found in the air from various sources.
one pesticide, chlordane, was detected by UC/EC in two of the dust samples. The
presence of chlordane in the samples could not be confirmed by bC/MS analysis.
Even if present, the levels of cnlordane detected in the two samples did not
appear to be high, une study (Starr, et. al. ±y74) found chlordane in 4b house-
hold dust samples at levels ranging from 1.7y to 41.30 ppm with an average of
7. by ppm.
Twenty-two additional compounds were tentatively identified in the dust samples.
These are listed in Table b, on pages 21 and kid. As was discussed on pages 33 and
34 in the section on the soil samples, these compounds most likely originated from
natural organic matter present in these types of samples. Chemical compounds indi-
cating a contamination problem were not detected.
b. Sediment and sludge
Sediment samples were collected from two wells, no longer in use, from sites
numbered b and /. Also a sludge sample was collected from the septic tank
located at site number 12. The results from these samples are listed in
Table d, on page 39. The levels of inorganics found in these samples do not
appear to be significant or to indicate a contamination problem.
3d
-------
Table 8
SEDIMENT AND SLUDGE SAMPLES FROM RESIDENTIAL SITES
Site Number
PARAMETERS
(ppm)
Well Sediment
6 7
Septic Tank Sludge
12
Arsenic L (Leachable)
T (Total)
Cadmium L
T
Lead L
T
Calcium L
T
Sodium L
T
Bromi de L
T
Chloride L
T
Fluoride L
T
Iodide L
T
Organic Compounds
Bis(2-ethylhexyl) phthalate
Butyl benzyl phthalate
Diethyl phthalate
Di-n-butyl phthalate
Acenaphthalene
Benzo ( a ) anthracene
Chrysene
Fluoranthene
Benzo (b+k) fluoranthene
Pyrene
Benzo(a)pyrene
Pesticides
PCBs
PBBs
Tentatively Identified Compounds
(quantity identified)
10.8
9.2
0.33
0.28
12.7
18
46,000
59,200
890
6900
<3.3
<80
42,800
42,800
<80
<80
0.89
<0.8
ND
ND
ND
ND
3
8.2
44
3.3
2.9
58
56
12^000
13,800
960
6460
<110
<110
470
10,100
40
89
<1
<1
0.49
0.24
1.6
2.3
1
1.1
2
1.3
ND
ND
ND
4
4.6
9.8
1.2
3.1
11
67
5670
82,500
1180
2100
<10
-------
ot the organic compounds detected in the samples, 4 phthalates were found
in the septic tank sludge. The presence of these compounds in sludge is probably
due to the fact that phthalates are contained in many household products.
In the well sediment from site number 1, a pnthalate and seven other compounds,
known as polycyclic aromatic hydrocarbons, were found. The most likely source
of these compounds is contamination from surface water runoff. A cracked con-
crete cap was used to cover the well. This would be inadequate protection
because tne crack would have permitted water runoff to enter the well.
Additional compounds identified by the JC/i'iS scan, listed in Table b, on pages
-------
C. Samples From Hemlock's Municipal Water And Sewage Treatment Systems
1. Water
The results of the water sample collected from the Hemlock municipal system
are listed in Table 9, on page 42.
The levels of the inorganics present in the water sample did not exceed existing
EPA standards. The inorganic parameters, for which EPA does not have standards,
were not at levels known to be harmful to health. Each of these inorganics
in water were discussed previously beginning on page 14.
The organic compound, methylene chloride, was detected at a level of 25 parts
per billion. This compound was detected in several of the other water samples
collected (See page 21). Quality assurance procedures used indicate that its
presence is due to laboratory contamination.
Three organic compounds, which sometimes indicate fecal contamination of the
water, were detected by a GC/MS scan. However, fecal contamination was not in-
dicated by fecal coliform tests performed by MDPH.
2. Sludge
A sludge sample was collected from the sewage treatment facility in Hemlock.
The results from this sample are listed in Table 9, on page 42.
The results of the analysis of the sludge sample does not indicate evidence
of a contamination problem. Levels of arsenic and lead appear to be low
according to Lisk's data on sewage sludge (1972), which found the average con-
centration of arsenic to be 90 ppm and lead to be 1,690 ppm. The cadmium level
also appears to be low since cadmium levels in sewage have been found to average
16 ppm, and range from 3 to 3,UOO ppm (CAST, 1976). The levels detected for the
remaining inorganics are believed to be typical.
41
-------
Table 9
WATER AND SLUDGE SAMPLES FROM
HEMLOCK'S MUNICIPAL SYSTEMS
WATER
Inorganics
Arsenic
Cadmi urn
Lead
Calcium
Sodium
Bromide
Chloride
Fluoride
Iodide
Inorganics
Arsenic
Cadmium
Lead
Calcium
Sodium
Bromide
Chloride
Fluoride
Iodide
7 ppb
0.6 ppb
2 ppb
1 04 ppm
75 ppm
< 0.01 ppm
48 ppm
0.5 ppm
< 0.0003 ppm
SLUDGE
(All values are in parts
Leachable Total
7.4 11
1.3 0.83
16 27
43,000 45,800
890 7,700
< 2.9 < 2.9
192 42,400
< 14 < 14
< 0.3 < 0.3
Organics
methyl ene chloride 25 ppb*
Pesticides ND
PCBs ND
PBBs ND
Tentatively Identified Compounds
cholestanol
9-octadecen-l-ol
hexadecanoic acid, 2-oxo, methyl ester
per million)
Organics
bis(2-ethylhexyl) phthalate 4.9
di-n-butyl phthalate 0.86
Pesticides ND
PCBs ND
PBBs ND
Tentatively Identified Compounds
eicosanol
heptadecanol
undecane
2-butanone
tetradecanoic acid
hexadecanoic acid
heneicosanoic acid
benzeneacetic acid
ppm - parts per million
ppb - parts per billion
ND - Not detected in the sample
* - Value may be attributed to laboratory contamination.
< - Actual value, if present, is less than stated value.
Leachable - Amount extracted from the sample by dilute acid washing.
Total - Total amount in the sample (Teachable plus remaining).
42
-------
The detected organic compounds included two phthalates. Because of their wide-
spread use in many products, their presence is expected. The remaining com-
pounds which were identified by the GC/MS scan are thought to originate from bio-
logical sources and are not known to indicate a contamination problem.
43
-------
D. Biological Samples
Plant, fish and other animal tissue samples were collected from the Hemlock
area and analyzed. The results from these samples are listed in Table 10,
on pages 45 and 46. Also in this table are numbers or letters which designate the
location from which the samples were collected. These numbers or letters are
plotted on the map located on page 15.
1. Msh
Fish samples were collected from Marsh Creek, Williams Creek, and McClellan
Run in the Hemlock area. Additional fish samples were taken from two loca-
tions on the Tittabawassee River near Midland, Michigan. The fish collected
from the Tittabawassee were caged fish which were being monitored by the MDNR.
Une location on the Tittabawassee was upstream from the Dow Company plant site
and the other was downstream from Dow. The caged fish from the Tittabawasee
were included in this study to determine if chemical contamination was still
present in fish downstream of Dow Chemical. The MDPH previously had issued a
warning against consuming fish from this river because of chemical contamination.
The levels of arsenic, cadmium and lead found in the fish were not excessive.
Lisk (1972) reports average levels of these metals in fresh water fish to be
O.U3-0.5 ppm, 0.02-0.15 ppm, and 0.5-2 ppm, respectively.
Phenol, an organic compound, was found in 4 of the samples. The presence
of phenol in fish tissues could be due to a number of reasons since it is
a compound widely found in the environment from both natural and anthropogenic
sources. One source of phenol is the microbial breakdown of certain pesticides.
However, this appears to be an unlikely source, in this case, because no pesticides
were found in the fish. If pesticide degradation was the source of phenol, the
pesticides would have accumulated and been found in the fish tissues. Other pos-
sible sources of phenol in fish are natural sources. The natural sources
44
-------
Table 10
PLANT AND ANIMAL TISSUE SAMPLES
Ffsn Fish Fish Fish Fish Fish Goose
II #2 13 14 15 16 Fat
Location
Parameters
Arsenic
Cadmium
Lead
Organic Compounds (DDITI)
phenol
di-n-butyl phthalate
diethyl phthalate
Pesticides (ppm)
PCBs (ppm)
PBBs (ppmj
Oioxin fpotl
Tentatively Identified Compounds
(Alcohols)
heptadecano!
4-methyl phenol
glycerol
cholest-5-en-3-ol
(Aldehydes)
tetradecanal
octadecanal
octadecenal
2,4,nonadienal
(Carboxylic Acids)
butanoic acid
3-methyl butanoic acid
octanoic acid, methyl ester
8-methyl decanoic acid, methyl ester
undecanoic acid, methyl ester
cyclopentane undecanoic acid, methyl ester
tetradecanoic acid
tetradecanoic acid, methyl ester
pentadecanoic acid, methyl ester
14-methylpentadecanoic acid, methyl ester
hexadecanoic acid
hexadecanoic acid, methyl ester
14-methyl hexadecanoic acid, metnyl ester
15-methyl hexadecanoic acid, methyl ester
heptadecanoic acid, methyl ester
16-metnvlheptadecanoic acid, methyl ester
octacecenoic acid
octaisce^oic acid, methyl ester
9-octadecenoic acid
10-octadecenoic acia, .retnvi este^
14-octadecenoic acid, methyl ester
17-octadecenoic acid, methyl ester
10,13-octadecadienoic acid, methyl ester
eicosanoic acid, metnyl ester
heneicosanoic acid, methyl ester
Other Compounds
LJ-dodecanediol diacetate
nonadecane
1 ,1-dimethoxyhexane
benzoic acid
NA
NA
NA
NA
0.17
NO
NO
NO
NA
X
X
X
X
X
A
<0.15
0.06
IC0.02
0.18
0.07
NO
NO
NO
NO
X
X
X
X
B
<0.15
0.023
< 0.02
0.02
NO
NO
NO
NO
X
X
X
X
1
X
c
<0.15
0.026
<0.02
0.9
NC
NO
NO
NO
X
X
X
X
X
X
X
X
X
c
<0.15
0.037
o.n
2
MO
NO
NO
ND
X
X
*
-------
Table 10
(continued)
Goose
Liver
Location
Parameters
Arsenic
Cadmium
Lead
Organic Compounds (com)
phenol
di-n-butyl phthalate
diethyl phthalate
Pesticides (ppm)
PCBs (ppm)
PBBs [ppm)
Dioxin (put,)
Tentatively Identified Compounds
(Alcohols)
heptadecanol
4-methyl phenol
4
<0.15
0.36
0.10
no
1
ND
NO
HO
NO
Chicken Chicken Chicken Chicken Cow Cow Cow Deer Deer Squirrel Squirrel
Fat t Fat Liver Liver Fat Liver Kidney Fat Liver Muscle Oraans
5 i 5
<0.5
<0.004
0.29
ND
<0.5
0.006
0.25
ND
ND ND
ND ND
ND , ND
ND ND
glycerol
cholest-5-en-3-ol X
X X
(Aldehydes)
tetradecanal
octadecanal
octadecenal
2,4,nonadienal
(Carboxylic Acids)
butanoic acid
3-methyl butanoic acid
octanoic acid, methyl ester
8-methyl decanoic acid, methyl ester
undecanoic acid, methyl ester
cyclopentane undecanoic acid, methyl ester
tetradecanoic acid
tetradecanoic acid, methyl ester
X
X
i
5 5 6
<0.15 <0.15
0.34 i 0.62
0.15 0.12
ND ND
ND NO
ND ND
ND ND
ND ND
X X
1
x ; x
pentadecanoic acid, methyl ester
14-methylpentadecanoic acid, methyl ester X
hexadecanoic acid X
hexadecanoic acid, methyl ester
1 4-methyl hexadecanoic acid, methyl ester
15-methyl hexadecanoic acid, methyl ester
heptadecanoic acid, methyl ester
1 6-methyl heptadecanoi c acid, methyl ester
octadecenoic acid
octadecenoic acid, methyl ester
9-octadecenoic acid
lO-octadecenoic acid, methyl ester
17-gctadecenoic acid, methyl ester
eicosanoic acid, methyl ester
heneicosanoic acid, methyl ester
Other Compounds
1 ,1-dodecanediol diacetate
nonadecane
1 .1 -dimethoxyhexane
benzole acid
X X
i
X
X
! X X
X
"1 X
: i
<0.15
0.037
0.11
0.13
ND
ND
NO
6 ! 6 R R
1
<0.15
2.2
<0.15
0.70
0.23 0.13
.75
ND
ND
ND '
ND ! ND
ND
ND
ND
ND
ND
<0.5
<.004
< .05
0.06
0.02
ND
ND
ND
ND
<0.15
0.11
0.13
.02
ND
ND
ND
ND
X 1 X
X
S S
<0.15
0.008
0.14
<0.15
0.22
1.1
1 .86
ND
ND
ND
ND'
X
X
X
X
X , X
j_ X
X X
X X
X
X
1
' X
X X
X j
X
X
X |
X X
X X 1 X
X X
X X
X
X XX
i X
X X
X
X
X
X
X
X
x
ppm - parts per million.
ppt - parts per trill ton.
NA - Not analyzed for in the sample.
ND - Not detected in the sample.
< - Actual value, if present, is less than stated value.
* - Caged fish from upstream Tittabawassee.
** - Caged fish from downstream Tittabawassee.
X - Indicates the compound was tentatively identified in the sample.
46
-------
include certain aquatic plants which produce phenol and the decomposition
of vegetation such as oak leaves, which enter the water (Hoak, 1957). Another
potential source may be industrial discharges.
Two phthalates were also present in three fish samples. As mentioned previously,
phthalates are used in many products and have been found throughout the environ-
ment and are commonly present in waterways and aquatic life (EPA Initial Report
of the TSCA Interagency Testing Committee, 1978). The levels of phthalates
found in these fish samples are not believed to be unusual. One type of
phthalate was reported to be present in fish, collected from various locations
across the country, at levels ranging from 0.2 to 10 ppm (Stalling et. al., 1973),
Dioxin was detected in a fish sample from site number 6, at a level of 23 parts
per trillion. This sample was from the caged fish located downstream from
Dow on the Tittabawassee. In 1978, 2b out of 35 fish samples, collected from
the Tittabawassee, Saginaw, and Grand Rivers, contained dioxin (EPA Memo on
Dioxin, December 20, 1978). Levels detected in the fish from the 1978 study
ranged from 4 to 695 parts per trillion. Dioxin was not detected in the other
fish samples.
A number of additional organic compounds were identified in the fish samples
by the GC/MS .scan. These compounds, are believed to be part of the normally
occurring background compounds which one would expect to be present in these
samples. They are not known to indicate chemical contamination of the fish.
i
2. Animal Tissues
Tissue samples were analyzed from a goose, two chickens, three cows, a deer
and a squirrel, taken from the Hemlock area. Arsenic was not detected in
any of the samples. Based on personal communication (July 2, 1980) with Val
Beasley, D.V.M., of the University of Illinois, College of Veterinary Medicine,
47
-------
the levels of cadmium and lead found were at background levels and do not indicate
a contamination problem. Dr. Beasley commented that the highest cadmium level
detected in cow liver (2.2 ppm) was slightly above levels normally found. However,
he believed that it was not high enough to cause health problems in cattle.
The organic compound, phenol, was detected in 5 of the tissue samples. The
phenol levels detected are not uncommon according to Dr. Fred Oehme (personal
communication on July 2, 1980), who is a veterinarian-toxicologist at Kansas
State University. Dr. Oehme stated that phenolic compounds are naturally
found in animal and plant tissues. Also, levels of phenol can increase in
animal tissues if the animal feeds on lush vegetation.
Diethyl phthalate was also found, in trace amounts, in the deer fat sample.
Because of the widespread distribution of phthalates in the environment, it
does not seem unusual that this compound would be found in some animal tissues.
Other chemical compounds which were detected in the samples appear to be back-
ground compounds which naturally occur in tissues. Pesticides, PCBs, PBBs, and
dioxin were not detected in these samples.
In addition to those samples listed in Table 10 on pages 45 and 46, two more
cow fat samples from site number 6 were analyzed for dioxin. Dioxin was not
found in either of these samples.
3. Plant
Algae, growing at the base of a water well at residence number 6, were analyzed
for organic chemical compounds. The results of the analysis are contained in
Table 10, on pages 45 and 46.
Evidence of chemical contamination was not found in the plant sample. The
phenolic and carboxylic acid compounds detected are believed to be naturally
present in plant tissues.
48
-------
E. Evaluation of Goose Wing Deformity
A goose from site number four was born with a deformity of one wing which made
it appear to be attached backwards to the body. The goose was autopsied and
tissue samples were analyzed (Table ID, pages 45 and 46) to determine if a
chemical contaminant may have caused this deformity. The goose was examined
by a Michigan State University veterinarian, and according to the autopsy
report (November 28, 1979), the deformity was actually an increased flexibility
of the joints close to the base of the wing. This allowed the wing to rotate
to the extent that the feathers appeared to point in the wrong direction when
the wing was flexed. Based on personal communications (July 2, 1980) with
Dr. Val Beasley (a veterinarian from the University of Illinois, College of
Veterinary Medicine), this type of deformity may have been caused by a mal-
position of the unhatched bird during the development process, a genetic dis-
order, or a chemical contaminant. However, chemical contaminants were not
observed in the tissues analyzed. According to Dr. Roland Winterfield (personal
communication, July 2, 1980), a specialist in avian diseases at Purdue University,
the wing disorder is most likely due to a genetic defect. Dr. Winterfield stated
that it was most likely not related to a chemical contaminant since other
deformities or problems were not found in the goose.
49
-------
F. Fish Bloconcentration Studies
Two fish bioconcentration studies were conducted in the Hemlock area. These
studies involved placing fathead minnows in tanks, filled with water from the
test sites, for a 30-day period. Two fish samples were collected before the
studies began (day 0) to serve as controls, and on day 15 and day 30. These
samples were analyzed to determine if trace chemical contaminants were present,
which would accumulate in the fish from the water.
The first bioconcentration study was performed during October, 1979, using the
well water from site number 4. The second fish bioconcentration study was
conducted during April, 1980, using water from the Hemlock municipal supply.
The results of the fish sample analyses are listed in Table 11(a), on page 51,
and in Table 11(b), on page 52. Based on the chemical compounds identified
and the levels detected in the control and the duplicate fish samples, there
appear to be no significant differences in the levels or types of chemicals
identified. The bioconcentration studies did not indicate contamination of the
water.
50
-------
Table ll(a)
FISH TISSUE SAMPLES FROM 30 DAY
BI CONCENTRATION STUDY USING
WATER FROM RESIDENTIAL SITE #4
Parameters
Control Control
Day 0 Day 0 Day 15 Day 15 Day 30 Day 30
Inorganics (ppm)
Arsenic
Cadmium
Lead
Organic Compounds (ppm)
phenol
diethyl jJhthalate
di-n-butyl phthalate
Pesticides (ppm)
PCBs (ppm)
PBBs (DOITI)
Dioxin (ppt)
Tentatively Identified Compounds
glycerol
cholest-5-en-3-ol
tetradecanal
octadecanal
decanoic acid, methyl ester
tetradecanoic acid
tetradecanoic acid, methyl ester
12-methyl tetradecanoic acid, methyl ester
14-methyl pentadecanoic acid, methyl ester
hexadecanoic acid
hexadecanoic acid, methyl ester
14-methyl hexadecanoic acid, methyl ester
15-methyl hexadecanoic acid, methyl ester
hexadecenoic acid, methyl ester
10-methyl .heptadecanoic acid, methyl ester
16-methyl heptadecanoic acid, methyl ester
10-octadecenoic acid, methyl ester
11-eicosenoic acid, methyl ester
heneicosanoic acid, methyl ester
tridecane
0.32
0.068
0.81
1.0
ND
ND
ND
ND
X
X
X
40.15
0.044
0.18
0.6
ND
ND
ND
ND
X
X
X
X
X
X
0.37
0.032
0.13
0.23
ND
ND
ND
ND
X
X
X
X
X
1.0
0.080
0.13
ND
ND
ND
ND
ND
X
X
X
X
X
'
0.53
0.13
0.12
ND
ND
ND
ND
ND
X
X
X
X
0.83
0.077
0.15
0.17
0.3
2.7
ND
ND
ND
ND
X
X
X
X
X
X
X
X
X
ppm - parts per million.
ppt - parts per trillion.
ND - Not detected in the sample.
< - Actual value, if present, is less than stated value.
X - Indicates the compound was tentatively identified in the sample.
51
-------
Table ll(b)
FISH TISSUE SAMPLES FROM 30 DAY
BI CONCENTRATION STUDY USING WATER FROM
HEMLOCK MUNICIPAL WATER SYSTEM .
Parameters
Control
Day 0
Control
Day 0
Day 15 Day 30
Blank
Inorganics (ppm)
Arsenic
Cadmi urn
Lead
Thallium
Organic Compounds (ppm)
phenol
diethyl phthalate
di-n-butyl phthalate
Pesticides (ppm)
PCBs (ppm)
PBBs (ppm)
Dioxin (ppt)
Tentatively Identified
Compounds
tetradecanoic acid
hexadecanoic acid
9-hexadecanoic acid
octadecanoic acid
0.16
0.33
0.88
<0.6
ND
ND
ND
ND
ND
X
X
X
X
0.21
0.45
1.3
<0.6
ND
ND
ND
ND
ND
X
X
X
X
0.20
1.1
0.94
<0.7
ND
0.4
0.7
ND
ND
ND
ND
X
X
X
X
0.33
0.33
-CO. 4
<0.4
ND
ND
ND
ND
ND
X
X
X
X
0.66
0.01
0.14
<0.07
ND
ND
ND
ND
ND
ppm - parts per million.
ppt - parts per trillion.
ND - Not detected in the sample.
< - Actual value, if present, is less than stated value.
X - Indicates the compound was tentatively identified in the sample.'
52
-------
G. Samples From Dow's Brine System And Areas Located Near It
Samples were collected from the brine system and areas located near it which
may have been affected by past spills or leaks in the system. The locations
and descriptions of the samples collected are indicated in Table 12, page 54,
and the sample results are listed in Table 13, pages 55 and 56.
Four brine samples were collected from different locations along the system.
The levels of the inorganic parameters detected, appear to be characteristic
of the brine and are not considered significant. Two organic compounds, which
probably resulted from background or laboratory contamination, were detected
in the brine samples. Methylene chloride, detected in one of the brine samples,
was also found in control samples. This indicates that sample contamination
had occurred. The other organic compound, tentatively identified by the GC/MS
scan, was diethylene glycol. This is a chemical compound which is used as a
solvent, plasticizer, and surfactant, and can be found in plastics, synthetic
sponges, paper products, cork compositions, adhesives, and dyes (Hawley, 1977).
The sample in which diethylene glycol was identified was originally collected
by a local resident and was stored in a glass jar. Trace contamination of the
sample could have occurred from the jar or lid before the sample was transferred
to an approved sample container. Diethylene glycol was not detected in the
soil sample from the area where the spilled brine had been collected.
Soil and sediment samples were collected from areas near the brine system and
analyzed for specific inorganic chemicals. Significant levels of these inorganics
were not found. The organic compounds identified in these samples, with the ex-
ception of di-n-butyl phthalate, are believed to have originated from natural
background sources. Di-n-butyl phthalate and similar compounds, are commonly
found in the environment and could have originated from many sources.
53
-------
Table 12
DESCRIPTION OF SAMPLES COLLECTED FROM THE DOW BRINE SYSTEM
AND AREAS LOCATED NEAR IT.
Type of Sample
brine
brine
brine
brine
soil
soil
sediment
sediment
sediment
sediment
sediment
sediment
Location *
F
F
G
**
F
G
H
E
D
I
J
K
Description
Waste brine collected from a
valve near the injection well.
Waste brine which had leaked
from the pipe and had collected
at the base of the well.
Brine which had been collected
earlier by one of the local
residents from a spill site.
Brine released from the filter-
ing system of the #6 lagoon loca-
ted on the Dow plant site in
Midland. (Location not marked on
the map'.)
Surface soil near the injection
well.
Surface soil from a spill site
located along the brine system.
Sediment from a sandy area next
to the brine system.
Creek bed sediment taken from the
area known as Smith Drain.
Sediment from the injection well
pond .
Sediment collected from an injec-
tion well head which was being
dismantled.
Sediment which was inside a brine
injection pipe at the horizontal
orifice plate near the well head.
Pipe scrapings from under a gas
separator located on a brine pro-
duction well. This well is loca-
ted in Midland County, one mile
South of Midland. The exact
location of the well is not marked
on the map.
* - Locations indicated on map (page 15).
** - Location not marked on the map.
1 - Addition sample collected at Location F
54
-------
CO
>-
CO
CQ
C£
CO
o o
o c
r» co
ft ft
it Ol
<5f VO
o o
o o
a- vo
CM en
o o
o o
en en
en co
r- LO
SO
^ 0
0 CO
o
f r
C O
C O
o o
CO VO
CO *f
0
o
CO
00
o
o
o
o
o
o
o
o
CM
o
o
CM
ft
1^
_l 1-
§§
en en
ft ft
if fO-
o o
o o
O
o
V
LO
CM
CM
CM
i \-~
Z
^£
Z
z
o o
o o
ro 1C
<« Ol
c o
31 CO
r P~
o
o o
3 r-
.(£)
^ ^
O
c o
O> CO
Cl
o
o o
r>- oo
CM CM
ft ft
co co
0
o
o
o
fm
CM
o
o
o
VO
CM
§
o
g"
CM
S
o
f..
CM
1 1
z
^J
z
z
0 CO
CM CM
V
o
CO CO
I-. O
Ol CO
^_
V
o
CO r
V
°°.co
LO CM
VO
r
0
o
o
r~-
1 1
z
^3^
y
Z
CO CVJ
r~" r~-
V
CM 00
o' o'
V
CO CO
CO
cr **.*
i CM
o
. ^_
o . .
V
V
VO
o
o'1"
CO
co'
r
.
LO
,
CO
<
J h*~
VV W
U E E O> "O -O
i-3 3 E -0 i- !- IV
C -r- -r- 3 1- t- S- T3
WE'OO-'-EOO'r-
W T3 IO. "a O> 3T3
i- 10 cu 10 o i- -c i o
«X O J CJ CO CO O Lu i i
z
«c
z
z
f
dj
E
^*
Z
Q
Z
a
a
z
a
z
o
z
a
z
a
z
o
z
t/1
0
c
10
0
w
IO
)
J
0)
1
CM
*_^
l/>
J3
VO
CO
co'
0)
+J
IO
S
L>
a
^
fj
3
J3
1
r^
D
CO
°
O
in
W
s_
u
i
p
5
u
1
a-
CM
CM'
^
o
c
a.
a
c
V
^
«
;j
1
CN.
O
z
e*
z
c
z
c
2T
o
z
a
z
a
z
o
z
o
z
f^
z
a
0
z
in
w
T3
O
+J
(/I
W
CL,
a
z
o
z
z
a
z
o
z
o
z
a
z
o
z
^1
3?
"
z
a
z
a
z
y)
ca
j
a.
0
Z
0
z
a
z
0
z
o
z
Q
Z
o
5*
"
z
"~^
z
JJ
o
z
in
ca
ca
a.
o
z
a
z
a
z
a
z
z
^c
z
z
<
z
^c
z
^c
z
^
z
z
c
X
o
T-
C3
55
------- An error occurred while trying to OCR this image.
-------
EPA requested permission from Dow to collect sediment samples from the waste
brine lagoon located at the Midland plant site. Dow refused EPA's request
based on their rationale that the sediment samples would not be representative
of the chemical composition of the brine. Dow indicated that the waste brine
was sand-filtered, thereby removing chemical contaminants, before being released
from the lagoon to brine transmission lines. However, Dow did permit sampling
of pipe sediment within the brine well injection system. As a result, three
pipe sediment samples, two from brine injection wells and one from a brine
production well, were collected.
Four synthetic organic chemical compounds were detected in sediments within
two of the pipes. These pipes are located at the brine production and injection
wells near Midland. Non-natural chemical compounds were not detected in the in-
jection well near Hemlock. The four organic chemicals detected were:
1) bis(2-ethylhexy1) phthalate
2) di-n-butyl phthalate
3) 4-chloro-m-cresol
4) 2-chlorophenol
In addition, a duplicate sample from location J was analyzed at a second
laboratory. From these analyses, two additional synthetic chemical compounds,
hexachlorobenzene (2.5 ppb) and pentachloroanisole (4(J ppb), were detected.
However, when analyzed by GC/MS, the presence of these compounds could not
be confirmed.
The three pipe sediment samples were also analyzed for dioxin. Using current
state-of-the-art methodology, dioxin was not found in these samples.'
57
-------
H. Evaluation of Uow's Brine Well Design and Construction
EPA staff reviewed technical data on the brine production and injection wells
operated by the Dow Chemical Company of Midland, Michigan. These data were
reviewed to determine what impact Dow's brine wells might have on shallower
aquifers used for water supplies in the Hemlock area. Part of this information
was originally requested through the Freedom of Information Act and was submitted
by Dow to the Congressional Subcommittee on Oversight of Government Management
on June 25, 1979. This information was later forwarded to EPA on October 30,
1979 (Appendix 1, page 63). EPA staff reviewed these data and were unable
to make a valid engineering evaluation of the mechanical conditions of the
wells because insufficient information was provided.
During a meeting with Dow representatives on January 24, 1980, additional
information was requested by EPA staff (Appendix 2, page 67). In response to
the request, Uow provided the necessary information to EPA (Appendix 3, page 69).
The information included details on injection well number 42 (location "J" on
the map) which is located near Hemlock.
Based on the available data, EPA staff concluded that the program followed by
Dow in the construction of their brine wells exceed current U.S. Geological
Survey, Division of Oil and Gas, On Land Injection Standards. In addition,
no evidence was found to suspect that these wells would be responsible for
cross-contamination between the geological formations.
A limitation to EPA's evaluation was that insufficient data were available on
wells constructed before 1970. Consequently, EPA was unable to make an evaluation
of these wells. It is Dow's position that, before 1970, the best technology
available at that time was used to construct the wells.
58
-------
V. CONCLUSION
Chemical contamination was not detected by analysis of the environmental and
tissue samples collected in the Hemlock, Michigan area. Based on available
information, the levels of the inorganic and organic chemicals in these samples
appeared to be within normal bounds. Most of the organic chemical compounds
identified are naturally occurring in the environment. Several synthetic
chemical compounds were also present. However, these compounds are frequently
found in the home and ambient environment due to the presence of consumer
products which contain them. The quality of water from certain wells in the
area contained elevated levels of sodium chloride salt which can affect the
taste of the water.
In summary, based on results from a large number and wide variety of samples,
the United States Environmental Protection Agency has found no evidence of
chemical contamination of the environment in the Hemlock, Michigan area.
59
-------
VI. REFERENCES CITED
Allison, F. E., Soil Organic Flatter and its Role In Crop Production, Elsevier
Scientific Publishing Co., N.Y., 1973.
Bear, K. £., editor, Chemistry of the Soil, American Chemical Society Monograph
Series #12b, Reinhold Publishing Co., 1955.
Bear, F. £., editor, Chemistry of the Soil, American Chemical Society Monograph
Series rflfaU, Reinhold Publishing Co., 19b4.
Beasley, Val, U.V.M.; University of Illinois, Veterinary Diagnostic Medicine
Dept. Personal communication on July 2, 198U.
Bohn, McNeal , U' Connor; Soil Chemistry, Wiley-Interscience Publication, John
Wiley and Sons, 1979.
Buckrnan, H. U. and Brady, N. C. ; The Nature and Properties of Soils bth Edition,
Macmillan Co., u.Y. I960.
Carey, Gowen, Forehand, Tai , Wiersma; "Heavy Metal Concentrations in Soils of Five
United States Cities, 1972 Urban Soils Monitoring Program."; Pesticides Monitoring
Journal. Vol. 13, No. 4, March, 198U; p. lbO-154.
CAST; Application of Sewage Sludge to Cropland: Appraisal of Potential Hazards
of the Heavy Metals to Plants and Animals, Council for Agricultural Science and
Tecnnology, Report #b4, I97b.
Conklin, P. and Fox, F. in Pentachlorophenol edited by K. Ranga Rao, Plenum
Publishing Co., NY 1978.
Eliassen, R. Domestic and Municipal Sources of Air Pollution in U.S P.H.S.
Proceedings for the National Conference on Air Polllution Nov. lb-2U, 1958;
tiHU; Washington, D.C., 1959.
EPA, Ambient Water Duality Criteria; Criterion Document for Fluranthene, Washington D.C.
EPA, Ambient Water quality Criteria; Criterion Document for Polynuclear Aromatic
Hydrocarbons; Washington D.C.
EPA, Health Assessment Document for Cadmium, Uffice of Research and Development,
EPA-bUU/8-79-Uu3, January 1979.
EPA, Initial Report of the TSCA Interagency Testing Committee, January 1978, EPA
5bU-lU-78/UUl, Washington, D.C.
EPA, iviemo of December 20, 1978 from Dr. Edward Oswald to Karl Bremer. Subject:
Summary of Results for Analyses of Samples of Fish from Michigan - EPA Region V -
for 2,3,78-Tetrachlorodibenzoparadioxin (TCDD).
LPA, quality Criteria for Water, 1976.
Ermolenko, N.K.; Trace Elements and Colloids in Soils, Israel Program for Scientific
Translations, 1972.
6U
-------
Federal Register. Rebuttable Presumption Against Registration; EPA, Office of
Pesticides Programs; Vol. 43, No. 4, August 2, 1978; p. 34U3U.
Hausenbui Her, R. L.; Soil Science Principles and Practices, W.M.C. Brown Co.,
Haw ley, G. G., Condensed Chemical Dictionary, 9th Edition, Litton Educational
Publishing, Inc., 1977.
Hesse, P. R., A Textbook of Soil Chemical Analysis, Chemical Publishing Co., NY, 1971.
Hessl, Stephen, M.D.; Head Section of Occupational Medicine at Cook County Hospital and
Assistant Professor of Environmental and Occupational Health Sciences, University
of Illinois, School of Public Health; Personal Communication on July 23, 1980.
Hoak, R. D.; "The Causes of Tastes and Odors in Drinking Water." Proc. llth Ind.
Waste Conf. Purdue University. Eng. Bull. 4:229.
Hoffman, U. and Wynder, E.; in Air Pollution, 2nd edition, edited by Arthur C. Stern,
Academic Press, N.Y. 19b8.
Interdepartmental Task force on PCBs, Polychlorinated Biphenyls and the Environment,
Washington, U.C., 1972.
Kimbrough, Renate, Ur., CDC toxicologist, Atlanta, Georgia; Personal communication
on July 7, 1980.
Klemmer, Leitis, Pfenninger; "Arsenic Content of House Dusts in Hawaii," Bulletin,
of Environmental Contamination and Toxicology, Vol. 14, No. 4, 1975, p. 449-4b2.
Lisk, U. J., "Trace Metals in Soils, Plants, and Animals," Adv. Agron., 24:267-311.
I'lacLeod, K.; Sources £f Emissions of PCBs into the Ambient Atmosphere and Indoor
Ai>. EPA report EPA-bOO/4-79-022, March~Ty79.
MDA, Final report from the laboratory examination of a goose and two chickens,
Nov. 28, 1979, Accession no. 2005fa4-5bb.
MNDR, Investigation of Groundwater Quality j_n the Hem!ock Area of Saginaw County,
Water Quality Division, Groundwater Compliance and Special Studies Section, April,
1979.
MDPH, The Hemlock Area Study - ^ Investigation Into Reported Health Problems and
Possible Water Contamination, Chemicals and Health Center, March,, 19/9.
Heikle, R. W.; "Decomposition of Organic Compounds" in Organic Chemicals in the Soil
Environment, edited by Gorhing, C. A. and Hamaker, J., Marcel Dekker, Inc., NY, 1972.
HWRC; Water Resource Conditions and Uses in the Shiawassee River Basin; Michigan
Water Resources Commission, October, 19b3.
Murphy, T. J., Ph.D., Associate Professor of Chemistry, DePaul Univeristy, Chicago;
Letter dated April 25, 1980, and attached unpublished report sent to Region V,
U.S. EPA, Office of Toxic Substances.
61
-------
National Resource Council; The Contribution of Drinking Wateir to_ Mineral Nurtition
_i_n Humans; prepared for EPA by the Nutrition Subcommittee of the Safe Drinking
Water Committee, NRC; National Academy of Sciences, Washington, D.C. 1979.
Oehine, Fred, Dr.; Veterinarian-lexicologist, Kansas State University, Toxicology
Laboratory; Personal Communication on Junly 2, 198U.
Shaw, hi.; "Lipid Composition as a Guide to the Classification of Bacteria" in Advances
in Applied Microbiology, Vol. 17, edited by D. Perlman, Academic Press, N.Y. 1974.
Sneed, Maynard, Brasted; Comprehensive Organic Chemistry, Vo 1. _3 The Halogens;
D. Van Nostrand Co., NY, 1954.
Solomon, R. L. and Hartford, J. W. "Lead and Cadmium in Dusts and Soils in a
Small urban Community," Environmental Science and Technology, Vol. 10, No. 8,
August, 197b, p. 71'3-777.
Stalling, Hogan, Johnson; "Phthalate Ester Residues - Their Metabolism and
Analysis in Fish"; Environmental Health Persp., 5:159-173 (1973).
Starr, Aldrich, McDougall, Mounce; "Contribution of Household Dust to the Human
Exposure to Pesticides"; Pesticides Monitoring Journal, Vol. 8, No. 3, Dec. 1974,
p. Z09-212.
Swaine, D. J.; The Trace Element Content of Soil, Commonwealth Agriculture Bureau,
Commonwealth Bur. Soil Technology Commun. No. 48, Harold Printing Works, England,
1955.
Townsend, W. N.; An Introduction to the Scientific Study of the Soil; St. Martin's
Press, NY, 1973.
Twenter, Floyd, geologist, U.S. G.S., Lansing, Mich., Personal Communication on
July 9, 1980.
Vinogradov's data in Soil Chemistry, by Bonn, McNead, O'Connor,; Wiley -Interscience
Publication, 1979.
Winterfield, Roland, Dr.; Professor of Avian Diseases, Veterinary Diagnostic
Laboratory, Purdue University; Personal communication on July 2, 1980.
-------
APPENDIX 1
ABRAHAM nmtarr, CONN, CHAIRMAN
HO«T M. JACXMN. WAW.
THOMAI r. EAOUrreM, MO.
LAWTON CHILC*. njk.
JOHNHirNN. OHIO
JIM SAMCM, TXMN.
.JJAVID ntYOR, ARK.
MO.UMM, MKH.
CHARLEB H. POKY, ILL.
JACOB K. JAVm, N.Y.
WILLIAM V. MOTH, JR., O*l_
TtD STCVCNS, ALASKA
CHARLES MC C. MATWAS, JR, MO.
JOHMC. OANrOHTM, MO.
WILUAM *. COHfH, MAIN*
DAVID C
CAM. UmH, MICH, CHAIRMAN
OAVID ntvoR. ARK. WILUAM s. COHOI, MA
-------
WASH.
», MO.
JOHN £t^"NN OHIO
JtM &ASSCM. TCNM.
DAt.B P*YC«. ARK.
CAAU LXVIH, MtOi.
,*Cib * JAV'T* N.Y.
v*r_L)AM V. ROTH. Jft., OC1_
TD 5TV£NS. Af-ASKA
CHA»i_C5 MC C MATHIAS. JK^
JOHN C. OAM'-ORTH. MO.
WIU_1AM C. COHCN, MAINC
DAVID DURCNVCAOU, MINM.
K.'--ZC>'**' ITFCi
LCVIN. MtCH., C-i*in*
WILL!* M S
DAVID SJKt
TA/r Ol»£CTD»
RtCHAJ9O A_ WECMAM
COUNSEL AND 5TA/T Dl
COMMITTEE ON
GOVERNMENTAL. AFFAIRS
SUBCOMMITTEE ON
OVERSIGHT OF GOVERNMENT MANAGEMENT
WASHINGTON. D.C. ZOSIO
June 25, 1979
SPECIAL DELIVERY
RETURN RECEIPT REQUESTED
Mr. Paul F. Oreffice
President and Chief Executive Officer
Dow Chemical Company
2030 Dow Center
Midland, Michigan 48640
Dear Mr. Oreffice:
The Subcommittee on Oversight of Government Management is
conducting an investigation into governmental implementation
of hazardous waste management programs by the Environmental
Protection Agency (EPA), the EPA regions and state agencies.
Pursuant to this investigation, we are interested in how govern-
mental units have responded to possible problems caused by the
brine injection deep wells you are operating or have operated
in the Midland, Michigan area. The Subcommittee has been
informed that some of these wells were constructed to extract
sodium chloride from the earth for your chemical manufacturing
processes in Midland, and that the brine sludge residue was
pumped back into reinjection wells.
It is also our understanding that the Michigan Department
of Natural Resources (DNR) has performed chemical analyses of
surface pools in the Fremont-Richland area formed as a result
of leaks and/or cracks in your company's brine reinjection lines.
The studies performed by DNR indicate that substances other than
brine residues were present in these surface pools, such as
flourides, ammonia, certain phenolic compounds and other
materials.
The Subcommittee is interested in the government's res-
ponse to possible chemical contamination of the groundwater by
such substances in Midland, Bay, Saginaw and Gratiot counties.
In order to assist the Subcommittee in its investigation, we
would appreciate your providing the following materials by
July 9, 1979:
64
-------
Mr. Paul F. Oreffice
President and Chief Executive Officer
Dow Chemical Company
June 25, 1979
Page 2
1. Copies of all groundwater and geological studies per-
formed by or for Dow, or under the direction of Dow, covering
any part of the area within a 25-mile radius of your Midland
facilities. Please also supply any other documents which may
indicate geological groundwater flow patterns or geologic
subsurface conditions relative to the areas of deep well in-
jection systems.
2. Copies of all chemical laboratory analyses, performed
since 1950, of spent brines and brine spills, as well as other
liquids, from Dow pipelines in the four counties mentioned
above, which Dow may have injected or is injecting into the
ground.
3. A copy of all well logs for deep well injection sys-
tems, showing the volume of materials injected into those wells
to date, as well as injection pressures. Please include with
these logs a list and map of all waste injection wells, as
well as landfills and surface impoundments owned or once owned
by Dow, whether or not these wells, landfills or impoundments
are presently in use. How long does Dow maintain such well
log information in its files? What record keeping is required
of Dow by federal law with respect to these wells, landfills
and impoundments?
4. Copies of all inorganic analyses and summary data
performed by or for Dow, or under the direction of Dow, which
indicate the presence of any heavy and/or pseudo metals in
Dow's injected waste streams. »
5. Copies of all analyses of organic constituents in the
injected wastes, specifically identifying their levels of
concentration.
In addition, please respond to the following specific
questions:
1. How much waste from the manufacture of trichlorophenol
and/or 2,4,5-T has been pumped into Dow's reinjection wells
and/or placed in Dow landfills or impoundments? Which wells,
landfills and impoundments?
2. At what rates and pressures does Dow dispose of waste
material through deep well reinjection systems? Is this rate
continuous.or intermittant, and are there seasonal fluctuations
in these rates? Are these rates low enough to insure the sta-
bility of the geologic formations?
65
-------
Mr. Paul F. Oreffice
President and Chief Executive Officer
Dow Chemical Company
June 25, 1979
Page 3
3. What types of high molecular weight and/or halogenated
organics are present in the waste stream of materials being
pumped to your reinjection wells? Please specifically iden-
tify types and concentrations of these materials.
4. Have EPA Priority Pollutant Analyses been performed
on Dow's waste streams? If so, please provide a copy to the
Subcommittee.
5. Has Dow performed any groundwater or well water analyses
of wells and water supplies within a 25-mile radius of your
Midland facilities within the last two years? If so, please
supply these to the Subcommittee.
6. Does Dow have any information which would indicate
that groundwater contamination may be caused by abandoned or
existing landfills and surface impoundments? If so, please
supply such information to the Subcommittee.
7. Has Dow ever encountered any problems in operating
injection systems, for example, failure of the injection
string within the groundwater aquifer? If so, please describe
fully.
8. Does Dow have any information which would indicate
that abandoned oil, gas or brine wells or other sources may
have contributed or are contributing to groundwater contami-
nation? If so, please supply this information to the Subcom-
mittee.
Y
Thanks for your cooperation in this matter. We realize
that this request is detailed. However, the Subcommittee has
scheduled hearings on governmental implementation of hazar-
dous waste management programs for later in July and we would
appreciate a prompt response. If there are any questions con-
cerning these requests, please contact Richard Tallman of the
Subcommittee staff at (202) 224-3682.
'CL/rt
66
-------
APPENDIX 2
/ UNITED STATES
\ ENVIRONMENTAL PROTECTION AGENCY
0 REGION V
^. 230 SOUTH DEARBORN ST.
^ CHICAGO. ILLINOIS 60604
JAM 3 i 1980
Mr. D. Dick DeLine
Manufacturing Manager
Dow Chemical U.S.A., Michigan Division
47 Building
Midland, Michigan 48640
Dear Mr. DeLine:
Thank you for your cooperation on January 24, 1980 regarding our
meeting and the samples obtained from the brine production and injec-
tion system. The day proved to be quite productive and will assist
in our evaluation of the Hemlock, Michigan area.
As you recall, Mr. Russell Diefenbach requested information during
the meeting on typical well records for the brine well injection
system. Per your request, Mr. Diefenbach has determined that the
following information is needed to continue his evaluation of the
Dow injection system (records from the Hemlock, Michigan area are
more appropriate):
1. Caliper Surveys for typical injection well(s) that are typical
for area.
2. Comparison of calculated cement volumes (w/o Caliper) v.s. cement
volume from bond logs.
3. Average % wash out used for cement calculations. How established.
. Results of Cement Bond Log's cement top v.s. calculated cement
tops. (Temp, survey's for CBL if used in lieu of).
5. Is the long casing string kept in tension during cement setting
time? After cement sets, is all casing weight set on braden-
head or is it slacked-off?
6. What casing accessories are used on long string of casing? Is
there pressure test after WOC?
7. Is casing dressed with centralizers to center allow 360° cement
sheath?
67
-------
s
" ' " 2
8. Describe typical cement program for long string.
Your response may be forwarded to my office at your first convenience.
I anticipate completion of this part of the evaluation during February,
1980.
Thank you for your consideration.
Sincerely,
Karl E. Bremer
Toxic Substances Coordinator
cc: Senator Carl Levin
68
-------
APPENDIX 3
DOW CHEMICAL U.S.A.
February 29, 1980 MICHIGAN DIVISION
MIDLAND, MICHIGAN 48640
Karl E. Bremer
Toxic Substances Coordinator
US EPA Region V
230 South Dearborn Street
Chicago, Illinois 60604
Dear Karl:
Enclosed is the information requested by Mr. Russell Diefenbach through
your letter of January 31, 1980. I was pleased that your visit to the
Michigan Division was quite productive and will assist you in your
evaluation of the Hemlock area.
This information was supplied by Mr. David Cella, Manager of the Brine
Chemicals Section and our technical expert in the brine well area.
Mr. Cella has been with the Michigan Division since 1970 and his
responses to Mr. Diefenbach*s inquiries relate to the procedures and
technology used over the past ten years. Prior to 1970, the best
technology available at the time was used.
1. Caliper surveys for typical injection well(s) that are typical
for the area.
Caliper survey and associated bond logs of three wells in the
Hemlock area are enclosed.
Production Wells #76 and #82
Injection Well #42
2. Comparison of calculated cement volumes (w/o Caliper) versus
cement volume from bond logs.
The number of sacks of cement and type of cement used in each
well are shown on the enclosed "Cementing Service Reports."
Production Wells #76 and #82
Injection Well #42
AN OPERATING UNIT OF THE DOW CHEMICAL COMPANY
69
-------
Karl E. Bremer
February 29, 1980
/f-^ Page 2
\ .
3. Average % wash out used for cement~calculations. How established.
/
Since 1973, we have calipered the open hole. Prior to that time,
the normal practice was to use the volumetric difference between
bit size and casing size plus 20-30% excess cement.
4. Results of Cement Bond Log's cement top versus calculated cement
tops. (Temp, survey's for CBL if used in lieu of).
A comparison of the data recorded on the Cementing Service Report
and the 3-Dimensional Velocity Log will give the actual results
versus calculated. Please see enclosed reports for production
wells #76 and #82 and injection well #42.
5. Is the long casing string kept in tension during cement setting time?
After cement sets, is all casing weight set on bradenhead or is it
slacked-off?
The long string of casing is held in tension at the surface during
cementing operations. After the cement has set, we pull one-half the
weight of the free pipe and set the slips.
6. What casing accessories are used on long string of casing? Is there
' pressure test after WOC?
The following accessories are used on long strings of casing:
(a) regular pattern guide shoe,
(b) cement float collar,
(c) centralizers (3-5/well).
Surface casing and long string are tested at a minimum of 500 psig.
7. Is casing dressed with centralizers to center allow 360" cement
x sheath?
Centralizers are used on the long string and are normally placed one
joint above casing shoe, below and above the Dundee zone, with one
or two between the Dundee and casing shoe. A caliper log or the
geologists log is used for judgment when placing the centralizers.
8. Describe typical cement program for long string.
A typical long string casing cement program is the same for both
production and injection wells, except for casing size. We consider
a cement fillup from 200' minimum to 500' above the Dundee zone an
adequate cement job. Normally, the target cement top is about
2000' to 2500' from the surface, which is approximately 500-1000' above
the Dundee formation. The supervisor on the job is given the latitude
to modify the cement program to fit the downhole conditions. The
brine field is in the Sylvania formation which is about 1000' to
1500' below the Dundee top.
70
-------
Karl E. Bremer
February 29, 1980
f~~ Page 3
8, Describe typical cement program for long string, (continued)
The hole is drilled down to the top of the production zone, circu-
lated and cleaned up. A caliper log is run and determination is
made as to where to place centralizers and what cement volume is
required to bring the cement top to 2000* from the surface. Depending
on-the condition of the hole, 10-20% excess cement over computed
volume is used.
The casing is then run by a casing crew using power tongs. The bottom
of the hole is tagged with the casing and the casing is then picked
up two feet. After the casing has been landed, circulation is
broken by pumping down casing and getting returns to surface.
A fresh water pad (50 barrels) is pumped ahead of the cement and the
cement is pumped in. When the cement reaches the bottom of the casing,
the pumping rate is slowed down so as not to exceed 2-1/2 barrels/minute.
The cement is displaced with water and drilling fluid, leaving 20-30 feet
of cement in the casing.
The well is shut in (WOC) and allowed to set 48 hours. The hole is
then nippled up to drill out below the casing. After nippling up,
the hole is run in with a drill pipe and top of cement tagged, drilled
up to casing seat, and pressure tested to a minimum of 500 psig.
If the test is approved, drilling is continued to total depth. After
the hole has been drilled to total depth and cleaned up, logs are run
on the hole. Almost all of our Sylvania wells are open-hole completion.
Generally, the wells are "in gage," and no major problems have been
encountered in the completion of our brine wells. There have been
very few changes in the cement program over the years, except for
type of cements used and/or the method of cement placement. We
prefer the "slo-flo" technique of cementing rather than "turbulent
flow," the merits of each being debatable. Naturally, we rely on our
Dowell Division for vendor technology in this field.
Karl, I am sorry for the delay in getting this information to you. If you
have additional questions or comments, please give me a call.
Sincerely,
D. Dick DeLine
Manufacturing Manager
(517)636-0150
lej
Enclosures (9)
cc: (Letter only) Senator Levin, B. G. Caldwell, Dr. H. Tanner,
Dr. M. Reizen, R. Ellison
71
------- |