OCCURRENCE OF PESTICIDES
IN DRINKING WATER, FOOD, AND AIR
Revised Draft Report
July 1987
Submitted to:
Mr. Paul Price, Project Officer
U.S. Environmental Protection-;Agency
Office of Drinking Water (WH-550D)
401 M Street, S.W.
Washington, DC 20460
-------
1. ALACHLOR
1.1 GENERAL CHARACTERISTICS
1.1.1 Physical/Chemical Properties
Alachlor [2-chloro-2',61-diethyl-N-(methoxymethyl) acetanilide] is a
pre-emergent amide herbicide. Synonyms and identifiers for alachlor are
Alanex, Alanox, and Lasso® (Berg 1986; SRI 1983). Alachlor is often applied
as a mixture with atrazine; this product is also known as Alazine (Berg 1986).
Alachlor is a white crystalline solid at 25°C (USEPA 1984d). It has a
molecular weight of 269.8, a molecular formula of C-| 4H2oN02c^' anc^ a melting
point of 39.5 to 41.5°C (USEPA 1984; Berg 1986). The aqueous solubility and
vapor pressure of alachlor at 25°C are 242 mg/1 (8.97 x 10~4 mol/1) and
2.2 x 10"5 torr (2.89 x 10~8 atm), respectively (Cohen et al. 1984).
Based on the reported vapor pressure and aqueous solubility, Henry's
constant for alachlor at 25°C is estimated to be 3.2 x 10-8 atm-m3/mol.
Kenaga and Goring (1978) report a Koc value of 190 for alachlor.
1.1.2 Use
Alachlor is a herbicide recommended for control of yellow nut sedge,
annual grasses, and broadleaf weeds. The product can be applied either as a
surface application after planting or shallowly incorporated before planting
in the upper 1 to 2 inches of soil. The recommended rate of application is
2.5 to 4 quarts per acre (CPCR 1986). Total U.S. alachlor usage ranges
between 80 and 85 million pounds of active ingredient annually (Kuch 1986).
Nearly all (96 to 98%) of the domestically supplied alachlor is applied
to field corn and soybeans (Glaze 1982). The major field corn producing
states are located in the Corn Belt (Illinois, Indiana, Iowa, Kentucky,
Missouri, Ohio); the Northern Plains (Kansas, Nebraska, North Dakota); the
Lake States (Michigan, Minnesota, Wisconsin); and the Southeast (Georgia,,
North Carolina). Pennsylvania and Texas are also considered major field corn
-------
producing states. Approximately 52 million pounds of alachlor active ingredient
were used on field corn in these states and regions in 1982 (USDA 1983).
Approximately 33 million pounds of alachlor active ingredient were used
on soybeans in the major soybean producing states in 1980. These states are
located in the North Central region (Illinois, Indiana, Iowa, Kansas,
Minnesota, Nebraska, Ohio); the Mississippi Valley (Arkansas, Kentucky,
Louisiana, Mississippi, Tennessee); and the Southeast (Alabama, Georgia, North
Carolina, South Carolina) (Hanthorn et al. 1982). Florida also reported the
use of alachlor on soybeans (Lipsey 1981).
The remaining 2 to 4 percent of alachlor production is used on other
crops, including: peanuts, potatoes, ornamentals, grain sorghum, dry edible
beans, cotton, tobacco, sugar cane, and strawberries (Glaze 1982, USDA 1983,
Lipsey 1981, Parks 1983). Some of these uses are restricted by location
(USEPA 1982a).
1.1.3 Environmental Transport and Transformation
Hie discussion of the environmental fate of alachlor is divided into the
following subsections: 1.1.3.1 Volatilization; 1.1.3.2 Sorption and Leaching
Potential; 1.1.3.3 Abiotic Transformations; 1.1.3.4 Biodegradation; 1.1.3.5
Fate in Water Treatment Plants; and 1.1.3.6 Summary. The discussion will
emphasize the environmental fate of alachlor in soil and water.
1.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as alachlor (H = 3.2 x
10-8 atm»m3/mol) with Henry's constants less than 3.0 x 10" 7 atm«m3/mol are
unlikely to undergo significant rates of volatilization from natural waters
under any conditions normally found in the environment. Volatilization
half-lives estimated by SAIC (see Appendix A) for alachlor in shallow water
under turbulent conditions support that conclusion. The estimated volatil-
ization half-lives for alachlor in a turbulent lake 1 m deep and in a stream
1 m deep with a mean current of 2 m/s and exposed to a mean wind velocity of
5 m/s are 7.9 years and 2.9 years, respectively.
-------
Volatilization rates of alachlor from groundwater to the soil column
above are projected to be substantially less than volatilization rates from
surface waters to the atmosphere. This is due primarily to the laminar,
nonturbulent nature of groundwater flow. Transport from groundwater by
volatilization may be further reduced by a build-up of chemicals in the pore
air at the pore air/groundwater interface, and an associated decrease in the
concentration gradient across the interface.
SAIC estimated a volatilization half-life for alachlor applied to the
soil surface of only 33 days (see Appendix A). That suggests that volatil-
ization could be substantially involved in the removal of alachlor from
surface soils. This conclusion is supported by the experimental results
published by Hargrove and Merkle (1971) and Beestman and Deming (1974, as
cited in CJSEPA 1984d. They reported that substantial losses of alachlor from
moist silt and sandy loam soils occurred through volatilization (USEPA 1984d).
The relatively short estimated volatilization half-life for alachlor from
the soil surface compared to that for volatilization from water is due to the
relatively small reported value of the organic carbon normalized sediment or
soil/water equilibrium partition coefficient (KqC = 190). Volatilization
rates for alachlor below the surface of the soil would probably be much slower
than at the soil surface.
1.1.3.2 Sorption and Leaching Potential
Based on the reported Koc value of 190 for alachlor, the estimated
sediment or soil/water equilibrium partition coefficients (Kg/W) range from
1.9 to 15 (see Appendix A). As discussed in Appendix A, the Ks/W values would
probably have to exceed 100 in most aquatic systems for the ratio of the
chemical mass adsorbed to suspended and exposed bottom sediment to the
chemical mass dissolved in the water column to exceed 0.1. The reason is that
in most aquatic systems most of the time, the ratio of the water mass to the
mass of suspended and exposed bottom sediment exceeds 103 (USGS 1983). There-
fore, transport by adsorption to suspended sediments and removal by adsorption
to bottom sediments are probably generally not very important processes for
alachlor in most aquatic systems.
-------
Based on its KQC value, alachlor would be classified as moderately
resistant to leaching from surface soils using five mobility classes defined
by Helling and Turner (1968 and cited in Hamaker 1975), The results of
various monitoring studies have supported that conclusion by showing that most
of the alachlor remains in the top 0-5 cm or 0-10 cm of soil layers and that
surface runoff does not generally remove substantial amounts of alachlor from
soil (USEPA 1984d). However, low levels (e.g., .04 ug/1) of alachlor have
been detected in 2 out of 14 wells sampled in a corn growing area of Nebraska
with sandy soils and shallow unconfined aquifers (Cohen et al. 1984). Also,
Wauchope (1978) reported that 3 to 14 percent of applied alachlor was lost to
surface runoff during a catastrophic flood (greater than 100-year frequency).
1.1.3.3 Abiotic Transformations
No information was found on the hydrolysis of alachlor in the literature
reviewed by USEPA (1984d) or SAIC. Alachlor is apparently susceptible to some
form of photodecomposition in water with a surfactant present and on soil
(Fang 1977 and Tanaka et al. 1981, as cited in USEPA 1984d). Tanaka et al. (1981,
as cited in USEPA 1984d) reported that alachlor in solution at an initial
concentration of 100 ppm was 85 percent degraded in the presence of a surfactant
after 138 minutes exosure to 300 nm UV light. In the absence of the Surfactant,
only 1 percent was degraded. Therefore, most of the photodecomposition does
not appear to be due to direct photolysis.
1.1.3.4 Biodegradation and Persistence in Soil
Biodegradation appears to be the primary mechanism of removal for
alachlor in soil. The reported half-life for alachlor in a laboratory
incubated silt soil was 8 days compared to 470 days in an autoclaved soil
(Beestman and Deming 1974, as cited in USEPA 1984d). Reported half-lives for
alachlor in soils under laboratory and field conditions have ranged from
4 days to 10 weeks (Cohen et al. 1984; USEPA 1984d). Therefore, it would be
classified as nonpersistent (half-life <20 days) to moderately persistent
(20 days _
-------
1.1.3.5 Fate in Water Treatment Plants
Baker et al. (1981), as cited in USEPA 1985a), reported that a conven-
tional water treatment process consisting of alum coagulation, flocculation,
sedimentation, and filtration was generally less than 50 percent effective in
removing alachlor from influents at initial concentrations of <0.2 to
2.0 ug/1. in addition, activated carbon adsorption may not be efficient due
to low reported break-through volumes for alachlor (De Filippi et al. 1980, as
cited in USEPA 1985a). Studies of levels of alachlor in raw and treated
surface water confirm that conventional treatment practices are not effective
in removing alachlor (Monsanto 1986).
1.1.3.6 Summary
Based on the above discussion and the literature reviews by Cohen et al.
(1984) and USEPA (1984d), the following conclusions can be made concerning the
most likely behavior of alachlor in soil and water:
o Biodegradation appears to be the primary mechanism for removal of
alachlor from soils. Reported half-lives for soil range from 4 days
to 10 weeks.
o Alachlor is unlikely to undergo significant rates of volatilization
from surface water or groundwater under any conditions normally found
in the environment.
o Volatilization appears to be a significant removal process for
alachlor on the soil surface. SAIC estimated a volatilization
half-life of 33 days for alachlor on the soil surface.
o Alachlor is expected to remain in solution in surface waters and not
bind to suspended solids. This is consistent with the findings that
alachlor is not removed by conventional drinking water treatment
practices such as filtration.
o Based on both theoretical and experimental results, alachlor can be
classified as at least moderately resistant to leaching from soils.
o No information was available on the hydrolysis of alachlor. Alachlor
appears to be susceptible to some form of photodecomposition in the
environment, but not to direct photolysis.
o Due to its moderate immobility, moderate resistance to leaching, and
low to moderate persistence in soils, alachlor is unlikely to signif-
icantly contaminate large numbers of surface water or groundwater
supplies. However, some limited contamination of shallow unconfined
aquifers or streams located near areas of alachlor use may occur.
-------
o Due to alachlor's low volatility and high solubility, it has
considerable potential for the contamination of surface water.
1.2 OCCURRENCE IN THE ENVIRONMENT
1.2.1 Water
This section presents available data for monitoring studies and surveys
to determine the extent of occurrence of alachlor in public drinking water
supplies and water other than drinking water.
1.2.1.1 Occurrence in Drinking Water
A number of studies provided data on the occurrence of alachlor in public
drinking water supplies. These studies, which were conducted on both regional
and national scales, are summarized in the following section.
Groundwater Sources — National Study
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) was conducted by SRI from June 1977 to March 1981. Samples of
finished drinking water were collected from 12 groundwater systems of varyin<
size throughout the United States and analyzed for alachlor. None of the
drinking water samples from the 12 groundwater systems contained levels of
alachlor in excess of the quantification limit of 0.1 ug/1.
Groundwater Sources — Regional Studies
Drinking water wells from eight counties in Maryland were examined by the
State of Maryland's Office of Environmental Programs, Department of Health and
Mental Hygiene, in the fall of 1983 (State of Maryland 1983). Thirteen
samples were collected from 11 locations and analyzed for alachlor. Two
samples were positive at concentrations of 0.1 and 0.8 ug/1. The detection
limit was 0.1 ug/1.
The Suffolk County Department of Health Services analyzed drinking water
wells in Long Island, New York, during 1984 (Holden 1986). Alachlor was not
detected in any of 24 samples collected (detection limit not reported).
-------
Drinking water wells throughout the State of Wisconsin were analyzed by
Union Carbide during 1983-1984 as part of a program by the Wisconsin DNR (Holden
1986). Possible contamination may have occurred from both point and nonpoint
sources. Of the 377 samples analyzed, 47 were found positive for alachlor
with a maximum concentration of 88 ug/1. The mean concentration, range of
values, and detection limit were not reported.
Two studies of drinking water wells in Iowa were available for
information on the occurrence of alachlor; one conducted by the Iowa State
University and the other by the Iowa Department of Water, Air, and Waste
Management (Baker and Austin 1983; Kelley and Wnuk 1986). Combined, the
studies included six Iowa counties sampled between 1981 and 1985. Fifty-nine
samples were analyzed from 25 sites, with all but one positive sample coming
from Humboldt County. The overall range was 0 to 2.7 ug/1 (the detection
limit for this high value was 0.01 ug/1). The one positive sample from
outside Humboldt County was 0.18 ug/1. The mean for Humboldt County samples,
all taken at one site, was 0.08 ug/1 (the detection limit was 0.01 ug/1). The
total number of positives and other detection limits were not reported.
Surface Water Sources — National Study
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) also contained information on alachlor contamination in drinking
water from surface water sources. Finished drinking water samples, collected
from 104 surface water systems of varying size throughout the United States,
were analyzed for alachlor. Drinking water samples from four very large
systems (serving graeter than 100,000 individuals), were found to contain
levels of alachor in excess of the quantification limit of 0.1 ug/1, ranging
between 0.1 and 0.9 ug/1, with an average value of 0.38 ug/1.
Surface Water Sources — Regional Studies
Alachlor was monitored for in known high-use areas of Illinois, Indiana,
Iowa, Michigan, Missouri, North Carolina, and Ohio (Monsanto 1986). Sampling
occurred at 24 community water treatment plants in these seven states, which
primarily or exclusively utilize surface water supplies. Populations of the
-------
24 communities ranged from 3 56 to 388,000 and, overall, the study is
representative of over 1.3 million people using public drinking water. Both
raw and finished water samples were collected daily from each location for 1
year, and were turned into weekly composites (as are presented in the report).
Sampling took place between April 1985 and April 1986. The lower limit of
method validation (LLMV) was 0.02 ug/1.
Alachlor was not detected in any weekly sample of finished water from 10
of the 24 plants. Of the other 14 plants, the annualized mean concentration
(AMC - i.e., a time-weighted average concentration) at 12 plants was less than
0.50 ug/1. The other two plants had AMCs of 0.69 ug/1 and 1.4 ug/1. Rarely
was alachlor found above 2.0 ug/1 in a weekly composite; only 2.6 percent of
the weekly composites exceeded 2.0 ug/1 during the sampling year (Monsanto
1986). The weekly composite maximum concentrations ranged from <0.20 to 10.7
ug/1 for raw water samples and from <0.20 to 10.9 ug/1 for finished water
samples. The AMCs ranged from 0 to 1.5 ug/1 for raw water samples, and from
0 to 1.4 ug/1 for finished water samples. [Note: For the AMC range, the
lower value assumes that all non-detected = 0 ug/1 and the higher value
assumes that all non-detected = 0.20 ug/1.) Individual sample results were
not presented in the report.
Baker (1983) provided the results of analysis of finished drinking water
samples collected from a water supply in Ohio from 1981-1982. The supply
obtained water from a river draining an agricultural area. Between June 1981
and July 1982, 15 finished drinking water samples were collected from this
plant. Although no detection limits were given, 14 of the samples had
positive concentrations of alachlor that ranged between 0.03 and 14.3 ug/1,
with an average of 4.5 ug/1.
In another study by Baker (1983), 49 samples were collected from 3 water
supplies in Ohio and analyzed for alachlor. The supplies obtained their raw
water from two rivers that drain agricultural areas. Average concentrations
for samples collected at each of the supplies between May 28 and July
27, 1983 were 1.08 (18 samples), 0.22 (15 samples), and 1.87 ug/1 (16 samples).
Peak concentrations observed in 1983 were 2.73, 0.47, and 5.91 ug/1, respec-
tively. Datta (no date) reported an overall mean concentration of 1.07 ug/1
-------
for these supplies. The detection limit and number of positive samples were
not reported.
Finished drinking water samples from New Orleans, Louisiana were analyzed
by Keith et al. (1976, as cited in USEPA 1984d). The range of positive samples
for alachlor was 0.17 to 2.9 ug/1. The number of samples analyzed, number of
positive samples, mean, and detection "limit were not reported.
1.2.1.2 Occurrence in Water Other Than Drinking Water
Eleven studies provided data on concentrations of alachlor in water other
than drinking water. Three studies assessed levels of alachlor in ground-
water, and eight studies provided information on levels of alachlor in surface
water other than drinking water.
Ground Water Sources
Spalding et al. (1980, as cited in Cohen et al. 1984) presented data from
a study conducted in atrazine high-use areas of Nebraska. Water samples from
14 wells, out of a total of more than 1,000 wells that were studied in the
area, were selected as representative of water in the area. Water samples
from 2 of the 14 wells contained an average alachlor concentration of approx-
imately 0.04 ug/1 (range = 0.018 to 0.071 ug/1). No detection limit was
reported.
In addition, ground water wells from Nebraska, as well as from Iowa,
Maryland, and Pennsylvania, have been found to have concentrations of alachlor
typically in the range of 0.1 to 10.0 ug/1 (Cohen et al. 1986). No other
information on these wells was available.
Ground water wells and spring water in the Big Spring Basin, Clayton
County, Iowa, were sampled during 1981-1982 for the occurrence of alachlor
(Datta no date). Ninety-five samples from 21 locations were analyzed for
alachlor, and 4 positive samples were found. An overall mean of 0.082 ug/1
(range = 0.05 to 0.15 ug/1) was calculated.
-------
A survey of alachlor in rural private wells was performed (Monsanto
1986). The survey analyzed for alachlor in 246 wells in 9 states. The wells
sampled were selected from private wells in areas where alachlor was
extensively used. Of the 246 wells sampled, 10 wells had detectable levels of
alachlor. The highest level reported was 5.8 ug/1. The remaining nine wells
had levels below 1 ug/1. The detection limit was 0.1 ug/1.
Surface Water Sources
Baker et al. (1981) presented data from a study that examined concen-
trations of alachlor in streamwater in Ohio. A total of 292 samples were
collected from 12 different streams during the spring and summer of 1981, and
analyzed for alachlor. The analysis identified 235 (80%) positive samples.
The maximum concentration observed was 104.6 ug/1. No detection limit for the
analysis was given.
Baker (1983) reported on levels of alachlor in water samples collected
from two Ohio rivers. Between May 28, 1983 and July 27, 1983, a total of
46 samples was collected (23 samples from each river). The average alachlor
concentrations for samples collected and analyzed from each of the rivers was
1.24 ug/1 and 3.11 ug/1, respectively. The number of positive values and the
detection limit for alachlor were not reported.
Another study examining surface waters in Ohio was reviewed for alachlor
occurrence data. Datta (no date) reported analyses of a creek in southwest
Ohio during 1981, and again in 1982. Mean concentrations of alachlor for the
two years were 13.9 ug/1 and 7.6 ug/1, respectively. No other information was
reported. The same source (Datta no date) also reported that for 5 northwest
Ohio rivers, 233 samples were found to have a mean of peak concentrations of
23.2 ug/1 (maximum concentration = 69.6 ug/1). The number of positives and
detection limit were not reported.
River samples from the Little Sioux River in northwest Iowa and Big
Spring Basin, Iowa, were analyzed for alachlor by the Iowa Department of
Water, Air, and Waste Management and as part of the Review of Hydrogeology,
Water Quality, and Land Management in the Big Spring Basin, respectively
-------
(Kelly and Wnuk 1986; Datta no date). During the overall study period from
1981 to 1985, 18 samples from 5 locations were taken and 10 proved positive
for alachlor. These positive samples represent a mean of 2.59 ug/1 (range =
0.06-20.0 ug/1). No detection limit was given. A reservoir on the Des Moines
River was sampled for alachlor during 1977-1978 by Leung et al. (1982, as cited
in USEPA 1984d). Three sites were sampled (upstream, within, and downstream of
tne reservoir). A mean concentration of 0.089 ug/1 (range = 0 to 0.82 ug/1) was
reported for positive samples. The number of positives, number of samples,
and detection limit were not reported.
Twenty-five samples from River Raisin, Michigan, were taken at U.S.
Geological Survey stations during 1982 (Datta no date). A maximum
concentration of 8.16 ug/1 was reported for alachlor. No other information
was reported.
Dudley and Karr (1980, as cited in USEPA 1984d) analyzed levels of
atrazine and alachlor in 45 samples of water, sediment, and fish. The samples
were collected in mid-July from a stream draining an agricultural watershed
(Black Creek) in Allen County, Indiana. No samples of water showed levels of
alachlor in excess of the detection limit of 100 ug/1. The exact number of
water samples was not reported.
Schepers et al. (1980, as cited in Baker et al. 1981) assessed concen-
trations of alachlor in 30 samples of water collected from a watershed in
Nebraska. Concentrations of alachlor in the 30 samples ranged between
"non-detectable" and 1.41 ug/1. However, the period of collection, detection
limit, and number of positive samples for alachlor were not reported.
Lake Erie water samples were collected during a study by Konasewich
et al. (1978, as cited in USEPA 1984d). Three samples were positive for
alachlor at a mean concentration of 3.05 ug/1 (range = 0.07-9.0 ug/1). The
number of samples and detection limit were not reported.
-------
1.2.2 Occurrence in Ambient Air
No information was found on concentrations of alachlor in ambient air.
Based on alachlor*s relatively low vapor pressure and widely dispersed
releases from agricultural application, alachlor is not expected to occur in
air at significant levels.
1.2.3 Soil/Sediments
Two studies were identified that reported alachlor levels in sediments.
There were no studies identified that examined alachlor levels in soils.
Dudley and Karr (1980 as cited in USEPA 1984d) analyzed levels of
alachlor in 45 sediment samples, of which none were found to contain levels
above the detection limit of 100 ug/1 (ppb).
Baker et al. (1979, as cited in USEPA 1984d) reported that 6 weeks after
alachlor wa9 applied ( 2.4 kg/ha) to an Iowa watershed, 1.0 ppm of its residues
were found in sediment from surface runoff events.
1.2.4 Food
Very little information was available on the occurrence of alachlor in
food in the United States. Three studies were identified; however, their
resulting data were limited. Hunt et al. (1980, as cited in USEPA 1984d)
reported that cabbage grown in soil sprayed with alachlor at the rate of 3.4
kg/ha and 6.8 kg/ha had no detectable residues when harvested 6 weeks later.
Brookhart and Johnson (1977) analyzed fish tissue from fish caught in
nine rivers that., empty into the Chesapeake Bay. All samples contained less
than 0.1 ug/g alachlor* Lastly, Dudley and Karr (1980, as cited in USEPA
1984d) reported no detectable alachlor in fish in the Black Creek watershed in
Allen County, Indiana, in 1977. In conclusion, because the information .obtained
during these studies is not representative of most food groups that constitute
an average diet, it was impossible to calculate the typical dietary intake.
-------
1.3 EXPOSURE SUMMARY
Limited studies have been conducted that provide useful data on the extent
of occurrence of alachlor in drinking water, food, and air. National monitoring
surveys of public water supplies from both ground and surface water sources
have reported alachlor levels to be <1.0 ug/1. Regional studies of surface
water sources from areas of high alachlor usage have reported levels greater
than 10 ug/1. However, typical concentrations tend to be much lower. No data
are available on alachlor levels in food or ambient air.
The available data on alachlor in drinking water are insufficient to
determine national exposure levels. However, based on a concentration of
10 ug/1, daily intake would be 20 ug/day. Alachlor concentrations in food and
air are unknown. Consequently, it is not possible to determine the total
exposure to alachlor or to determine which of the three sources of exposure is
the major contributor to total daily intake.
-------
2. ALDICARB
2.0 SUMMARY
Aldicarb is a widely used insecticide. Both aldicarb and its degradation
products are relatively stable and mobile in groundwater. Data are not
available to characterize levels of aldicarb in drinking water. Low levels of
aldicarb have been reported in wells near its use.
2.1 GENERAL CHARACTERISTICS
2.1.1 Physical/Chemical Properties
Aldicarb [2-methyl-2(methylthio)propionaldehyde-o-(methylcarbamoyl)oxime]
is a carbamate pesticide, a synthetic relative of the alkaloid physostigmine.
It is a systemic insecticide, acaricide, and nematicide. Synonyms for
aldicarb are Temik, OMS 771, and UC21149 (Berg 1986).
Aldicarb is a white crystalline solid at 25°C (Verschueren 1983). It has
a molecular weight of 190.25', a molecular formula of C17H14N2O2S, and a melting
point of 99 to 100°C (Windholz 1976). The aqueous solubility and vapor pressure
of aldicarb at 25°C are 6.0 g/1 (3.2 x 10~2 mol/1) and 1.0 x 10-4 torr
(1.3 x 10~7 atm), respectively (Verschueren 1983).
The ratio of the vapor pressure to the aqueous solubility gives an
estimated Henry's constant at 250C for aldicarb of 4.06 x 1 0~9 atm«m3/mol.
Rao and Davidson (1980) report an octanol/water partition coefficient (KqW) of
5 for aldicarb. Aldicarb has an estimated organic carbon normalized soil/
water equilibrium coefficient (Koc) of 36 (Cohen et al. 1984).
Under aerobic conditions in aqueous solution, aldicarb is rapidly
transformed to aldicarb sulfone and aldicarb sulfoxide, which have aqueous
solubilities of 7.8 and 43 g/1, respectively (Cohen et al. 1984).
-------
2.1.2 Use
Aldicarb is registered as a restricted use insecticide/nematicide in 10
or 15 percent granular formulations (Berg 1986). Registered uses of aldicarb
include cotton, potatoes, peanuts, soybeans, pecans, sugar beets, citrus
fruit, sweet potatoes, edible beans, sugarcane, sorghum, and ornamentals
(commercial field grown and nursery plantings, greenhouse crops, and potted
plants) (Holtorf 1982).
EPA estimated that domestic usage of aldicarb ranged from 5 to 6 million
pounds annually as of 1986 (Kuch 1986). In 1982, approximately 54 percent of
annual domestic usage was applied to cotton and potatoes. Other major uses of
aldicarb include application to peanuts, soybeans, and pecans which, combined
with usage on cotton and potatoes, constitute 90 percent of the estimated
annual domestic usage (Holtorf 1982).
A number of restrictions govern the use of aldicarb, including site,
volume, and insect specific restrictions, and guidelines for methods of
application. Since 1979, registrations and label restrictions have prohibited-
any aldicarb use on Long Island, New York, after the discovery of aldicarb in
groundwater in that area resulted in the closing of several wells (USEPA
1983a).* In 1983, Florida suspended the use of aldicarb on citrus fruits; that
suspension is still in effect. Wisconsin and Maine restricted application to
potatoes to emergence stage only. Fifteen counties, from Massachusetts to
Virginia, have similar label restrictions prohibiting use on potatoes (CPCR
1986).
For moat agricultural crops, use of aldicarb is restricted to one
application per crop. For cotton and sugar beets, aldicarb use is restricted
to two and three applications per crop, respectively (USEPA 1982b). Aldicarb
granules are applied in 4 to 6 inch bands over a row on top of the hill. They
~Personal communication between Dennis Edwards, Office of Pesticide Programs,
U.S. Environmental Protection Agency, Washington, DC, and Chris Rioux, SAIC
JRB Associates), May 1984.
-------
can also be drilled into the soil or planted with the seed (CPCR 1986).
Working the product into the soil to 2 to 4 inches is recommended.
Restrictions on volume of application have also been defined for aldicarb
and range from 2 pounds per acre for the treatment of aphids on dried-type
beans to 67 pounds per acre for the treatment of aphids and mites on pecans.
The average application rate for nematodes is 14 pounds per acre; leafhoppers
average 7-14 pounds per acre (CPCR 1986).
2.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of aldicarb, aldicarb sulfoxide,
and aldicarb sulfone is divided into the following subsections: 2.1.3.1
Volatilization; 2.1.3.2 Sorption and Leaching Potential; 2.1.3.3 Abiotic
Transformations; 2.1.3.4 Biodegradation and Persistence in Soil and Water; and
2.1.3.5 Summary. The discussion will emphasize the environmental fate of
aldicarb and to a much lesser extent aldicarb sulfoxide and aldicarb sulfone
in soil and water. Aldicarb sulfoxide and aldicarb sulfone are products of
aldicarb degradation in water and soil under aerobic conditions.
2.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as aldicarb (H = 4.1 x
10-9 atm«m3/mol) with Henry's constants less than 3.0 x 10~7 atm*m3/mol are
unlikely to undergo significant rates of volatilization from natural waters
under any conditions normally found in the environment. The volatilization
half-lives SAIC estimated for aldicarb in shallow waters under turbulent
conditions using equations A-1 and A-3 through A-8 support that
conclusion. Hie estimated volatilization half-lives for aldicarb in a
turbulent lake 1 m deep and in a shallow turbulent river or stream 1 m deep
are 51 years and 19 years, respectively (see Appendix A).
By substituting the reported Koc value for aldicarb of 36 into equation
A-9 along with the compound's vapor pressure (1.0 x 10-4 torr) and aqueous
solubility (6.0 x 103 mg/1), SAIC estimates a volatilization half-life for
aldicarb on soil of 34 days. Therefore, volatilization may be an important
-------
removal process for aldicarb on the soil surface, depending on the rates of
other removal processes. Although volatilization rates of aldicarb beneath
the soil surface would probably be much slower, decreasing rapidly with
increasing soil depth, Bromilow et al. (1980, as cited in Cohen et al. 1984)
suggested that volatilization could substantially contribute to the loss of
aldicarb from soil.
No information was found concerning the vapor pressures or volatilization
rates of aldicarb sulfone and sulfoxide.
2.1.3.2 Sorption and Leaching Potential
Substituting the estimated KoC value of 36 into equation A-11 gives
estimated sediment or soil/water partition coefficients (Kg/W) for aldicarb
ranging from 0.36 to 2.9 for sediments or soils with organic carbon fractions
ranging from 0.01 to 0.08, respectively. Aldicarb sulfoxide has reported Ks/W
values of 0.34 and 3.3 for soils with an organic fraction of 0.14 (Cohen et
al. 1984). Since Kg/W values for both aldicarb and aldicarb sulfoxide appear
to generally be less than 10, it is unlikely that the ratio of the total mass
of aldicarb residues adsorbed to suspended and exposed bottom sediment to the
total mass of aldicarb residues dissolved in the water column will exceed 0.01
in most surface waters. The reason, as discussed in Appendix A, is that the
ratio of the water mass to the suspended and exposed bottom sediment mass
exceed 10^ in most surface waters (USGS 1983). Therefore, it is unlikely that
transport by adsorption to suspended sediment and removal by adsorption to
exposed bottom sediments are important processes for either aldicarb or
aldicarb sulfoxide in surface waters.
Substitution of the estimated KqC value of 36 into equation A-13 gives
an estimated soil TLC Rf value (see Appendix A) of 0.71 for aldicarb adsorbed
to a soil with an organic fraction of 0.014, a pore fraction of 0.5, and a soil
density of 2.5 g/cm3. Therefore, based on the five mobility classes defined
by Helling and Turner (1968 and cited in Hamaker 1975) for a soil with the
same properties (Appendix A), aldicarb would be expected to be moderately
mobile (Class 4) in, and be moderately susceptible to leaching from, surface
soil. Since reported Ks/W values for aldicarb sulfoxide are similar to those
-------
from aldicarb, and since the aqueous solubility of aldicarb sulfoxide (43 g/1)
is several times greater than aldicarb, it would be expected that aldicarb
sulfoxide would be more susceptible to leaching than aldicarb. The suscepti-
bility of not only aldicarb and aldicarb sulfoxide, but also aldicarb sulfone
to leaching from soil, is demonstrated by reported aldicarb residues typically
ranging from 1 to 50 ppb in numerous groundwater samples taken from several
states (Cohen et al. 1984).
2.1.3.3 Abiotic Transformations
Aldicarb has reported hydrolysis half-lives of 12.5 years at pH 5.5 and
5°C, 5 years at pH 7.5 and 15°C, and 3.5 years at pH 8.5 and 5oc (Hansen and
Spiegel 1983, as cited in Cohen et al. 1984). Aldicarb sulfoxide has reported
hydrolysis half-lives in groundwater at 15QC of 22 years at pH 6, 2.2 years at
pH 7, 82 days at pH 8, and 8.2 days (by extrapolation) at pH 9 (Lemley and
Zhong 1984). Aldicarb sulfone has reported hydrolysis half-lives in ground-
water at 15°C of 6.7 years at pH 6, 243 days at pH 7, 24.3 days at pH 8, and
2.4 days (by extrapolation) at pH 9 (Lemley and Zhong 1984). Based on these
results, it appears that hydrolysis would not be an important removal process
for aldicarb over the normal pH range of 6 to 9 in natural waters or for
aldicarb sulfoxide and aldicarb sulfone at pHs less than 8. However, at pHs
greater than 8, hydrolysis may become an important removal process for
aldicarb sulfoxide and aldicarb sulfone.
No information was available in the literature reviewed by Cohen et al.
(1984) or SAIC concerning the photolysis or photo-oxidation of any of the
aldicarb compounds.
2.1.3.4 Biodegradation and Persistence in Soil and Water
Under aerobic conditions, oxidative half-lives for the oxidation of
aldicarb to aldicarb sulfoxide and aldicarb sulfone in soil and water are
reported to generally be less than 2 weeks (Bromilow et al. 1980, as cited
in Cohen et al. 1984). Under anaerobic conditions, aldicarb sulfoxide and
aldicarb sulfone are susceptible to reduction to aldicarb sulfide (Cohen
et al. 1984). Therefore, aldicarb in aerobic soils and possibly aldicarb
-------
sulfoxide and aldicarb sulfone in anaerobic soils could be classified as
nonpersistent (half-lives <20 days) in soils by Rao and Davidson (1980). There
was no information available on the persistence of aldicarb in anaerobic soils
or the persistence of aldicarb sulfide and aldicarb sulfone in aerobic soils.
2.1.3.5 Summary
Based on the above discussion and on the literature review by Cohen
et al. (1984), the following conclusions can be tentatively made with respect to
the most likely behavior of aldicarb, aldicarb sulfoxide, and aldicarb sulfone
in soil and water:
o Based on theoretical considerations, aldicarb is unlikely to undergo
significant rates of volatilization from natural waters under any
conditions. No information is available on volatilization rates for
aldicarb sulfoxide or aldicarb sulfone.
o Based on theoretical considerations and limited data, volatilization
may contribute significantly to the removal of aldicarb from the soil
surface and from soil at shallow depths.
o Based primarily on theoretical considerations, it is unlikely that
transport by adsorption to suspended sediments or removal by
adsorption to bottom sediments are important processes for either
aldicarb or aldicarb sulfoxide.
o Based upon theoretical considerations, leaching data and groundwater
monitoring, it appears that aldicarb, aldicarb sulfoxide, and aldicarb
sulfone are all susceptible to leaching from soils.
o Hydrolysis does not appear to be an important removal process for
aldicarb within the normal pH range of 6 to 9 (for natural waters) or
for aldicarb sulfoxide and aldicarb sulfone at pHs less than 8.
However, at pHs greater than 8, hydrolysis may become an important
removal process for aldicarb sulfoxide and aldicarb sulfone.
o Aldicarb appears to readily undergo oxidation to aldicarb sulfoxide
and aldicarb sulfone in soils under aerobic conditions with a reported
half•life of less than 1 week. Aldicarb sulfoxide and aldicarb
sulfone are reported to readily undergo reduction to aldicarb sulfide
in anaerobic soils. There was no information available on the
persistence of aldicarb in anaerobic soils or the persistence of
aldicarb sulfoxide and aldicarb sulfone in aerobic soils.
-------
2.2 OCCURRENCE IN THE ENVIRONMENT
2.2.1 Water
This section presents available data from monitoring studies and surveys
to determine the extent of occurrence of aldicarb and its transformation
products, aldicarb sulfoxide and aldicarb sulfone, in public drinking water
supplies and water other than drinking water.
2.2.1.1 Occurrence in Drinking Water
USEPA (1983a) summarized data on aldicarb in ground water samples collected
near citrus groves in Florida. One sample from a possible drinking water source
had a reported level of 3 to 5 ug/1 aldicarb. The detection limit, number of
samples, and number of positive samples were not reported*
2.2.1.2 Occurrence in Water Other Than Drinking Water
Three reports were identified that summarized data on aldicarb detected
in ground water samples* Sample collection was conducted in several regions
of the United States, mostly in high use areas.
Zaki et al. (1982, as cited in Cohen et al. 1984) reported on levels of
aldicarb in ground water samples collected in 1982 from wells on Long Island,
New York. Total aldicarb residues greater than the detection limit of 1 ug/1
were found in ground water samples of 27 percent of 8,404 wells surveyed.
Cohen et al* (1984) reported data from unpublished EPA pesticide registra-
tion files, which showed aldicarb residues of 1 to 4 ug/1 in ground water
samples from wells in Oregon, Washington, Texas, and North Carolina. Samples
of water from wells near a lily bulb farm in northern California showed
aldicarb residues of up to 24 ug/1. Hie detection limit, number of samples,
and number of positive samples were not reported for the sampling conducted in
Oregon, Washington, Texas, North Carolina, and California. In an analysis of
groundwater from nine wells in southern New Jersey, samples from three of the
wells had aldicarb concentrations of 3, 4, and 50 ug/1; no detection limit was
reported.
-------
USEPA (1983a) summarized data on water samples collected since 1960 from
wells in Wisconsin, Florida, Maine, Virginia, and North Carolina. Samples
from approximately 4 percent of the wells studied had concentrations of
aldicarb in excess of 10 ug/1. The detection limit, number of samples, and
number of positive samples were not reported*
2.2.2 Occurrence in Ambient Air
No data were obtained on levels of aldicarb in ambient air.
2.2.3 Soil/Sediments
Only one study was identified that examined the occurrence of aldicarb
in soils. Carey and Kutz (1983) presented data on soil samples collected at
various depth intervals in areas treated with aldicarb in Florida and
Mississippi. Of eight samples collected in Florida at depth intervals between
0 and 213 cm, all four samples collected at depths of 122 cm and greater
contained detectable levels of aldicarb varying from 6,000 to 87,000 ug/kg, dry
weight. None of the seven samples collected in Mississippi at depth intervals
between 0 and 183 cm contained detectable levels of aldicarb. Hie minimum
detection limit reported was 1,000 ug/kg aldicarb.
There were no studies identified that examined the occurrence of aldicarb
in sediments.
2.2.4 Food
The Food and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the intake of various substances,
including aldicarb, from foods consumed by adults, toddlers, and infants. Die
FDA provided mean daily intakes of aldicarb reflecting detections of aldicarb
in 12 total diet studies conducted from April 1982 to April 1985 (FDA 1986).
For the 6- to 11-month old infant, daily intake of aldicarb was 0.002 ug/day.
For the 2-year old toddler, aldicarb intake was 0.006 ug/day. Aldicarb intakes
for adult males and females are presented in Table 2-1. For the adult male,
aldicarb intakes ranged between 0.008 and 0.018 ug/day* Daily aldicarb intakes
for adult females ranged between 0.008 and 0.014 ug/day. Intakes were highest
for the 60- to 65-year old age group of both sexes.
-------
Table 2-1. Summary of FDA Total Diet Study Estimates For
Adult Male and Female Aldicarb Intake
Sex/Age Group Intake (ug/day)*
14-16
year
old
female
0.009
14-16
year
old
male
0.008
25-30
year
old
female
0.008
25-30
year
old
male
0.011
60-65
year
old
female
0.014
60-65
year
old
male
0.018
Source: FDA 1986.
2.3 EXPOSURE SUMMARY
While little or no data are available on the extent of aldicarb in
drinking water, food, and air, a few limited surveys provide some useful
information on occurrence. Monitoring surveys of groundwater (not specified
as drinking water sources) in areas of high aldicarb use have reported levels
of aldicarb and its metabolites in the range of 1 to50 ug/1. These relatively
high levels are the direct result of aldicarb's susceptibility to leaching
from the soil column. Recent dietary studies report that the daily intake of
aldicarb for 25- to 30-year old males ranged from 0.008 to 0.018 ug/day, with a
mean daily intake of 0.011 ug/day. No data are available on aldicarb levels
in ambient air. However, based on aldicarb's low vapor pressure, levels in
air are expected to be negligible.
Because of the lack of monitoring data on aldicarb, and its degradation
products in public water supplies, EPA is unable to make any estimates of
exposure front drinking water. Based on monitoring data and physical and chemical
properties of the compounds, dietary exposure is expected to be very low.
Although no monitoring data are available, air levels of the compounds are also
expected to be low. As a result of the expected low levels in food and air, if
aldicarb occurs in drinking water, consumption of the water will be the major
route of exposure.
-------
3. CARBOFURAN
3.0 SUMMARY
Carbofuran is a widely used pesticide that has been shown to contaminate
shallow ground water and surface water as a result of typical agricultural
practices. Hie compound has been found in several states at levels as high
as 65 ug/1.
3.1 GENERAL CHARACTERISTICS
3.1.1 Physical/Chemical Properties
Carbofuran (2,3-dihydro-2,2-dimethyl-7-benzofuranyl methylcarbamate) is
an insecticide/nematocide. Synonyms for carbofuran are Furadan, Yaltox,
NIA-10242 ENT 27164, Bay 70143, Brifur, Crisfuran, Curaterr, D 1221, and
FMC 10242 (Berg 1986).
Carbofuran is a white crystalline solid at 25°C (Windholz 1976). It has
a molecular weight of 221.3, a molecular formula of Ci2H15N03' an(* a
point of 150 to 153°C (Windholz 1976). The aqueous solubility of carbofuran
at 25°c is 700 mg/1 (3.2 x 10" 3 mol/1) and its vapor pressure at 33°C is
2 x 10-5 torr (2.6 x 10-8 atm) (Verschueren 1983).
The ratio of the vapor pressure at 33°C to the aqueous solubility at
25°C gives an estimated Henry's constant for carbofuran at 25° to 33°C of
8.1 x 10~9 atm*m^/mol (2.9 x 10~® torr 1/mg). Reported Koc values for
carbofuran range from 29 to 63 (Cohen et al. 1984).
3.1.2 Use
Carbofuran is used as a broad spectrum carbamate insecticide, nematocide,
and miticide (Berg 1986). Gianessi (1986) estimates that current use of
carbofuran is about 8 million pounds.
-------
Glaze estimated that in 1980 approximately 11 million pounds oC carbo-
furan active ingredient were available for domestic usage and approximately 8
million pounds were exported (Glaze 1982).
Gianessi (1986) estimated that of the approximately 30 million pounds of
carbofuran used nationwide from 1978 to 1982, 20 million pounds were applied
to field corn and 4 million pounds were applied to sorghum. The remaining
production was used on a large number of crops, including alfalfa, tobacco,
peanuts, rice, sugarcane, potatoes, soybeans, sweet corn, cotton, grapes, and
small grains.
3.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of carbofuran is divided into
the following subsections: 3.1.3.1 Volatilization; 3.1.3.2 Sorption and
Leaching Potential; 3.1.3.3 Abiotic Transformations; 3.1.3.4 Biodegradation;
3.1.3.5 Fate in Water Treatment Plants; and 3.1.3.6 Summary. The discussion
will emphasize the environmental fate of carbofuran in soil and water.
3.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as carbofuran
(H = 8.1 x 10~9 atm«m3/mol) with Henry's constants less than 3 x 10"7 atm*m3/mol
are unlikely to undergo significant rates of volatilization from natural waters
under any conditions normally found in the environment. Volatilization half-
lives estimated by SAIC (see Appendix A) for carbofuran in shallow water under
turbulent conditions support that conclusion. The estimated volatilization
half-life for carbofuran in a turbulent lake 1 m deep with a mean current of
2 m/s and exposed to a mean wind velocity of 5 m/s are 27 years and 10 years,
respectively. Volatilization rates of carbofuran from ground water to the soil
column above are believed to be substantially less than volatilization rates
from surface waters to the atmosphere. This is due primarily to the laminar,
nonturbulent nature of ground water flow. Transport from ground water by
volatilization may be further reduced by a build-up of chemicals in the pore
air at the ground water/pore air interface, and an associated decrease in the
concentration gradient across the interface.
-------
The reported experimental volatilization half-life for carbofuran applied
to a moist, loamy soil at 25oc and exposed to a wind speed of 1 km/s was 24
days (Swann et al. 1979, as cited in Cohen et al. 1984). This suggests that
volatilization could be substantially involved in the removal of carbofuran
from surface soils. The relatively low estimated volatilization half-life for
carbofuran from soil surface compared to that for volatilization from water is
due to the small value of the organic carbon normalized sediment or soil/
water equilibrium partition coefficient (KoC). Volatilization rates for
carbofuran below the surface of the soil would probably be much slower than at
the soil surface.
3.1.3.2 Sorption and Leaching Potential
Once released to surface water, carbofuran is expected to remain in
solution and not to bind to suspended solids. The basis for this conclusion
is provided in the following analysis.
Reported Koc values for carbofuran range from 29 to 63 (Cohen et al.
1984). Substituting the median of the range of KqC values (46) into equation
A-11 of Appendix A gives estimated sediments or soil/water equilibrium
partition coefficients (Ks/W) for carbofuran ranging from 0.46 to 3.7 for sedi-
ments or soils with organic fractions ranging from 0.01 to 0.08. The estimated
Ks/w values suggest that at equilibrium, the concentration of carbofuran in
suspended and exposed bottom sediment may be comparable or even several times
greater than that in water. However, as discussed in Appendix A, the Ks/W
value would probably have to be greater than 100 in most aquatic systems for
the ratio of the chemical mass adsorbed to suspended and bottom sediments to
the chemical mass dissolved in the water column to exceed 0.1. The reason is
that in most aquatic systems most of the time, the ratio of the water mass to
the mass of suspended and exposed bottom sediment exceeds 103.
Various experimental studies have shown that carbofuran is susceptible to
leaching from a wide range of soils (USEPA 1984e). In addition, carbofuran has
been detected at concentrations typically between 1 and 50 ug/1 in a number of
unconfined aquifer groundwater samples taken in Wisconsin and New York (Cohen
et al. 1984).
-------
3.1.3.3 Abiotic Transformations
Carbofuran has been shown to hydrolyze in water. The rate of hydrolysis
is very dependent on pH. Reported hydrolytic half-lives for carbofuran in
water are approximately 1 year at pH 6, 4 weeks at pH ?, and 1 day at pH 9
(Cohen et al. 1984). Based on these data, Cohen et al. (1984) suggest that
the hydrolytic half-lives for carbofuran in groundwater would generally range
from 2 to 50 weeks.
Seiber et al. (1978, as cited in USEPA 1984e) reported that the presence
of sunlight increased the loss of carbofuran from rice paddy water and drainage
ponds. No other information on the possible photolysis or oxidation by
photochemically generated oxidants in water, soil, or air was found in the
literature searched by USEPA (1984e).
3.1.3*4 Persistence in Soil
Based on a review of the literature, Cohen et al. (1984) indicate that
half-lives of carbofuran in soils typically range from 1 to 37 weeks. It is
classified as moderately persistent in soils (Rao and Davidson 1980) as opposed
to persistent or nonpersistent. Degradation rates in soil appear to generally
be greater in alkaline than in acidic soils and are generally greater in anaerobic
or flooded soils them in aerobic soils (USEPA 1984e; Cohen et al. 1984).
3.1.3.5 Bioaccumulation
While no data have been identified on carbofuran's potential for bioaccumu-
lation, the pesticide is not ejected to bioaccumulate based on its physical
and chemical properties. Carbofuran has been shown to hydrolyze in water and
is not expected to persist in the environment for a sufficient period of time
to bioaccumulate.
3.1.3.6 Fate in Water Treatment Plants
Based on limited data, USEPA (1985b) suggests that the following processes
would be effective in removing carbofuran from drinking water: adsorption to
activated carbon, reverse osmosis, and ozonation.
-------
3.1.3.7 Summary
Based on the above discussion and the literature reviews by Cohen et al.
(1984) and USEPA (1984e), the following conclusions can be made concerning the
most likely behavior of carbofuran in soil and water:
o Carbofuran is unlikely to undergo significant rates of volatilization
from surface water or groundwater under any conditions normally
encountered in the environment.
o Volatilization appears to be a significant removal process for
carbofuran on the soil surface. An experimental half-life of 24 days
has been reported for carbofuran applied to a soil surface.
o Based on theoretical considerations, the proportion of total carbo-
furan present in an aquatic system that is adsorbed to suspended and
exposed bottom sediment is probably generally much less than the
proportion dissolved in the water column.
o Based on experimental results, carbofuran may be classified as at
least moderately susceptible to leaching due to its low Koc value.
o Hydrolysis appears to be an important removal process for carbofuran
in alkaline waters. Reported hydrolytic half-lives for carbofuran in
water decrease with increasing pH from 1 year at pH 6 to 1 day at
pH 9.
o Reported and estimated half-lives for carbofuran in soils range from
1 to 37 weeks. Removal rates are generally greater in alkaline or
anaerobic (flooded) soils than in acidic or aerobic soils.
o Due to its moderate persistence and mobility in soils, carbofuran may
contaminate some shallow unconfined aquifers or streams located near
areas of carbofuran use.
o The persistence of carbofuran in soil does not appear to be long
enough to pose any long-term threat to water supplies in previously
contaminated areas that are no longer being contaminated.
3.2 OCCURRENCE IN THE ENVIRONMENT
3.2.1 Water
This section presents available data from monitoring studies and surveys
to determine the extent of occurrence of carbofuran in public drinking water
supplies and water other than drinking water.
-------
3.2.1.1 Occurrence in Drinking Water
No national studies were obtained addressing the occurrence of carbofuran
in drinking water. However, three state studies were reviewed and are
summarized below.
Ground Water Sources — Regional Studies
In 1984, the Suffolk County Department of Health Services (Holden 1986)
examined drinking water wells in Long Island, New York, for various pesticides.
The survey sampled both public and private wells in close proximity to fields
where carbofuran, aldicarb, and other pesticides were used. Of the 5,083 wells
sampled, 1,535 contained detectable levels of carbofuran and 250 to 300 wells
contained levels greater than 15 ug/1. The maximum level reported was 65 ug/1.
As part of a program of the Wisconsin Department of Natural Resources
(Holden 1986), drinking water wells were also analyzed for carbofuran
during 1983-1984. This statewide Wisconsin study examined wells suspected of
contamination by both point and nonpoint sources. Of 78 samples analyzed, 2
were positive with a high concentration of 7 ug/1. No other information was
reported.
Ground water wells were sampled near Richmond, Rhode Island, as part of
a cooperative project between the CJ.S. Geological Survey and the Rhode Island
Water Resources Board to identify potential drinking water sources for future
water supply use (Offutt 1984). Eleven samples were collected during 1984
from five locations and analyzed for carbofuran. Seven were positive with a
mean concentration of 3.7 ug/1 {range = 2 to 7 ug/1). The detection limit was
1.0 ug/1.
3.2.1.2 Occurrence in Water Other Than Drinking Water
One national study of unidentified sources is reported for the occurrence
of carbofuran. Four regional studies addressing levels of carbofuran in ground
water and three surface water studies are also discussed here.
-------
Ground Water Sources
Unpublished data summarized from EPA registration files and reported in
Cohen et al. (1984) indicated that concentrations of carbofuran were present
in samples of groundwater collected from private wells in areas with sandy
soils and water table aquifers in Wisconsin and New York. Concentrations
ranged from 1 to 50 ug/1 carbofuran. The number of samples, number of
positive samples, and detection limit were not reported.
Three studies were obtained examining groundwater wells in California.
These studies were conducted by the California Department of Food and
Agriculture in 1981, and again in 1982, and an evaluation in 1984 during the
California State Board's Toxics Special Project {Ramlit Associates,. Inc.
1983; Holden 1986; Cohen and Bowes 1984, respectively). Overall, approxi-
mately 30 counties were sampled for the occurrence of carbofuran. Over 200
samples were collected from as many sites, with only 2 proving positive. One
concentration was 0.5 ug/1, the other was not reported. "Hie detection limit
for the sample concentration given above was also not reported.
surface Water Sources
The Army Corps of Engineers sampled U.S. Geological Survey water stations
along rivers of the Honey Creek watershed in northwest Ohio during 1981 (Datta
no date). Carbofuran was sampled for at 12 locations and a maximum concen-
tration of 45 ug/1 was found. The number of samples, number of positives, and
detection limit were not reported.
Dudley and Karr (1980, as cited in USEPA 1984d) presented data on levels
of carbofuran in water, fish, and sediment samples collected from a stream
draining the Black Creek agricultural watershed in Allen County, Indiana,
during 1977-1978. Although the detection limit and the number of water samples
collected and tested were not reported, none of the samples contained carbofuran
in excess of the detection limit.
Woodham et al. (1975) presented monitoring data from a study to determine
whether significant pesticide accumulation had occurred in two counties in
North Carolina. Samples of pond water collected both inside and outside the
-------
study area contained no carbofuran in excess of the detection limit of
0.05 ug/1.
Unidentified Sources
Water samples of unidentified sources were collected nationally from
various studies and entered in the U.S. EPA's STORET data base (as reported in
USEPA 1984e) during 1979 to 1982. Of the 21 stations sampled for carbofuran,
11 had undetectable levels. However, 58 samples were reported as positive.
The total number of samples, detection limit(s), and range of positive values
were not reported.
3.2.2 Occurrence in Ambient Air
No monitoring data that addressed levels of carbofuran in ambient air
were found.
3.2.3 Soi1/Sedimen ts
Several studies have been identified that examine the occurrence of
carbofuran in soil and sediments. Soil and sediment samples were collected
from two counties in North Carolina where tobacco cropland had been treated.
One soil sample contained 2,000 ug/kg of carbofuran; other soil samples
contained less than 50 ug/kg of carbofuran. No residues of carbofuran were
found in pond sediment samples in excess of the detection limit of 50 ug/kg
(Woodham et al. 1975).
Samples of sediment collected from a stream draining an agricultural
watershed in Indiana had no detectable levels of carbofuran (Dudley and Karr
1980, as cited in USEPA 1984d). The number of samples tested and the detection
limit were not reported.
Kadoum and Mock (1978 as cited in USEPA 1984e) monitored sediment in
tailwater pits collecting irrigation runoff in Haskell County, Kansas, in
1974. Carbofuran was detected in pit bottom sediments in four of 54 samples
collected from 31 pits (four pits implicated) at mean levels between 30.6 and
50.0 ppb (max. 759 ppb).
-------
3.2.4 Food
The Food and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the intake of various substances,
including total carbofuran, from foods consumed by adults, toddlers, and
infants. The FDA provided mean daily intakes of carbofuran reflecting
detections of carbofuran in 12 total diet studies conducted from April 1982
to April 1985 (FDA 1986). For the 6- to 11-month old infant, daily intake of
carbofuran was 0.0011 ug/day. For the 2-year old toddler, carbofuran intake
was 0.005 ug/day. Total carbofuran intakes for adult males and females are
presented in Table 3-1. For the adult male, carbofuran intakes ranged between
0.005 and 0.010 ug/day. Daily carbofuran intakes for adult females ranged
between 0.006 and 0.013 ug/day. Intakes were highest for the 25- to 30-year
old age group of both sexes.
TABLE 3-1. Summary of FDA Total Diet Study
Estimate for Adult Male and Female
Total Carbofuran Intake
Sex/Age Group Intake (ug/day)*
14-16
year
old
female
0.006
14-16
year
old
male
0.005
25-30
year
old
female
0.013
25-30
year
old
male
0.010
60-65
year
old
female
0.006
60-65
year
old
male
0.008
Source: FDA 1986.
3.3 SUMMARY
Limited studies have been conducted that provide useful data on the
extent of occurrence of carbofuran in drinking water, food, and air.
Monitoring surveys of public water supplies have reported carbofuran levels as
high as 65 ug/1 in some high-use agricultural areas. Other regional studies
typically reported maximum concentrations only as high as 7 ug/1. While some
recent dietary studies report that the daily intake of carbofuran for 25- to
-------
30-year old males and females is 0.010 and 0.013 ug/day, respectively, other
studies have estimated that dietary levels may be as high as 460 ug/day. The
actual range of intake of carbofuran in the diet is unknown. No data were
found on carbofuran levels in ambient air. However, based on carbofuran's
low vapor pressure and high water solubility, levels in air are expected to
be very low.
According to the available monitoring data, the following limited
conclusions can be made on the total exposure to carbofuran. Drinking water
exposure in areas where carbofuran is used can result in intakes of greater
than 100 ug/day. Typically, however, intakes of 10-15 ug/day may be expected.
While dietary levels are not clearly defined, actual measurements suggest that
for the majority of the population, intake levels are low. Inhalation intake
is expected to be minimal.
-------
4. CHLORDANE
4.0 SUMMARY
Chlordane is a nonpoiar liquid that is currently used as a soil insec-
ticide. In the environment, chlordane tends to bind to soil and to degrade
slowly. Limited monitoring data are available for chlordane in drinking
water. Because of its current use and its tendency to bind to soil, chlordane
is expected to occur only at very low concentrations in drinking water, less
than 0.01 ug/1. Because of the wide use of chlordane, it could potentially occur
in all parts of the United States.
4.1 GENERAL CHARACTERISTICS
4.1.1 Physical/Chemical Properties
Chlordane (1,2,4,5,6,7,8,8-octachloro-2,3,3a,4,7,7a-hexahydro-4,7-meth-
anoindene) is a broad spectrum insecticide belonging to the group of chlori-
nated hydrocarbons known as cyclodiene insecticides (USEPA 1983b). Synonyms and
identifiers for chlordane include the following (Berg 1986):
Belt Kilex Lindane Synklor
Chlor-Kil Kypchlor Termi-Ded
Chlortox Niran Topiclor 20
Corodane Octachlor Velsicol 1068
Gold Crest C-100 Octa-Klor
Pure chlordane is a pale yellow liquid at 25°C composed of a 75:25 ratio
of the cis and trans isomers (USEPA 1980a; Verschueren 1983). It is completely
miscible in most organic solvents (Berg 1986). Chlordane has a molecular
weight of 409.8 and a molecular structure of CigHgC^g (Windholz 1976). The
boiling points of the cis and trans isomers are 107 to 109°C and 103 to 105°C,
respectively (Callahan et al. 1979). The pure chlordane mixture has an aqueous
solubility and vapor pressure at 25°C of 5.6 x 10~2 mg/1 (1.37 x 10" 7 mol/1)
and 1 x 10~5 torr (1.3 x 10~8 atm), respectively (Callahan et al. 1979).
Technical chlordane is a viscous amber liquid with a chlorine odor con-
sisting of a mixture of 60 to 75 percent chlordane isomers and 25 to 40 percent
-------
of 24 other organochlorine compounds (Berg 1986; Verschueren 1983). Technical
chlordane has an aqueous solubility of approximately 9 ug/1 (Verschueren 1983).
The ratio of the vapor pressure to the aqueous solubility gives an
estimated Henry's constant at 25°c for chlordane of 9.5 x 10-5 atm*m3/mol.
The ratio of the estimated Henry's constant to the product of the gas constant
times the temperature in degrees Kelvin gives an estimated dimensionless
Henry's constant of 3.89 x 10"^. Chlordane has an estimated K__ value of
oc
1.4 x 105 (Mabey et al. 1981).
4.1.2 Use
An EPA review of chlordane, begun in March 1971, concluded that this
chemical is carcinogenic and that its use produces widespread contamination of
the environment based on residues found in soil, air, water, food, wildlife,
and man. On November 26, 1974, EPA published a notice of intent to cancel all
registered uses of chlordane, except for subsurface ground insertion for termite
control and dipping of roots and tops of nonfood plants. The EPA Administrator
issued a notice of intent to suspend the registrations of certain products
containing chlordane on July 25, 1975, and the final notice of suspension
was issued on December 24, 1975. Cancellation proceedings continued through
November 1977, at which time settlement negotiations began between EPA and the
Velsicol Chemical Corporation.
The final cancellation order putting into effect the terms of settlement
was issued on March 6, 1978 (USEPA 1983b; Kirk-Othmer 1979). In this agreement,
the use of chlordane on food and other crops was to be phased out over a 5-year
period. Underground termite control would be the only EPA-approved use of
chlordane after 1980. In addition, all nontermite uses of chlordane during the
phase-out period was restricted to treatment by certified applicators or profes-
sional commercial seed treatment companies (Chemical Regulation Reporter 1978).
Prior to its suspension and subsequent cancellation for most uses, chlor-
dane was widely used as a pre-emergent insecticide for the control of corn
rootworms, wireworms, and cutworms; as a soil treatment for the control of
termites in structures; and as an insecticide for control of a variety of
-------
pests on citrus, potatoes, strawberries, and tomatoes, and on lawns, gardens,
turf, and ornamentals. Minor agricultural uses included applications to hay,
tobacco, soybeans, vegetables, and peanuts (USEPA 1976).
Domestic production and use of chlordane has declined nearly 50 percent
since 1974, when the EPA issued the notice of intent to cancel all agricultural
uses of chlordane. However, the use of chlordane as a termiticide has continued,
and chlordane is currently the most widely used insecticide for the control of
subterranean termites in the United States according to the results of a risk/
benefit analysis of chemicals used for subterranean termite control (USEPA
1983b). EPA estimated that 6.5 million pounds of chlordane were used for
industrial/commercial purposes (primarily termite control) in 1972. The break-
down of use by region was: North Central, 1.6 million pounds; South Central,
1.3 million pounds; Southeast, 1.3 million pounds; Northeast, 1.0 million
pounds; Southwest, 1 million pounds; and Northwest, 0.3 million pounds. In
1986 approximately 4 million pounds of chlordane were estimated to be used
(Gianessi 1986). This figure is based on data obtained from various state,
regional, and national pesticide usage surveys conducted by, USDA, USEPA, and
the Department of Food and Agriculture of the State of California. USEPA
(1983b) reported that the largest quantity was initially distributed in Region IV,
which includes Alabama, Georgia, Florida, Mississippi, North Carolina, South
Carolina, Tennessee, and Kentucky. Because chlordane is applied by subsurface
ground injection, the potential for ground water contamination is high.
4.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of chlordane is divided into the
following subsectionst 4.1.3.1 Volatilization; 4.1.3.2 Sorption and Leaching
Potential; 4.1.3.4 Abiotic Transformations; 4.1.3.4 Biodegradatlon and
Persistence in Soil and Water; and 4.1.3.5 Summary. The discussion will
emphasize the environmental fate of chlordane in soil and water.
4.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as chlordane (H = 9.5 x
10-5 atm«m3/mol) with Henry's constants exceeding 10~5 atm«m3/mol are likely
-------
to undergo significant rates of volatilization from surface waters under any
conditions normally found in the environment. Volatilization half-lives
estimated for chlordane in rivers or streams and turbulent lakes by SAIC using
equations A-1 and A-3 through A-8 indicate that volatilization may be an
important removal process for chlordane in shallow, turbulent waters, but may
not be for some deep, stagnant waters depending on the relative rates of other
removal processes such as adsorption to sediments with respect to volatilization.
Estimated volatilization half-lives for chlordane in rivers or streams range
from 12 hours in a shallow turbulent river or stream to 140 days in a deep low
flow river or stream (see Appendix A). Estimated volatilization half-lives for
chlordane in a turbulent lake range from 28 hours in a lake 1 m deep to 12 days
in a lake 10 m deep. This analysis, however, does not reflect that chlordane
will rapidly migrate to sediments where they will be adsorbed and thus will not
be available for volatilization.
The volatilization rates of chlordane from groundwater to the soil column
above are projected to be substantially less than volatilization rates from
surface waters to the atmosphere. This is due primarily to the laminar,
nonturbulent nature of groundwater flow. Volatilization rates from ground water
may be further reduced by a build-up of chemical in the pore air at the ground
water/pore air interface, and an associated decrease in the concentration
gradient across the interface.
By substituting the estimated KQC value for chlordane of 1.4 x 105
(Section 4.1) into equation A-9 along with the compound's vapor pressure
(1 x 10-5 torr) and aqueous solubility (5.6 x 10-2 mg/1), SAIC estimates the
volatilization half-life for chlordane on soil to be 12 days. Therefore,
given the resistance of chlordane to chemical and biological degradation,
volatilization could be an important removal process for chlordane on soil.
4.1.3.2 Sorption and Leaching Potential
Substituting the estimated KQC value of 1.4 x 105 into equation A-11
gives estimated sediment or soil/water equilibrium partition coefficients
(Kg/W) for chlordane ranging from 1.4 x 103 to 1.1 x 104 for sediments or
soils with organic carbon fractions ranging from 0.01 to 0.08. Iherefore,
-------
even though the ratio of the water mass to the mass of suspended and exposed
bottom sediments exceeds 103 in most surface waters (USGS 1983), a substantial
proportion of the total mass of chlordane may be adsorbed to suspended or
exposed bottom sediments in many surface waters. Therefore, transport by
adsorption to suspended sediment and removal by adsorption to bottom sediment
are predicted generally to be important processes for chlordane in surface
waters.
Substitution of the estimated KQC value of 1.4 x 105 into equation A-13
gives an estimated soil TLC Rf value (Appendix A) of <0.10. Therefore, based
on the five mobility classes defined by Helling and Turner (1968) and cited in
Hamaker (1975) for a soil with the same properties (Appendix A), chlordane would
be expected to be immobile (Class 1) in, and extremely resistant to leaching
from, surface soil.
4.1.3.3 Abiotic Transformations
Eichelberger and Lichtenberg (1971, as cited in Callahan et al. 1979)
reported that greater than 97 percent of added chlordane was recovered from
river water samples incubated at room temperature for 8 weeks. Biis suggests
that neither hydrolysis nor biodegradation are significant removal processes
for chlordane in natural waters.
Callahan et al. (1979) summarize reports that indicate that chlordane may
be susceptible to photochemical sensitization that involves the decomposition
of a compound via transfer of absorbed solar energy for naturally occurring
organics. However, insufficient data are presented to estimate photodecom-
position rates*
4.1.3.4 Biodegradation and Persistence in Soil and Water
Chlordane appears to be resistant to biodegradation not only in river
water (Subsection 4.1.3.3), but in soil as well. Castro and Yoshida (1971)
and Watanbe (1973), both as cited in Callahan et al. (1979), reported that
chlordane is persistent in both flooded (anaerobic) and nonflooded (aerobic)
soils. Rao and Davidson (1980) reported a degradative half-life for chlordane
-------
in soil of approximately 3.3 years. Residues of chlordane may persist for 14
or more years after application at detectable levels, depending on the appli
cation rate and soil environment (USEPA 1980c). Therefore, chlordane appears
to be extremely persistent in soils.
4.1.3.5 Summary
Based on the above discussion and the literature review by Callahan
et al. (1979), the following tentative conclusions can be made concerning the
most likely behavior of chlordane in soil and water:
o Based on theoretical considerations, limited data, and the apparent
resistance of chlordane to hydrolysis and biodegradation, volatil-
ization along with adsorption to suspended and bottom sediments are
predicted to be the major removal mechanisms for chlordane in surface
waters.
o Based on theoretical considerations, chlordane is predicted to be
highly resistant to leaching from surface soils.
o Based upon limited data, chlordane appears to be resisant to both
hydrolysis and biodegradation in river water. It may be. somewhat
susceptible to photodecomposition via photochemical sensitization, but
no rates are available.
o Based on several soil studies and on the observed persistence of
chlordane residues in soils at detectable levels for many years after
application, chlordane can be classified as extremely persistent in
soils.
4.2 OCCURRENCE IN THE ENVIRONMENT
4.2.1 Water
The following section presents the data available from monitoring studies
and surveys to determine the extent of occurrence of chlordane in public
drinking water supplies and water other than drinking water.
-------
4.2.1.1 Occurrence in Drinking Water
Ten regional studies of chlordane in drinking water are addressed in this
section. Where possible, groundwater sources and surface water sources have
been discussed separately.
Ground Water Sources
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. None of the 100
ground water supplies sampled, representative of the 5 aquifers in Florida,
contained measurable levels of chlordane (no detection limit was reported).
Less than 8 percent of samples from 96 locations utilizing the Floridan
aquifer were found positive for chlordane, in a 1984 study by the Florida
Department of Environmental Resources and the U.S. Geological Survey (Holden
1986). These supplies serve over three million people. No other information
from the study was reported.
No positive samples were found for chlordane in the following two studies;
one involved the 1984-1985 sampling of 42 sites from 12 towns in Connecticut
and the other involved the analysis of 67 samples from Long Island, New York,
in 1984 (Waggoner 1985; Holden 1986). The Connecticut drinking water wells
serve a population of over 570,000, and the detection limit for that study was
3.3 ug/1. No other information about either study was reported.
Two positive samples (one for alpha chlordane and one for gamma-chlordane)
out of 107 samples analyzed were found for groundwater wells in Idaho (Idaho
Department of Health and Welfare 1984). Monitoring for pesticides in drinking
water wells is not routinely done; the sampling performed was in response to a
particular contamination incident, not for any comprehensive monitoring program.
A relatively low mean of 0.0002 ug/1 was reported for these two samples, with
one sample having a chlordane concentration of 0.038 ug/1 (no detection limit
was reported). It appears that "negative" samples, possibly assigned a
detection limit value, were included in the calculation of the mean.
-------
Benvenue et al. (1972b, as cited in Cirelli 1978) conducted a study to
determine the extent of organochlorine pesticide contamination of drinking
water in Hawaii. A total of 45 finished drinking water samples were collected
from February 1971 to May 1971. Concentrations of chlordane were detected in
four of the samples, with a range of positive values between 0.0005 and 0.005
ug/1, and an average concentration of 0.001 ug/1 (no detection limit was reported).
Surface Water Sources
Irwin and Healy (1978), summarizing data collected during a water quality
reconnaissance of public water supplies in Florida, reported that none of the
16 surface water supplies sampled contained chlordane in excess of the detection
limit. No detection limit was reported.
In the New Orleans Water Supply Study conducted by USEPA Region VI (USEPA
1975a), samples of drinking water were analyzed for levels of halogenated
organics. Although the number of positive samples was not reported, the con-
centrations of chlordane in three samples analyzed ranged from "non-detected"
to less than 0.1 ug/1. No detection limit for the analyses was reported.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. The
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. A research team, appointed by
the Association, reviewed data on the quality of water at 31 representative
Canadian and Ubited States communities and 24 offshore sites to evaluate the
human health implications.
For each of the 31 communities, data consisted of: 1) background infor-
mation on the community; 2) treatment plant schematics and associated treatment
process information; and 3) water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological and radiological parameters, inorganic
-------
parameters, and organic parameters (including volatiles, base/neutrals,
pesticides and PCBs, and phenols and acids). For each parameter, the water
type, time period, concentration (mean and range), number of samples, and
detection limit are presented.
For most of the volatile organics, including chlordane, the available
data indicated that there were very low levels of these contaminants in the
raw, treated, or tap water. Most of the values found were "not detected" or
near the detection limit (Canadian Public Health Association 1986).
Unspecified Sources
Two regional studies identified levels of pollutants in drinking water
from unspecified sources. A Region V Survey (USEPA 1975b), conducted during
the first three months of 1975, assessed levels of organic chemicals in
samples of raw and finished drinking water collected from 83 utilities in
Indiana, Illinois, Michigan, Minnesota, Ohio, and Wisconsin. Samples were
analyzed for various pollutants including gamma-chlordane. One sample of
finished drinking water contained gamma-chlordane at a level of 0.004 ug/1.
No limits of detection for pesticides were reported.
In a study to identify the sources of pollutants entering a sewage
treatment plant. Levins et al. (1979a) collected samples from two drinking
water sources. Although no detection limit was reported, the analyses
detected no chlordane in any of the samples.
Schafer et al. (1969, as cited in USEPA 1980a) reported concentrations of
chlordane up to 8 ug/1 for finished drinking waters, although no location(s)
were given. No other Information on this study was reported.
4.2.1.2 Occurrence in Water Other Than Drinking Water
Two national studies and seven regional studies analyzed concentrations
of chlordane in water other than drinking water.
-------
Ground Water Sources — Regional Study
Tucker and Burke (1978) reported the results of a statewide ground water
monitoring project conducted in New Jersey. For this study, samples of water
were collected from 163 wells in nine counties. The analysis showed that
samples of water from five wells contained levels of chlordane in excess of
the minimum reportable concentrations of 0.01 ug/1. The range of positive
values was 0.01 to 0.02 ug/1 chlordane.
One California study of ground water wells found four samples, from as
many counties, positive for chlordane with a maximum value of 22 ug/1. No
other data were presented for this study. One of three well samples collected
in the Barstow, California, area was found to have a chlordane concentration
of 0.4 ug/1 by the Victorville Regional Water Quality Board in 1980 (Raralit
Associates, Inc. 1983).
Surface Water Sources — National Study
The National Pesticide Monitoring Network (Gilliom et al. 1985) examined
surface water samples from rivers nationwide during 1975-1980, and found no
positive samples of chlordane out of 2,943 samples analyzed from 177 loca-
tions. The detection limit was 0.15 ug/1.
The National Surface Water Monitoring Program (Carey and Kutz 1983)
presented data on levels of chlordane in surface water samples collected
throughout the United States between 1976 and 1980. Although no detection
limit for chlordane was reported, 1.1 percent of the samples analyzed had
detectable concentrations of chlordane, with a maximum reported value of 0.23
ug/1. The number of samples taken was not reported. It is not known whether
the samples were filtered or unfiltered.
Surface Water Sources — Regional Studies
Truhlar and Reed (1976) reported the results of analysis of water samples
taken from four streams in Pennsylvania. The streams drained four types of
land use areas: forests, general farms, orchards, and residential areas.
-------
None of the 19 samples collected and analyzed from April 1970 to February 1971
contained chlordane in excess of the detection limit. Hie detection limit was
not reported.
Benvenue et al. (1972b, as cited in Cirelli 1978) conducted extensive
sampling of two canals on Qahu, Hawaii, to determine the extent of organo-
chlorine pesticide contamination. A total of nine samples were collected from
two canals and analyzed for residues of chlordane. The analyses showed an
average concentration of chlordane of 0.007 ug/1 and a range of positive values
of 0.003 to 0.018 ug/1. Hie number of positive samples and the detection
limit were not reported.
Barks (1978) presented the results of a USGS water quality study con-
ducted from April 1973 to July 1974 in the Ozark National Scenic Riverways,
Missouri. During the study, 20 surface water samples were collected from
3 sites on the Current River and 1 site on Jacks Fork and analyzed for pesticide
content. None of the unfiltered samples contained concentrations of chlordane
in excess of the detection limit (no detection limit was reported).
Englande et al. (1978) presented the results of extensive chemical analyses
of six Advanced Wastewater Treatment (AWT) plant effluents. The plants were
located in California, the District of Columbia, and Texas. A mean concentration
of less than 0.039 ug/1 chlordane was identified in one plant effluent; the
other systems had no detectable levels of chlordane. Die number of positive
samples and the detection limit were not reported.
4.2.2 Occurrence in Ambient Air
Results of the Suburban Air Sampling Program of the National Pesticide
Monitoring Program for 1975 were reported in Kutz et al. (1976) and SRI
(1983). Samples were collected at three suburban locations during April (two
samples), May (four samples), and June (four samples). Concentrations of
chlordane were detected in eight samples from a city in Florida, with a mean
of 0.015 ug/m3 and a range of "not detected" to 0.026 ug/m3. A mean of 0.035
ug/m3 and a range of trace to 0.059 ug/m3 were reported for samples collected
from a city in Mississippi. All ten samples were positive. No chlordane was
-------
of the detection limit (the detection limit was not reported). Only one
sample of the 21 sediment samples collected and analyzed contained chlordane.
The sample was collected from a stream draining a residential area and con-
tained 250 ug/kg chlordane. No detection limit for the analyses was reported.
Several reports addressed levels of chlordane in stream bed sediments.
Britton et al. (1983) reported on levels of pesticides in water-sediment
mixtures (unfiltered samples) and in bottom material samples collected by the
National Stream Quality Accounting Network (NASQAN) in 1976. Throughout the
United States, 151 permanent stations plus stations added as part of local
programs were sampled for pesticides, including chlordane. Water-sediment
samples were collected quarterly; bottom material samples were collected
semiannually. Chlordane was detected in water-sediment samples collected at
three of six stations in California; the maximum chlordane level found at
these stations was 1.7 ug/kg. Concentrations of chlordane in excess of the
detection limit were reported in 24 of 153 bottom material samples collected
from 13 different regions (the detection limit was not reported). The maximum
level found at these stations was 333 ug/kg.
Carey and Kutz (1983) summarized data on pesticide residues in sediment
samples collected by the National Surface Water Monitoring Program (NWMP) from
1976 to 1980. Chlordane was detected in 15.3 percent of the sediment samples
analyzed; the maximum reported value was 2,964 ug/kg. No information was
available regarding the detection limits of analysis*
4.2*5 Food
The Food'and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the intake of various contaminants,
including chlordane, from foods consumed by adults, toddlers, and infants.
The FDA provided mean daily intakes of chlordane reflecting detections of
chlordane in 12 total diet studies conducted from April 1982 to April 1985
(FDA 1986). For the 6- to 11-month infant, the daily intake of chlordane was
0.016 ug/day. For the 2-year old toddler, chlordane intake was 0.019 ug/day.
Intakes of chlordane by adult males and females are presented in Table 4-1.
For the adult male, chlordane intakes ranged between 0.024 and 0.081 ug/day.
-------
Daily chlordane intakes for adult females ranged between 0.025 and 0.070 ug/day.
Intakes were highest for the 60- to 65-year old age group of both sexes.
Table 4-1. Summary of FDA Total Diet Study Estimates for Chlordane
Intakes from Adult Males and Females.
Sex/Age Group Intake (ug/day)
14-16
year
old
female
0.025
14-16
year
old
male
0.024
25-30
year
old
female
0.048
25-30
year
old
male
0.044
60-65
year
old
female
0.070
60-65
year
old
male
0.081
Source: FDA 1986.
4.3 EXPOSURE SUMMARY
While few current data are available on the extent of occurrence of chlor-
dane in drinking water, air, and food, several limited surveys provide some
useful information. Chlordane is very persistent in the soil column and conse-
quently, can pose a long-term but low-intensity threat to groundwater supplies.
As a result of this low-intensity threat, the maximum reported level of chlordane
in drinking water is only 0.005 ug/1. Data on the occurrence of chlordane in
ambient air are also limited. Even during the early and mid-1970's when chlordane
was widely used as an agricultural insecticide, outdoor air levels were low,
with a reported maximum value of 0.059 ug/m3. Current ambient air levels are
likely to be less because today's uses have less potential for release into the
air. However, potential does exist for indoor air exposure because chlordane
is often injected directly into a house's foundation to fight termites.
Unfortunately, current data reflecting this practice are not available.
Recently dietary studies report the mean daily intake of chlordane for
25-30 year old males to be 0.044 ug/day. Earlier dietary studies reflected
the presence of pesticide residues on the samples. However, with the ban of
chlordane in 1978, this source of contamination was eliminated. Consequently,
dietary levels of chlordane are less today than in 1978.
-------
TABLE 4-2
Reported
Estimated
Route
Exposure Levels
Adult Intakes
Drinking Water
<0.1 ug/1
<0.2 ug/day
Diet
—
0.044 ug/day
Air
*
~Ambient air levels of the compounds are believed to be very low. Indoor air
levels, although unknown, are expected to be higher than outdoor.
Hie information that is currently available on the occurrence of chlordane
in the environment and the potential for exposure are insufficient to determine
the national distribution of intake of chlordane by any of the three routes.
Because data reported for drinking water diet and air exposure levels were
derived from studies conducted during the early 1970s to the mid 1980's (prior
to the restriction of most uses of chlordane), current levels from all three
sources are expected to be minimal. However, additional data are needed to
evaluate current exposure levels and to provide a comparable basis for estimating
the total combined intake from each media.
-------
5. DBCP
5.0 SUMMARY
DBCP is a chemical that has been widely used, as a soil fumigant. As a
result of this use, DBCP contamination has been detected in a number of
groundwater wells at .levels of several ug/1. Due to the cancellation of most
of DBCP's uses in the last 10 years, production of this chemical has greatly
declined. Because of the current low level of use, DBCP is not expected to be
a common contaminant of drinking water supplies.
5.1 GENERAL CHARACTERISTICS
5.1.1 Physical/Chemical Properties
DBCP (1,2-dibromo-3-chloropropane) is a soil fumigant. Synonyms for DBCP
included BBC 12, Fumazone, Nemagon, Nemanax, Nemaset, Nematocide, and Oxy DBCP
(Berg 1986).
DBCP is an amber to dark brown liquid at 25°C (Verschueren 1983). It has
a molecular weight of 236.36, a melting point of 6°C, and a boiling pdint of
196°C (Windholz 1976). The aqueous solubility and vapor pressure of DBCP at
21°C are 1,000 mg/1 (4.2 x 10-3 mol/1) and 0.8 torr (1.1 x 10-3 atm), respec-
tively.
The ratio of the vapor pressure to the aqueous solubility gives an esti-
mated Henry's constant for DBCP at 21 °C of 2.6 x 10-4 atm*mVmol (8.0 x 1 0~4
torr 1/mg). The ratio of the estimated Henry's constant to the product of the
gas constant times the temperature in degrees Kelvin gives an estimated dirnen-
sionless Henry's constant of 1.08 x 10~2. Reported Koc values for DBCP
include 55, 119, 129, and 149 (Cohen et al. 1984).
5.1.2 'Use
DBCP was used as a soil fumigant for nematode control (Berg 1986).
Although domestic production figures could not be obtained, an estimated
-------
32,658,000 pounds of DBCP active ingredient were used in 1977. in that year,
the three major domestic producers of DBCP voluntarily curtailed production
(USDA 1978). This action on the part of the three domestic producers came as
a result of the EPA's decision on November 3, 1977 to greatly curtail the uses
of DBCP on 19 crops (42 FR 57543). On October 29, 1979, the EPA Administrator
issued the Agency's final decision to suspend unconditionally the registrations
of DBCP for all uses except on pineapple fields in Hawaii (44 FR 65161; USEPA
1984h).
As noted above, DBCP registrations for all uses except on pineapple fields
in Hawaii were suspended by EPA on October 29, 1979. Of the approximately
43,000 acres devoted to the production of pineapples in Hawaii, 5,000 acres
were treated in 1977 with approximately 665,200 kg of DBCP at a rate of 133 kg
of active ingredient per acre (USDA 1978). Cancellation proceedings are
currently under consideration for the use of DBCP on pineapple fields in Hawaii,
on the basis of recently acquired information indicating that the use of DBCP
as a soil fumigant has led to the contamination of ground water aquifers used
as sources of drinking water (USEPA 1984h).
5.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of DBCP is divided into the
following subsections: 5.1.3.1 Volatilization; 5.1.3.2 Sorption and Leaching
Potential; 5.1.3.4 Abiotic Transformations; 5.1.3.4 Biodegradation and
Persistence in Soil and Water; and 5.1.3.5 Summary. The discussion will
emphasize the environmental fate of DBCP in soil and water.
5.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as DBCP (H = 2.6 x 10~4
atm«m3/mol) with Henry'3 constants greater than 10~4 atm*m3/mol are likely to
undergo significant rates of volatilization from surface waters under any
conditions normally found in the environment. Volatilization half-lives
estimated for DBCP in rivers or streams and turbulent lakes by SAIC using
equations A-1 and A-3 through A-8 indicate that volatilization is an important
process for DBCP in shallow turbulent waters, but may not be for some deep
-------
stagnant waters, depending on the relative rates of other removal processes
with respect to volatilization. Estimated volatilization half-lives for DBCP
in rivers or streams range from 3.7 hours in a shallow turbulent river or
stream to 86 days in a deep low flow river or stream (see Appendix A). Estimated
volatilization half-lives for DBCP in turbulent lakes range from 16 hours in a
lake 1 m deep to 6.6 days in a lake 10 m deep.
The volatilization rates of DBCP from groundwater to the soil column above
are projected to be substantially less than volatilization rates from surface
waters to the atmosphere. This is due primarily to the laminar, nonturbulent
nature of groundwater flow. Volatilization rate from ground water may be
further reduced by a build-up of chemical in the pore air at the ground
water/pore air interface, and an associated decrease in the concentration
gradient across the interface.
By substituting a mean KqC value for DBCP of 113 (Section 5.1.1) into
equation A-9 along with the compound's vapor pressure (0.80 torr) and aqueous
solubility (1,000 mg/1), SAIC estimates the volatilization half-life for DBCP
on the soil surface to be 3.2 minutes. Therefore, volatilization from the soil
surface is predicted to be extremely rapid. Even though volatilization rates
of DBCP in soil beneath the surface would be much slower, the extremely rapid
estimated volatilization rates from the soil surface suggest that volatilization
could be an important removal process for DBCP beneath the soil surface as well.
5.1.3.2 Sorption and Leaching Potential
Substituting the mean KqC value of 113 into equation A-11 gives estimated
sediment or soil/water partition coefficients (Kg/W) for DBCP ranging from 1.1
to 9.0 for sediments or soils with organic carbon fractions ranging from 0.01
to 0.08, respectively. Since the estimated Ks/W values for DBCP are less than
10, it is unlikely that the ratio of the total mass of DBCP adsorbed to suspended
and exposed sediments to the total mass of DBCP dissolved in the water column
will exceed 0.01 in most surface waters. The reason is that in most surface
waters most of the time, the ratio of the water mass to the mass of suspended
and exposed bottom sediment exceeds 1,000 (USGS 1983). Therefore, transport by
adsorption to suspended sediment and removal by adsorption to bottom sediment
are probably not important processes for DBCP in surface waters.
-------
Substitution of the mean Koc value of 113 into equation A-1 3 gives an
estimated soil TLC Rf value (Appendix A) of 0.44 for DBCP adsorbed to a soil
with an organic carbon fraction of 0.014, a pore fraction of 0.5, and a soil
density of 2.5 g/cm3 . Therefore, based on the five mobility classes defined
by Helling and Turner (1968) and cited by Hamaker (1975) for a soil with the
same properties (Appendix A), DBCP would be expected to have an intermediate
mobility (Class 3) in, and to be intermediate between being moderately sus-
ceptible and moderately resistant to leaching from, surface soil. However,
DBCP has been detected in groundwater samples from several states including
Hawaii, California, Arizona, South Carolina, Maryland, and Alabama (Cohen
et al. 1984; Carey and Kutz 1983). In Hawaii, an aquifer that lay several
hundred feet beneath the surface was contaminated. In one California study,
DBCP was reported to have leached 15 m through the unsaturated zone (Nelson
et al. 1981, as cited in Cohen et al. 1984).
5.1.3.3 Abiotic Transformations
Hydrolysis half-lives for DBCP at pH 7 are reported to be 1 41 years at
150C and 38 years at 25°C (Burlinson et al. 1982). Estimated hydrolysis
half-lives for DBCP at pH 9 range from 5 years at 1S^C to 0.6 years at 25°C
(Deeley 1986). Therefore, hydrolysis does not appear to be a significant
removal process for DBCP in natural waters. No information on photolysis or
oxidation is available. However, since DBCP does not have any chromophores
that absorb light strongly above the approximate solar radiation cutoff at the
earth's surface of 290 nm nor a functional group susceptible to oxidation, it
is unlikely that DBCP undergoes significant rates of direct photolysis or
oxidation in the environment.
5.1.3.4 Biodegradation and Persistence in Soil and Water
Under optimal laboratory conditions in a bioactive agricultural soil, DBCP
was reported to have a degradation half-life of approximately 10 weeks (Castro
and Belser 1968, as cited in Cohen et al. 1984). Based on that, DBCP would be
classified as moderately persistent (half-life between 20 and 100 days) in soil
under optimal conditions (Rao and Davidson 1980). No other information on
biodegradation was available in the literature reviewed by Cohen et al. (1984).
-------
5.1.3.5 Summary
Based on the above discussion and on the. literature review by Cohen
et al. (1984), the following tentative conclusions can be made concerning the
most likely behavior of DBCP in soil and water:
0 Based upon theoretical considerations and reportedly low hydrolysis
rates, volatilization is expected to be the primary mechanism of DBCP
removal in surface waters and on the soil.
0 Based upon theoretical considerations, transport by adsorption to
suspended sediment and removal by adsorption to bottom sediment are
not expected to be important processes for DBCP in aquatic systems.
° Based upon theoretical considerations, leaching data, and groundwater
monitoring, DBCP appears to be at least moderately susceptible to
leaching.
0 Reported hydrolysis half-lives for DBCP at pH 7 exceed 10 years.
Therefore, hydrolysis does not appear to be an important removal
process for DBCP in water. No information is available on photolysis
or oxidation. However, based on its chemical structure, DBCP is not
expected to undergo significant rates of photolysis or oxidation in
the environment.
0 Under optimal conditions in a bioactive agricultural soil, DBCP is
reported to have a degradation half-life of 10 weeks, which cor-
responds to a classification of relatively persistent in soils-.
0 Due to its estimated rapid volatilization from most surface waters,
DBCP is not expected to significantly contaminate surface water supplies.
5.2 OCCURRENCE IN THE ENVIRONMENT
5.2.1 Water
This section presents available data from monitoring studies and surveys
to determine the extent of occurrence of DBCP in public drinking water supplies
and water other than drinking water.
5.2.1.1 Occurrence in Drinking Water
Only one regional study was found that addressed levels of DBCP in
samples of drinking water obtained from groundwater and surface water sources.
-------
Groundwater Sources
Carey and Kutz (1983) presented data collected in 1979 on levels of DBCP
in ground water samples collected in California, Arizona, Texas, South Carolina,
and Alabama. Eight positive ground water samples collected from California and
South Carolina showed DBCP concentrations ranging from 0.01 ug/1 to 10.8 ug/1,
and an average concentration of 2.8 ug/1. Seven of the positive samples were
from private wells; one sample containing 0.01 ug/1 was from a municipal well.
The number of ground water samples tested was not reported.
Surface Water Sources
Carey and Kutz (1983) also reported on levels of DBCP in samples of surface
water collected in California, Arizona, Texas, South Carolina, and Alabama in
1979. Two positive samples collected from municipal water supplies in California
showed DBCP concentrations of 0.09 ug/1 and 0.1 ug/1. The number of surface
water samples tested was not reported.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. The
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. The data collected cover the
period from the mid-1970*s to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background informa-
tion on the community; 2) treatment plant schematics and associated treatment
process information; and 3) water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological and radiolpgical parameters, inorganic
parameters, and organic parameters (including volatiles, base/neutrals,
-------
pesticides and PCBs, and phenols and acids). For each parameter, the water
type, time period, concentration (mean and range), number of samples, and
detection limit are presented.
For most of the volatile organics, including DBCP, the available data
indicated that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
5.2.1.2 Occurrence in Water Other Than Drinking Water
Several regional studies (Peoples et al. 1980; Nelson et al. 1981; Pinto
1980; Cohen 1981; Carter and Riley 1981; and Mink 1981, all as cited in
Cohen et al. 1984) and the EPA pesticide registration files, as reported in
Cohen et al. (1984), provided monitoring data on levels of DBCP in ground water.
Positive concentrations of DBCP ranging from 0.02 ug/1 to 20 ug/1 were detected
in samples of ground water in Hawaii, California, Arizona, South Carolina, and
Maryland. Areas with the highest concentrations were the San Joaquin Valley in
California and the region southwest of Phoenix, Arizona. Hie number of samples
analyzed, the number of positive samples detected, and the detection limit were
not reported.
Holden (196j») reported that, in a massive survey of more than 8,000 public
and private wells in the San Joaquin Valley of California, detectable levels of
DBCP were found in approximately 25% of the wells tested* The highest level of
DBCP found in the survey was 1,240 ug/1.
Samples of groundwater collected from seven previously unsampled sites in
Hawaii showed'detectable levels of DBCP in the range of 0.05 ug/1 to 0.5 ug/1
(USEPA 1984h)V7 Some of the sites have been used as drinking water supplies.
The number of samples tested, the number of positive samples, and the detec-
tion limit were not reported.
-------
5.2.2 Occurrence in Ambient Air
Pellizzari and Bunch (1979) analyzed seven air samples in El Dorado,
Arkansas, for DBCP. Six of the seven samples were DBCP positive, with mean and
median concentrations of 0.0064 and 0.0018 ug/m3r respectively, and a maximum
detected concentration of 0.014 ug/m^.
5.2.3 Soil/Sediments
Minimal information was available on studies that had been conducted to
determine the occurrence of DBCP in soil and sediments. Two such studies were
identified, both involving soil sample collection. No sediment studies were
identified.
Analysis of California soil cores collected by Nelson et al. (1981, as
cited in Cohen et al. 1984) showed ug/kg amounts of DBCP that had leached 15 m
through the unsaturated zone. Analysis of additional California soil cores
collected by Zalkin et al. (1983 as cited in Cohen et al. 1984) detected no
DBCP in soil cores collected as deep as 10m, 4 years after the last DBCP
application, indicating rapid downward movement of DBCP in the soil. In both
studies, the number of samples tested, the number of positive samples detected,
and the detection limit were not reported.
5.2.4 Food
Several studies were identified that examined the occurrence of DBCP in
food in the United States* These studies are summarized here.
Monitoring data provided by Del Monte Corporation and the PDA showed no
detectable residues of DBCP in a variety of samples analyzed (USEPA 1984h).
However, these residue data are "limited in scope." The number of samples
analyzed and the detection limit were not reported.
USDA (1978) presented data from several sources on DBCP residues in a
variety of raw agricultural commodities. Newsome et al. (1977, as cited in
USDA 1978) detected DBCP residues ranging from 20 to 1,500 ug/kg in carrots
-------
grown in fumigated soil. The detection limit, the number of samples analyzed,
and the number of positive samples were not reported.
The Shell Chemical Company provided data on levels of DBCP in raw agricultural
commodities. Residues of DBCP in the range of 10 to 1,120 ug/kg were detected
in broccoli, cabbage, cauliflower, and cucumbers. Additional monitoring data
provided by Dow Chemical Company identified residues of DBCP in peanut kernels
in the range of 10 to 140 ug/kg. The number of samples analyzed, the number of
positive samples detected, and the detection limit were not provided (USDA 1978).
In its compliance program report for FY 79 on pesticides and metals, the
FDA (1982c) presented data on levels of DBCP in samples of domestic and imported
fish. A total of seven of 1,515 fish samples analyzed contained levels of DBCP
in excess of the detection limit. The seven positive values were from domestic
fish samples; the maximum value reported was "trace" (detection limit not given).
No detectable levels of DBCP were found in peaches or pineapples grown in
DBCP-treated areas during the FDA's FY79 surveys (FDA 1981b). The detection
limit for these studies was not reported.
The data obtained on levels of DBCP in food were insufficient for use in
estimating the typical dietary intake of DBCP.
5.3 EXPOSURE SUMMARY
Current data are unavailable on the extent of occurrence of DBCP in
drinking water, food, and air. According to the monitoring data collected
during the late 1970's when DBCP use was still widespread, contamination of
surface water supplies was minimal because of the compounds high volatility.
However, due to its moderate susceptibility to leaching from the soil column,
DBCP levels in some groundwater supplies ranged from 0.01 to as high as 10.8
ug/1. Because these surveys, while limited, were performed in areas of high
DBCP usage and because very little DBCP has been released to the environment
since 1977, current levels are not expected to exceed those of the earlier
surveys. The available food and air data are no longer valid for estimating
exposure to DBCP because most of the data were collected before DBCP was
restricted. However, current intake levels are expected to be minimal.
-------
6. 2,4-D
6.0 SUMMARY
2,4-D (2,4-dichlorophenoxy) acetic acid) is a widely used systemic
herbicide. Large amounts of 2,4-D and its esters are used in household and
agricultural herbicide products. Because of 2,4-D's tendency to bind to soils
and to degrade relatively rapidly, levels in drinking water are low. Surveys
of 2,4-D in drinking water have found that when the compound occurs at
detectable levels the concentration does not exceed 1 to 2 ug/1.
6.1 GENERAL CHARACTERISTICS
6.1.1 Physical/Chemical Properties
2,4-d (2,4-dichlorophenoxy) acetic acid) is a systemic herbicide. The term
2,4-D includes the parent acid as well as the 35 derivatives (esters and salts)
(USEPA 1982d). Only a small amount of the parent acid is used commercially.
Synonyms and identifiers for 2,4-D, which are numerous, include Agrotect,
Aqua Kleen, Dinoxol, Estone, Herbidal, Salvo, Weedone, Weed-B-Gon, Tributon,
Transamine, and Miracle (Berg 1986).
2,4-D is a white powder at 25°C (Verschueren 1983). It has a molecular
weight of 221.04, a molecular formula of CeHgCI^C^, and a melting point of
138°C (Windholz 1976). The aqueous solubility of 2,4-D is 890 mg/1 at 25°C
(Verschueren 1983) and Its vapor pressure at 25°C is 6.0 x 10-7 torr (Laskowski
et al. 1982). the ratio of the vapor pressure to the aqueous solubility gives
an estimated Henry's constant at 25°C for 2,4-D of 2 x 10-10 atm«m3/mol.
Reported KqC values for 2,4-D include 20 (Rao and Davidson 1980) and 32
(Hamaker 1975). The pKa value for 2,4-D is 2.80.
The aqueous solubility and vapor pressure at 258C of the n-butyl ester of
2,4-D are 1 mg/1 and 3.9 x 10-4 torr, respectively (Zepp et al. 1975). The
ratio of the vapor pressure to the aqueous solubility gives an estimated
Henry's constant for the n-butyl ester of 2,4-D of 1.6 x 10-4 atm*mVmol. No
information on the physical/chemical properties of other esters of 2,4-D could
be found in the literature.
-------
6.1.2 Use
An estimated 60 to 75 million pounds of 2,4-D active ingredient were
produced annually in the United States. Imports and exports of 2,4-D are
estimated at 3 to 5 million and 10 to 15 million pounds, respectively. Actual
domestic usage of 2,4-D in 1986 was 60 million pounds (Kuch 1986). Gianessi
(1986) reported approximately 40 million pounds used in a 33-state area.
Registered uses of 2,4-D include post-emergent weed control in grasses,
wheat, barley, oats, sorghum, corn, sugarcane, and rice. Certain formulations
are registered for pine release, water hyacinth control, and prevention of
seed formation, and other formulations are registered for broadleaf control
in cereal grains (Berg 1986).
An estimated 15 to 20 million pounds of 2,4-D active ingredient (approxi-
mately 31 to 32% of the total amount of 2,4-D estimated to have been used in
1980) was applied to wheat (Kuch 1980). Hie CJSDA reported that 5.1 million
pounds of 2,4-D were used on field corn nationwide in 1982 (USDA 1983).
Rangeland and pastureland throughout the United States were treated with
between 10 and 12 million pounds of 2,4-D in 1980 (Kuch 1980); 46,500 pounds
of 2,4-D were used in 1982 on California rangelands (California 1982).
In 1980, between 6 and 7 million pounds of 2,4-D were applied to sites
such as airfields, driveways, equipment yards, fencerows, implement storage
yards, incineration areas, industrial yards, loading areas and ramps, mills,
oil tank farms, parking areas, pole yards, sidewalks, warehouse plant sites,
vacant lots, and around buildings and water meters.
EPA reported the use of 3 to 3.5 million pounds of 2,4-D nationwide on
lawns and turf in 1980 (Kuch 1980). EPA has registered the use of 2,4-D on
bahaigrass, bentgrass, bermudagrass, bluegrass, carpetgrass, centipedegrass,
dichondra, fescue, red top, St. Augustine grass, zoysia .grass, and golf
courses.
-------
An estimated 4 to 6 percent of the total domestic usage of 2,4-D in 1980
(2-3 million pounds) was applied to aquatic areas throughout the United States.
EPA has registered 2,4-D as an herbicide to control weeds in and around lakes
and ponds {impounded water); in marshes, estuaries, and shore lines; along
ditch banks, drainage ditch banks, and irrigation ditch banks (adjacent to
water); and in nonpotable water, drainage ditches, drainage and irrigation
systems, streams, and waterways (Kuch 1980).
According to EPA, the following uses of 2,4-D in 1980 represented only
4 percent of the total domestic usage for that year: sugar cane, rice, barley,
soybeans, almonds, sweet corn, sunflowers, apples, pecans, pears, peanuts,
cherries, citrus, peaches, plums and prunes, walnuts, and forestry (Kuch 1980).
6.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of 2,4-D is divided into the
following subsections: 6.1.3.1 Volatilization; 6.1.3.2 Sorption and Leaching
Potential; 6.1.3.3 Abiotic Transformations; 6.1.3.4 Biodegradation and
Persistence in Soil and Water; and 6.1.3.5 Summary.
2,4-D is applied as an acid and as various esters and amine salts of the
acid compound. In the environment, the n-butyl ester appears to undergo rapid
hydrolysis to 2,4-D (Zepp et al. 1975). This discussion will emphasize the
environmental fate of 2,4-D in soil and water.
6.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as 2,4-D (H = 2.0 x
10-10 atm«m3/mol) with a Henry's constant less than 3 x 10~10 atm*m3/mol are
less volatile than water. Therefore, the concentration of 2,4-D in water is
not expected to decrease at all due to volatilization.
Substitution of the mean reported Koc value of 26 for 2,4-D into equation
A-9 along with the compound's aqueous solubility (890 mg at 25°C) and vapor
pressure (6.0 x 10-7 torr at 25°C) gives an estimated volatilization half-life
of 1.7 years for 2,4-D on the soil surface. Volatilization rates for 2,4-D
-------
beneath the soil surface are expected to be even slower, decreasing rapidly
with increasing soil depth (Lyman et al. 1982).
As stated above, the n-butyl ester of 2,4-D readily hydrolyzes to the
parent acid under environmental conditions. However, the n-butyl ester also
readily volatilizes from surface water and soil. No other information on the
fate of the esters of 2,4-D could be found in the literature (Zepp et al. 1975).
6.1.3.2 Sorption and Leaching Potential
The pKa value for 2,4-D is 2.80. Because the pH of most soils is greater
than 4.5 and that of most natural waters is greater than 6.0 (Tinsley 1979;
Stumm and Morgan 1970), 2,4-D will probably exist primarily in the anionic,
dissociated form in the environment. Soils and sediments generally ,have over
all negative potentials except at low pH, thus the adsorption of anions such
as the 2,4-D anion to soils and sediments will be weak unless the soils and
sediments have large anion exchange capacities or are at a low pH. This would
help to explain a low mean reported KqC value of 26 for 2,4-D.
Substitution of the mean reported Koc value of 26 into equation A-11 gives
estimated sediment or soil/water equilibrium partition coefficients (Kg/W) of
0.26 to 2.1 for 2,4-D adsorbed to sediments or soils with organic carbon
fractions ranging from 0.01 to 0.08, respectively. Because the estimated Ks^w
values are less than 10 for an assumed Koc value of 26, it is not expected that
the ratio of the total mass of adsorbed 2,4-D to dissolved 2,4-D would exceed
0.01 in surface waters for which the mean KqC value of the suspended and bottom
sediment is <0.26. The reason is that the ratio of the water mass to the mass
of suspended and exposed bottom sediments exceeds 103 in most surface waters
(USGS 1983). However, in cases where the sediment has a large anion exchange
capacity, transport by adsorption to suspended sediment and removal by adsorption
to bottom sediment may be important processes for 2,4-D in surface waters.
Substitution of the mean reported KqC value of 26 into equation A-13 gives
an estimated soil TLC Rf value of 0.79 for 2,4-D adsorbed to a soil with an
organic carbon fraction of 0.014, a pore fraction of 0.5, and a soil density of
2.5 g/cm3. Therefore, based on the five mobility classes defined by Helling
-------
and Turner (1968) and cited by Hamaker (1975) for a soil with the same properties
(Appendix A), 2,4-D would be expected to be at least moderately mobile (Class
4) in, and to be at least moderately susceptible to leaching from, surface soils
with a Koc value of <26. However, 2,4-D would be expected to be far less mobile
in soils with a large anion exchange, capacity.
Although several field studies have indicated that 2,4-D is susceptible
to moderate leaching, the higher concentrations of the compound in those
studies remained in the upper layers of the soil (Bovey and Young 1980). For
example, in tests conducted by the U.S. Air Force, 4,000 pounds per acre of
Herbicide Orange were applied at a depth of 15 cm in soil plots in Utah.
Although 2,4-D residues were detected to a depth of 90 cm after 282 days, more
than 90 percent of the residues remained in the top 30 cm of soil (NRCC 1978).
Studies by Barnett et al. {1967 as cited in NRCC 1978) on the movement of
2,4-D residues in soil under simulated rainfall showed that most of the 2,4-D
remained at a depth of 0 to 8 cm, although some was present at a depth of
8 to 15 cm. Only negligible amounts were detected below 15 cm. The fact that
2,4-D does not appear to be as susceptible to leaching from soils in the
studies cited above as would be predicted by a Koc value of <26 may be due to
a major contribution to adsorption by anion exchange.
There have been a number of studies on the removal of 2,4-D from soils by
surface runoff water, either through leaching or by erosion of soil particulates
to which the herbicides are adsorbed (Barnett et al. 1967; Bovey and Young
1980; Lutz et al. 1973; Lawson 1976). Although substantial quantities of the
herbicide are often detected in runoff occurring soon after herbicide applica-
tions, concentrations rapidly decline in succeeding runoffs. Total losses of
herbicide due to runoff over substantial periods of time do not generally
account for more than 5 percent of the applied herbicide. For example, although
concentrations of 2,4-D in initial runoff from recently sprayed experimental
plots in North Carolina were often relatively high (0.3-4.2 ppm), concentrations
in succeeding runoffs rapidly declined (Bovey and Young 1980). Total herbicide
loss after several runoffs typically accounted for less than 1 percent of the
applied herbicide. Nevertheless, substantial quantities of herbicide can be
transported by surface runoff from watersheds receiving heavy herbicide spraying
and heavy rainfall.
-------
6.1.3.3 Abiotic Transformations
Although some of the esters of 2,4-D are susceptible to rapid hydrolysis
in the environment to 2,4-D (SAIC 1981b), 2,4-D itself is unlikely to undergo
significant rates of abiotic hydrolysis in the environment due to its lack of
functional groups that are readily susceptible to hydrolysis (Lyman et al.
1982).
Phenoxy herbicides such as 2,4-D can undergo several different photo-
reactions in water: photo-oxidation of the phenoxy side chain to form
chlorophenols; photonucleophilic displacement of CI by OH to form chloro-
phenols; or a photo-reductive dechlorination involving the replacement of Cl
with H to form phenoxyacetic acid (Crosby 1976? Akermark 1978).
Zepp et al. (1975) estimated the average photolytic half-life of 2,4-D in
shallow clear water as follows. The rate of photolysis was assumed to be
given by the following equation:
d[C] = - kA a [C]
dt
Where
kA = average rate of sunlight absorption
4 = quantum yield.
The average rate of sunlight absorption was calculated by integrating sunlight
intensity data over a t2-hour period of sunlight for September on a clear day
at latitude 34°N. If the quantum yield is assumed to be independent of wave
length or is an average value, the rate of photolysis becomes pseudo first
order with a half-life given by;
/2 3 2/ka a
Substituting the value of kA and a reported quantum yield for 2,4-D in water
into equation 2-5, it was estimated that the photolytic half-life of 2,4-D in
-------
clear, shallow water in September at latitude 34°N, exposed to 12 hours of
unobstructed sunlight, would be 20 days (Zepp et al. 1975). Estimates of
half-lives under cloud cover or in deeper water would be much greater.
6.1.3.4 Biodegradation and Persistence in Soil and Water
Several studies have indicated that 2,4-D is susceptible to relatively
rapid rates cf degradation in both water and soil. Schultz and Herman (1971)
determined the persistence of 2,4-D in the water and mud of nine ponds located
in Florida, Georgia, and Missouri. The ponds were sprayed with the dimethyl-
amine salt of 2,4-D at the rate of 2, 4, or 8 pounds pec acre acid equivalent
of 2,4-D. Residues of 2,4-D declined in Florida and Georgia ponds, from
maximums of 0.35 and 0.69 mg/1, respectively, observed 3 days after spraying,
to less than 0.005 mg/1 within 14 and 28 days, respectively, after spraying.
2,4-D residues in the mud of the ponds at these locations never exceeded 0.05
mg/kg and declined to less than 0.005 mg/kg within 56 days after spraying.
Residues in Missouri ponds declined from a maximum of 0.63 mg/1 to less than
0.005 mg/1 within 56 days after spraying. A maximum value of 0.170 mg/kg
2,4-D residue wa3 found in the mud of these ponds, but no 2,4-D residues could
be detected in the mud of the ponds past 28 days after spraying. Altom and
Stritzke (1973) reported that the average half-life of 2,4-D in three Oklahoma
soils was 4 days.
Norris (1966) studied the persistence of 2,4-D in forest floor litter
under laboratory conditions. Approximately 85 percent of 2,4-D applied to red
alder forest floor litter was decarboxylated within 300 hours.
6.1.3.5 Summary
Based upon the above discussion the following tentative conclusions can
be made concerning the most likely behavior of 2,4-D in soil and water:
o Most esters of 2,4-D are expected to be rapidly hydrolyzed to the
parent acid.
o Based upon theoretical considerations, volatilization is not expected
to be an important removal process for 2,4-D in water or on soil.
-------
o Based upon theoretical considerations, transport by adsorption to
suspended sediment and removal by adsorption to bottom sediments are
not expected to be important for sediments with low anion exchange
capacity but may be important for sediments with large anion exchange
capacities.
o Based upon reported Koc values for 2,4-D, the compound is expected to
be mobile in soil and susceptible to leaching. However, the mobility
of 2,4-D would be expected to be much lower in soils with large anion
exchange capacities. Although monitoring data have indicated tnat
2,4-D is moderately susceptible to leaching, leaching does not appear
to be as an important process for 2,4-D in some soils as was predicted
based on the reported Koc value in surface soils. The reason may be
due to possibly strong anion exchange capacities of subsurface soils.
o Based upon theoretical considerations, 2,4-D is unlikely to undergo
significant rates of hydrolysis in the environment. However, 2,4-D
does appear to be susceptible to direct photolysis and photooxidation
in the environment. A daytime photolysis half-life of 20 days has
been estimated for 2,4-D in shallow water in September at latitude
34°N.
o Based on several studies, 2,4-D appears to be susceptible to almost
complete biodegradation within 2 to 8 weeks in soil and surface water.
6.2 OCCURRENCE IN THE ENVIRONMENT
6.2.1 Water
The following section presents the available data from monitoring studies
and surveys to determine the extent of occurrence of 2,4-D in public drinking
water supplies and water other than drinking water.
6.2.1.1 Occurrence in Drinking Water
Studies at both the national and regional levels have addressed concen-
trations of 2,4-D in drinking water. The results of three national studies
and two regional studies are discussed in this section. Reported levels of
2,4-D in drinking water obtained from groundwater sources and surface water
sources are discussed separately.
-------
Ground Water Sources — National Studies
A detailed survey of the contaminants in the water supplies of 10 cities
(selected on the basis of their source of raw water) was conducted as part of
the National Organics Reconnaissance Survey (NORS) (USEPA 1975b). Two of the
systems sampled, located in Florida and Arizona and serving more than 100,000
individuals, used ground water as their water source. Water samples taken from
these water supplies had concentrations of 2,4-D below the minimum quantifiable
concentration, although the minimum quantifiable concentration was not reported.
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) was conducted from June 1977 to March 1981. Samples of finished
drinking water were collected from 12 ground water systems of varying sizes
throughout the United States and analyzed for 2,4-D. None of the drinking
water samples from the 12 ground water systems contained levels of 2,4-D in
excess of the quantification limit of 0.5 ug/1.
As part of the 1978 Rural Water Survey (USEPA 1984i), samples were
collected from 267 households in rural locations throughout the United States
for 2,4-D analyses. A total of 71 public ground water systems of varying sizes
were covered by the survey. None of the samples collected from these public
ground water systems contained concentrations of 2,4-D in excess of the minimum
quantification limit of 0.01 ug/1.
Information provided by the Federal Reporting Data System (FRDS 1984),
which contains information on public water supplies that are found to be in
violation of current Maximum Contaminant Levels (MCLs), indicated that no
violations of the 2,4-D MCL of 100 ug/1 were reported during the years 1979-1983.
Ground Water Sources — Regional Study
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. Analysis of samples
of finished drinking water taken from 100 water supplies utilizing five aquifers
in Florida identified four supplies with positive values of 2,4-D of 0.02 to
0.28 ug/1, well below the MCL of 100 ug/1.
-------
Surface Water Sources — National Studies
In the NORS survey of 10 cities (USEPA 1975b), samples of drinking water
were collected from eight systems having surface waters affected by different
types of pollution. The systems were selected based on the type of pollution
affecting their water supplies: 1) uncontaminated upland waters; 2) raw water
contaminated with agricultural runoff; 3) raw water contaminated with munici-
pal wastes; and 4) raw water contaminated with industrial discharges. Samples
of water from one system contaminated with agricultural runoff contained
detectable levels of 2,4-D (0.04 ug/1). The minimum quantifiable concentra-
tion was not reported.
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) also contained information on levels of 2,4-D in drinking water
obtained from surface water sources. Between June 1977 and March 1981,
finished drinking water samples were collected from 105 surface water systems
of varying sizes throughout the United States and analyzed for 2,4-D. Only
one of the 105 surface water systems sampled contained 2,4-0 in excess of the
quantification limit of 0.5 ug/1. This system was located in Texas, and con-
tained 1.1 ug/1 of 2,4-D.
The 1978 Rural Water Survey (USEPA 1984i) also presented data on levels
of 2,4-D in drinking water obtained from surface water sources. Samples were
collected from 21 public surface water systems of varying sizes and analyzed
for 2,4-D. None of the samples collected from these systems contained concen-
trations of 2,4-D in excess of the minimum quantification limit of 0.01 ug/1.
Drinking water from surface water supplies in the United States must be
monitored for 2,4-D at least once every 3 years in accordance with the National
Interim Primary Drinking Water Regulations. Data obtained from the FRDS for
the period from 1979 to 1983 indicated that no violations of the 2,4-D MCL of
100 ug/1 had been reported (FRDS 1984).
-------
Surface Water Sources — Regional Studies
Irwin and Healy (1978) summarized data collected during a water quality
reconnaissance of public water supplies in Florida and reported that finished
drinking water samples collected from 9 of 16 surface water supplies contained
2,4-D in the range of 0.04 to 0.94 ug/1, well below the MCL of 100 ug/1.
In a study on the effects of forest runoff on the quality of a public
water supply in Oregon, Elliott (1979) observed an ambient concentration of
2,4-D of 50 ug/1.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October, 1984 through August, 1985 (Canadian Public Health Association
1986). The data collected covered the period from the mid 1970s to early 1985.
The study was funded by the Health Protection Branch of Health and Welfare
Canada, and the Ontario Ministry of the Environment. The data collected cover
the period from the mid-1970's to early 1985. A research team, appointed by
the Association, reviewed data on the quality of water at 31 representative
Canadian and United States communities and 24 offshore sites to evaluate the
human health implications.
For each of the 31 communities, data consisted of: 1) background informa-
tion on the community; 2) treatment plant schematics and associated treatment
process information; and 3) water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological and radiological parameters, inorganic
parameters, and organic parameters (including volatiles, base neutrals, pesti-
cides and PCBs, and phenols and acids). For each parameter, the water type,
time period, concentration (mean and range), number of samples, and detection
limit are presented.
For most of the volatile organics, including 2,4-D, the available data
indicated that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
-------
6.2.1.2 Occurrence in Water Other Than Drinking Water
One national study and six regional studies presented data on monitored
levels of 2,4-D in water other than drinking water. All seven studies addressed
levels of 2,4-D in surface water.
National Study
The National Surface Water Monitoring Program (NWMP) (1982) presented data
on levels of 2,4-D in surface water samples collected throughout the United
States during the years 1975-1978. During this period, 1,557 surface water
samples were collected and analyzed. Although no detection limit was reported
for 2,4-D, 26 samples contained concentrations of 2,4-D ranging up to 1.91 ug/1.
Regional Studies
Truhlar and Reed (1976) reported on water samples collected from four
streams in Pennsylvania and analyzed for chlorinated hydrocarbon pesticides
during the period from April 1970 to February 1971. The streams drained four
different types of land use areas: forest, general farms, orchards, and resi-
dential areas. Of the 19 samples collected and analyzed, 10 contained concen-
trations of 2,4-D in excess of the detection limit. Two-thirds of the samples
collected from the stream draining the residential area and half of the samples
collected from the streams draining the general farms and orchard areas contained
residues of 2,4-D at levels ranging from 0.0 to 0.08 ug/1. No detection limit
was reported for the study.
Joyce and Sikka (1977) conducted a study for the U.S. Corps of Engineers
to determine the levels of 2,4-D in St. Johns River water as a result of 2,4-D
use during routine water hyacinth control operations. A total of 45 water
samples were collected from nine locations on the St. Johns River from July
1975 to January 1976. Concentrations of 2,4-D ranging from "non-detectable"
to 1.3 ug/1 were reported, with an average concentration of 0.17 ug/1 (no
detection limit was reported).
-------
Barks (1978) presented the results of a USGS water quality study conducted
from April 1973 to July 1974 in the Ozark National Scenic Riverways, Missouri.
During the study, 20 surface water samples were collected from three sites on
the Current River and one site on Jacks Fork. Analysis of unfiltered samples
showed a concentration of 2,4-D of 0.03 ug/1 in a sample collected on the
Current River below Montauk State Park during a storin. No detection limit was
reported.
Englande et al. (1973) presented the results of extensive chemical analy-
sis of six Advance Wastewater Treatment (AWT) plant effluents. Four of the
sampled plants were located in California, and one each in the District of
Columbia and Texas. Although none of the 63 AWT effluent samples contained
concentrations of 2,4-D in excess of the 2,4-D MCL of 100 ug/1, mean concen-
trations of less than 0.023 ug/1 and 0.095 ug/1 were detected in two effluent
samples from two California plants. A mean concentration of less than 0.032
ug/1 was detected in samples from the Texas plant. 2,4-D was not detected in
the other systems sampled. The number of positive samples and the detection
limit were not reported.
Wall et al. (1978) conducted a study in November 1973 to determine the
levels of selected herbicides in filtered stream water samples from the Maumee
River Basin in northwestern Ohio. Water samples were collected from three
rivers in the Maumee River Basin. None of the samples contained concentrations
of 2,4-D in excess of the detection limit of 1 ug/1. The number of samples was
not reported.
Hiatt (1976, as cited in NAS 1977) detected concentrations of 2,4-D as
high as 70 ug/1 in samples collected from streams in Oregon after aerial
application of the pesticide to forestland. The number of samples analyzed
and the detection limit were not reported.
6.2.2 Ambient Air
Stanley et al. (1971, as cited in Grover et al. 1976) and Compton et al.
(1972, ibid.) reported the results of studies that assessed 2,4-D ambient air
levels in 16 U.S. cities over a 1-year period. Two of the three positive
-------
samples reported were from cities in New York (0.00115 and 0.00154 ug/m3) and
the third positive sample was from a city in Utah {0.004 ug/rn^).
6.2.3 Soil/Sediments
Several studies were identified that examined the occurrence of 2,4-D in
soils and sediments. Britton et al. (1983) reported on levels of pesticides
in water-sediment mixtures (unfiltered samples) and in bottom material samples
collected by the National Stream Quality Accounting Network (NASQAN) in 1976.
Throughout the United States, 151 permanent stations plus stations added as
part of local programs were sampled for pesticides, including 2,4-D. Water-
sediment mixtures were collected quarterly? bottom materials were collected
semiannually. 2,4-D was detected in water-sediment samples at one of 14
stations in the Ohio region, four of 12 stations in the Lower Mississippi
region, three of 10 stations in the Arkansas-White-Red region, five of 13
stations in the Texas-Gulf region, two of five stations in the Lower Colorado
region, and two of six stations in the California region, the maximum level
°f 2,4-D found at these stations was 1.6 ug/1. No samples of bottom materials
contained detectable levels of 2,4-D. The detection limit for 2,4-D in the
water-sediment and the bottom material samples was not reported.
The National Surface Water Monitoring Program (NWMP 1982) presented data
on levels of 2,4-D in sediment samples collected between 1975 and 1979. Of
542 sediment samples analyzed during that period, only one sample contained a
detectable concentration of 2,4-D, at 14.88 ug/kg (collected during the spring
of 1978). No detection limit for 2,4-D was reported.
Wall et al. (1978) conducted a study in November 1973 to determine the
levels of selected herbicides in bottom sediment material samples from the
Maumee River Basin in northwestern Ohio. A total of 21 bottom sediment
samples were collected at 6 sites on 3 rivers in the Maumee River Basin.
Concentrations of 2,4-D ranging from less than 5 to 90 ug/kg were detected in
unconsolidated bottom sediment samples. The detection limit and the number of
positive samples were not reported.
-------
6.2.4 Food
The data on levels of 2,4-D in foods were obtained primarily from studies
conducted throughout the 1970*s. Sources of information on 2,4-D in foods
include FDA market basket surveys and other FDA compliance program reports.
Data are limited on the dietary intake of 2,4-D in the United States.
Dietary exposure to 2,4-D appears to be low; there have been no findings of
2,4-D in FDA adult market basket surveys since 1973. The average total daily
intake of 2,4-D, based on detectable levels of 2,4-D in market basket studies
performed between 1965 and 197 3, were calculated to range from 0.0006 to
0.07 ug/kg/day (FDA 1981b).
For the period June 1970 to April 1971, residues of 2,4-D ranging from 10
to 130 ug/kg were detected in three leafy vegetable samples. In addition, one
residue of 10 ug/kg was identified in leafy vegetable samples taken during the
period June 1971 to July 1972. A residue of 2,4-D of 14 ug/kg was detected in
a potato sample collected during the period August 1972 to July 1973 (Johnson
and Manske 1976; Manske and Corneliussen 1974; Manske and Johnson 1975).
Johnson et al. (1981a) presented data on residues of 2,4-D in infant and
toddler foods for the period August 1975 to July 1976. During the infant/
toddler studies, 2,4-D was detected in one of the toddler diet samples. None
of the infant food samples contained concentrations of 2,4-D in excess of the
approximate quantitation limit of 20 ug/kg.'
The FDA has also monitored 2,4-D (primarily in small grains) as part of
its surveillance and compliance program (FDA 1981b). Sampling results from
1978 to 1982 indicated that only 20 percent of the samples analyzed showed
levels of 2,4-D* Concentrations ranged from trace quantities to 2.1 ppm.
It is expected that dietary levels of 2,4-D vary somewhat with geographical
location, with higher levels occurring in foods from areas of the sources of
2,4-D exposure. However, because of insufficient data, no estimates could be
made of variations in intake by geographical region.
-------
6.3 EXPOSURE SUMMARY
Several comprehensive studies exist that provide data on the extent of
occurrence of 2,4-D in drinking water and food. Data on 2,4-D in air are more
limited. According to monitoring data collected as part of various Federal
surveys conducted from 1975 to 1983, no public water supplies contained 2,4-0
at levels exceeding the current Maximum Contaminant Level (MCL) of 100 ug/1.
indeed, of the three Federal surveys reviewed, 2,4-D was detected in only two
out of over 200 groundwater and surface water supplies (less than 1 percent)
at concentrations of 0.04 ug/1 and 1.1 ug/1. Similarly, levels of 2,4-D
reported in regional surveys were also low, with concentrations ranging from
0.02 to 0.94 ug/1. These levels are low considering 2,4-D's relatively
extensive use in weed control, as well as use on food crops.
Data are limited on the dietary intake of 2,4-D in the United States.
Information on 2,4-D in foods is available through FDA market basket surveys
and FDA compliance program reports, both of which comprehensively analyze the
various classes of food products. The data on levels of 2,4-D in foods were
obtained primarily from studies conducted throughout the 1970s. The average
total daily intake of 2,4-D, based on detectable levels of 2,4-D in market
basket studies performed between 1965 and 1973, ranged from 0.04 to 5.0 ug/
day. There have been no findings of 2,4-D in FDA adult market basket surveys
since 1973.
Currently available information on the occurrence of 2,4-D is insufficient
to determine the national distribution of intake by any of the three routes.
The table below indicates that food exposure can result in higher intakes than
air or water and that the total non-drinking water intake will be on the order
of a few ug/day. The number of people who actually receive such exposures is
unknown. EPA has little information on present levels of 2,4-D in ambient and
indoor air and in food; therefore, the intakes by these routes cannot be
determined with any certainty.
-------
Table 1. Exposure Estimates for 2,4-D
Reported Exposure Estimated
Source Levels (low-high) Adult Intake
Drinking Water
Diet
Air
0-1.1 ug/1
0 - 0.004 ug/3i3
0 - 2.2 ug/day
0#04 - 5 ug/day
0 - 0.8 ug/day
-------
7. GOB
7.0 SUMMARY
EDB is a highly volatile liquid that until recently has been used as a
soil fumigant and insecticide. EDB is a contaminant of groundwater in areas
where it is used agriculturally. It has been found in private drinking water
supplies at levels up to several hundred ug/1. Due to the recent cancellation
of EDB uses, occurrences of EDB in air, food, and water are expected to
decline in the future.
7.1 GENERAL CHARACTERISTICS
7.1.1 Physical/Chemical Properties
1,2-Dibromoethane (ethylene dibromide-EDB) is a colorless liquid at 25°C
(Verschueren 1983). Trade names for products containing EDB include Bromofurae,
Celmide, Dowfume, EDB-85, E-D-Bee, KopFume, Nephis, and EDB (Berg 1986). It
has a molecular weight of 187.9, a molecular formula of C2H2Br2, a melting
point of 9.97®C, and a boiling point of 131.6'C. The aqueous solubility and
vapor pressure of EDB at 30°C are 4,310 mg/1 (2.3 x 10""3 mol/1) and 17 torr
(2.2 x 10~2 atm»m3/raol), respectively (Verschueren 1983).
The ratio of the vapor pressure to the aqueous solubility gives an esti-
mated Henry's constant for EDB at 30°C of 9.6 x 10~3 atm^mVmol (3.9 x 10~3
torr*l/rag). The ratio of the estimated Henry's constant to the product of the
gas constant times the temperature in degrees Kelvin gives an estimated
dimensionless Henry's constant at 30°C for EDB of 0.39. Reported Koc values
for EDB range from 36 to 158 (Cohen et al. 1984).
EDB is stable when stored at room temperature, but hydrolyzes in the
presence of heat and moisture to hydrobromic acid and ethylene glycol. The
rate of this degradation is variable, depending on pH, temperature, and the
microbiological populations present (FDA 1981b).
-------
7.1.2 Use
Annual domestic production of EDB was reported by EPA to exceed 280 million
pounds. More than 90 percent of the total quantity used, or 260 million pounds
of EDB, was estimated to be used as an additive in leaded gasoline, while more
than 20 million pounds was used for agricultural purposes (USEPA 1984a). EPA
estimated in 1983 that 23 million pounds of EDB were applied to 1.0 million
acres of farmland annually as a soil fumigant alone (USEPA 1983c). Soil fumigant
use accounted for over 90 percent of total domestic agricultural use of EDB in
1978, and use as a soil fumigant had increased significantly between 1978 and
1983. In 1984-1985, use of EDB decreased to 500,000 pounds due to cancellation
of most uses (Kuch 1986). The use of EDB as an additive in gasoline is declining
as a result of EPA's regulations phasing out leaded gasoline (USEPA 1983c).
EDB was used in the production of soybeans, cotton, peanuts, pineapples,
and 30 other fruit and vegetable crpps for protection from attack by nematodes
and tropical fruit flies (USEPA 1984a). EDB is used in the fumigation of
imported fruit and grain, and spot fumigation of grain milling machinery
(USEPA 1983c).
Other minor uses of EDB included application to felled logs, termite
control, beehive supers and honeycombs, storage vault fumigation, and the
fumigation of bench soil, balled plants, and grass sod to control the Japanese
beetle under the Japanese Beetle Domestic Quarantine Program (USEPA 1984a).
Major geographic areas where EDB is used include the Southeast and California.
On September 28, 1983, EPA issued a notice of intent to cancel the major
uses of EDB. This action was initiated on the basis of the potential onco-
genic, mutagenic, and reproductive effects of EDB. Uses of EDB for soil
fumigation, fumigation of stored grain, spot fumigation of grain milling
machinery, and fumigation of felled logs were scheduled for cancellation
within 30 days of the notice. Cancellation of EDB use for quarantine fumi-
gation was scheduled for September 1, 1984 (48 FR 46234). In addition, an
emergency suspension order for the use of EDB for soil fumigation was made
effective on September 28, 1983 (48 FR 46228). EPA continued registrations of
EDB products used for termite control, fumigation of beehive supers, honey
-------
combs, vault fumigation, and Japanese beetle control, with the stipulation
that specified changes were made on labels to reduce the level of risk
resulting from use (48 FR 46234).
As a result of the cancellation of the majority of uses of EDB, it is
expected that the releases of EDB to air and groundwater from agricultural
uses will be greatly reduced. While release of EDB to the atmosphere from the
use of leaded gasoline can contribute significant airborne concentrations of
EDB, these levels are also expected to decrease slowly over the course of the
next 10 years as the use of leaded gasoline declines.
7.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of EDB is divided into the fol-
lowing subsections: 7.1.3.1 Volatilization; 7.1.3.2 Sorption and Leaching
Potential; 7.1.3.3 Abiotic Transformations; 7.1.3.4 Biodegradation and
Persistence in Soil and Water; and 7.1.3.5 Summary. The discussion will
emphasize the environmental fate of EDB in soil and water.
7.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as EDB (H = 9.6 x 10-3
atra.m3/mol) with Henry'3 constants greater than 10-5 atm'm^/mol are likely to
undergo significant rates of volatilization from surface waters under any
conditions normally found in the environment. Volatilization half-lives esti-
mated for EDB in rivers or streams and lakes by SAIC using equations A-1 and
A-3 through A-8 support this conclusion. Estimated volatilization half-lives
for EDB in rivers or streams range from 0.79 hour in a turbulent river or
stream 1 m deep to 6.7 days for a stagnant river or stream 10 m deep (see
Appendix A). Estimated volatilization half-lives for EDB in turbulent lakes
range from 7.4 hours in a lake 1 m deep to 74 hours in a lake 10 m deep.
The rate of volatilization of EDB from ground water to the soil column
above is projected to be substantially less than volatilization rates from
surface waters to the atmosphere. This is due primarily to the laminar,
nonturbulent nature of ground water flow. Transport from ground water by
-------
volatilization may be further reduced by a build-up of chemical in the pore
air at the ground water/pore air interface, and an associated decrease in the
concentration gradient across the interface.
Hie use of equation A-9 would be inappropriate for estimating volatili-
zation half-lives for EDB on soils because the equation was derived from data
on compounds with vapor pressures several orders of magnitude .lower than that
of EDB. However, due to a relatively high Henry's constant (9.6 x 10-3
atm'm^/mol) and a relatively low median KQC value of 97, volatilization is
probably a significant removal process for EDB on soil.
7.1.3.2 Sorption and Leaching Potential
Substituting the median (97) of the range of reported KqC values from 36
to 158 into equation A-11 gives estimated sediment or soil/water equilibrium
partition coefficients (Ks/W) for EDB ranging from 0.97 to 7.8 for sediments
or soils with organic fractions ranging from 0.01 to 0.08. The estimated Kg/W
values suggest that at equilibrium, the concentration of EDB in suspended and
exposed bottom sediment may be several times greater than the concentration in
the water column. However, since the estimated Kg/W values for EDB are less
than 10, it is unlikely that the ratio of the total mass of EDB adsorbed to
suspended and exposed bottom sediments to the total mass of EDB dissolved in
the water column will exceed 0.01 in most surface waters. The reason is that
in most surface waters, most of the time, the ratio of the water mass to the
mass of suspended and exposed bottom sediment exceeds 1,000 (USGS 1983).
Therefore, transport by adsorption to suspended sediment and removal by
adsorption to bottom sediments are probably not important processes for EDB
in surface waters.
Substitution of the median (97) of reported KqC values into equation A-13
gives an estimated soil TLC Rf value (Appendix A) of 0.47 for EDB adsorbed to
a soil with an organic carbon fraction of 0.014, a pore fraction of 0.5, and
a soil density of 2.5 g/cm^. Therefore, based on the five mobility classes
defined by Helling and Turner (1968) and cited by Hamaker (1975) for a soil
with the same properties (Appendix A), EDB would be expected to have an inter-
mediate mobility (Class 3) in, and be in between being moderately susceptible
-------
and moderately resistant to leaching from, surface soil. EDB has been detected
in several samples of ground water from several states (Cohen et al. 1984). In
one study, EDB was reported to have leached to a depth of greater than 10 meters
under the soil surface (Zalkin et al. 1983, as cited in Cohen et al. 1984).
7.1.3.3 Abiotic Transformations
Johns (1976, as cited in USEPA 1984f) reports a hydrolysis half-life of
5 to 10 days for EDB in neutral aquatic media at ambient temperature. However,
preliminary results of another study have led to an estimated hydrolytic
half-life for EDB in groundwater of >6 years (Cohen et al. 1984). The dis-
crepancy in hydrolytic half-lives may be due to biologically mediated or
enzyme catalyzed hydrolysis reaction in surface water.
There was no information in the literature reviewed by USEPA (1984f) or
Cohen et al. (1984) concerning any possible photolytic or oxidative degradation
of EDB. However, EDB does not have any chromophores that absorb strongly above
290 nm nor any functional groups susceptible to oxidation. Therefore, EDB is
unlikely to undergo significant rates of photolysis or oxidation in the
environment.
7.1.3.4 Biodegradation and Persistence in Soil and Water
There is no information in the literature reviewed by USEPA (1984f) or
Cohen et al. (1984) concerning the biodegradation of EDB in natural waters and
only limited information concerning biodegradation in soils. Castro and Belser
(1968, as cited in Cohen et al. 1984) reported that approximate half-lives for
EDB in a "bioactive" agricultural soil ranged from 2 to 18 weeks depending only
on the soil aliquot used. McKenry and Thomason (1974, as cited in USEPA 1984f)
performed a mass balance analysis for EDB in a soil/water system and reported
unaccounted for losses of 15 percent after 3 days and 40 percent after 10 days.
They attributed the losses to a possible combination of biodegradation,
volatilization, and leaching, but did not attempt to determine the relative
contributions of each.
-------
7.1.3.5 Summary
Based on the above discussion and the literature reviewed by Cohen et al.
(1984) and CJSEPA (1984f), the following tentative conclusions can be made con-
cerning the behavior of EDB in soil and water:
o Based primarily on theoretical considerations, volatilization is
expected to be the primary removal process for EDB in surface waters
and on soils.
o Based on theoretical considerations and on limited data, transport by
adsorption to sediment and removal by adsorption to bottom sediments
are not expected to be important processes for EOB in aquatic systems.
o Based on theoretical considerations and on some leaching and
ground water monitoring data, EDB appears to be at least moderately
susceptible to leaching from soils.
o There is a large discrepancy between the reported hydrolytic half-life
of 5-10 days in a neutral aquatic medium, and estimated hydrolytic
half-lives of >6 years in groundwater. The discrepancy may be due to
biological mediation (enzyme catalysis) in the first case.
o There is no information available concerning any possible photolytic
or oxidative degradation of EOB. However, due to its chemical
structure, EDB is unlikely to undergo significant rates of photolysis
or oxidation in the environment.
o Estimated degradation half-lives for EDB in the same soil varied from
2 to 18 weeks depending on the soil aliquot, which corresponds to a
persistence in soil classification ranging from nonpersistent (half-
life <30 days) to persistent (half-life >100 days).
o Ground water samples from several states have been reported to be
contaminated with EDB* However, due to its estimated high rate of
volatilization from surface waters, EDB is not expected to signif-
icantly contaminate surface water supplies, nor has it been identified
in surface water supplies.
7.2 OCCURRENCE IN THE ENVIRONMENT
7.2.1 Water
The following section presents the data available from monitoring studies
and surveys to determine the extent of occurrence of EDB in public drinking
water supplies and water other than drinking water.
-------
7.2.1.1 Occurrence in Drinking Water
No studies were identified that addressed EDB levels in drinking water
from surface water sources; however, several regional studies have addressed
levels of EDB in drinking water from ground water sources. Ground water samples
were collected between 1980 and 1983 by a number of California State agencies
with the assistance of EPA. Concentrations of EDB were detected in samples of
water from two small community drinking water supplies in the San Joaquin
Valley area. The number of drinking water supplies sampled, the number of
samples collected, values for the positive systems, and the detection limit
were not reported (USEPA 1983c).
As a result of positive findings in California, EPA requested that the
State of Hawaii perform EDB monitoring. Samples of drinking water from more
than 100 wells located near areas of EDB use in Hawaii were collected between
June and September 1983 (Wong et al. in progress, as cited in USEPA 1983c).
Samples of water from four wells at one site contained levels of EDB ranging
from 0.02 to 0.10 ug/1. NO detection limit for the study was reported.
SAIC/JRB Associates (1981a) identified locations in the United States
where organic chemicals were found in groundwater supplies. Levels of EDB
ranging from 0.02 to 560 ug/1 were detected in 25 samples of water collected
from wells in several counties in New Jersey. Hie total number of samples and
the detection limit were not provided. It is not possible to determine the
source of the high values from the information presented.
The State of Connecticut began a well water testing program (Chemical
Regulation Reporter 1984) in early February 1984, after EDB residues were found
in grain and citrus products throughout the United States, as well as in Florida
drinking water. Wells located near present or former tobacco fields where EDB
had been used until 1982 were tested. A total of 240 public and private wells
were sampled. Hie analysis showed that 32 wells in north central Connecticut
contained levels of EDB ranging from 0.1 to 1.4 ug/1. ttie detection limit was
not reported.
-------
7.2.1.2 Occurrence in Water Other Than Drinking Water
Several studies have reported levels of EDB in ground water and surface
water other than drinking water. Levels in ground water and surface water are
discussed separately below.
Ground Water Sources
The primary sources of information on levels of EDB in groundwater are
state surveys conducted to determine whether there is any correlation between
EDB use and ground water contamination by EDB. Five states (South Carolina,
Florida, Georgia, California, and Washington) have detected EDB in ground water
in the range of 0.05 to 20 ug/1 (Cohen et al. 1984). In water samples collected
from 19 wells in South Carolina, samples from three wells contained levels of
EDB ranging from 0.036 to 0.24 ug/1 (Senn 1983, as cited in Cohen et al. 1984).
In Florida, water samples have been collected from 334 wells since
mid-1983 by the State. Analysis showed that samples of water from 86 of the
33 4 wells contained EDB concentrations in excess of the detection limit of
0.02 ug/1; 75 of the 86 positive samples contained concentrations greater than
0.1 ug/1 (Bigler 1983, as cited in USEPA 1983c).
As a result of notification to EPA by EDB registrants that levels of EDB
were as high as 100 ug/1 in three wells sampled in Georgia, further monitoring
of the wells in the area was initiated by EPA and the U.S. Geological Survey.
Analysis showed that samples of water from 6 out of 19 wells were contaminated
with EDB in the range of 0.03 to 11.8 ug/1 (Jovanovich and Cohen in review as
cited in USEPA 1983c).
A California study conducted in 1980 found that samples of ground water
from 2 out of more than 200 wells sampled were contaminated by EDB. The two
positive samples contained 0.1 and 0.2 ug/1 EDB (Zalkin et al. in progress, as
cited in USEPA 1983c). Additional ground water monitoring detected EDB in 15
to 20 samples of more than 250 water samples analyzed. The typical values of
the positive samples ranged from 0.1 to 2.0 ug/1 EDB (USEPA 1983c). With the
exception of the Florida survey, no detection limit for EDB was reported for
these state surveys.
-------
The Washington Department of Agriculture, in cooperation with the Department
of Social and Health Services (DSHS) and local health departments, conducted a
study of ground water in five western Washington counties for EDB contamination
(Washington State Department of Social and Health Services 1985). Shallow and
deep wells of 95 public and private drinking water supplies were sampled from
June to October 1984. Private wells were chosen if sampling would provide
information to protect the public. Sample analysis was performed by the U.S.
Environmental Protection Agency, Region X Laboratory. In four counties, a
total of 13 supplies (10 public, 3 private) were contaminated with EDB above
the health advisory level of 0.02 ug/1, with a population served of approxi-
mately 550 persons. Out of a total of 131 samples analyzed (including an
unspecified number of blanks), 34 were positive for EDB with a mean concentration
of 0.76 ug/1 (range = 0.018 to 5.7 ug/1). Hie detection limit was 0.01 ug/1.
Surface Water Sources
One sample of surface water was collected from an agricultural area in
Georgia that had been treated with EDB. The sample did not contain EDB in
excess of the detection limit, although the detection limit was not reported
(Jovanovich and Cohen in review, as cited in USEPA 1983c).
Five samples of rainfall and runoff water collected near an EDB fumigation
center in Florida contained 0.9-2.0 ug/1 EDB (Going and Spigarelli 1976, as
cited in Brodzinsky and Singh 1982).
7.2.2 Ambient Air
Brodzinsky and Singh (1982) compiled data on levels of EDB in ambient air
from published and unpublished sources into a master file. They analyzed and
assessed data quality, reliability, and representativeness. Quality codes
were assigned based on a quantitative assessment or on the authors' best
professional judgment. Concentrations ranged from "undetectable" in clean
atmospheres to 240 ug/m3. In urban areas, typical EDB concentrations ranged
from 0.08 to .460 ug/m3. Using data with quality codes of excellent, good, or
acceptable, the authors calculated the following median EDB concentrations:
rural/urban areas, 0.0 ug/m3; urban/suburban areas, 0.2 ug/m3; and source-
dominated areas (undefined), 1.5 ug/m3.
-------
Due to the decline of the use of EDB in gasoline and EPA's cancellation of
the use of EDB as a soil, grain, and quarantine (fruit and vegetable) fumigant,
current ambient air levels of EDB are expected to be lower than the levels in
the above study.
7.2.3 Soil/Sediments
Very little information was available on studies that examined EDB occur-
rence in soil and sediments. Only one soil study was identified, which involved
soil sample collection near an agricultural fumigation center in the State of
Florida. Analysis results indicated low levels of EDB (0.4 to 3.4 ug/kg) in
the soil samples. No detection limit was reported (Going and Spigarelli 1976,
as cited in Brodzinsky and Singh 1982).
7.2.4 Food
The dietary intake of EDB in the United States was estimated by EPA in
Position Documents 2/3 and 4 and in two technical support documents (USEPA
1983c, 1984c,f). Dietary intakes were calculated for the following uses of
EDB: soil fumigation, post-harvest fumigation of grain and grain milling
machinery, and quarantine fumigation of citrus and tropical fruits. Total
intake estimates of EDB resulting from these uses for the U.S. population
generally ranged between <0.01 and 0.18 ug/kg/day.
Dietary levels of EDB may vary with geographic location (see Table 7-1
below), ihe highest dietary intakes were estimated to occur in the South,
where the combined intake of EDB from citrus and grain totals 0.0062 ug/kg/day
(USEPA 1984b,c).
Most of the information identified on the occurrence of EDB in foods in
the United States was from studies that monitored EDB residues in grains
(USEPA 1983c). USEPA summarized monitoring data for EDB residues in raw
grains, uncooked and cooked grain products, and in baby foods in a scientific
support and decision document for grain and grain milling fumigation uses
(USEPA 1984b).
-------
Table 7-1. Dietary Levels of EDB by Geographic Region
Dietary intake (uq/kq/day)
Region
Citrusa
Grainb
Northeast
North Central
South
West
0.00052
0.0005
0.0003
0.00074
0.0051
0.0052
0.0059
0.0052
a USEPA 1984c
b USEPA 1984b
In raw grains, the highest average residue reported was 2,000 ug/kg in
other grains, which includes rice, barley, oats, rye, and sorghum (FDA 1983, as
cited in USEPA 1984b). Levels in raw wheat ranged from 4.9 ug/kg (USDA 1984a,
as cited in USEPA 1984b) to 43 ug/kg (Grocery Manufacturers of America 1984a,b,
as cited in USEPA 1984b). EDB levels in raw corn ranged from not detected
(North Carolina Department of Agriculture 1984, as cited in USEPA 1984b) to
220 ug/kg (Grocery Manufacturers of America 1984a,b, as cited in USEPA 1984b).
The detection limits and number of positive samples were not presented for any
of these studies.
For uncooked grain products, the highest level of EDB detected was
140 ug/kg in corn (Grocery Manufacturers of America 1984a,b, as cited in USEPA
1984b). Levels of EDB in wheat ranged from 2.9 ug/kg (Florida Department of
Agriculture 1984, as cited in USEPA 1984b) to 22 ug/kg (MRI 1984, as cited in
USEPA 1984). EDB concentrations in other grains ranged from 0.75 ug/kg
(Georgia Departinent of Agriculture 1984, as cited in USEPA 1984b) to 2.3 ug/kg
(North Carolina Department of Agriculture 1984, as cited in USEPA 1984b).
Cooked grain products had the overall lowest EDB concentrations. Hie
highest concentration was 7.7 ug/kg in wheat (North Carolina Department of
Agriculture 1984, as cited in USEPA 1984b). The highest reported level in
cooked corn was 4.7 ugA? (MRI 1984, as cited in USEPA 1984b).
-------
Of 86 baby foods sampled, only two were positive for EDB. These were in
wheat and rice with a range of <1-1.3 and <1-1.5 ug/kg, respectively (Grocery
Manufacturers of America 1984b, as cited in USEPA 1984b).
Wallace et al. (1984) reported results from the Total Exposure Assessment
Methodology (TEAM) survey analysis for EDB in composite samples of selected
FDA food classes. The food classes chosen for the studies (dairy, meat, fatty
food, and beverages) were those expected to contain detectable levels of
volatile organic chemicals. EDB was not quantified in any of the composite
samples analyzed.
Monitoring data for EDB in citrus was submitted to EPA from numerous
sources (USEPA 1984f). In all, EDB residues in 450 samples of whole citrus
fruit and 130 samples of citrus pulp have been determined. EDB residues in
whole fruit ranged from "not detected" (<1 ug/kg) to 16,000 ug/kg, with a
median of 250 ug/kg and an average of 1,633 ug/kg. Residues of EDB in citrus
pulp ranged from <1 ug/kg to 6,250 ug/kg, with a median of 265 ug/kg and an
average of 373 ug/kg•
These estimates reflect uses of EDB (soil and fruit fumigation) which have
been cancelled or greatly reduced. Because of these restrictions, current
intakes are expected to be less than the above estimates.
7.3 EXPOSURE SUMMARY
The data currently available on the occurrence of 1,2-dibromoethane or
ethylene dibromide (EDB) in drinking water, air, and food are limited, but
suggest low levels of contamination from each route of exposure. EDB, a highly
volatile liquid that was used until recently as an agricultural fumigant, has
been detected as a contaminant of ground water supplies in agricultural areas.
Its use as an additive to leaded gasoline has also produced contamination of
ambient air, especially in urban districts. Due to the 1983 cancellation of
agricultural uses and to a decline in the use of leaded gasoline, EDB occurrence
in each medium is expected to decline.
-------
The detection of EDB at levels ranging from 0-20 ug/1 in drinking water
studies conducted in various regions of the United States indicates the potential
for exposure in agricultural areas where EDB use was more extensive. Because
EDB does not tend to bind to soils/ it is readily available for leaching into
ground water supplies. It is susceptible to volatilization from surface soils
and thus to contamination of ambient air in agricultural areas. EDB's use in
gasoline also provides a source for contamination of ground water supplies in
urban areas through the leaking of gasoline storage tanks and the leaching
of the confounds. The volatilization of EDB from gasoline uses provides a
significant release of this pollutant to air in urban areas where a greater
number of automobiles and gasoline engines can be found. Ambient air concen-
trations range from "undetectable" to 240 \ig/a?, with typical levels in urban/
suburban areas of 0.20 ug/m3. However/ EDB releases from gasoline emissions
are expected to be decreasing as a result of decreasing use of EDB in gasoline.
The estimated daily intakes from food were derived from dietary levels
presented in EPA's Position Document 2/3, Position Document 4/ and two technical
support documents (USEPA 1983c, 1984b/C). These intake levels were calculated
from results of studies conducted by EPA to evaluate the use of EDB as a soil
fumigant, a post-harvest fumigant of grain, a fumigant of grain milling machinery,
and a quarantine fumigant of citrus and tropical fruits. As indicated before,
current levels are expected to be lower than this estimate. Results are sum-
marized in Table 7-2.
Table 7-2. EXPOSURE ESTIMATES FOR EDB
Source
Reported Exposure
Levels (low to high)
Estimated
Adult Intake
Drinking Water
0-20 ug/1
0-40 ug/day
Diet
<0.05 ug/day
Air
0 - 0.20 ug/1
0-4 ug/day
The information currently available on the occurrence of EDB in the environ-
ment and the potential for exposure is insufficient to determine the distribution
-------
of exposure on a national scale from any of the three sources. The estimated
daily intake levels provided in the above table suggest that similar ranges of
intake occur from ambient air and drinking water, and that the potential for
intake from consumption of foods is very small.
However, the number of people who are actually exposed to such high exposure
levels from food or air sources is unknown. Because the ranges of possible
exposure and intake levels from all three sources are so wide and the available
data do not reflect recent declines in uses, an attempt to provide a combined
or cumulative daily intake estimate should not be made at this time.
-------
8. HEPTACHLOR/HEPTACHLOR EPOXIDE
8.0 SUMMARY
Eieptachlor is an insecticide that was extensively used prior to 1974. EPA
has limited the use of heptachlor to subterranean termite control. The current
use of heptachlor continues to provide a low-intensity threat of contamination.
8.1 GENERAL CHARACTERISTICS
8.1.1 Physical/Chemical Properties
Heptachlor (1,4,5,6,7,8,8-heptachloro-3a,4,7,7a-tetrahydro-4,7-methanoindene)
is a cyclodiene insecticide (NAS 1977). Synonyms and identifiers for heptachlor
include Drinox H-34®, Gold Crest H-60, Heptamul®, and Heptox (Berg 1986).
Termide, a mixture of heptachlor and chlordane, is a termiticide (USEPA 1983b).
Heptachlor is a white crystalline substance at 25°C. It has a molecular
weight of 373.75, a molecular formula of C.10H5C27, and a melting point of
95-96°C (Windholz 1976). The aqueous solubility and vapor pressure of
heptachlor at 25°C are 0.18 mg/1 (4.8 x 10~7 mol/1) and 3.0 x 10"4 torr (3.9 x
10~7 atm), respectively (Mabey et al. 1981).
The ratio of the vapor pressure to the aqueous solubility gives an
estimated Henry's constant for heptachlor at 25°C of 8.1 x 10~4 atm*m3/mol.
The ratio of the estimated Henry's constant to the product of the gas consternt
times the temperature in degrees Kelvin gives an estimated dimensionless
Henry's constant for heptachlor at 25°C of 0.033. Heptachlor has an estimated
KQC value of 1.2 x 104 (Mabey et al. 1981).
Heptachlor expoxide is a biodegradation product of heptachlor that is
frequently detected in areas of heptachlor use. It has a molecular weight of
389.2, a molecular formula of C10H5C170, and a melting point of 157-160°C
(Mabey et al. 1981; Callahan et al. 1979). At 25°C, heptachlor epoxide has an
aqueous solubility of 0.36 mg/1 (9.0 x 10~7 mol/1) and a vapor pressure of
3.0 x 10~4 torr (3.9 x 10~7 atm) (Mabey et al. 1981).
-------
The ratio of the vapor pressure to the aqueous solubility gives an
estimated Henry's constant for heptachlor epoxide at 25°C of 4.3 x 10-4
atm*m3/mol. The ratio of the estimated Henry's constant to the product of the
gas constant times the temperature in degrees Kelvin gives an estimated dimen-
sionless Henry's constant for heptachlor epoxide at 25°C of 0.018. Heptachlor
epoxide has an estimated KQC value of 2.2 x 102 (Mabey et al. 1981).
8.1.2 Use
Prior to 1974, heptachlor was widely used as a pre-emergent insecticide
for the control of corn rootworms, wireworms, and cutworms; as a seed treat-
ment; as a soil treatment for the control of termites and ants; and on
gardens, lawns, turf, and ornamentals (USEPA 1976). In 1974, EPA issued a
notice of intent limiting registered uses of heptachlor to subsurface control
of termites and dipping of roots and tops of nonfood plants. In 1975, the EPA
Administrator issued a notice of intent to suspend the registrations of
certain pesticide products containing heptachlor (USEPA 1983b).
On March 6, 1978, EPA issued a final cancellation order putting into
effect the terms of settlement for the cancellation proceedings (USEPA 1983b).
Under this cancellation order, an agreement was outlined to phase out the
¦nontermiticide uses of heptachlor over a 5-year period. All uses permitted
during the phase-out period were restricted to treatment by certified appli-
cators or professional commercial seed treatment companies.
Currently, heptachlor is used to control subterranean termites. Between
0.75 and 1 million pounds of heptachlor were used in 1986 (Kuch 1986). Hep-
tachlor has limited use as a single active ingredient and usually is applied
in combination with chlordane (USEPA 1983b). A second approved use of hep-
tachlor is to control fire ants in sugarcane and pineapple fields in Hawaii.
Approximately 30 percent (or 550,000 pounds) of the heptachlor used in
1971 was applied to commercial and residential structures for protection
against termites and to nurseries, lawns, and gardens. By 1974, domestic use
of heptachlor for termite control had increased to 1.4 million pounds (USEPA
1976). In 1983, between 1 and 2 million pounds of heptachlor were used in the
United States for termite control (USEPA 1983b).
-------
Heptachlor is present as an impurity in chlordane; typically, heptachlor
constitutes 10 percent of technical grade chlordane (USEPA 1980f). Nearly 10
million pounds of chlordane were applied in the United States in 1980;
approximately 1 million pounds of heptachlor would have been present as an
impurity in chlordane in that year. EPA (USEPA 1983b) reported that the
largest quantity of chlordane was initially distributed to EPA Region IV,
which includes Alabama, Georgia, Florida, Mississippi, North Carolina, South
Carolina, Tennessee, and Kentucky. Because chlordane is applied by subsurface
ground injection, the potential for water contamination in some areas may be
high.
8.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of heptachlor and heptachlor
epoxide is divided into the following subsections: 8.1.3.1 Volatilization;
8.1.3.2 Sorption and Leaching Potential; 8.1.3.3 Abiotic Transformations;
8.1.3.4 Biodegradation and Persistence in Soil and Water; and 8.1.3.5 Summary.
The discussion will emphasize the environmental fate in soil and water.
8.1.3.1 Volatilization
Lyman et al. (1982) indicate that compounds such as heptachlor (H = 8.1 x
10-4 atm«m3/mol) and heptachlor epoxide (H = 4.3 x 10~4 atm«m3/mol) are likely
to undergo significant rates of volatilization from surface waters under any
conditions normally found in the environment. Volatilization half-lives for
heptachlor and heptachlor epoxide in rivers or streams and in turbulent lakes
were estimated by SAIC using equations A-1 and A-3 through A-8 (see Appendix A).
The estimated volatilization half-lives indicate that volatilization is probably
an important removal process for heptachlor epoxide in virtually all surface
waters and for heptachlor in many surface waters. However, in the case of
heptachlor, which is much more susceptible than heptachlor epoxide to
competing removal processes such as hydrolysis, biodegradation, and
adsorption, volatilization may not be an important removal process in some
deep and/or stagnant waters. Estimated volatilization half-lives for
heptachlor and heptachlor epoxide in rivers or streams range from 2.2 hours
and 3.3 hours, respectively, in a turbulent river or stream 1 m deep to 3.3
months and 3.5 months, respectively, in a stagnant river or stream 10 m deep.
-------
Estimated volatilization half-lives for heptachlor and heptachlor epoxide in
turbulent lakes range from 13 hours and 16 hours, respectively, in a lake 1 m
deep to 5.6 days and 6.8 days, respectively, in a lake 10 m deep.
The volatilization rates of heptachlor and heptachlor epoxide from ground
water to the soil column are projected to be substantially less than those
from surface waters to the atmosphere. This is due primarily to the laminar
(nonturbulent) nature of ground water flow. Volatilization rates from ground
water may be further reduced by a build-up of the chemicals in the pore air at
the pore air/ground water interface, which would decrease the concentration
gradients across the interface.
Substitution of the estimated KQC values of 1.2 x 104 for heptachlor and
2.2 x 102 for heptachlor epoxide (Mabey et al. 1981) into equation A-9 along
with the compounds' vapor pressures of 3 x 1 0~4 torr and aqueous solubilities
of 0.18 mg/1 and 0.36 mg/1, respectively, gives estimated volatilization
half-lives of 2.7 hours and 5.8 minutes for heptachlor and heptachlor epoxide,
respectively, adsorbed to the soil surface. Volatilization rates for
heptachlor and heptachlor epoxide beneath the soil surface would be much
slower, decreasing rapidly with increasing soil depth (Lyman et al. 1982).
8.1.3.2 Sorption and Leaching Potential
Substitution of the estimated KQC values of 1.2 x 1 04 for heptachlor and
2.2 x 102 for heptachlor epoxide (Mabey et al. 1981) into equation A-11 (see
Appendix A) gives estimated sediment or soil/water equilibrium partition
coefficients (Ka/W) ranging from 1.2 x 102 to 9.6 x 102 for heptachlor and
from 2.2 to 9.6 for heptachlor epoxide adsorbed to sediments or soils with
organic carbon fractions ranging from 0.01 to 0.08, respectively. The estimated
Kg^w values indicate that at equilibrium the concentration of heptachlor in
suspended and bottom sediment may be several hundred times the concentration in
the water column and that the concentration of heptachlor epoxide in suspended
and bottom sediment may be several times the concentration in the water column.
However, because the estimated Ks/W values for heptachlor epoxide are all less
than 10, it is unlikely that the ratio of the total mass of heptachlor epoxide
adsorbed to suspended and bottom sediment to the total mass of heptachlor
epoxide dissolved in the water column will exceed 0.01 in most surface waters.
-------
The reason is that the ratio of the water mass to the mass of suspended and
exposed bottom sediment exceeds 103 in most surface waters most of the time
(USGS 1983). Therefore, transport by adsorption to suspended-sediment and
removal by adsorption to bottom sediment are not expected to be important
processes for heptachlor epoxide in surface waters. However, because the
estimated Kg/W values for heptachlor exceed 100, the ratio of adsorbed to dis-
solved heptachlor will probably exceed 0.1 in some surface waters, thereby
making such processes at least occasionally important for heptachlor.
Substitution of the estimated KQC values of 1.2 x 104 for heptachlor and
2.2 x 102 for heptachlor epoxide into equation A-1 3 gives estimated soil TLC
Rf values (Appendix A) of <0.1 and 0.32 for heptachlor and heptachlor epoxide,
respectively, adsorbed to a soil with an organic carbon fraction of 0.014, a
pore fraction of 0.5, and a soil density of 2.5 g/cm^. Therefore, based on
the five mobility classes defined by Helling and Turner (1968) and cited by
Hamaker (1975) for a soil with the same properties, heptachlor would be
expected to be immobile (Class 1) in, and resistant to leaching from, surface
soil. Heptachlor epoxide would be expected to be moderately immobile (Class 2)
in, and moderately resistant to leaching from, surface soil.
3.1.3.3 Abiotic Transformations
Demayo (1972, as cited in Callahan et al. 1979) has reported a neutral
first-order hydrolysis rate constant of 3.0 x 10~2 hr~' for heptachlor in
distilled water at 30°C. Assuming that acid and base catalyzed hydrolysis
does not contribute significantly to the overall rate of hydrolysis over the
normal pH range of 6-9 in natural waters, substitution of that rate constant
into equation A-1 (Appendix A) gives an estimated hydrolysis half-life of 23
hours for heptachlor in natural waters at 30°C. Eichelberger and Lichtenberg
(1971, as cited in Callahan et al. 1979) reported a hydrolysis rate for
heptachlor in Ohio river water at 25°C that roughly corresponds to a hydrolysis
half-life of 3.5 days. They reported that no heptachlor remained 2 weeks after
introducing the compound to river water and that the only hydrolysis product
identified during that time was 1-hydroxychlordene. However, after 2 weeks,
they began to detect heptachlor epoxide, which constituted approximately 40
percent of the initial heptachlor present after 4 weeks. Their work suggests
-------
that heptachlor is hydrolyzed in natural waters to 1-hydroxychlordene, which
is then biologically transformed to heptachlor epoxide. Based on the report
by Eichelberger and Lichtenberg (1971, as cited in Callahan et al. 1979) that
heptachlor epoxide remained stable in Ohio River water for 3 weeks, heptachlor
epoxide appears to be resistant to hydrolysis.
Heptachlor and heptachlor epoxide in organic solvents and as solid films
are susceptible to direct and sensitized photolysis when exposed to light with
wavelengths above the approximate solar radiation cutoff at the earth's surface
of 290 nra. Therefore, both compounds may be susceptible to significant rates
of direct and/or sensitized photolysis in shallow surface waters or on the soil
surface. However, no information is available that would allow estimates to be
made of photolysis rates in water or soil (Callahan et al. 1979).
There are no experimental data available on the photooxidation of hepta-
chlor or heptachlor epoxide in surface waters. However, based on structural-
activity relationships, Mabey et al. (1981) estimate second-order rate constants
of 3 x 1010 hr~1 and 2,500 M~1 hr~1 for the oxidation of heptachlor by
singlet oxygen (^C^) and peroxy radicals (RO2'), respectively, in natural waters.
Die substitution into equation A-15 of those rate constants and assumed concen-
trations of 10-12m and 1 0-^M for singlet oxygen and peroxy radicals, respectively,
in sunlit natural waters, and substitution of the resulting pseudo first-order
rate constant into equation A-1, gives an estimated photo-oxidation half-life
of 23 hours for heptachlor in sunlit shallow surface waters. This indicates
that photo-oxidation may be competitive with hydrolysis in removing heptachlor
from sunlit, shallow surface waters. However, the estimated second-order rate
constants of <3,600 M"' hr"' and 20 M~1 hr~1 for the photo-oxidation of heptachlor
epoxide by 10^ aad RO^*, respectively, indicate that photo-oxidation is probably
not an important process for the removal of heptachlor epoxide.
8.1.3.4 Biodegradation and Persistence in Soil and Water
As discussed earlier, Eichelberger and Lichtenberg (1971, as cited in
Callahan et al. 1979) reported that following the rapid abiotic hydrolysis of
heptachlor to 1-hydroxychlordene, the 1-hydroxychlordene was more slowly bio-
transformed over a several-week period to primarily heptachlor epoxide. In
-------
contrast, Lu et al. (1975, as cited in Callahan et al. 1979) studying a model
aquatic microcosm concluded that after the rapid abiotic hydrolysis of hepta-
chlor to 1-hydroxychlordene, the 1-hydroxychlordene was rapidly biotransformed
to primarily 1-hydroxy-2,3-chlordene epoxide, but that the heptachlor epoxide
formed constituted no more than 5 percent of the original heptachlor added.
Miles et al. (1969, as cited in Callahan et al. 1979) reported that soil
microorganisms under aerobic conditions converted heptachlor in solution over
a several-week period to various metabolic products, including heptachlor
epoxide and 1-hydroxy-2,3-chlordene epoxide. Miles et al. (1971 as cited in
Callahan et al. 1979) reported that the major metabolic product formed from
the anaerobic degradation of heptachlor in solution by soil microorganisms was
chlordene.
Rao and Davidson (1980) reported a mean half-life for heptachlor in soils
of 63 days under aerobic conditions in the laboratory. However, under field
conditions, they reported a much higher mean half-life of 426 days for hepta-
chlor in soils. Heptachlor epoxide is reported to be much more resistant to
hydrolysis than heptachlor and to biodegradation than 1-hydroxychlordene and
is very persistent in both soil and water (Callahan et al. 1979).
8.1.3.5 Summary
Based on the above discussion and the literature review by Callahan
et al. (1979), the following tentative conclusions can be made concerning the
most likely behavior of heptachlor and heptachlor epoxide in soil or water:
o Based on theoretical considerations, volatilization is expected to be
an important removal process for heptachlor in shallow, turbulent
surface waters and on the soil surface. However, volatilization is
not expected to be competitive with the rapid hydrolysis of heptachlor
in many deeper and/or stagnant surface waters. Estimated volatiliza-
tion rates for heptachlor epoxide are comparable to those of hepta-
chlor, but due to the resistance of heptachlor epoxide to hydrolysis
and biodegradation, volatilization is expected to be an important
removal process for heptachlor epoxide in most surface waters and on
soil.
o Based on theoretical considerations, transport by adsorption to
suspended sediment and removal by adsorption to bottom sediment are
-------
expected to be important processes for heptachlor in some surface
waters, but not for heptachlor epoxide.
o Based on theoretical considerations, heptachlor is expected to be
resistant and heptachlor epoxide moderately resistant to leaching from
soils.
o Heptachlor appears to have an abiotic hydrolysis half-life of about
1 to 3 days in natural waters. The major product of hydrolysis is
1-hydroxychlordene. Heptachlor epoxide appears to be resistant to
abiotic hydrolysis and has an estimated hydrolysis half-life exceeding
several years in natural waters.
o Both heptachlor and heptachlor epoxide appear to be susceptible to
direct and sensitized photolysis in the environment, but the available
information is too limited to estimate rates.
o Photooxidation primarily by singlet oxygen is expected to be competi-
tive with hydrolysis in sunlit shallow surface waters for heptachlor
but not heptachlor epoxide.
o Following the rapid abiotic hydrolysis of heptachlor to 1-hydroxy
chlordene in soil and water, the 1-hydroxychlordene undergoes much
slower rates of biotransformation to such products as heptachlor
epoxide, 1-hydroxy-2,3-chlordene epoxide, and chlordene. Based on
limited data, 1-hydroxy-2,3-chlordene epoxide appears to be the
primary metabolic product under aerobic conditions, and chlordene
appears to be the primary metabolic product under anaerobic condi-
tions. Heptachlor epoxide is formed under aerobic conditions, but
to a far lesser degree than 1-hydroxy-2,3-chlordene epoxide.
o Heptachlor epoxide appears to be extremely resistant to both hydrol-
ysis and biodegradation and is very persistent in soil and water.
8.2 OCCURRENCE IN THE ENVIRONMENT
8.2.1 Water
The following section presents data available from monitoring studies and
surveys to determine the extent of occurrence of heptachlor and its transfor-
mation product, heptachlor epoxide, in public drinking water supplies and
water other than drinking water.
-------
8.2.1.1 Occurrence in Drinking Water
Several studies at the regional level have addressed concentrations of •
heptachlor and heptachlor epoxide in drinking water. The results of 10 regional
studies are discussed in this section. Where possible, reported levels of
heptachlor and heptachlor epoxide in drinking water from ground water sources
and surface water sources have been discussed separately.
Ground Water Sources
Twelve towns in Connecticut were sampled during 1984-1985 for heptachlor
and heptachlor epoxide by the Connecticut Agricultural Experiment Station, New
Haven, Connecticut (Waggoner 1985). These towns, combined, serve a population
of over 570,000 people. Drinking water wells were sampled at 42 locations and
no samples were positive for either heptachlor or heptachlor epoxide (detec-
tion limits were 0.2 ug/1 and 0.47 ug/1, respectively).
Shallow drinking water wells of 10 counties in northwest Mississippi were
analyzed in 1983-1984 during a Mississippi State University study on pesticide
hazard assessment (MSU 1984). No positive samples were found for either
heptachlor or heptachlor epoxide out of 143 samples analyzed (detection
limit = 0.001 ug/1).
Similarly, no positive results were obtained from 67 samples analyzed in
Long island, New York, for either heptachlor or heptachlor epoxide. Hiese
results were from a 1984 study reported by the Suffolk County Department of
Health Services (Holden 1986). No detection limit(s) was reported.
Drinking water supplies of the Floridian aquifer were analyzed at 96
locations in 1984 by the Florida Department of Environmental Regulation and
the U.S. Geological Survey (Holden 1986). These supplies serve a combined
population of over three million people. Less than 8 percent of the samples
were positive for any of the pesticides sampled for, including heptachlor and
heptachlor epoxide. No other information was reported.
-------
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. None of the 100
water supplies sampled using the 5 aquifers in Florida contained heptachlor
and heptachlor epoxide in excess of the detection limit3. The detection
limits were not reported.
Drinking water wells were sampled in Idaho for heptachlor epoxide by the
Idaho Department of Health and Welfare (1984). One sample out of 107 was
positive with a concentration of 0.015 ug/1 (no detection limit was reported).
Sandhu et al. (1978) summarized the results from a study conducted in two
rural counties in South Carolina. Water supplies sampled were outside
incorporated municipalities. Samples were taken randomly from wells where
there was no pretreatment prior to use. It was not noted whether these
supplies were public or private. Also, data were collected on drinking water
from different land use areas in each county (i.e., agricultural, forest, and
residential). The number of samples taken and the number of positive samples
were not reported; however, the percent of samples exceeding the detection
limit of 0.01 ug/1 was reported by the county.
Results indicated that 62.5 percent of the samples taken in Chesterfield
County had detectable levels of heptachlor .and 42 percent had detectable levels
of heptachlor epoxide. The sample means were 0.015 and 0.008 ug/1 for heptachlor
and heptachlor epoxide, respectively. In Hampton County, 45.5 percent of the
samples contained heptachlor and 64 percent of the samples contained heptachlor
epoxide in excess of the detection limit. The sample means for the Hampton
County samples were 0.009 and 0.018 ug/1 for heptachlor and heptachlor epoxide,
respectively. (These percentages included samples taken from forest land use
areas.) Concentrations of heptachlor in drinking water samples from agricultural
areas in Chesterfield and Hampton counties ranged from not detected to 0.16 ug/1
with a mean of <0.01 ug/1 (not detected). For heptachlor epoxide, concentrations
ranged from not detected to 0.09 ug/1 with a mean of <0.01 ug/1 (not detected).
In residential areas in both counties heptachlor concentrations of heptachlor
and heptachlor epoxide in drinking water samples ranged from not detected to
0.045 ug/1 and not detected to 0.01 ug/1, respectively. Sample means were
<0.01 ug/1 (not detected) in all cases.
-------
Tucker and Burke (1978) reported that heptachlor epoxide was detected at
a level of 0.6 ug/1 in water from a public drinking'water supply well in
Camden County, New Jersey.
Surface Water Sources
Irwin and Healy (1978) reported that none of 16 surface water supplies
sampled in Florida contained heptachlor or heptachlor epoxide in excess of the
detection limits. The detection limits were not reported.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected "covered the period from the raid 1970s to early 1985. Hie
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. Ohe data collected cover the
period from the mid-1970*s to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and Uhited States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background
information on the community; 2) treatment plant schematics and associated
treatment process information; and 3) water quality data. Water sample types
included raw water (treatment plant intake), distribution water (treated
water), and tap water. Water quality data collected included general param-
eters (e.g., alkalinity, turbidity), microbiological and radiological param-
eters, inorganic parameters, and organic parameters (including volatiles,
base/neutrals, pesticides and PCBs, and phenols and acids). For each param-
eter, the water type, time period, concentration (mean and range), number of
samples, and detection limit are presented.
For most of the volatile organics, including heptachlor and heptachlor
epoxide, the available data indicated that there were very low levels of these
contaminants in the raw, treated, or tap water. Most of the values found were
"not detected" or near the detection limit (Canadian Public Health Association
1986).
-------
Unspecified Sources
Pellizzari et al. (1979) presented monitoring data on halogenated hydro-
carbons in drinking water from the northern New Jersey area and the Buffalo/
Niagara, New York area. The limits of detection ranged from 0.005 to 0.025
u9/l for both studies. Of 22 drinking water samples taken from the^ northern
New Jersey area, two samples (9%) contained heptachlor at levels of 0.007 and
0.02 ug/1, and none contained heptachlor epoxide. None of the 16 drinking
water samples from the Buffalo/Niagara area were found to contain heptachlor
or heptachlor epoxide in excess of the detection limits.
In a report on source identification of pollutants entering a sewage
treatment plant, Levins et al. (1979a) sampled two drinking water sources in
a drainage basin in Georgia. Although detection limits were not reported,
heptachlor and heptachlor epoxide were not detected in either of the drinking
water samples.
8.2.1.2 Occurrence in Water Other Than Drinking Water
Three national studies are summarized for the occurrence of heptachlor
and heptachlor epoxide in surface waters. Eight regional studies addressed
levels of heptachlor and heptachlor epoxide in water other than drinking water.
Two of the regional studies addressed levels of heptachlor and heptachlor
epoxide in ground water, and six studies provided data on levels of heptachlor
and heptachlor epoxide in surface water.
Ground Water Sources
Ground water wells in 25 California counties were analyzed as part of the
California State Board's Toxics Special Project during 1984 (Cohen and Bowes
1984). Heptachlor was found in three samples with a maximum concentration of
0.3 ug/1. Die other positive values, total number of samples, and detection
limit were not reported.
Tucker and Burke (1978) presented data on levels of heptachlor and
heptachlor epoxide in samples of water collected from 163 wells, including
-------
private and public drinking water supplies, industrial sites, and wells in the
vicinity of landfills in nine New Jersey counties. The analysis showed that
samples of water from three supply wells contained heptachlor and samples from
seven wells contained heptachlor epoxide in excess of the minimum reportable
concentration of 0.01 ug/1. The maximum values for heptachlor and heptachlor
epoxide were 1.0 and 0.6 ug/1, respectively. The maximum concentration for
heptachlor epoxide occurred in a public drinking water supply well.
Surface Water Sources
National studies of surface water analyses for heptachlor and heptachlor
epoxide were conducted by USEPA (1976) and Breidenbach et al. (1967, as
cited in USEPA 1980f). The overall mean as reported for both pesticides
was 0.0063 ug/1. The minimum values determined for each were both 0.001 ug/1,
and the maximum values were 0.035 ug/1 and 0.02 ug/1 for heptachlor and
heptachlor epoxide, respectively. The number of positive values, total number
of samples, and detection limit were not reported.
The National Pesticide Monitoring Network examined rivers nationwide from
1975 to 1980 (Gilliom et al. 1985). Heptachlor epoxide was reported as being
detected in 9 out of 2,946 samples analyzed from 177 locations (detection
limit = 0.01 ug/1). The mean and range of values were not reported.
Heptachlor and heptachlor epoxide were analyzed for in rivers and streams
in upstate New York (Estabrooks no date) in 1982-1983. No samples for either
pesticide were found to be positive out of 252 samples analyzed. The detec-
tion limit was 10.0 ug/1.
Barks (1978) presented the results of a USGS water quality study con-
ducted from April 1973 to July 1974 in the Ozark National Scenic Riverways,
Missouri. During the study, 20 surface water samples were collected from
3 sites on the Current River and 1 site on Jacks Fork and analyzed for
pesticide content. The analysis of unfiltered samples found no heptachlor or
heptachlor epoxide in excess of the detection limit (the detection limit was
not reported).
-------
Englande et al. (1978) presented the results of extensive chemical
analysis of six Advanced Wastewater Treatment (AWT) plant effluents. Four
plants were located in California, and one each in the District of Columbia
and Texas. None of the 63 AWT effluent samples contained concentrations of
heptachlor in excess of the detection limit (the detection limit was not
reported).
Truhlar and Reed (1976) reported on water samples collected from four
streams in Pennsylvania and analyzed for chlorinated hydrocarbon pesticides
during the period from April 1970 to February 1971. The streams drained four
different types of land use areas. Concentrations of heptachlor and hepta-
chlor epoxide were not detected in any of the 25 stream samples.
Schacht (1974) presented the results of a study to determine the levels
of pesticides in the surface water of Lake Michigan and its tributaries.
During the period 1970 to 1972, a total of 45 water samples were collected.
Concentrations of heptachlor epoxide ranging from "non-detected" to 0.017
ug/1 were found in the samples. The detection limit for heptachlor epoxide
was reported as less than 0.0002 ug/1.
Dappen (1974) reported the results of a study to determine the pesticide
content of urban storm runoff in Nebraska. Runoff samples were collected at
three stations in a Nebraska city during and after storms. A total of 80
samples were collected at the first station during 16 different storms.
Concentrations of heptachlor and heptachlor epoxide in samples from the first
station ranged from 0 to 0.059 ug/1 and 0 to 0.2 ug/1, respectively. At the
second sampling station, 55 samples were collected during 9 different storms.
Concentrations of heptachlor and heptachlor epoxide in samples from the second
station ranged from 0 to 0.176 ug/1 and 0 to 0.194 ug/1, respectively. A total
of 14 samples were collected at the third station during 3 storms. Concen-
trations of heptachlor in samples from the third station ranged from 0 to
0.055 ug/1. No heptachlor epoxide was detected. No detection limits were
reported for the study.
-------
8.2.2 Ambient Air
Several studies have been performed by EPA in recent years to determine
levels of pesticides in ambient air. Kutz et al. (1976) reported on levels of
heptachlor in the ambient air of 16 states between 1970 and 1972. The maximum
level measured was 0.0278 ug/m3 in Tennessee; the mean value for all positive
samples was 0.001 ug/m3. Heptachlor was detected in 42 percent of the 2,479
samples collected during the monitoring period (no detection limit was
reported). Although the frequency of detection increased slightly during the
study period (38-44%), the mean of positive samples declined from 0.0014 to
0.0007 ug/m3. Kutz et al. (1976) cautioned that the sampling locations in
these studies were selected for potentially high concentrations of pesticides
in ambient air.
Kutz et al. (1976) also presented the results of a pilot suburban air
monitoring program that was conducted between April and June 1975. During
that period, five duplicate samples were collected in suburban areas of Miami,
Florida; Jackson, Mississippi; and Fort Collins, Colorado. Heptachlor was
detected in all of the samples collected near Miami and Jackson at mean values
of 0.0021 and 0.011 ug/m3, respectively. None of the samples collected from
Fort Collins, Colorado, contained detectable levels of heptachlor (no
detection limit was reported). The maximum, value of heptachlor measured in
the 30 samples was 0.0221 ug/m3. Heptachlor epoxide was detected in 20
percent of the samples from Jackson, Mississippi, with a maximum reported
value of 0.0013 ug/m3. NO heptachlor epoxide was detected in samples from
Florida or Colorado.
Arthur et al. (1976) collected and analyzed air samples from the
Mississippi Delta, one of the highest pesticide usage areas in the united
States due to the intensive cotton production. Between 1972 and 1974, samples
were collected at weekly intervals; a total of 156 samples was obtained. The
maximum heptachlor level reported was 0.008 ug/m3. No other information was
reported regarding levels of pesticides monitored during the sampling period.
Kutz et al. (1976) presented the results of a pilot suburban air moni-
toring program that was conducted between April and June 1975, reporting a
-------
maximum heptachlor epoxide concentration of 0.0013 ug/m3 and that heptachlor
epoxide was detected in 20 percent of the samples from only one of the three
sites. No detection limit was reported.
Arthur et al. (1976) reported finding a maximum heptachlor epoxide level
of 0.0093 ug/m3 among 156 samples collected weekly from the Mississippi Delta
between 1972 and 1974.
8.2.3 Soil/Sediments
One soil study and two sediment studies were identified that examined the
occurrence of heptachlor and heptachlor epoxide. Carey et al. (1978, 1979)
presented data collected during the 1971 and 1972 National Soils Monitoring
Program. During the late summer and fall of 1971, composite soil samples were
obtained from 1,486 10-acre sites in 3 7 states. Data were collected from
1,483 sites in the same states in the summer and fall of 1972. During the
1971 sampling program, heptachlor was detected in 73 (or 5%) of the 1,486
samples, in the range of 10 to 1,370 ug/kg. Heptachlor epoxide was detected
in 103 (or 7%) of the 1,486 samples, in the range of 10 to 430 ug/kg. During
the 1972 collection program, heptachlor was detected in 57 (or 4%) of the
1,483 samples, in the range of 10 to 600 ug/kg. Heptachlor epoxide was
detected in 97 (or 7%) of 1,483 samples, in the range of 10 to 720 ug/kg. "Hie
minimum detection limits for organochlorines ranged from 2 to 30 ug/kg.
Britton et al. (1983) reported on levels of pesticides in water-sediment
mixtures (unfiltered samples) and in bottom material samples collected by the
National Stream Quality Accounting Network (NASQAN) in 1976. Throughout the
United States, 151 permanent stations, plus stations added as part of local
programs, were sampled for pesticides, including heptachlor. Water-sediment
mixtures were collected quarterly; bottom materials were collected semi-
annually. During 1976, 169 water-sediment samples and 153 bottom material
samples were collected. Heptachlor was detected in water-sediment samples at
1 of 13 stations in the Texas-Gulf region and 1 of 5 stations in the Lower
Colorado region. Hie maximum level of heptachlor found at these stations was
0.05 ug/1. Only one of six bottom material samples collected in the Arkansas-
White-Red region had a detectable level of heptachlor, at 4.8 ug/kg. Heptachlor
-------
epoxide was detected in one of six water-sediment samples from the California
region, at a level of 0.01 ug/1. The detection limits for heptachlor and
heptachlor epoxide in the water-sediment and the bottom material samples were
not reported.
Schacht (1974) presented the results of a study to assess the levels of
pesticides in the sediments of Lake Michigan and its tributaries. During the
1970-1971 collection period, a total of 50 sediment samples were collected
from tributaries to Lake Michigan, at stations 40 to 80 yards offshore of
sewage treatment plant discharges into Lake Michigan and at open water
stations 1 to 3 miles offshore in Lake Michigan. Levels of heptachlor in
samples collected in the tributaries ranged from "non-detected" to 0.24
u9/kg, with an average concentration of 0.06 ug/kg. No heptachlor was
detected in samples collected at the stations in Lake Michigan. Heptachlor
epoxide was identified in samples collected in Lake Michigan, in the range of
"non-detected" to 1.35 ug/kg, with the maximum concentration of 1.35 ugAg
found in a sample collected 40 to 80 yards offshore of a sewage treatment
plant discharge. In addition, heptachlor epoxide was found in samples
collected in the tributaries to Lake Michigan, in the range of 0.02 to 57
u9/kg. The detection limits for heptachlor and heptachlor epoxide in
sediments were 0.0025 and less than 0.0025 ug/kg, respectively.
8.2.4 Food
Several estimates of dietary intake of heptachlor and heptachlor epoxide
have been reported for recent years, the Food and Drug Administration (FDA)
includes heptachlor and heptachlor epoxide among the contaminants analyzed
for in its "total Diet Studies (also known as the Market Basket Surveys) to
evaluate the daily intake estimates for the adult male, 6-oonth old infant,
and 2-year old toddler (Gartrell et al. 1986a,b).
The total heptachlor intake for the adult male in FY 81/82 was 0.0757
ug/day. Hie grain and cereal products group contributed 100 percent of the
intake. Hie total heptachlor epoxide intake for the adult male in FY 81/82 was
0.467 ug/day. The meat, fish, and poultry and the dairy products categories
accounted for 39.6 percent of the total intake of heptachlor epoxide by the
adult male.
-------
V,
J.
COy?^y °^t **«, v*,
* /'So
heptachlor epoxide, its transformation
<3 been reported in several studies,
-°cticide and seed and soil
~"er, the cancellation
e heptachlor
<*©
1 cations
®.f
*>
'0;
v«*.
*<*
Oe
<9.
Of
;9t
Of
•0*
'®*e* to
'3.. *<^£
^ "
<*V e*>^ ' <*„ 9sej. "" t,
^Or
'S*J.
Of
Of
, ***»
*»• „ ^
*e*t.
V*.
e®
ror
o* *-fit*OA ^6 n,
^ eA>*, ^ ^
S" A. *«• ,. * ^
Oy *6j
S-e Vr '
®« .
"¦h ,
0 „ ^Oi
* °&o
to"^;a;
**<> 4 "
0^-l Q Qfit*0*l0
*bm - **fc ** -
**¦©
e^.
re,
r«vj,
r«^c
eQ
?eCtA
Xepf. °f j ^Of '
,0« ^ ^>o, X ^
® , 0 0., <""> „ *>*U f'
^ ° °'7U 0'°0°1 ^ tJ^'°°0?
r*0,
'°5a
i
ce
0.02
sed in
c<».
co>- "<>
¦fn^T ^ ..
¦°^
°^
'*<3
*+e
°f
%t 6
'O^.
e,„. ""J
*C°4N
0V°0'
0.'0°7
°°:0°°°>
<%
isket Surveys)
s of heptachlor
tales) are
.184 ug/day
xes that the
solely from foods
heptachlor residues
:'rom insects or that
rmination that heptachlor
. meat, fish, and poultry
tken in is substantially
iat bioaccumulation occurs
been treated or contaminated.
•Go
'8
-------
The low-level detection of heptachlor in 42 percent of almost 2,500 air
samples taken from 16 states between 1970 and 1972 provides an indication of
removal occurring from water and soil by volatilization. Although the
locations sampled had been selected for the potential of experiencing high
levels of pesticides present in the ambient air and thus the percent of
detectable measures was greater than would have resulted from a truly random
survey, the maximum concentration measured only 0.0278 ug/m3. Again, due to
restrictions imposed, current levels in the ambient air are expected to be
lower.
Hie information currently available on the occurrence of heptachlor and
heptachlor epoxide in the environment and the potential for exposure is
insufficient to determine the national distribution of intake by any of the
three routes, for either compound (Tables 8-2 and 8-3). However, taking into
consideration that the data reported for drinking water and air exposure were
Table 8-2* Exposure Estimates for Heptachlor
Reported Exposure Estimated Adult
Source Levels Intake
Drinking Water <0.1 ug/m3 <0.2 ug/day
Diet — 0.0007 ug/day
Air ? ?
Table 8-3. Exposure Estimates for Heptachlor Epoxide
Reported Exposure Estimated Adult
Source Levels Intake
Drinking Water <0.1 ug/m3 <0.2 ug/day
Diet — 0.184 ug/day
Air ? ?
-------
derived' from studies conducted during the early to mid-1970s (prior to
restriction of the use of heptachlor) and that data for food exposure were
collected during the early to mid-1980s? therefore, it is likely that current
inhalation and drinking water exposure levels will be much less than the
reported levels and that current dietary exposure levels will be similar to the
reported levels. While exposure to heptachlor is expected to be equally minimal
from all three sources, diet is likely to be the greatest source of exposure
to heptachlor epoxide.
Because the ranges of potential intake levels from drinking water and air
exposure are so wide and varied for both compounds and because the estimated
intake levels provided for food are single values and were collected under
different usage conditions, it is not possible at this time to generate an
accurate estimate of the total exposure to either heptachlor or heptachlor
epoxide. Additional data are needed to evaluate current exposure levels and to
provide a comparable basis for estimating the total combined intake from each
media.
-------
9. LINDANE
9.0 SUMMARY
Lindane is a widely used insecticide. Due to its rapid volatilization
from soils and surface waters, levels of lindane in drinking water are
expected to be very low for most systems. Monitoring of lindane, while
limited, suggests that it will not occur at levels greater than 0.1 ug/1.
9.1 GENERAL CHARACTERISTICS
9.1,1 Physical/Chemical Properties
Lindane gamma-hexachlorocyclohexane, gamma-benzene hexachloride) is the
active ingredient of benzene hexachloride (BHC) (NAS 1977). Synonyms and
identifiers for lindane include the following (Berg 1986):
Agronexit
Exagamma
Forlin
Gallogamma
Gamaphex
Gammacol
Gammalin
Gamma BHC
Gamma HCH
Gammex
Gammexane
Inexit
Isotox
Lacco Hi Lin
Lacco Lin-O-Mulsion
Lindafor
Lindagam
Lindagrain
Lindagranox
Lindalo
Lindamul
Lindapoudre
Lindaterra
Lin-O-Sol
Nexit
Novigam
Silvanol
Lindane is a crystalline substance at 25°C and has a musty odor (Windholz
1976). It has a molecular weight of 290.85, a molecular formula of CgHgCig,
and a melting point of 112.50C. The aqueous solubility of lindane at 25oC is
7.8 mg/1 (2.7 x 10~5 mol/1) and it has reported vapor pressures at 20°C of 1.6
x 10-4 torr (2.1 x 10-7 atm) (Mabey et al. 1981) and 9.4 x 10~6 torr (Callahan
et al. 1979).
The ratio of the highest reported vapor pressure at 2CPC to the aqueous
solubility at 25°C gives an estimated Henry's constant for lindane at 20-25OC
of 7.8 x 10-6 atm«m3/mol. Hie ratio of the estimated Henry's constant to the
product of the gas constant times the temperature in degrees Kelvin gives an
-------
estimated dimensionless Henry's constant for lindane at 20O-25OC of 3.2 x 10~4.
Reported values for lindane include 911 (Kenaga and Goring 1978) and
1.08 x 103 (Rao and Davidson 1980).
9.1.2 Use
The USDA (1980a) reports 555 Federal registrations for products containing
lindane and 86 state-registered products for which Federal registrations have
been requested. The registrations are for the use of lindane-containing
products on a large number of fruit and vegetable crops, including pineapples,
pecans, and avocados; on lawns, turf, and commercial ornamentals; on commercial
and government forests and timberlands; on Christmas trees; as a seed treatment
on field crops; as dips and sprays for livestock, pets, and their premises; and
as a treatment for controlling parasites that infest humans (USDA 1980a; USEPA
1980e).
In February 1977, EPA issued a notice of a rebuttable presumption against
registration and continued registration of pesticide products containing lindane.
In 1983, EPA's Office of Pesticide Programs issued Lindane Position Document 4
(USEPA 1983d). In a summary of its regulatory position, EPA (USEPA 1983d)
listed restrictions for the use of lindane on commercial ornamentals, avocados,
pineapples, pecans, livestock, Christmas trees, structural treatments, forestry,
and dog dusts, and washes and in household products including flea collars,
shelf paper, and household sprays. Additional household uses were cancelled in
1984 (Kuch 1986).
Lindane is used nationwide as an insecticide. EPA reports 0.8 to 1 million
pounds of lindane used in 1986 (Kuch 1986). Major use areas are Hawaii, the
Midwest, Federal forest areas, the South, and California.
9.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of lindane is divided into the
following subsections! 9.1.3.1 Volatilization; 9.1.3.2 Sorption and leaching
Potential; 9.1.3.3 Abiotic Transformations; 9.1.3.4 Biodegradation and
Persistence in Soil and Water; and 9.1.3.5 Summary. The discussion will
emphasize the environmental fate of lindane in soil and water.
-------
9«1.3.*1 Volatilization
Lyman et al. (1982) indicate that compounds such as lindane (H = 7.8 x
10-6 atm«m3/mol) with Henry's constants between 1 0~6 an
-------
case,-volatilization appears to be an important removal process for lindane on
the soil surface.
Volatilization rates for lindane beneath the soil surface would be expected
to decrease rapidly with depth. Jury et al. (1983) developed a mathematical
model for describing transport and loss of soil-applied organic chemicals.
Using this model, the authors calculated cumulative volatilization of lindane
after 0 days, expressed as a percent of the 1 kg/ha amount initially incorporated.
Assuming no water evaporation, the results indicated a maximum of 32.6 percent
volatilization (depth of incorporation = 1 cm, organic carbon fraction = 0.0125,
boundary layer thickness = 0.5 mm) and a minimum of 1.9 percent volatilization
(depth of incorporation = 10 cm, organic carbon fraction 0.0250, boundary layer
thickness = 5 mm). The authors concluded that for shallow incorporation of
lindane under well-mixed surface conditions, volatilization represents a
significant loss pathway. However, the importance of volatilization decreases
rapidly with soil depth.
9.1.3.2 Sorption and Leaching Potential
Substituting the mean KQC value of 1.0 x 10^ (Section 9.1.1) into
equation A-11 gives estimated sediment or soil/water equilibrium partition
coefficients (Ks/W) for lindane ranging from 10 to 80 for sediments or soils,
with organic carbon fractions ranging from 0.01 to 0.08, respectively. Hie
estimated Ks/W values indicate that at equilibrium, the concentration of
lindane in suspended or exposed bottom sediment may be more than 10 times the
concentration in the water column. However, since the estimated Kg/W values
are less than 100, it is unlikely that the ratio of the total mass of lindane
adsorbed to suspended and exposed bottom sediment to the total mass of lindane
dissolved in the water column will exceed 0.1 in most surface waters. The
reason is that in most surface waters most of the time, the ratio of the water
mass to the mass of suspended and exposed bottom sediment exceeds 1,000 (CJSGS
1983). Therefore, transport by adsorption to suspended sediment and removal
by adsorption to bottom sediment are probably not important processes for
lindane in most surface waters. That conclusion is supported by studies on
the lindane distribution between water and sediment in a flooded quarry and in
an oligotrophic lake (Callahan et al. 1979).
-------
Substitution of the mean KQC value of 1.0 x 10^ into equation A-13 gives
an estimated soil TLC Rf value (Appendix A) of 0.19 for lindane adsorbed to a
soil with an organic carbon fraction of 0.014, a pore fraction of 0.5, and a
soil density of 2.5 g/cm3. Therefore, based on the five mobility classes
defined by Helling and Turner (1968) and cited by Hamaker (1975) for a soil
with the same properties (Appendix A), lindane would be expected to be at least
moderately immobile (Class 2) in, and to be at least moderately resistant to
leaching from, surface soil.
9.1.3.3 Abiotic Transformations
There is no information available concerning the hydrolysis of lindane
under environmental conditions. However, only negligible losses of lindane
over an 8-week period from river water samples with pH values varying between
7 and 8 indicate that hydrolysis probably is not an important removal process
for lindane (Eichelberger and Lichtenberg 1971, as cited in Callahan et al.
1979). If it is assumed that the reported neutral first-order hydrolysis rate
constant at 20OC for lindane in a 75-percent ethanol mixture with water is the
approximate value in water, a hydrolysis half-life of 180 days can be
estimated for lindane in water at a pH of 8 (Callahan et al. 1979).
There is no information on the photo-oxidation of lindane in natural
waters. However, the compound does not have functional groups that are
readily susceptible to oxidation, and attempts to oxidize lindane with strong
oxidants such as ozone, chlorine, and potassium permanganate have failed
(Leigh 1969, as cited in Callahan et al. 1979).
Although lindane is reported to undergo some limited photolysis in sun
light, it is not expected to undergo significant rates of photolysis in the
environment (Callahan et al. 1979). The reason is that it does not have any
chromophores that absorb light strongly at wavelengths above the approximate
solar radiation cutoff at the earth's surface of 290 nm (Callahan et al. 1979;
Lyman et al. 1982).
-------
9.1.3*4 Biodegradation and Persistence in Soil and Water
The report that only negligible amounts of lindane were lost over an
8-week period in river water (Eichelberger and Lichtenberg 1971, as cited in
Callahan et al. 1979) suggests that biodegradation may not be an important
removal process for lindane in at least some natural waters. Estimated mean
degradation half-lives of 266 days and 151 days have been reported for lindane
in laboratory incubated aerobic and anaerobic soils, respectively (Rao and
Davidson 1980). Therefore, lindane has been classified as persistent in soil
(e.g., half-life >100 days) by Rao and Davidson (1980).
9.1.3.5 Summary
Based upon the above discussion and the literature review by Callahan
et al. (1979), the following tentative conclusions can be made concerning the
most likely behavior of lindane in soil and water:
o Based upon theoretical cnsiderations assuming that lindane has a vapor
pressure of 1.6 x 10~4 torr, volatilization may be an important removal
process for lindane in some shallow and/or turbulent surface waters,
but not in deep and/or stagnant waters or groundwater. If the lower
reported vapor pressure of 9.4 x 10~® torr is assumed, volatilization
is unlikely to be an important removal process for lindane in natural
waters under any conditions.
o Based upon theoretical considerations, volatilization is expected to
be an important removal process for lindane on the soil surface and
beneath the soil surface at shallow depths, but not for lindane in soil
at deeper depths.
o Based upon theoretical considerations and some limited supporting data,
transport by adsorption to suspended sediment and removal by adsorption
to bottom sediment are not expected to be important processes for
lindane in surface waters.
o Based upon theoretical considerations and groundwater monitoring,
lindane is predicted to be fairly resistant to leaching from soils.
o Based upon theoretical considerations a.nd some limited supporting
data, lindane is not expected to undergo significant rates of
hydrolysis, oxidation, or photolysis in the environment.
o Based upon its negligible loss from river water over an 8-week period,
it appears that biodegradation may not be an important removal process
for lindane in at least some natural waters.
-------
o Lindane has been classified as persistent in soils (e.g./ half-life
>100 days) based on reported degradative half-lives of 266 and 151
days in aerobic and anaerobic soils, respectively.
o Due to its apparent resistance to leaching but fairly long persistence
in water and soil/ lindane is expected to pose a medium- to long-range
but low-intensity threat to ground water and surface waters.
9.2 OCCURRENCE IN THE ENVIRONMENT
9.2.1 Drinking Water Sources
This section presents the available data from monitoring studies and
surveys to determine the extent of occurrence of lindane in public drinking
water supplies and water other than drinking water.
9.2.1.1 Occurrence in Drinking Water
National and regional studies have addressed concentrations of lindane
in drinking water. The results of four national studies and seven regional
studies are discussed in this section. Where possible, reported levels of
lindane in drinking water from groundwater sources and surface water sources
have been discussed separately.
Ground Water Sources — National Studies
The Federal Reporting Data System (FRDS 1984) provides information on
public water supplies found to be in violation of current maximum contaminant
levels (MCLs). The system is representative of approximately all 47,700 U.S.
public ground water supplies. Although the number of ground water systems in
the United States that have monitoring requirements for lindane is not provided
by the FRDS/ the data indicated that there were no violations of the MCL of
4 ug/1 for lindane between 1979 and 1983.
A detailed survey of the contaminants in the water supplies of 10 cities
waB conducted as part of the National Organics Reconnaissance Survey (USEPA
1975b). Two of the systems sampled, located in Florida and Arizona/ utilized
ground water as their water source. Lindane was not observed above the minimum
-------
quantifiable concentration in samples taken from these water supplies. The
minimum quantifiable concentration was not reported.
The 1978 Rural Water Survey (USEPA 1984i) involved the collection of
samples from 267 households in rural locations throughout the United States.
The majority of the households tested utilized private water supplies; however,
samples of water from 71 public drinking water systems of varying sizes using
ground water were tested for lindane. Of these, only one public ground water
system exceeded the minimum quantification limit of 0.002 ug/1 for lindane.
The lindane concentration in the water from that system was 0.006 ug/1.
Ground Water Sources — Regional Studies
Two positive lindane samples out of 107 samples analyzed were found for
ground water wells in Idaho (Idahe Department of Health and Welfare 1984).
Monitoring for pesticides in drinking water wells is not routinely done; the
sampling performed was in response to a particular contamination incident, not
for any comprehensive monitoring program. A relatively low mean of 0.0007 ug/1
was reported for these two samples, with one sample having a lindane concen-
tration of 0.076 ug/1 (no detection limit was reported). It appears that
"negative" samples, possibly assigned a detection limit value, were included
in the calculation of the mean.
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. Of the 100 ground
water supplies sampled utilizing the five aquifers in Florida, none contained
lindane in the finished water in excess of the detection limit (detection limit
not reported).
Sandhu et al. (1978) summarized results from a rural water supply study
in two counties in South Carolina. Waste supplies sampled were outside incor-
porated municipalities. Samples were taken randomly from wells where there was
no pretreatment prior to use. It was not noted whether these supplies were
public or private. Also, data were collected of drinking waters from different
land use areas in each county (i.e., agricultural, forest, and residential).
Data are presented below for agricultrual and residential areas. Although the
-------
number of samples and the number of positive samples were not reported, the
sample mean and the percent of samples exceeding the detection limit of
0.01 ug/1 were reported.
The results of the study by Sandhu and coworkers showed that 58.3 percent
of the samples taken in Chesterfield County and 100 percent of the samples
taken in Hampton County had concentrations of lindane ir. excess of the detection
limit of 0.01 ug/1 (these percentages include forest land use samples). The
range, mean, and median values for agricultural samples in Chesterfield County
were not; detected - 0.14, 0.037, and 0.005 ug/1, respectively. For residential
samples in Chesterfield County, these same values were: not detected - 0.054,
0.009, and 0.016 ug/1, respectively. Concentrations or samples from Hampton
County were higher. The range, mean, and median values for the agricultural
area samples were 0.010 - 0.27, 0.10, and 0.16 ug/1, respectively. The same
values for residential area samples in Hampton County were: 0.14 - 0.32, 0.18,
and 0.22 ug/1, respectively.
Achari et al. (1975) reported the results of a study conducted in South
Carolina to determine the ambient levels of lindane in ground water. During
the study, 27 samples were collected and analyzed. Analysis of the samples
showed a range of "non-detected" to 0.021 ug/1 lindane, with an average of
0.0012 ug/1. However, the detection limit for lindane and the number of
positive samples were not reported.
Benvenue et al. (1972a) conducted a study to determine the extent of
organochlorine pesticide contamination of drinking water in Hawaii. A total
of 45 samples of finished drinking water was collected from February 1971 to
May 1971. Lindane was detected in 2 of the samples, with values of 0.00006 to
0.0004 ug/1 (detection limit not given).
Surface Water Sources — National Studies
Information was obtained from the Federal Reporting Data System (FRDS
1984) on violations of the current MCL of 4 ug/1 for lindane. Data were
obtained for the years 1979-1983 and include information for all of approx-
imately 11,600 surface water systems in the United States. No violations of
-------
the MCL of 4 ug/1 for lindane were reported for samples analyzed during this
time period.
In the NORS survey of 10 cities (USEPA 1975b), samples of drinking water
were collected from eight systems having surface waters affected by different
types of pollution. Samples of water from two of the eight systems analyzed
contained lindane. Mater from one system in Ohio that utilized raw water
contaminated with industrial discharges contained 0.01 ug/1 of lindane. The
other system located in North Dakota, utilized raw water contaminated with
agricultural runoff. Water sampled from this system showed a detectable but
nonquantifiable concentration of lindane, The minimum quantifiable concen-
tration for lindane in the MORS study was not reported.
The Rural Water Survey (USEPA 1984iJ-also presented data on samples of
drinking water obtained from surface water sources. Samples were taken from
21 drinking water sources of varying sizes throughout the United States. None
of the samples collected from these systems contained concentrations of lindane
in excess of the minimum quantification limit of 0.002 ug/1.
Surface Water Sources — Regional Studies
Irwin and Healy (1978), summarizing data collected during water quality
reconnaissance of public water supplies in Florida, reported that none of 16
surface water supplies sampled contained lindane in excess of the detection
limit. The detection limit was not reported*
In a study on the effects of forest runoff on the quality of a public
surface water supply in Oregon, Elliott (1979) observed an ambient concen-
tration of lindane of 2 ug/1. Sampling data such as the number of samples,
number of positive samples, and other information were not provided.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. The
study was funded by the Health Protection Branch of Health and Welfare Canada,
-------
and the Ontario Ministry of the Environment. The data collected cover the
period from the mid-1970's to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 ccrrnuni ties, data consisted of: 1) background
information on the community; 2) treatment plant schematics and associated
treatment process information; and 3) water quality data. Water sample types
included raw water (treatment plant intake), distribution water (treated
water), and tap water. Water quality data collected included general
parameters (e.g., alkalinity, turbidity), microbiological and radiological
parameters, inorganic parameters, and organic parameters (including volatiles,
base/neutrals, pesticides and PCBs, and phenols and acids). For each
parameter, the water type, time period, concentration (mean and range), number
of samples, and detection limit are presented.
For most of the volatile organics, including lindane, the available data
indicated that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
Unspecified Sources — National Studies
The EPA Water Supply Program (USEPA 1977a) tested 684 drinking water
samples from various sources across the country, for pesticide contamination.
Levels of lindane of less than 0.001 ug/1 were detected in samples from 37
sites, and levels of less than 0.10 ug/1 were found at 16 sites. At three
additional sites, samples contained lindane levels of 0.160, 0.200, and
0.900 ug/1. The detection limit was not reported. No further information on
this study were reported.
Unspecified Sources — Regional Studies
ttie Region V survey (USEPA 1975c), conducted in the first three months of
1975, sampled treated water supplies from 93 utilities in USEPA Region V for
-------
organic chemical content. Only one sample exceeded the detection limit foe
lindane; this sample contained 0.004 ug/1 of lindane.
In a report on source identification of pollutants entering a sewage
treatment plant, Levins et al. (1979a) tested two drinking water sources in a
drainage basin in Georgia. Although the detection limits were not reported,
lindane was not detected in either of the two samples.
9.2.1.2 Occurrence in Water Other Than Drinking Water
One national study and six regional studies presented the results of
monitored levels of lindane in water other than drinking water.
Ground Water Sources — Regional Study
Tucker and Burke (1978) presented data on levels of lindane in samples of
water collected from 163 wells in nine New Jersey counties. Only two samples
contained levels of lindane in excess of the minimum reportable concentration
of 0.01 ug/1. Hie concentrations of the positive samples were 0.01 and
0.9 ug/1, well below the EPA drinking water standard of 4 ug/1.
Surface Water Sources — National Study
The National Surface Water Monitoring Program {NWMP 1982) presented data
on levels of lindane in surface water samples collected throughout the United
States between 1975 and 1979. During the 5-year period, 2,480 samples were
collected. Although no detection level was reported for lindane, four posi-
tive samples were found during testing. The positive samples were collected
in California, Texas, and Hawaii, with concentrations ranging from 0.01-0.02
ug/1.
Surface Water Sources — Regional Studies
Truhlar and Reed (1976) reported the results of analysis of water samples
taken from four streams in Pennsylvania, between February 1969 and April 1971,
and analyzed for chlorinated hydrocarbon pesticides. Ttie streams drained four
-------
types of land use areas: forests, general farms, orchards, and residential
areas. Of the 83 samples collected, 24 were analyzed from the residential
area. Three of these samples were from base flow conditions, and the other 21
samples were taken during storm conditions. All three base flow examples
contained no lindane (reported as 0.00 ugl). Concentrations of the 21 storm
samples ranged from "trace" to 0.34 ug/1. The mean concentration of the
positive samples was 0.08 ug/1, although only one sample exceeded this
concentration. The remainder of the samples ranged from "trace1* to 0.02 ug/1.
No detection limit was reported for the study.
Two surface water samples were tested for lindane in the Beaumont, Texas/
Lake Charles, Louisiana area (USEPA 1980e). No lindane was detected in the
samples (detection limits not reported).
Dappen (1974) reported the results of a study to determine the pesticide
content of urban storm runoff in Nebraska. Runoff samples were collected
at three stations in a Nebraska city during and after storms. A total of
80 samples was collected during 16 different storms at the first station.
Concentrations of lindane ranged from 0 to 0.848 ug/1. At the second sampling
station, 55 samples were collected during 9 different storms. The range of
lindane concentrations in the samples was 0 to 1.02 ug/1. A total of 14 samples
were collected during 3 storms at the third sampling station. Concentrations
of lindane in samples from the third station ranged from "trace" to 0.479 ug/1.
The detection limits were not reported.
Benvenue et al. (1972a) conducted a study to determine the extent of
organochlorine pesticide contamination of nonpotable waters in Hawaii. A total
of 46 samples was collected from several locations on 4 of the islands. Although
the number of positive samples for lindane was not given, the concentrations
ranged from "not detected" to 0.0034 ug/1 (detection limit not reported).
Englande et al. (1978) presented the results of extensive chemical
analysis of six Advance Wastewater Treatment plant effluents. Four plants were
located in California, and one each in the District of Columbia and Texas.
None of the effluents contained detectable levels of lindane. The detection
limit was not reported.
-------
9.2.2 Occurrence in Ambient Air
Several studies have been performed by EPA in recent years to determine
levels of pesticides in ambient air (Kutz et al. 1976; Carey and Kutz 1983).
Kutz et al. (1976) reported on levels of lindane in the ambient air of 16
states between 1970 and 1972. The maximum level measured was 0.0117 ug/m3 in
Tennessee; the mean value for all positive samples was 0.C009 ug/rr,3. Lindane
was detected in 68 percent of the 2,479 samples collected during the monitoring
period (the detection limit was not reported). Although the mean levels of
lindane remained more or less" constant during the study, the frequency of
detection declined from 84 percent to 49 percent. Kutz et al. (1976) cautioned
that the sampling locations in these studies were selected for potentially high
concentrations of pesticides in ambient air.
Kutz et al. (1976) also presented the results of a pilot suburban air
monitoring program that was conducted between April and June of 1975. During
that period, five duplicate samples were collected in suburban areas of Miami,
Florida; Jackson, Mississippi; and Fort Collins, Colorado. Lindane was
detected in all of the samples collected near Miami and Jackson at mean values
of 0.0017 and 0.0023 ug/m3, respectively. Of the samples collected near Fort
Collins, 80 percent contained detectable levels of lindane/ the mean value was
0.0007 ug/m3. The maximum value measured in the 30 samples was 0.0054 ug/m3.
The detection limits for lindane were not reported.
Carey and Kutz (1983) presented a summary of ambient air monitoring data
obtained by EPA for selected pesticides at 10 locations in the United States
in 1980. The states sampled were South Carolina, Texas, Alabama, California,
Mississippi, Montana, and Illinois. Of the 123 samples collected, only
0.8 percent were positive for lindane (the detection limit was not reported).
The mean concentration reported was 0.0001 ug/m3; the maximum concentration
was 0.0015 ug/m3.
Arthur et al. (1976) collected and analyzed air samples from the
Mississippi Delta, one of the highest pesticide usage areas in the United
States due to the intensive cotton production. Between 1972 and 1974, samples
were collected at weekly intervals; a total of 156 samples was obtained. The
-------
maximum level of lindane reported was 0.0093 ug/m3. No other information was
reported regarding levels of lindane monitored during the sampling period.
9.2.3 Soil/Sediments
Several studies were identified that examined the occurrence of lindane
in soil and sediments. As expected from lindane's volatility and potential to
bind to sediments, lindane levels in soil and sediments are low. These studies
(one soil and four sediment studies) are summarized below.
Carey et al. (1978, 1979) presented data obtained during the 1971 and 1972
National Soils Monitoring Program. During late summer and fall of 1971, composite
soil samples were obtained from 1,486 10-acre sites in 37 states. Data were
collected from 1,483 sites in the same states in the summer and fall of 1972.
Although lindane is listed as one of the organochlorine compounds detectable by
their methodology, no positive detections for lindane were reported in the soil
samples. The minimum detection limits for organochlorines ranged from 2 to
3 0 ug/kg.
Britton et al. (1983) reported on levels of pesticides in water-sediment
mixtures (unfiltered samples) and in bottom material samples collected by the
National Stream Quality Accounting Network (NASQAN) in 1976. Oiroughout the
United States, 151 permanent stations, plus stations added as part of local
programs, were sampled for pesticides, including lindane. Water-sediment
mixtures were collected quarterly; bottom materials were collected semiannually.
Lindane was detected in water-sediment samples at 1 of 10 stations in the
Arkansas-White-Red region and at 1 of 13 stations in the Texas-Gulf region.
The maximum level of lindane found at these stations was 0.01 ug/1. water-
sediment samples from two of six stations in the California region had detectable
lindane levels with a maximum concentration of 0.05 ug/1. No samples of bottom
material contained detectable levels of lindane. The detection limits for
lindane in the water-sediment and the bottom material samples were not reported.
The National Surface Water Monitoring Program (NWMP 1982) presented data on
levels of lindane in sediment samples collected between 1975 and 1979. Of 939
sediment samples analyzed, only 2 samples contained detectable concentrations
-------
of lindane, at 0.09 and 0.30 ug/kg. The two positive samples were obtained in
Michigan and Pennsylvania, respectively, during the spring and summer of 1977.
The detection limits were not reported.
Schacht (1974) presented the results of a study to assess the levels of
pesticides in the sediments of Lake Michigan and its tributaries. During the
1970-1971 collection period, a total of 50 sediment samples were collected fro.n
tributaries to Lake Michigan, at stations 40 to 80 yards offshore of sewage
treatment plant discharges into Lake Michigan, and at open water stations 1 to
3 miles offshore in Lake Michigan. Levels of lindane in samples collected in
the tributaries ranged from "non-detected" to 0.15 ug/kg. No lindane was
detected in samples collected at the stations in Lake Michigan. The detection
limit for lindane in sediments was 0.0025 ug/kg.
USEPA (1980d) reported that 2 ug/kg--of lindane were detected in a sediment
sample taken from the Calcasieu River (Beaumont, Texas/Lake Charles, Louisiana
area) in March 1978.
9.2.4 Food
The Food and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the occurrence of various substances,
including lindane, in food consumed by adults, toddlers, and infants. According
to the results of a recently published survey for FY 81/82, lindane was present
in the following food categories: neat, fish, and poultry; sugar and adjuncts;
and grain and cereal products (Gartrell et al. 1986a,b). The highest levels of
lindane were detected in the sugar and adjuncts category for the adult survey,
and in the grain and cereal products category for the infant and toddler survey.
The FDA also estimates dietary intake of contaminants detected in food
consumed by adults, infants, and toddlers. In FY 81/82, the total intake of
lindane by the adult male was 0.150 ug/day (Gartrell et al. 1986a). Hie largest
sources of lindane were the meat, fish, and poultry group, which contributed
54 percent of total lindane intake, and the sugar and adjuncts group which
contributed 51.5 percent. The beverages category, which includes drinking
water, did not contribute any lindane to the dietary intake by the adult male.
-------
For the toddler, total intake of lindane in FY 81/82 was 0.0827 ug/day.
Total intake was provided by only two sources, the sugar and adjuncts group
(89.8%) and the meat, fish, and poultry category (10.2%). Hie total dietary
intake by the infant was 0.0145 ug/day. The grain and cereal products cate-
gory was the largest source of lindane, accounting for 80 percent of total
intake. Drinking water did not contribute any lindane to the dietary intake of
infants or toddlers (Gartrell 1986b).
The FDA provided mean daily intakes of lindane reflecting detections of
lindane in 12 total diet studies conducted from April 1982 to April 1985 (FDA
1986). For the 6- to 11-month old infant, daily intake of lindane was 0.014
ug/day. For the 2-year old toddler, lindane intake Was 0.083 ug/day. Lindane
intakes for adult males and females are presented in Table 9-1. For the adult
male, lindane intakes ranged between 0.108 and 0.167 ug/day. Daily lindane
intakes for adult females ranged between 0.084 and 0.135 ug/day* Intakes were
highest for the 14- to 16-year old age group for both sexes.
Source: FDA 1986
9.3 EXPOSURE SUMMARY
Several studies have been conducted which provide useful data on the
extent of occurrence of lindane in drinking water, air, and food. Monitoring
surveys of public water supplies usually report lindane levels ranging from
"non-detected" to 1.0 ug/1. However, actual levels of lindane in public water
supplies are generally much lower than this. These low levels are probably the
result of lindane's rapid volatilization from soil and surface water and its
TABLE 9-1. Summary of FDA Total Diet Study Estimates
for Adult Male and Female Lindane Intake
Sex/Age Group
Intake (ug/day)
14-16 year old female
14-16 year old male
25-30 year old female
25-30 year old male
60-65 year old female
60-65 year old male
0.1 35
0.167
0.097
0.157
0.084
0.108
-------
persistence in the soil column. Unfortunately, due to its extensive use, low
levels of lindane in natural waters may occur in all areas of the united States.
In areas of high usage, lindane may also be a common contaminant in air. According
to available monitoring data, ambient air levels, in areas selected for potentially
high concentrations of pesticides, range from 0.0001 to 0.0117 ug/m3. Recent
dietary studies report that the daily intake of lindane for 25-30 year old males
ranged from 0.108 to 0.167 ug/day with a mean daily intake of 0.157 ug/day.
Table 9.2. Exposure Estimates for Lindane
Reported
Estimated
Source
Exposure Levels
Adult Intake
Drinking Water
<1.0 ug/1
<2.0 ug/day
Diet
—
0.2 ug/day*
Air
0 - 0.01 ug/m3
<0.2 ug/day
*Note: 25-30 year old adult male.
The above table summarizes the exposure levels of lindane in drinking
water, food, and air. The drinking water and ambient air exposure levels are
presented as the maximum levels expected based on the limited amount of
monitoring data available. The dietary exposure is taken from an average of
several total diet studies taken from different regions of the United States
over several years. (The number is rounded to one significant figure.) The
estimated daily intake levels were derived using several exposure assumptions,
including an average intake of 2 liters per day of drinking water and 20
cubic meters per day of air.
The currently available information on occurrence of lindane is insuf-
ficient to determine the national distribution of Intake by any of the three
routes. The table indicates that the total non-drinking water intake of
lindane is quite low, on the order of a few tenths of a ug/day, and if lindane
occurs in drinking water at levels more than a few tenths of a ug/1, it will
be the major source of exposure. The number of people who actually receive
drinking water exposures greater than a few ug/1 is unknown. Restrictions
governing the use of lindane, which were imposed by EPA in 1983, may result
in a decreased daily intake by all exposure routes in the future.
-------
10. METHOXYCHLOR
10.0 SUMMARY
Methoxychlor is a chlorinated insecticide that is widely used on a number
of crops and around livestock. However, because methoxychlor binds tightly to
soil, only low levels are found in . rinfcing water. Based on limited monitoring
data, levels of methoxychlor in drinking water are expected to be less than
0.1 ug/1. Compliance data submitted to EPA under the interim standard for
methoxychlor show no drinking water supply with levels greater than 100 ug/1.
10.1 GENERAL CHARACTERISTICS
10.1.1 Physical/Chemical Properties
Methoxychlor [2,2-bis(p-methoxyphenyl)-1,1,1-trichloroethane] is an insecti-
cide used for the control of insects on fruit and shade trees, vegetables, dairy
and beef cattle, home gardens, around farm buildings, and in streams (Berg 1986;
NAS 1977). Synonyms and identifiers for methoxychlor include Evershield CM and T,
Flo Pro McSeed Protectant, and Marlate (Berg 1986).
Technical grade methoxychlor consists of approximately 88 percent of the
P,p-isomer with the remaining 12 percent being primarily the o,p-isomer. Both
isomers are white crystalline solids at 25°C, have a molecular weight of 345.65,
and a molecular formula of C^H^jCljO (Windholz 1976? Berg 1986; Kirk-Othmer
1981). Ihe p,p-isomer has a melting point of 98°C and an aqueous solubility of
4.0 x 10-2 mg/1 at 24°C. It is very soluble in aromatic, chlorinated, and ketonic
solvents (Berg 1986? Windholz 1976). No vapor pressure for methoxychlor could
be found in the literature. A Koc value of 8.0 x 104 has been reported for
methoxychlor (Kenaga and Goring 1978).
10.1.2 Use
Methoxychlor is registered for use as an insecticide on about 87 crops
(including alfalfa? nearly all fruits and vegetables; and corn, wheat, rice,
and other grains); beef and dairy cattle; swine, goats, and sheep; and for
-------
agricultural premises, home gardens, and outdoor fogging (NAS 1977). EPA also
has registered methoxychlor as a selective chemical for controlling slugs and
snails on apricots, cherries, plums, peaches, and ornamental plants (USDA 1980b).
Methoxychlor is widely used because of its long residual action against insects
and low toxicity to humans (Berg 1986).
In 1976, approximately 1.4 million pounds of methoxychlor lactive ingredient)
were usea on agricultural crops, including alfalfa, tobacco, corn, and hay and
forage (Eichers et al. 1978). In 1932, areas producing corn, sorghum, soybeans,
wheat, barley, oats, cotton, tobacco, rice, peanuts, and alfalfa in the United
States received approximately 650,000 pounds of methoxychlor* (California 1982).
Livestock applications accounted for 4,736,000 pounds of methoxychlor (Eichers
et al. 1978).
More recent data show that 5 million pounds of methoxychlor were used in
1986 (Kuch 1986). In a survey of 33 states, Gianessi reports that 1.5 million
of methoxychlor were used for agricultural purposes (Gianessi 1986).
10.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of methoxychlor is divided into
the following subsections: 10.1.3.1 Volatilization; 10.1.3.2 Sorption a-nd
Leaching Potential; 10.1.3.3 Abiotic Transformations; 10.1.3.4 Biodegradation;
and 10.1.3.5 Summary. The discussion will emphasize the environmental fate of
methoxychlor in soil and water.
10.1.3.1 Volatilization
No information was available concerning the vapor pressure of methoxychlor.
Therefore, SAIC was unable to estimate volatilization half-lives for methoxychlor
in surface waters or on soils. Methoxychlor has been detected in rain and snow,
which suggests that volatilization occurs, but no data on volatilization rates
are available. Volatilization rates of a number of pesticides on soils have been
shown to be proportional to the magnitude of the Koc value (Swan et al. 1979, as
* Personal communication between Herman w. Delvo, Economic Research Service,
U.S. Department of Agriculture, and Corinne Macaluso, JRB Associates, June 5,
1984. Data quoted from a computer printout of the 1982 Crop and Livestock
Pesticide Usage Survey.
-------
cited in Lyman et al. 1982). Therefore, the relatively large Koc value for
methoxychlor of 8.0 x 104 suggests that volatilization may not be an important
removal process for methoxychlor on soil.
10.1.3.2 Sorption and Leaching Potential
Substituting the reported KQC value for methoxychlor of 8.0 x 1 04 into
equation A-11 gives estimated sediment or soil/water equilibrium partition
coefficients (Kg/W) for methoxychlor ranging .jm 800 to 6.0 x 10^ for sediments
or soils with organic carbon fractions ranging from 0.01 to 0.08, respectively.
As discussed in Appendix A, transport by adsorption to suspended sediments and
removal by adsorption to bottom sediments may be important processes in some
aquatic systems for chemicals with Ks/W values as low as 100. Therefore, such
processes are likely to be generally important for methoxychlor with estimated
Ks^w values ranging from 800 to 6.0 x 103. Freeden et al. (1975, as cited in
USEPA 1984g) reported that approximately 50 percent of the methoxychlor in
Saskatchewan River water samples were bound to suspended sediment.
Substituting the reported KQC value of 8.0 x 104 into equation A-1 3 gives
an estimated soil TLC Rf value (Appendix A) of <0.1 for methoxychlor adsorbed to
a soil with a relatively low organic fraction of 0.014, a pore fraction of 0.5,
and a soil density of 2.5 g/cm3. Therefore, based on the five mobility classes
defined by Helling and Turner (1968) and cited by Hamaker (1975) for a soil with
the same properties (Appendix A), methoxychlor would be expected to be immobile
and resistant to leaching from soils. The study by Hunt and Sachs (1969, as
cited in CJSEPA 1984g) supports that conclusion. They reported that all of the
methoxychlor sprayed once a year over a several year period remained within the
top 6 inches of soil and most remained within the top 1 inch of soil.
10.1.3.3 Abiotic Transformations
Wolfe et al. (1977, as cited in USEPA 1984g) list a neutral first-order
hydrolysis rate constant and a second-order base catalyzed hydrolysis rate constant
at 27oc of 2.2 x 10-8 M-Vs-1 and 3.8 x 1O-4 M-1s-1, respectively, for methoxychlor.
Substitution of those rate constants into equation A-14 (Appendix A) and the
resulting pseudo first-order rate constants into equation A-1 gives estimated
-------
hydrolysis half-lives for methoxychlor at 27oc of approximately 1 year over the
normal pH range of 6 to 9 for natural waters. Therefore, hydrolysis does not
appear to be an important removal process for methoxychlor in natural waters.
Methoxychlor appears to undergo rapid rates of some type of photodecomposition
in the presence of naturally occurring organics in some waters but not in others.
Wolfe et al. (1977, as cited in USEPA I984g) reported photodecomposition half-lives
in shallow waters for methoxychlor of 2.2, 5.4, and 2.9 hours in three river
waters, but detected no decomposition after 2 hours in two other river waters.
The photodecomposition is not primarily due to direct photolysis since the photo-
decomposition half-life for methoxychlor in distilled water is greater than
300 hours compared to 7.3 hours in the presence of added humic acid. Indirect
sensitized photolysis and oxidation by singlet oxygen have also been ruled out as
the primary mechanism of photodecomposition (USEPA 1984g). Zepp et al. (1976, as
cited in USEPA 1984g) suggested that the photodecomposition may involve oxidation
by free radicals generated by the photolysis of naturally occurring organics.
Even though photodecomposition rates decrease rapidly with depth, it has been
suggested that photodecomposition is an important removal process for methoxychlor
in natural surface waters since low solubility compounds such as methoxychlor
tend to concentrate in organic surface films (USEPA 1984g).
10.1.3.5 Biodegradation and Persistence in Soil and Water
Brockway et al. (1979, as cited in USEPA 1984g) reported an approximately
80 percent decomposition of methoxychlor in a model ecosystem over a 17-week
period. Luczak (1969, as cited in USEPA 1984g) reported a greater than 90 percent
decomposition of methoxychlor in a polluted surface water over a 28-week period.
Biodegradation nay have been responsible for at least some of the decomposition,
but it is not known whether it was the primary mechanism of removal (USEPA I984g).
As discussed above, photodecomposition may be an important removal process for
methoxychlor in surface waters.
Fogel et al. (1982, as cited in USEPA 1984g) reported a "significant"
degradation of methoxychlor initially present at 1,000 ppm in a soil under
anaerobic conditions after a 100-day incubation, but not under aerobic or sterile
conditions. This indicates that biodegradation is primarily responsible for
-------
degradation under anaerobic conditions and that significant rates of biodegra-
dation did not occur under the aerobic conditions. Similar results were reported
by Castro and Yoshida (1971, as cited in Callahan et al. 1979). Castro and
Yoshida reported that 27 ppm methoxychlor was completely degraded in three out of
four flooded (anaerobic) soils after 90 days incubation, but that concentrations
of methoxychlor between 11 and 27 ppm remained after 90 days incubation in
unflooded (aerobic) soiis. liiey estimated the times required for 50 percent
removal of the methoxychlor in the flooded soils to range from 10 days for flooded
(anaerobic) soils with high organic content to 45 days for flooded (anaerobic)
soils with low organic content. The estimated times required for 50 percent
disappearance in upland unflooded (aerobic) soils were all >90 days. Based on
the above results, methoxychlor would be classified as non-persistent (half-life
<20 days) to moderately persistent (half-life between 20 and 100 days) in flooded
anaerobic soil but persistent (half-life greater than 100 days) in aerobic soils
(Rao and Davidson 1980).
10.1.3.5 Summary
Based on the above discussion and the literature review by USEPA (1985e),
the following conclusions can be made concerning the most likely behavior of
methoxychlor in soils and water:
0 No information could be found or estimates made concerning the volatili-
zation rates of methoxychlor from water. Based on its relatively large
reported Koc value (8.0 x 104), volatilization rates from the soil
surface may be low.
° Based on both theoretical considerations and limited experimental results,
transport by adsorption to sediments and removal by adsorption to bottom
sediments may be important processes for methoxychlor in many aquatic
systerns•
" Based on both theoretical considerations and limited experimental results,
methoxychlor appears to be resistant to leaching from soils.
° Based on experimental results, hydrolysis does not appear to be an
important removal process for methoxychlor in water. Methoxychlor is
susceptible to rapid rates of some type of photodecomposition in some
natural waters but not in others.
0 Biodegradation appears to be the primary removal process for methoxychlor
in soils. Removal rates are much greater in anaerobic soils than in
aerobic soils. In one study, the times required to remove 50 percent of
-------
1,000 ppm in flooded anaerobic soils ranged from 10 days in soils with
high organic content to 45 days for soils with low organic content. The
times required to remove 50 percent of 1,000 ppm methoxychlor in aerobic
soils were all well above 90 days.
0 Due to its apparent resistance to leaching, significant contaminations
of ground water supplies by methoxychlor are unlikely.
10.2 OCCURRENCE IN THE ENVIRONMENT
10.2.1 Water
The following section presents data available from monitoring studies and
surveys to determine the extent of occurrence of methoxychlor in public drinking
water supplies and water other than drinking water.
10.2.1.1 Occurrence in Drinking Water
Studies at both the national and regional levels provided monitoring data
on the occurrence of methoxychlor in public drinking water supplies. Where
possible, reported levels of methoxychlor in drinking water obtained from
ground water sources and surface water sources have been discussed separately.
Ground Water Sources — National Studies
The Federal Reporting Data System (FRDS 1984) provides information on
public water supplies found to be in violation of current maximum contaminant
level (MCLs). No violations of the current MCL of 100 ug/1 were reported for
methoxychlor between 1979 and 1983. The National Interim Primary Drinking
Water Regulations require all states and primary agencies to report violations
of the interim drinking water standards. The reported violations are recorded
in the FRDS. EPA believes that the vast majority of supplies have complied
with the Interim reporting requirements; therefore, it is unlikely that any
drinking water supplies currently exceed 100 ug/1.
The 1978 Rural Water Survey (RWS) (USEPA 1984i) involved the collection
of samples from 267 households in rural locations throughout the United States.
There were a total of 71 public groundwater supplies of varying sizes providing
-------
drinking water to households sampled in the RWS, although the majority of
households sampled used private water supplies. Water from only one public
ground water system exceeded the minimum quantification limit of 0.02 ug/1 for
methoxychlor. Hie concentration of methoxychlor in the water from that system
was 0.09 ug/1.
The National Screening Program for Organics in Drinking Water (NSP) (Boland
1981) was conducted from June 1977 to March 1981. Finished drinking water samples,
collected from 12 ground water systems, were analyzed for methoxychlor. None of
the finished drinking water samples contained methoxychlor in excess of the
quantification limit of 0.1 ug/1.
Ground Water Sources — Regional Studies
Irwin and Healy (1978) analyzed data collected in 1976 during a water quality
reconnaissance of public water supplies in Florida. None of the finished drinking
water samples taken from 100 water supplies using the 5 major aquifers in Florida
contained methoxychlor in excess of the detection limit. No detection limit for
methoxychlor was reported.
Tucker and Burke (1978) presented data on levels of methoxychlor iri samples
of water collected from drinking water wells located near industrial zones and
landfills in nine counties in New Jersey. None of the samples contained
methoxychlor in excess of the minimum reportable concentrations of 0.08 ug/1.
In another study by the Florida Department of Environmental Resources and
the U.S. Geological Survey (Holden 1986), drinking water supplies that tap the
Floridian aquifer were sampled. The population served by this aquifer exceeds
3 million people. Of the 96 sites sampled, less than eight samples were positive
for methoxychlor. No other information was reported.
Twelve towns in Connecticut were sampled by the Connecticut Agricultural
Experiment Station in 1984-1985 (Waggoner 1985). Drinking water wells sampled
served, collectively, a population of over 573,000. Methoxychlor was not detected
at any of 42 locations sampled (total number of samples was not reported). The
detection limit was 3.3 ug/1. Similarly, no methoxychlor was found in 88 samples
-------
taken from drinking water wells on Long Island/ New York, during 1984 (Holden
1986). The limit of detection was not reported*
Sandhu et al. (1978) summarized results from a rural water supply study in
two counties in South Carolina. Water supplies sampled were outside incorporated
municipalities. Samples were taken randomly from wells where there was no
pretreatment prior to use. It was not noted whether these supplies were public
or private. Also# data were collected of drinking waters from different land use
areas in each county (i.e., agricultural, forest, and residential). Data are
presented below for agricultural and residential areas. Although the number of
samples and the number of positives samples were not reported, information on the
sample mean and the percent of samples exceeding the detection limit of 0.01 ug/1
is available. The results showed that 46 percent of the samples taken in
Chesterfield County and 64 percent of the samples taken in Hampton County had
concentrations of methoxychlor in excess of the detection limit of 0.01 ug/1
(These percentages include forest land use samples). The range, mean, and median
values for agricultural samples in Chesterfield County were: not detected -
0.032, 0.003, and <0.01 ug/1 (not detected), respectively. For residential
samples in Chesterfield County, these same values were: not detected - 0.15,
0.049, and <0.01 ug/1 (not detected), respectively. Concentrations of methoxychlor
in samples from Hampton County were generally lower. The range, mean and median
values for the agricultural area samples were: not detected - 0.027, 0.01, and
0.018 ug/1, respectively. The same values for residential area samples in Hauqpton
County were: not detected - 0.069, 0.032, and 0.005 ug/1, respectively.
Surface Water Sources — National Studies
Information was collected from the Federal Reporting Data System (FRDS)
1984) on violations of the current methoxychlor MCL of 100 ug/1 in all surface
water systems in the United States. None of the systems sampled during 1979
to 1983 were found to contain methoxychlor in excess of the current MCL.
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) also presented data on levels of methoxychlor in drinking water
from surface water sources. Finished drinking water samples, collected from
-------
104 surface water systems throughout the United States, were analyzed for
methoxychlor. None of the finished drinking water samples collected from the
104 surface water systems contained methoxychlor in excess of the quantifi-
cation limit of 0.1 ug/1.
The 1978 Rural Water Survey (CJSEPA 1984i) also presented data on levels
of methoxychlor in drinking water obtained from surface water sources.
Finished drinking water samples were collected from 2 2 public drinking water
systems of various sizes. None of the samples from the 22 systems contained
methoxychlor in excess of the minimum quantification limit of 0.02 ug/1.
Surface Water Sources — Regional Studies
Irwin and Healy (1978) reported on levels of methoxychlor in finished
drinking water samples collected from 16 public water supply systems in
Florida that use surface water as their source. In this study, none of the
samples from the 16 systems contained methoxychlor in excess of the detection
limit. The detection limit for methoxychlor was not reported.
Finished drinking water samples collected in 1970 from both the East and
West Jefferson Water Works, Mississippi, also showed that methoxychlor was not
present (Brodtmann 1976, as cited in USEPA 1984g). The detection limit was less
than 0.1 ug/1 (the number of total samples was not reported).
In a study on the effects of forest runoff on the quality of a public
water supply of Oregon, Elliott (1979) observed an ambient concentration of
methoxychlor in 50 ug/1.
The National Academy of Sciences (NAS 1977) reviewed information on a
variety of organic chemicals and their presence in public water supplies.
No methoxychlor was detected in any samples collected during a New Orleans
drinking water survey (NAS 1977). The minimum quantification limit and the
number of samples taken were not given in this report.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
-------
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. Hie
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. The data collected cover the
period from the mid-1970's to early 1935. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United .States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background informa-
tion on the community; 2) treatment plant schematics and associated treatment
process information; and 3) water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological arid radiological parameters, inorganic
parameters, and organic parameters (including volatiles, base/neutrals, pesti-
cides and PCBs, and phenols and acids). For each parameter, the water type#
time period, concentration (mean and range), number of samples, and detection
limit are presented.
For most of the volatile organics, including methoxychlor, the available
data indicated that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
10.2.1.2 Occurrence in Water Other Than Drinking Water
One national study and several regional studies addressed levels of meth-
oxychlor in surface water other than drinking water; two sources addressed
methoxychlor in groundwater.
Ground Water Sources
Ground water wells in three California counties were sampled by Ramlit
Associates, Inc. (1983). The only information reported was a maximum value of
100 ug/1 of methoxychlor for 11 positive samples. Other ground water sources
-------
in California and New Jersey were sampled for methoxychlor, where it was detected,
although no specific data were reported (Greenberg et al. 1982 and Maddy et al.
1982, as cited in USEPA 1984g).
Surface Water Sources
The National Surface Water Monitoring Program (NiWMP 1982) presented data
on concentrations of methoxychlor in surface water samples collected through
out the United States between 1975 and 1979. During the 5-year period, 2,480
samples were collected and analyzed. None of the samples contained methoxychlor
in excess of the minimum quantification limit, although the limit was not given
in the report.
Hie National Pesticide Monitoring Network examined results of river samples
between 1975 and 1980 (Gilliom et al. 1985). Methoxychlor was not detected in
2,761 samples at 172 locations nationwide {detection limit = 0.1 ug/1).
Englande et al. (1978) presented the results of extensive chemical analyses
of six Advanced Wastewater Treatment (AWT) plant effluents. Samples of effluent
were collected from four plants located in California, and one each in the
District of Columbia and Texas. None of the 63 AWT effluent samples contained
methoxychlor in excess of the detection limit (no detection limit was reported).
Schacht (1974) presented the results of a study to determine the levels
of pesticides in the surface water of Lake Michigan and its tributaries. From
1970 to 1972, a total of 45 water samples were collected from Lake Michigan
and its tributaries. Levels of methoxychlor ranging from "not detectable" to
0.106 ug/1 were detected in the samples. The detection limit for methoxychlor
in water was reported as 0.001 ug/1.
Bradshaw et al. (1972) conducted a study to determine the seasonal vari-
ations in pesticide residues in tributary waters to a small lake in Utah. Water
samples were collected from 16 stations biweekly from March 1 through July 1,
1970 and weekly or semiweekly through February 1971. Elevated levels of
methoxychlor were detected in samples of water entering the lake in late spring
(May-June) and late fall (October-November). The maximum level of methoxychlor
-------
was 5.2 ug/1, detected in June. The number of samples collected and the number
of positive samples detected were not reported.
Ambient surface water sources from various locations in upstate New York
and the San Joaquin Valley, California, were sampled during 1982-1983 and 1978,
respectively (Estabrooks, no date; Ramlit Associates, Inc. 1983). Only one of a
total of 72 sair.pias was positive for methoxychlor at a concentration of 70 ug/1
(detection limit was not reported). This was in the San Joaquin Valley, California.
10.2.2 Ambient Air
Although the presence of methoxychlor in air might be inferred from its
presence in rainwater and snow (Strachan et al. 1980, as cited in USEPA 1984g),
no data directly addressing levels of methoxychlor in ambient air were obtained.
10.2.3 Soils/Sediments
From reviewing available soil and sediment study results, it appears that
methoxychlor is more prevalent in aquatic sediments than in soils.
Seven studies were identified that addressed methoxychlor in soils of the
United States (USEPA 1984g). Only one of these studies indicated levels above
the detection limit. This study involved collection of soils from croplands in
43 states in 1969; one sample out of 1,729 soils showed a methoxychlor level
(0.28 ppm) above the detection limit (0.01 ppm) (Wiersma et al. 1972 as cited in
USEPA 1984g). The remaining 3ix studies were con ducted between 1968 and 1979 in
as many as 35 states, in cropland and urban areas. Methoxychlor was not detected
in these studies above the detection limits (USEPA 1984g). The number of samples
for each study was not reported.
Four studies that address levels of methoxychlor in aquatic sediments were
identified. First, the National Surface Water Monitoring Program (NWMP) presented
data on levels of the insecticide in 939 sediment samples collected from 1975 to
1979 (NWMP 1982). The study results indicated that none of the samples analyzed
contained methoxychlor in excess of the detection limit. The detection limit was
not reported.
-------
A second study (Britton et al. 1983), conducted in 1976 reported levels of
pesticides in water-sediment mixtures and in bottom material samples collected
at National Stream Quality Accounting Network (NASQAN) stations. Throughout the
United States, 151 permanent stations, plus additional stations identified as
part of local programs, were sampled for pesticides, including methoxychlor.
No methoxychlor was detected in any of the 169 water-sediment mixture samples or
ir, the 153 sediment samples analyzed. The detection limits for methoxychlor in
the water sediment and the bottom material samples were not reported.
Schacht (1974) presented the results of a study to determine the levels of
pesticides in the sediments of Lake Michigan and its tributaries. During the
1970-1971 study period, a total of 50 sediment samples were collected from tribu-
taries to Lake Michigan, at stations 40-80 yards offshore of sewage treatment
plant discharges into Lake Michigan, and in open water stations 1 to 3 miles
offshore in Lake Michigan. Levels of methoxychlor were detected in Lake Michigan
sediment samples in the range of 0.13 to 6.6 ug/kg, with an average concentration
of 1.24 ugAg» Levels of methoxychlor ranging from 0.19 to 175 ug/kg, with an
average concentration of 21.1 ug/kg, were detected in samples collected in the
tributaries to Lake Michigan. The detection limit for methoxychlor in the study
was 0.01 ug/kg.
The fourth and last study identified (Warry and Chan 1981, as cited in TXNT
1986) involved the collection of sediment samples from the Niagara River. Samples
were collected and analyzed approximately every 2 weeks during the 1979-1980
study period. Methoxychlor residues were present in the samples, although the
number of samples and the levels of methoxychlor present were not indicated.
10.2.4 Food
The Food and Drug Administration conducts Total Diet Studies (also known as
Market Basket Surveys) to evaluate the occurrence of various substances, including
methoxychlor, in food consumed by adults, toddlers, and infants. According to
the results of a recently published survey for FY 81/82, methoxychlor was detected
in legume vegetables, dairy products including whole milk, and grain and cereal
products. Dairy products had the highest concentration of methoxychlor in the
adult, infant, and toddler diets (0.0002, 0.0011, and 0.0003 ppm, respectively)
(Gartrell et al. 1986a,b).
-------
The dairy products category was the largest source of methoxychlor for
the adult male, accounting for 5 5.4 percent of total intake. The legume vege-
tables category was also a significant contributor/ accounting for 32.6 percent
of total daily methoxychlor intake. The beverage category, which includes
drinking water, did not contribute any dietary intake of methoxychlor by the
adult male (Gartrell 1986a).
Only two sources, whole milk and other dairy products and dairy substitutes,
contributed to methoxychlor intake for the 6-month old infant. Again, whole
milk, other dairy products, and dairy substitutes plus grain and cereal products
were the sources of methoxychlor in the toddler diet. Grain and cereal products
accounted for the single largest source (about 46%) of total daily intake
(Gartrell et al. 1986b).
The FDA provided mean daily intakes of methoxychlor reflecting observed
levels of methoxychlor in 12 total diet studies conducted from April 1982 to
April 1985 (FDA 1986). For the 6- to 11-month old infant, daily intake of
methoxychlor was 0.016 ug/day. For the 2-year old toddler, methoxychlor intake
was 0.076 ug/day. Methoxychlor intakes for adult males and females are presented
in Table 10-1. For the adult male, methoxychlor intakes ranged between 0.099 and
0.110 ug/day. Daily methoxychlor intakes for adult females ranged between 0.075
and 0.085 ug/day. Intakes were highest for the 14- to 16-year old age group for
both sexes.
Table 10-1. Summary of FDA Total Diet Study Estimates for
Methoxychlor Intakes for Adult Males and Females.
Sex/Age Group
Intake (ug/day)
14-16 year old female
14-16 year old male
25-30 year old female
25-30 year old male
60-65 year old female
60-65 year old male
0.081
0.110
0.075
0.099
0.085
0.101
Source: FDA 1986
-------
10.3 EXPOSURE SUMMARY
Limited data are available on the extent of methoxychlor in drinking water,
food, and air, a few surveys provide useful information on occurrence. The reported
concentrations of methoxychlor are listed in the table below by source of exposure.
The estimate of drinking water exposure is presented in a range to reflect the
uncertainty of the exposure data. This 'uncertainty arises from the limited amount
of monitoring data available. The high value was taken from the highest reported
concentration of methoxychlor in drinking water in a national survey. While the
actual range of occurrence is unknown, EPA believes that the absence of a single
violation of the interim standard of 100 ug/1 strongly suggests that all systems
are currently below this level. This discussion of estimates of exposure to
methoxychlor in adults assumes that an average adult male drinks 2 liters of
water per day.
Table 10-2. Exposure Estimates for Methoxychlor
Reported Exposure
Estimated
Source
Levels (low-high)
Adult Intake
Drinking Water
0-0.09 ug/1
0-0.2 ug/day
Diet
—
0.1 ug/day
Air
No data available
Dietary exposures to methoxychlor are expressed as a single number, ihis
number is the average of several FDA market basket surveys taken from different
regions of the United States between 1982 and 1985. As such, the number is a
reasonable estimate of the average dietary intake in the United States. However,
some individuals nay receive higher levels of exposure.
The currently available information on occurrence of methoxychlor is insuf-
ficient to determine the national distribution of intake by any of the three
routes. While the table indicates that drinking water exposure can result in
higher intakes than food, this should be taken only as a relative indication of
the potential for intake by drinking water. The number of people who actually
receive drinking water exposures that are greater than dietary exposures is
unknown. EPA has no information on present levels of methoxychlor in ambient
and indoor air and, therefore, cannot determine the relative significance of
intakes by this route for the population as a whole.
-------
11. PENTACHLOROPHENOL (POP)
11.0 SUMMARY
Pentachlorophenol (PCP) is widely used as a wood preservative, herbicide,
and defoliant. Oily limited data are available on occurrence in water and
food; no information was found on air levels. Levels of PCP reported to occur
in drinking water fall into the range of 0 to 10 ug/1.
11.1 GENERAL CHARACTERISTICS
11.1.1 Physical/Chemical Properties
Pentachlorophenol (PCP) is a preservative, herbicide, preharvest defo-
liant and desiccant, slimicide, and molluscicide. Synonyms and identifiers
for PCP are "penta," Dowicide EC-7, DOW Pentachlorophenol DP-2 Antimicrobial,
penchlorol, Pentacon, Penwar, Priltox, Sinituho, and Weedone (Berg 1986).
Technical grade PCP is a buff colored crystal at 25°C composed of 85 to 90
percent PCP and approximately 10 percent 2,3,4,6-tetrachlorophenol (FDA 1981b;
Callahan et al. 1979). PCP is a white crystalline solid at 25°C (Verschueren
1983). It has a molecular weight of 266.35, a molecular formula of CgCl^OH,
and a melting point of 190-191°C (Windholz 1976). The pKa of the phenolic
hydrogen is 4.74 (Callahan et al. 1979).
Reported aqueous solubilities for PCP at 20°C included 14 mg/1 (Verschueren
1983), 18 mg/1 (FDA 1981b) and 80 mg/1 (Windholz 1976). The range of reported
aqueous solubilities for PCP is probably due to the dependence of the aqueous
solubility on pH. The aqueous solubility of PCP should increase with increasing
pH over the approximate pH range of 2.7 to 6.7 due to an increasing ratio of
the more soluble anionic form to the undissociated form. The pH dependence of
the aqueous solubility should level off above a pH of approximately 6.7, since
greater than 99 percent of the PCP at pHs above 6.7 will be in the anionic
form.
-------
The vapor pressure of the solid crystalline PCP at 20°C is 1.1 x 10~4
torr (1.4 x 10-7 atm). If it is assumed that the aqueous solubility of undis-
sociated PCP in aqueous solutions is the 14 mg/1 reported by Verschueren
(1983), the ratio of the vapor pressure to that aqueous solubility gives an
estimated Henry's constant for undissociated PCP at 20°C of 2.6 x 10-6
atm-m^/mol. The ratio of the estimated Henry's constant to the product of the
jas constant times the temperature in degrees Kelvin gives an estimated dimen-
sionless Henry's constant for undissociated PCP at 20°C of 1.1 x 10~4. Within
the pH range of most natural waters (6-9), most of the PCP present will be in
the anionic form. The aqueous solubility of the anionic form is much greater
than the undissociated form so Henry's constant over the pH range of most
natural waters should be much smaller than the 2.6 x 10~s atm*m3/mol for
undissociated PCP.
Mabey et al. (1981) estimated a KqC value for undissociated PCP of 5.3 x
104 which is much larger than the experimental KQC value of 900 listed in
Kenaga and Goring (1978). Although the pH at which the experimental value
of 900 was determined is not listed, it was probably substantially higher than
the pKfl for PCP. Because the anionic form of PCP is adsorbed to a lesser
extent than the undissociated form, that would explain the large discrepancy
between the two values.
11.1.2 Use
On October 18, 1978, EPA published a notice of rebuttable presumption
against registration of pesticide products containing PCP based on the risk
criteria of teratogenicity and fetotoxicity. Comments were received on this
notice and addressed in Position Document 2/3, which was made available to the
public on February 19, 1981. In Position Document 2/3, the agency indicated
that the use of PCP also posed the risk of oncogenicity because of the presence
of the contaminants, hexachlorodibenzo-p-dioxin and hexachlorobenzene. The
agency proposed changes in registrations for PCP. Position Document 4, released
in July 1984, outlined their final regulatory position on PCP.
Ihe EPA estimate of the total amount of PCP used in 1986 was 45,000,000
pounds (Kuch 1986). The use of PCP as a wood preservative applied to products
-------
such as railroad ties, lumber, timber, plywood, pilings, posts, crossarms,
crossties, and poles accounted for 80 percent (or 40 million pounds) of the
estimated 1980 domestic annual production (FDA 1981b).
PCP's use on insulation board and in cooling towers, and as a slimicide
in pulp and paper mills, accounted for another 11 percent and 6 percent,
respectively, of annual domestic production (FDA 1981b).
The remaining uses of PCP include its applications as a preservative to
such diverse products as burlap, leather, paints, and petroleum; its use as a
termiticide for structural pest control; its use as an herbicide around rail-
road tracks, ballast railyards, farms and industrial areas, parking lots,
fence rows, driveways, highways, and walkways; its use as a mossicide on lawns
in western Oregon and Washington; its use as a preharvest defoliant on alfal-
fa; and its uses as a preharvest desiccant on beans (as a seed treatment),
clover, lespedeza, and vetch (FDA 1981b; USDA 1980b).
11.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of PCP is divided into the
following subsections: 11.1.3.1 Volatilization; 11.1.3.2 Sorption and
Leaching Potential; 11.1.3.3 Abiotic Transformations; 11.1.3.4 Biodegradationj
and 11.1.3.5 Summary. The discussion will emphasize the environmental fate of PCP
in soil and water.
11.1.3.1 Volatilization
The dimensionless Henry's constant for undissociated PCP cannot be used
to estimate volatilization half-lives of PCP in natural waters. The reason is
that within the pH range of 6 to 9 normally encountered in natural waters,
most of the PCP will be in the anionic form which is predicted to have a much
lower Henry's constant than undissociated PCP. Therefore, estimated volatili-
zation half-lives for undissociated PCP should be much lower than for the
primarily dissociated, anionic PCP in natural waters.
-------
Volatilization half-lives for undissociated PCP in streams on rivers were
estimated by SAIC using equations A-1 and A-3 through A-6. The estimated
volatilization half-lives for undissociated PCP in streams or rivers range from
13 days for a turbulent stream or river 1 m deep to 3.6 years in a stagnant
stream or river 10 m deep (see Appendix A). Volatilization half-lives for
undissociated PCP in turbulent lakes were estimated by SAIC using equations
A-1, A-7, and A-8. The estimated volatilization half-lives for PCP in turbulent
lakes range from 34 days in a lake 1 tti deep to 340 days in a lake 10 m deep.
Since the estimated volatilization half-lives for undissociated PCP in
streams or rivers and turbulent lakes are comparable to or longer than many of
the half-lives reported for photolysis and biodegradation and since volatili-
zation half-lives for the primarily dissociated PCP in natural waters are
expected to be much longer than those for undissociated PCP, volatilization is
probably not an important removal process for PCP in natural waters. That
conclusion is supported by experimental work. Moss et al. (1983) and
Pignatello et al. (1983), both as cited in USEPA (1986), reported negligible
losses of PCP from surface waters due to volatilization.
The rate of volatilization of PCP from groundwater to the soil column
above is projected to be much lower than the volatilization rate from surface
waters to the atmosphere. This is due primarily to the laminar (non-turbulent)
nature of ground water flow. Transport from ground water by volatilization may
be further reduced by a build-up of chemical in pore air at the pore air/ground
water interface and an associated decrease in the concentration across the
interface.
11.1.3.2 Sorption and Leaching Potential
It is difficult to predict the sorption and leaching potential of PCP due
to a lack of data on the dependence of Koc on pH. Substitution of the esti-
mated Kq,, value of 5.3 x 10^ for undissociated PCP into equation A-11 gives
estimated sediment or soil/water equilibrium partition coefficients (Kg/W) for
undissociated PCP ranging from 530 to A.2 x 1 03 for sediments or soils with
organic fractions ranging from 0.01 to 0.08, respectively. That suggests that
a substantial proportion of the total undissociated PCP present in an aquatic
-------
system could be adsorbed to suspended and exposed bottom sediment. However,
at pHs exceeding 6, almost all of the PCP present in an aquatic system would
be in the anionic form, which does not appear to adsorb to sediments or soils
as strongly as undissociated PCP (Callahan et al. 1979). Therefore, transport
by adsorption to suspended sediment and removal by adsorption to bottom sedi-
ment may be important processes for PCP in acidic waters with a pH less than
5.7 where a substantial proportion of the total PCP present is undissociated.
However, the processes are probably much less important for PCP in the great
majority of natural waters with a pH between 6 and 9 where most of the total
PCP present is in the anionic form.
Substitution of the estimated Koc value for undissociated PCP of 5.3 x 10^
into equation A-1 3 (Appendix A) gives an estimated soil TIC Rf value of <0.1
for undissociated PCP adsorbed to a soil with an organic fraction of 0.014, a
pore fraction of 0.5, and a soil density of 2.5 g/cm3. Therefore, based on the
five mobility classes defined by Helling and Turner (1968) and cited by Hamaker
(1975) for a soil with the same properties (Appendix A), undissociated PCP would
be classified as immobile and resistant to leaching from soils. However, the
decrease in adsorption of PCP by soils with increasing pH suggests that the
anionic form of PCP is adsorbed to a lesser extent by soils than the undissociated
form and is more susceptible to leaching (Callahan et al. 1979). Therefore,
under acidic conditions in which a substantial proportion of the total PCP is
undissociated, PCP may be resistant to leaching. However, under neutral or
alkaline conditions where most of the total PCP is dissociated, PCP could be
susceptible to leaching (Callahan et al. 1979; USEPA 1986).
11.1.3.3 Abiotic Transformations
PCP does not have any groups that are readily susceptible to hydrolysis
and it is therefore unlikely to undergo significant rates of hydrolysis in the
environment (Lyman et al. 1982; Callahan et al. 1979). Mabey et al. (1981)
estimated the second-order rate constants for the oxidation of PCP in aqueous
solution by singlet oxygen (r02) and peroxy radicals (R02*) to be <7 x 1 03
m-1hr-1 and 1 x 105 m"1hr_1. Substituting those rate constants and assumed
concentrations for ^2 and R(>2* in sunlit surface waters of 10"1 and 10~9M,
respectively, into equation A-15, and the resulting pseudo first- order rate
-------
constant into equation A-1, gives an estimated half-life for the oxidation
of PCP by '02 and R02* in sunlit surface waters of 290 days. Therefore, the
oxidation of PCP by ' 02 and R02* in sunlit surface waters does not appear to
be an important removal process for PCP in natural waters.
PCP strongly absorbs light with wavelengths above the approximate solar
radiation cutoff of 290 nm at the earth's surface and has been reported to
undergo rapid rates of direct photolysis in water. Reported photolytic half-
lives for PCP in extremely shallow waters at neutral or alkaline pH range from
less than 1 hour to 4.6 days (Callahan et al. 1979; USEPA 1986). Therefore,
even though photolytic rates decrease rapidly with increasing depth, photolysis
appears to be a major removal process for PCP in shallow waters and at the
surface of deeper waters.
11.1.3.4 Biodegradation
PCP appears to be susceptible to biodegradation in river water, but only
if the microorganisms have become acclimated. Banerjee et al. (1984) and
Baker et al. (1980), both as cited in USEPA (1986), reported that no signif-
icant rates of PCP biodegradation occurred in previously uncontaminated river
water. However, Banerjee et al. (1984) and Pignatello et al. (1983), both as
cited in USEPA (1986), reported that significant rates of biodegradation occurred
in river waters after a 2 to 3 week lag period and eventually became the dominant
removal mechanism for PCP in the waters.
Crosby et al. (1981), as cited in USEPA 1986) reported that degradation
half-lives for PCP in soils typically range from 2 to 4 weeks. Therefore, PCP
would be classified as nonpersistent in soils (e.g., ti/2 <2° days) by Rao and
Davidson) (1980). Degradation half-lives are lowest for soils with high organic
and/or moisture content, and with temperatures conducive to microbiological
activity. Ide et al. (1972), as cited in USEPA 1986) reported that biodegra-
dation was the primary removal process for PCP applied to a rice paddy, and
that PCP was almost completely decomposed within several weeks after application.
-------
11«1.3*5 Summary
Based upon the above discussion and the literature reviews by Callahan
et al. (1979) and USEPA (1986), the following conclusions can be made concerning
the most likely behavior of PCP in soil and water:
o PCP is unlikely to undergo significant ratss of volatilization from
surface or groundwater under any conditions normally found in the
environment.
o Transport by adsorption to suspended sediments and removal by adsorp-
tion to bottom sediments may be significant processes for PCP in some
acidic waters (pH <6) but are unlikely to be important in natural
waters within the normal pH range of 6-9.
o pcp is strongly adsorbed to acidic soils but is generally weakly
adsorbed to neutral and alkaline soils. Therefore, PCP may be sus-
ceptible to leaching from neutral or alkaline soils, but generally not
from acidic soils.
o Direct photolysis is an important removal process for PCP in shallow
waters and at the surface of deeper waters.
o PCP appears to be susceptible to biodegradation in river waters but
only after a 2 to 3 week lag period. After the lag period, biode-
gradation may become the dominant removal process in some natural
waters.
o Degradation half-lives for PCP in soil typically range from 2 to 4
weeks. Therefore, PCP can be classified as relatively non-persistent
in soil. The major removal process for PCP in soils appears to be
biodegradation.
o Due to its immobility in acidic soils and its low persistence in soil,
PCP adsorbed to acidic soils is unlikely to contaminate drinking water
supplies. However, under neutral or alkaline conditions, the mobility
of PCP may be great enough to contaminate some drinking water supplies
even though its persistence in soils is relatively low.
11.2 OCCURRENCE IN THE ENVIRONMENT
11.2.1 Water
This section presents the available data from monitoring studies and
surveys to determine the extent of occurrence of PCP in public drinking water
supplies and water other than drinking water.
-------
11.2.1.1 Occurrence in Drinking Water
Studies at both the national and regional levels have addressed con
centrations of PCP in finished drinking water. The results of two national
studies and two regional studies are discussed below. Where possible, ground
water sources and surface water sources are discussed separately.
Ground water Sources National Study
The National Screening Program for Organics in DrinJcing Water (NSP)
(Boland 1981) was conducted from June 1977 to March 1981. Finished drinking
water samples, collected from 12 ground water systems of varying sizes, were
analyzed for PCP. None of the finished drinking water samples contained PCP
in excess of the quantification limit of 1.0 ug/1.
Surface Water Sources — National Study
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) presented data on levels of PCP in drinking water withdrawn from
105 surface water systems located throughout the United States. Of the 105
water supply systems sampled, finished drinking water from only two systems had
concentrations of PCP in excess of the quantification limit of 1.0 ug/1. These
systems, located in Pennsylvania and Michigan, had concentrations of 12 ug/1
and 1.3 ug/1 of PCP, respectively.
Surface Water Sources — Regional Study
Buhler et al. (1973, as cited in Cirelli 1978) identified a PCP
concentration of 0.6 ug/1 in a sample of finished drinking water collected
from a water supply that obtains its raw water from a river in Oregon. The
number of samples analyzed, the number of positive samples identified, and the
detection limit were not reported.
Unspecified Sources — National Study
Cirelli (1978) presented unpublished data from the EPA Drinking Water
Program. The water supply systems of 108 cities were studied. Low levels of
-------
PCP were detected in samples of drinking water from 86 of the 108 systems
sampled. The positive samples had a mean concentration of 0.07 ug/1 of PCP;
the median concentration was 0.051 ug/1 of PCP.
Unspecified Sources - Regional Study
Levins et al. (1979b) sampled drinking water in the Muddy Creek drainage
basin in Cincinnati, Ohio. None of the four samples of finished drinking
water contained PCP in excess of the detection limit. The detection limit was
not reported.
11.2.1.2 Occurrence in Water Other Than Drinking Water
Several reports at both the national and regional levels have discussed
occurrences of PCP in water other than drinking water.
Surface Water Sources — National Study
Although PCP is not routinely monitored by the National Water Monitoring
Program (NWMP), PCP was identified in 15 whole water samples collected from
five sites in the California Aqueduct between 1974 and 1976. The range of
positive values was 0.01 to 16 ug/1 PCP, Ihe total number of samples analyzed
and the detection limit were not reported (Cirelli 1978).
Surface Water Sources — Regional Studies
Buhler et al. (1973, as cited in Cirelli 1978) reported on levels of PCP
in samples of voter collected from the Willamette River in Oregon. Levels of
PCP in the river water samples ranged from 0.1 to 0.7 ug/1. The number of
samples analyzed, the number of positive samples, and the detection limit were
not reported.
Benvenue et al. (1972a) collected a 24-hour composite effluent sample
from a sewage treatment plant on the island of Oahu, Hawaii/ to determine the
extent of organochlorine pesticide contamination. Analysis of the sample
showed a concentration of PCP of 2.6 ug/1.
-------
To assemble a database which wsuld reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. The
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. Hie data collected cover the
period from the mid-1970's to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background infor-
mation on the community; 2) treatment plant schematics and associated treatment
process information; and 3} water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological and radiological parameters, inorganic
parameters, and organic parameters (including volatiles, base/neutrals,
pesticides and PCBs, and phenols and acids). For each parameter, the water
type, time period, concentration (mean, range), number of samples, and detection
limit are presented.
For most of the volatile organics, including pentachlorophenol, the
available data indicated that there were very low levels of these contaminants
in the raw, treated, or tap water. Most of the values found were "not detected"
or near the detection limit (Canadian Public Health Association 1986).
Benvenue et al. (1972b) found levels of PCP of 0.01 ug/1 in lake water in
Hawaii.
Unspecified Sources — Regional Studies
Buhler et al. (1973, as cited in Cirelli 1978) analyzed levels of PCP in
effluent samples from sewage treatment plants in three cities in Oregon.
Concentrations of PCP in the positive samples ranged from 1.0 to 4.0 ug/1.
The number of samples analyzed, the number of positive samples identified, and
the detection limit were not reported.
-------
Levins et al. (1979b) sampled influent to a sewage treatment plant in Ohio
to determine the amount of PCP entering the system. One positive sample was
identified out of the six samples analyzed. The concentration of this sample
was 26 ug/1 of PCP. No detection limit was reported.
11.2.2 Ambient Air
The detection of pentachlorophenol in rain water and in snow (Paasivinta
et al. 1985; Benvenue et al. 1972b, as cited in Cirelli 1978) offers indirect
evidence of its presence in ambient air. Cautreels et al. (1977) reported
ambient air concentrations of 0.0057 to 0.0078 ug/m^ in Antwerp, Belgium, and
0.00025 to 0.00093 ug/m3 in Chacaltaya, Bolivia.
11.2.3 Soil/Sediments
Minimal information was available on the occurrence of PCP in soils and
sediments. One study involved the collection of sediment samples from nine
coastal sites in the vicinity of Portland, Maine. Reported PCP levels ranged
between 0.01 and 2.4 ppb (Ray et al. 1983, as cited in TXNT 1986).
11.2.4 Food
11.2.4.1 Dietary Intake
The Food and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the occurrence of various substances,
including pentachlorophenol, in food consumed by adults, toddlers, and infants.
According to the results of a recently published survey for FY 81/82, penta-
chlorophenol ws detected in several categories, including oils and fats;
grains and cereal products; meat, fish, and poultry; sugar and adjuncts; and
other dairy substitutes. Oils and fats had the highest concentrations of
pentachlorophenol in adult, infant, and toddler diets (0.0072, 0.0034, and
0.0073 ppm, respectively) (Gartrell et al. 1986a,b).
Several estimates of dietary intake of pentachlorophenol have been
reported in recent years. The largest source of pentachlorophenol for the
-------
adult was the grain and cereal products category, which accounted for 56.4
percent of the intake. Hie meat, fish, and poultry category also contributed
significantly (26.9%). ihe beverage category, which includes drinking water,
did not contribute any pentachlorophenol diet (Gartrell et al. 1986a).
The largest sources of pentachlorophenol intake for the 6-month-old
infant were the grain and cereal products and the oils and fats categories,
which contributed 47.8 and 36.5 percent, respectively. Meat, fish, and poultry
and the grain and cereal products categories were the largest sources of
pentachlorophenol intake for the 2-year old toddler (41.9 and 41.4 percent,
respectively) (Gartrell et al. 1986b).
The FDA provided mean daily intakes or" pentachlorophenol reflecting
detections of pentachlorophenol in 12 total diet studies conducted from April
1982 to April 1985 (FDA 1986). For the 6- to 11-month old infant, daily
intake of pentachlorophenol was <0.5 ug/day. For the 2-year old toddler,
pentachlorophenol intake was <0.7 ug/day.
Pentachlorophenol intakes for adult males and females are presented in
Table 11-1. For the adult male, pentachlorophenol intakes ranged between
<1.4 and <1*6 ug/day. Daily pentachlorophenol intakes for adult females
ranged between <1.0 and <1.1 ug/day. Intakes were highest for the 25- to
30-year old age groups for both sexes.
Table 11-1. Summary of FDA Total Diet Study Estimates for
Pentachlorophenol Intakes for Adult Males and Females
Sex/Age Group
Intake (ug/day)
25-30 year old female
2 5-30 year old female
60-65 year old female
60-65 year old male
14-16 year old female
14-16 year old male
<1.0
<1.5
<1.1
<1.6
<1.0
<1.4
•Intakes are shown as "<" because they are based on findings possibly affected
by sporadic reagent contamination, thus leading to possible high results.
Source: FDA 1986.
-------
11.3 EXPOSURE SUMMARY
The data currently available on the occurrence of pentachlorophenol in
drinking water and in food are very limited. Data on its presence in air are
almost entirely lacking. 3he continued use of pentachlorophenol as a wood
preservative, herbicide, and defoliant, even after EPA's issuance of a notice
of rebuttable presumption against the registration of pesticide products con-
taining PCP, provides a continued low-intensity threat of contamination from
each route. Hie low level presence of pentachlorophenol in drinking water is
probably due in part to PCP's tendency to become bound to soil and sediment,
especially under acidic conditions. Its tendency to bind as well as its
susceptibility to biodegrade leads to pentachlorophenol's low persistence and
immobility in soils and to the probability that PCP will not be a significant
contaminant of drinking water supplies. The potential of PCP contamination of
surface water supplies is further reduce^' by is rapid removal through photolytic
processes.
The presence of pentachlorophenol in foods and the mean daily intake of
1.6 ug/day and 1.1 ug/day by 25-30 year old males and females, respectively,
as determined by the FDA during surveys conducted between 1982 and 1985,
indicates that residues persist. The only data currently available on the
presence of pentachlorophenol in air were derived from studies conducted in
Antwerp, Belgium and Chacaltaya, Bolivia.
The available data on the occurrence of PCP in air, food, and water is
insufficient to characterize the range and distribution of intake by any of
these routes. In general, the limited occurrence data suggests that PCP will
occur in a small number of surface water systems at concentrations of a few ug/1.
PCP tends to bind to many soils, therefore levels in ground water are expected
to be lower than surface water; however, the available monitoring data are
insufficient to confirm this. Because PCP is used widely as a wood preservative,
contamination of drinking water supplies can occur in all areas of the United
States. Dietary exposure to PCP is expected to be common. !he available data
suggest that for the average person the intakes will be less than 1 ug/day.
While EPA has no data on the levels of PCP in air in the United States, based
on PCP's physical properties, little exposure is expected from this source.
-------
12. 2,4,5-TP
12.0 SUMMARY
2, 4,5-TP is an herbicide that was widely used in the 1960's and 1970's.
These uses resulted in low-level contamination of a few ground water supplies.
Because of the cancellation of 2,4,5-TP's us«s in the early 1980's, present
and future occurrence levels of the chemical are considered to be unlikely.
12.1 GENERAL CHARACTERISTICS
12.1.1 Physical/Chemical Properties
2,4,5-TP [2-(2,4,5-trichlorophenoxy)propionic acidj is a selective,
pre- and post-emergent herbicide that was introduced in 1952; it is available
as the amine as well as sodium salts and various esters (NAS 1977). Synonyms
and identifiers for 2,4,5-TP include silvex, Aqua-Vex, Ded-Weed, fenoprop,
Fruitone T, Kuron, Kurosal, Silvi-Rhap, and Weed-B-Gon (Berg 1986).
2,4,5-TP is a crystalline substance at 25°C (Windholz 1976). It has a
molecular weight of 269.53, a molecular formula of C9H7C3,303, and a melting
point of 181.6®C (Windholz 1976). 2,4,5-TP and technical grade 2,4,5-TP have
reported aqueous solubilities at 25®C of 140 mg/1 and 180 mg/1, respectively
(FDA 1981b; NAS 1977; Verschueren 1983). No vapor pressure for 2,4,5-TP could
be found in the literature. A KqC value of 2.6 x 103 has been reported for
2,4,5-TP (Kenaga and Goring 1978). No pKa value could be found in the
literature for 2,4,5-TP.
The aqueous solubilities of most of the salts of 2,4,5-TP are reported to
be greater than that of the acid, but no values are given (FDA 1981b). No
vapor pressure or aqueous solubilities for the various esters of 2,4,5-TP
could be found in the literature. However, by analogy to the structurally
similar 2,4,5-trichlorophenoxyacetic acid (2,4,5-T), the various esters of
2,4,5-T are expected to have substantially greater vapor pressures than the
acid, but lower aqueous solubilities (SAIC 1981a).
-------
12.I.J Use
In 1977, domestic production of 2,4,5-TP was estimated to range between
3.7 million and 4.1 million pounds (NAS 1977). Total annual domestic use of
the herbicide in 1979 was estimated at 3 million pounds (FDA 1981b; USEPA
1979a). The EPA Administrator suggested a range of use in Position Document
1/2/3 of between ?.8 million and 3.3 million pounds (USEPA 1979a).
In March, 1979, EPA issued an immediate suspension of the sale, use, or
distribution of 2,4,5-TP on forest land, rights-of-way, pastureland, for homes
and gardens, on commercial and ornamental turf, and on aquatic weeds and ditch
banks (USEPA 1979b). The use of 2,4,5-TP on rangeland, rice, sugarcane, orchards,
and other noncrop areas was allowed to continue. However, on March 14, 1980,
EPA convened hearings to gather additional data on the risks and benefits of
using 2,4,5-TP for these purposes. As of 1984, the remaining uses of 2,4,5-TP
were cancelled and production of the pesticide was discontinued (Berg 1986).
12.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of 2,4,5-TP is divided into the
following subsections: 12.1.3.1 Volatilization; 12.1*3.2 Sorption and
Leaching Potential; 12.1.3.3 Abiotic Transformations; 12.1.3.4 Biodegradation
and Persistence in Soil and Water; and 12.1.3.5 Summary. Ihe discussion will
emphasize the environmental fate of 2,4,5-TP in soil and water.
12.1.3.1 Volatilization
No vapor pressure could be found for 2,4,5-TP in the literature. However,
if it is assumed (based on its higher molecular weight and degree of chlorina-
tion) that the vapor pressure of 2,4,5-TP is less than the 6 x 10" 7 torr
reported for 2,4-D, the estimated Henry's constant for 2,4,5-TP (based on
the ratio of the vapor pressure to aqueous solubility) would be less than
1.3 x 10-7 atm*m3/mol. Lyman et al. (1982) indicate that compounds such as
2,4,5-TP with a Henry's constant less than 3 x 10-7 atm»m3/mol are less
volatile than water. Therefore, the concentration of 2,4,5-TP in water is not
expected to decrease due to volatilization. Because the vapor pressures of
-------
the various esters of 2,4,5-TP are expected to be greater and" the aqueous solu-
bilities lower than that of the acid, volatilization rates for the various esters
of 2,4,5-TP are expected to be substantially greater than those for the acid.
Substitution of the reported KQC value of 2.6 x 103 for 2,4,5-TP into
equation A-9 along with the compound's vapor pressure (<6 x 10~7 torr at 25°C)
and mean aqueous solubility 1160 mg/1 at 25°C) gives an estimated volatilization
half-life of >30 years for 2,4,5-TP on the soil surface. Volatilization rates
fcr 2,4,5-TP beneath the soil surface are expected to be even slower, decreasing
rapidly with increasing depth (Lyman et al. 1982).
12.1.3.2 Sorption and Leaching Potential
2,4,5-TP is a carboxylic acid herbicide. As an ionic chemical, the
behavior of 2,4,5-TP in soil is more variable than that of nonionic organic
compounds. Nonionics interact predictably with soil organic matter surfaces
and are not influenced greatly by the pH of the soil. Carboxylic acids in
ionic form, however, may be repelled from negatively charged soil surfaces at
high pH. As pH decreases, the amount of both nonionized organic matter and
nonionized compound increases. Under these conditions, 2,4,5,-TP would behave
as a neutral molecule, and thus adsorption would increase. This is reflected
in the work of Kenaga and Goring (1978) who reported that at low pH, 2,4,5-TP
is strongly adsorbed by hydrated iron and aluminum oxides.
A soil/water partition coefficient (KqC) of 2,600 is reported for
2,4,5-TP (Kenaga and Goring 1978), but the pH is not reported. Based on the
soil/water partition coefficient of 2,600, 2,4,5-TP would be expected to have
low mobility in soil. However, FDA (1981b) notes that the mobility of 2,4,5-TP
varies widely with formulation and soil type. As discussed above, the KqC
would be expected to decrease with increasing pH.
12.1.3.3 Abiotic Transformations
By analogy to the esters of 2,4,5-T, the esters of 2,4,5-TP are expected
to be susceptible to hydrolysis to 2,4,5-TP. However, it is unlikely that
2,4,5-TP itself will undergo significant rates of hydrolysis in the environment.
-------
It does not have any functional groups that are readily susceptible to hydrolysis
(Lyman et al. 1982). No information could be found in the literature concerning
the direct photolysis or photo-oxidation of 2,4,5-TP.
12.1.3.4 Biodegradation and Persistence in Soil and Water
Based on similarity of structure to 2,4,5-T, the half-life of 2,4,5-TP in
soil is estimated to be 3 to 4 months. The primary transformation product is
2,4,5-trichlorophenol. The half-life of residues of 2,4,5-TP in range grass is
reported to be 1 week (FDA 1981b).
12.1.3.5 Summary
Based on the above discussion, the following tentative conclusions can be
made concerning the most likely behavior of 2,4,5-TP, it salts, and esters in
soil and water:
o Based on theoretical considerations, volatilization is not expected
to be an important removal process for the acid or various salts of
2,4,5-TP. However, by analogy to the esters of 2,4,5-T, volatilization
could possibly be a significant removal process for the esters of
2,4,5-TP.
o Based on a reported KqC value of 2,600 for the acid, the mobility of
the acid in soil is predicted to be low. However, the mobility of
2,4,5-TP varies widely with soil type and pH, generally increasing
with increasing pH. No information on the mobility of the esters was
found in the literature.
o By analogy to the esters of 2,4,5-T, the esters of 2,4,5-TP are expected
to be at least moderately susceptible to hydrolysis to 2,4,5-TP. Based
on theoretical considerations, 2,4,5-TP acid is not expected to undergo
significant rates of hydrolysis in the environment.
o By analogy to 2,4,5-T, the half-life of 2,4,5-TP in soil is estimated
to be 3-4 months.
-------
12.2 OCCURRENCE IN THE ENVIRONMENT
12.2.1 Water
This section presents the available data from monitoring studies and
surveys to determine the extent of occurrence of 2,4,5-TP in public drinking
water supplies and water other than drinking water.
12.2.1.1 Occurrence in Drinking Water
Studies at both the national and regional levels have addressed concen-
trations of 2,4,5-TP in drinking water. The results of four national surveys
and six regional studies are discussed in this section. Reported levels of
2,4,5-TP in drinking water from ground water sources and surface water sources
have been discussed separately.
Ground Water Sources -- National Studies
The Federal Reporting Data System (FRDS 1984) provides information on
public water supplies found to be in violation of current maximum contaminant
levels (MCLs). Data are not available on the number of ground water systems in
the United States that have monitoring requirements for 2,4,5-TP. However, no
violations of the current MCL of 10 ug/1 were reported for 2,4,5-TP during the
years 1979-1983.
A detailed survey of the contaminants in the water supplies of 10 cities
was conducted as part of the National Organics Reconnaissance Survey (NORS)
(USEPA 1975b). TWo of the systems sampled, located in Florida and Arizona,
used ground water as their water source. . Water samples taken from these water
supplies had concentrations of 2,4,5-TP below the minimum quantifiable concen-
tration, although the minimum quantifiable concentration was not reported in
the survey.
The National Screening Program (NSP) for Organics in Drinking Water (Boland
1981) was conducted from June 1977 to March 1981. None of the finished drinking
water samples from the 12 systems were found to contain 2,4,5-TP in excess of
the quantification limit of 2.0 ug/1.
-------
The 1978 Rural Water Survey (RWS) (USEPA 1984i) involved the collection
arid analysis for 2,4,5-T of drinking water samples from 267 households located
in rural areas throughout the United States. A total of 71 public ground water
systems were covered by the survey. None of the samples from the 71 public
groundwater systems contained residues of 2,4,5-TP in excess of the minimum
quantification limit of 0.1 ug/1.
Ground Water Sources — Regional Study
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. Finished drinking
water samples were collected from 100 water supplies that use the 5 aquifers in
Florida. Water samples from 4 of the 100 water supplies contained residues of
2,4,5-TP ranging from 0.04 to 0.30 ug/1.
Another, study by the Florida Department of Environmental Resources and
the U.S. Geological Survey (Holden 1986) examined drinking water supplies
tapping the Floridian aquifer, which serves over 3 million people. The only
information available on this 1984 study was that a mean concentration of
0.02 ug/1 for 2,4,5-TP was found from 96 locations sampled.
A study known as the Mississippi Pesticide Hazard Assessment Project
(MSU 1984) examined drinking water from shallow wells of 10 counties in
northwest Mississippi. During 1983-1984, 143 samples were analyzed from the
10 counties and none were found positive for 2,4,5-TP. Some detection limits
varied; the highest was 0.015 ug/1, and the lowest was 0.03 ug/1 (the number
of wells sampled was not reported).
Similarly, 88 samples taken in 1984 from drinking water wells in Long
Island, New York showed no positive concentrations found for 2,4,5-TP (Holden
1986). The number of wells and the detection limit were not reported.
Low levels of 2,4,5-TP were found in drinking water wells in Idaho (Idaho
Department of Health and Welfare 1984). Monitoring for pesticides in drinking
water wells is not routinely conducted; the sampling performed was in response
to a particular incident, not for any comprehensive monitoring program.
-------
Eleven of 107 samples were positive with a mean concentration of 0.007 ug/1
(range = not detected to 0.62 ug/1). No other information was available.
Surface Water Sources — National Studies
Information was obtained from the Federal Reporting Data System (FRDS
1984) on violations of the current MCL of 10 ug/'l for 2,4,5-TP. The data were
obtained for the years 1979-1983, and include information for all surface
water systems in the United States. Monitoring results identified one surface
water system in North Carolina that violated the MCL of 10 ug/1 for 2,4,5-TP
for a period of 3 months in 1979.
In the NORS survey of 10 cities (USEPA 1975b)# the drinking water systems
of 8 cities with surface waters affected by different types of pollution were
sampled. Of the eight systems sampled, samples of water from one system whose
raw water supply was contaminated with industrial discharges were found to
contain 0.02 ug/1 2,4,5-TP. The minimum quantifiable concentration was not
reported.
The National Screening Program for Organics in Drinking Water (NSP)
(Boland 1981) also contained information on 2,4,5-TP in drinking water from
surface water systems. Finished drinking water samples were collected from
105 surface water supplies. None were found to contain residues of 2,4,5-TP
in excess of the quantification limit of 2.0 ug/1.
The 1978 Rural Water Survey (RWS) (USEPA 1984i) also presented information
on finished drinking water samples obtained from surface water sources.
Finished drinking water samples were collected from 21 surface water systems
of varying sizes. None of the samples from these systems was found to
contain residues of 2,4,5-TP in excess of the minimum quantification limit of
0.1 ug/1.
Surface Water Sources — Regional Studies
During the 1976 USGS survey of Florida public drinking water supplies,
samples of finished drinking water from 4 of the 16 surface water systems
-------
sampled were found to contain 2,4,5-TP at levels ranging from 0.03 to
0.08 ug/1 (Irwin and Healy 1978), -well below the MCL of 10 ug/1.
In a study on the effects of forest runoff on the quality of a public
water supply in Oregon, Elliott (1979) observed an ambient concentration of
2,4,5-TP of 5 ug/1.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1985 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. ttie
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. The data collected cover the
period from the mid-1970's to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background
information on the community; 2) treatment plant schematics and associated
treatment process information; and 3) water quality data. Water sample types
included raw water (treatment plant intake), distribution water (treated
water), and tap water. Water quality data collected included general param-
eters (e.g., alkalinity, turbidity), microbiological and radiological param-
eters, inorganic parameters, and organic parameters (including volatiles,
base/neutrals, pesticides and PCBs, and phenols and acids). For each param-
eter, the water type, time period, concentration (mean, range), number of
samples, and detection limit are presented.
For most of the volatile organics, including 2,4,5-TP, the available data
indicated that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
-------
12.2.1.2 Occurrence in Water Other than Drinking Water
Seven reports addressed levels of 2,4,5-TP in surface water other than
drinking water; only one report addressed occurrence of 2,4,5-TP in ground
water.
Ground water wells were sampled "ir. 1984 frcra 25 counties in California
during the California State Board's Toxic Special Project (Cohen and Bowes
1984). Three positive samples were reported for 2,4,5-TP, with a maximum value
of 1.0 ug/1. No other information was reported.
Monitoring data for 2,4,5-TP are available from the National Surface Water
Monitoring Program (NWMP 1982) for the years 1975-1978. Of the 1,556 samples
taken during the 4-year period, only one sample (0.1%) contained detectable
residues of 2,4,5-TP. The positive sample was collected during the summer of
1978 and contained 0.49 ug/1 of 2,4,5-TP. No detection limits were reported.
Monitoring data from the National Pesticides Monitoring Network (Gilliom
et al. 1985) indicated that between 1975-1980, 1,768 river samples were collected
from 167 sites nationwide and analyzed for 2,4,5-TP. Only two samples were
found positive, with a detection limit of 0.5 ug/1. No other information was
available.
River and stream samples from upstate New York were analyzed for 2,4,5-TP
by the New York State Department of Environmental Conservation (Estabrooks
no date) during 1982-1983. One sample of 36 was positive at a concentration
of 1.6 ug/1 (detection limit =0.12 ug/1).
Truhlar and Reed (1976) reported on water samples taken from four streams
in Pennsylvania and analyzed for chlorinated hydrocarbons. The streams drained
four different types of land-use areas. Out of the 25 samples collected from
April 1970 to February 1971, 2,4,5-TP was detected in only one sample at a level
of 0.02 ug/1. The positive sample was taken from the stream that drained the
residential area. No detection limit was reported for the study.
-------
The National Research Council of the National Academy of Sciences reported
that residues of 2,4,5-TP ranging from 0.01 to 0.21 ug/1 were detected in
samples of surface water from 15 western states. No further information was
supplied regarding the sampling locations or the analytical methods and detection
limits (NAS 1977).
Barks (1978) presented the results of a USGS water quality study conducted
from April 1973 to July 1974 in the OzarJc National Scenic Riverways, Missouri.
During the study, 20 surface water samples were collected from three sites on
the Current River and one site on Jacks Fork to be tested for pesticide content.
Analysis of unfiltered samples found no residues of 2,4,5-TP in excess of the
detection limit (the detection limit was not reported).
Englande et al. (1978) presented the results of extensive chemical analysis
of four Advanced Wastewater Treatment (AWT) plant effluents# The four plants
were located in Lake Tahoe, California; Blue Plains, District of Columbia; and
Pomona and Orange County, California. Although none of the 54 awt effluent
samples contained residues of 2,4,5-TP in excess of the drinking water standard
of 10 ug/1, mean concentrations of less than 0.016 ug/L and 0.083 ug/1 2,4,5-TP
were identified in System 1 and System 2 effluent samples, respectively, of the
Pomona AWT. 2,4,5-TP was not detected in the remaining systems. The number of
positive samples and the detection limit were not reported.
12.2.2 Ambient Air
No data were obtained on 2,4,5-TP ambient air concentrations. Because of
the low vapor pressure of 2,4,5-TP and its esters, air concentrations are
expected to be low.
12.2.3 Soil/Sediments
Two studies were identified that examined the occurrence of 2,4,5-TP in
sediments. No studies were identified on the 2,4,5-TP occurrence in soil.
The National Surface Water Monitoring Program (NWMP 1982) presented data
on levels of 2,4,5-TP in samples of sediment. From 1975 to 1978, 541 sediment
-------
samples were analyzed for residues of 2,4,5-TP; only one sample contained
detectable levels of 2,4,5-TP. This positive sample contained 9.11 ug/1 of
2,4,5-TP and was collected in the spring of 1978. No detection limits were
reported.
Britton et al. (19B3) reported on levels of pesticides detected in water-
sediment mixtures (unfiltered samples) and in hottora material samples collected
by the National Stream Quality Accounting Network (NASQAN) in 1976. Throughout
the United States, 151 permanent stations, plus stations added as part of local
programs, were sampled for pesticides, including 2,4,5-TP. Water-sediment
samples were collected quarterly; bottom material samples were collected
semiannually. 2,4,5-TP was detected in water-sediment samples at 1 of 12
stations in the Lower Mississippi region and 2 of 6 stations in the California
region. Die maximum level of 2,4,5-TP in water-sediment samples at these
stations was 2.2 ug/1. No samples of bottom materials contained detectable
levels of 2,4,5-TP. Ihe detection limits for 2,4,5-TP in water-sediment and
in the bottom material samples were not reported in the study.
12.2.4 Food
Little information was available on the occurrence of 2,4,5-TP in food in
the Uhited States; however, the studies that were identified in the literature
are summarized here. Levels in food are expected to decrease due to the
cancellation of uses of 2,4,5-TP on food crops.
Leidy et al. (1975, as cited in USEPA 1979a) did not detect residues of
2,4,5-TP in apples harvested 29-91 days after 2,4,5-TP was applied to the
ground cover under apple trees. However, Cochrane et al. (1976, as cited in
USEPA 1979a) reported that the direct application of a 20 mg/1 solution of
2,4,5-TP to apple trees (to prevent fruit drop) resulted in residues in unwashed
fruit of 97 ugAg initially, 46 ug/kg at harvest (10 days), and 35 ug/kg after
4 months in storage. Also, after storage, washed fruit contained 2,4,5-TP
residues of 15 ug/kg; washed and waxed fruit contained 14 ug/kg.
Hie available data were insufficient to estimate the typical human
dietary intake of 2,4,5-TP in foods today.
-------
12.3 EXPOSURE SUMMARY
The data available on the occurrence of 2,4,5-TP in drinking water and in
food are limited and information on its presence in ambient air could not be
obtained. However, because of the low vapor pressure of 2,4,5-TP and its esters,
air concentrations are expected to be low. 2,4,5-TP, a herbicide used widely
during the 1960's and 1970's, has been detected as a low-level contaminant of
drinking water supplies and of some foods. The cancellation of many of its
major uses in 1979 and the suspension of the remaining uses in 1984 should
result in decreasing levels of contamination from all three routes of exposure.
The absence of any widespread violation of the interim standard, along
with the available monitoring data, suggests that the extensive use of 2,4,5-TP
during the 1970s did not result in contamination of more than a few ug/1.
Based on the assumption that adults consume 2 liters of drinking water per day,
intake of 2,4,5-TP would not exceed more than 10 ug/day. With the discontinu-
ation of 2,4,5-TP, current levels and exposures of 2,4,5-TP are expected to be
less than the levels reported to occur in the 1970s studies.
-------
T^ble 12-1. Exposure Estimates for 2,4,5-TP
Range of Reported Exposure Estimated
Source Levels (low to high) Daily Intake
Drinking Water 0.02-5 ug/1* 0.0006-0.1429 ug/kg day
Air No data available
Food** No data available
*The data provided for drinking water exposure do not present the results of a
study conducted in Idaho reporting levels from "not detected" to 520,000 ug/1
with a mean of 11 samples of 6,800 ug/1. From the data provided, it appears
that the exposure to 2,4,5-TP will be very low and will be derived almost
entirely from drinking water. Although data are not available on the
occurrence of 2,4,5-TP in air or food, these sources are not expected to be
significant routes of exposure. The total combined exposure to 2,4,5-TP is
expected to decrease in the future.
-------
13. TOXAPHENE
13.0 SUMMARY
Toxaphene is an insecticide that was widely used on crops and livestock
in the 1960's and 1970's. However, due to its ability to volatilize and its
limited mobility to move in soil, it has rarely been detected in drinking water.
All uses of toxaphene were cancelled in the early 1980s. Current exposures to
toxaphene are expected to be minimal and to decrease further in the future.
13.1 GENERAL CHARACTERISTICS
13.1.1 Physical/Chemical Properties
Toxaphene is an insecticide that is produced from the chlorination of
camphene. Toxaphene is a mixture of at least 175 compounds, of which the
structures of fewer than 10 are known (NAS 1977). Synonyms and identifiers
for toxaphene are Attac 4-2, Camphoclor, Camphofene Huileux, Motox, Phenacide,
Phenatox, polychlorocamphene, strobane T-90, Toxakil, and several others
(Berg 1986).
Technical grade toxaphene is a yellowish semicrystalline gum with a melting
point range of 65°C-95°C (Kirk-Othmer 1981). It has an approximate overall
molecular formula of C^gHigCla (Windholz 1976) which corresponds to a mean
molecular weight of 414. Overall aqueous solubilities at 25°C of 0.50 mg/1 and
0.74 mg/1 (1.8 x 1 0~6 mol/1 assuming molecular weight of 414) have been reported
for toxaphene (Callahan et al. 1979). The overall vapor pressure of toxaphene
at 25°C is reported to be between 0.2 torr (2.6 x 10~4 atm) and 0.4 torr.
The ratio of the smallest reported vapor pressure to the largest reported
aqueous solubility gives an estimated minimum overall Henry's constant for
toxaphene at 25°C of 0.144 atm»m3/mol. The ratio of the estimated minimum
Henry's constant to the product of the gas constant times the temperature in
degrees Kelvin gives an estimated minimum dimensionless Henry's constant for
toxaphene at 25®C of 6.0. Toxaphene has an estimated Kq,, value of 964 (Mabey
et al. 1981).
-------
13.1.2 Use
In 1974, domestic production of the insecticide toxaphene was estimated
at 103 million pounds, with approximately 74 million pounds of that amount
applied for agricultural uses (USEPA 1977b). By 1976, the estimated production
level decreased to 3 3 million pounds of toxaphene End total domestic usage in
1982 vas 16 million pounds (USEPA 1982c). Of those 16 million pounds, 5.9
million pounds were estimated to have been used on field crops (USDA 1983).
In 1982, the Environmental Protection Agency (EPA) published its intent
to cancel or restrict registrations of pesticide products containing toxaphene
(Federal Register of November 29). All major uses of toxaphene were cancelled,
effective within 30 days of the publication of the notice. Some minor uses of
toxaphene were permitted to continue until 1986 (USEPA 1982c).
13.1.3 Environmental Transport and Transformation
The discussion of the environmental fate of toxaphene is divided into the
following subsections: 13.1.3.1 Volatilization; 13*1.3.2 Sorption and Leaching
Potential; 13.1.3.3 Abiotic Transformations; 13.1.3.4 Biodegradation and
Persistence in Soil and Water; and 13.1.3.5 Summary. The discussion will
emphasize the environmental fate of toxaphene in soil and water.
13.1.3.1 Volatilization
Toxaphene is a complex mixture of chlorinated hydrocarbons with a mean
molecular weight of approximately 414, a minimum overall vapor pressure of
0.2 torr, a maximum overall aqueous solubility of 0.74 mg/1, and an estimated
overall minimum Henry's constant of 0.144 atra»mVmol, at 25°C as discussed
in Section 13.1.1* SAIC used those numbers and equations A-1 and A-3 through
A-8 to estimate approximate overall volatilization half-lives for the toxaphene
mixture in rivers or streams and turbulent lakes. Estimated volatilization
half-lives for toxaphene in rivers or streams range from 1.1 hours in a turbulent
river or stream 1 m deep to 11 days in a stagnant river or stream 10 m deep
(see Appendix A). Estimated volatilization half-lives for toxaphene in turbulent
lakes range from 11 hours in a lake 1 m deep to 4.5 days in a lake 10 m deep.
-------
The estimated volatilization half-lives suggest that volatilization would
be an important removal process for most or possibly all of the major components
of the toxaphene mixture in surface waters under any conditions# However, even
though toxaphene is volatile enough to be effectively removed by air stripping
(USEPA 1985d), the persistence of low levels of toxaphene residues in aquatic
systems over several years (Callahan et al. 1979) suggests that volatilization
may not be an important removal process for some of the components of the
mixture. Estimated volatilization half-lives are difficult to interpret because
toxaphene is a complex mixture of many different polychlorinated compounds.
The persistent low levels of toxaphene residues in aquatic systems may be due
to the lower volatility components of the toxaphene mixture (Callahan et al.
1979).
Volatilization rates of toxaphene components from ground water to the soil
column above are projected to be substantially less than those from surface
waters to the atmosphere. This is due primarily to the laminar nonturbulent
nature of ground water flow. Transport from ground water by volatilization may
be further reduced by a build-up of toxaphene components in the pore air at the
pore air/ground water interface, and an associated decrease in the concentration
gradients across the interface.
By substituting the estimated overall Koc value of 964 {Mabey et al. 1981)
for the toxaphene mixture into equation A-9 along with the mixture's minimum
reported vapor pressure (0.2 torr) and maximum reported aqueous solubility
(0.74 mg/1), SAIC estimates the overall volatilization half-life for the
toxaphene mixture on soil to be less than 1 minute. However, volatilization
half-lives for the less volatile components of the mixture would be greater.
In addition, volatilization rates of toxaphene components beneath the soil
surface should be much slower than at the soil surface.
13.1.3.2 Sorption and Leaching Potential
Substituting the estimated KqC value of 964 into equation A-11 gives
estimated sediment or soil/water equilibrium partition coefficients (Kg/W) for
the overall toxaphene mixture ranging from 9.6 to 77 for sediments or soils
with organic carbon fractions ranging from 0.01 to 0.08. The estimated Ks/W
-------
values indicate that at equilibrium the concentration of many of the toxaphene
components in suspended or exposed bottom sediment may be 10 or more times
greater than the concentration of the same components in the water column.
However, since the estimated overall Ks/W values are less than 100, it is
unlikely that the ratio of the total mass of toxaphene components adsorbed
to suspended and bottom sediment to the total mass of toxaphene components
dissolved in the water column will exceed 0.1 in most surface waters. The
reason is that in most surface waters, most of the time, the ratio of the water
mass to the mass of suspended and bottom sediment exceeds 1,000 (USGS 1983).
Therefore, transport by adsorption to suspended sediments and removal by
adsorption to bottom sediments are probably not important processes for most
of the toxaphene components in most surface waters.
Substitution of the estimated KqC value of 964 into equation A-13 gives
an estimated overall soil TLC Rf value (Appendix A) of 0.19 for the toxaphene
mixture absorbed to a soil with an organic carbon fraction of 0.014, a pore
fraction of 0.5, and a soil density of 2.5 g/cm3. Therefore, based on the
five mobility classes defined by Helling and Turner (1968) and cited by Hamaker
(1975) for a soil with the same properties (Appendix A), the toxaphene mixture
overall would be expected to be moderately immobile (Class 2) in, and moderately
resistant to leaching from, surface soil. Although some components of the
toxaphene mixture may be more susceptible to leaching than the mixture overall,
detectable levels of toxaphene have generally not been found in either surface
or groundwater supplies (USEPA 1985d).
10.1.3.3 Abiotic Transformations
The toxaphene mixture overall has been estimated to have a hydrolysis
half-life exceeding 10 years in water with a pH between 5 and 8 (Callahan et al.
1979). Therefore, hydrolysis does not appear to be an important removal process
for toxaphene in natural waters.
No information is available concerning the oxidation of toxaphene in
natural waters. Based on the structures of known components of toxaphene,
Mabey et al. (1981) estimated a second order rate constant of 3 M"1 hr~1 for
the oxidation of the toxaphene mixture overall by sunlight-generated peroxy
-------
radicals. Substitution of that rate constant and an assumed concentration of
1 x 10~9m peroxy radicals in sunlit surface waters into equation A-15 and
the resulting pseudo first-order rate constant into equation A-1 gives an
estimated half-life of >10^ years for the oxidation of toxaphene by peroxy
radicals in surface waters.
The known components of the toxaphene mixture do not have chromophores
that absorb light strongly above the approximate solar radiation cutoff at the
earth's surface of 290 nm (Lyman et al. 1982). Therefore, the toxaphene
mixture is not expected to undergo significant rates of photolysis in the
environment. Wolfe et al. (1976), as cited in Callahan et al. (1979) reported
that the GC profile of a toxaphene mixture did not change after irradiation by
light with wavelengths >290 nm.
10.1.3.4 Biodegradation and Persistence in Soil and Water
At least some of the components of the toxaphene mixture are very per-
sistent in soil, sediments, and water. Callahan et al. (1979) reviewed a
number of lake studies in which low levels of toxaphene residues were reported
to persist in both sediments and the water column for several years after
application. Degradation half-lives of up to 20 years have been reported for
the toxaphene mixture overall in aerobic soils and sediments (FDA 1981a).
However, toxaphene appears to be susceptible to reduction and much less
persistent in anaerobic soils and sediments. Parr and Smith (1976), as
cited in Callahan et al. 1979) reported that biodegradation removed 50 percent
of toxaphene over a 6-week period in flooded (anaerobic) soil, but that no
degradation occurred over the same period in aerobic sediment.
1.3.6 summary
Based on the above discussion and the review by Callahan et al. (1979),
the following tentative conclusions can be made concerning the mo3t likely
behavior of toxaphene in soil and water:
o Based upon theoretical considerations, volatilization is expected to be
an important removal process for most or all of the major components of
the toxaphene mixture in surface waters and on soil*
-------
o Based upon theoretical considerations, transport by adsorption to
suspended sediment and removal by adsorption to bottom sediment are
not expected to be important processes for most or all .of the major
components of the toxaphene mixture in most surface waters.
o Based upon theoretical considerations and monitoring data, most or all
of the major components of the toxaphene mixture are expected to be at
least moderately resistant to leaching from surface soil.
o Based upon limited data, the major components of toxaphene are not
expected to undergo significant rates of hydrolysis nor direct photo-
lysis in the environment. Based upon theoretical considerations,
toxaphene is not expected to undergo significant rates of photo-
oxidation in the environment.
o Based upon monitoring data, many of the components of the toxaphene
mixture appear to be extremely resistant to biodegradation in aerobic
soils and sediments with estimated degradation half-lives of up to 20
years. However, toxaphene appears to be susceptible to both abiotic
and biologically mediated reduction in anaerobic soils and sediments.
13.2 OCCURRENCE IN THE ENVIRONMENT
13.2.1 Water
This section presents the available data from monitoring studies and
surveys to determine the extent of occurrence of toxaphene in public drinking
water supplies and water other than drinking water.
13.2.1.1 Occurrence in Drinking Water
Studies at both the national and regional levels have addressed concen-
trations of toxaphene in drinking water. Uie results of three national
studies and several regional studies are discussed in this section. Where
possible, reported levels of toxaphene in drinking water obtained from ground
water sources and surface water sources have been discussed separately.
Ground Water Sources — National Studies
The Federal Reporting Data System (FRDS 1984) provides information on public
water supplies found to be in violation of current Maximum Contaminant Levels
(MCLs). Data are not available on the number of ground water systems in the
U.S. that have monitoring requirements for toxaphene; however, no violations of
the current MCL of 5 ug/1 were reported for toxaphene during the years 1979-1983.
-------
The 1978 Rural Water Survey (USEPA 1984i) involved the collection of
samples from 267 households {the majority using private water supplies) in
rural locations throughout the U.S. and analyses for toxaphene. A total of 71
public drinking water systems of varying sizes using groundwater were covered
by the survey. None of the samples from the 71 groundwater systems studied
exceeded the minimum quantification limit of 0.17 ug/1 for toxaphene.
Ground Water Sources — Regional Studies
Irwin and Healy (1978) summarized data collected in 1976 during a water
quality reconnaissance of public water supplies in Florida. None of the 100
water supplies sampled utilizing the five aquifers in Florida contained toxa-
phene in excess of the detection limit (the detection limit was not reported).
Tucker and Burke (1978) presented the results of analyses of samples
collected from wells in nine New Jersey counties. None of the samples tested
contained concentrations of toxaphene in excess of the minimum reportable
concentration of 0.06 ug/1 for the study.
Surface Water Sources — National Studies
Data were obtained from the Federal Reporting Data System (FRDS 1984) on
violations of the current MCL of 5 ug/1 for toxaphene. The data were obtained
for the years 1979-1983 and include information for all surface water systems
in the United States. The analysis indicated that none of the samples from the
surface water systems studied contained toxaphene in excess of the current MCL.
The 1978 Rural Water Survey (USEPA 1984i) also presented data on drinking
water samples- obtained from surface water systems of varying sizes. None of
the samples from the 21 public drinking water systems contained concentrations
of toxaphene in excess of the minimum quantification limit of 0.17 ug/1.
Surface Water Sources — Regional Studies
Irwin and Healy (1978), summarizing data collected during a water quality
reconnaissance of public water supplies in Florida, reported that none of the
-------
samples from the 16 surface water supplies studied contained concentrations
of toxaphene in excess of the detection limit. The detection limit was not
reported.
In a study on the effects of forest runoff on the quality of a public
water supply in Oregon, Elliott (1979) observed a concentration of toxaphene
of 3 ug/1.
To assemble a database which would reflect the status of Great Lakes
drinking water quality, the Canadian Public Health Association gathered data
from October 1984 through August 1995 (Canadian Public Health Association 1986).
The data collected covered the period from the mid 1970s to early 1985. The
study was funded by the Health Protection Branch of Health and Welfare Canada,
and the Ontario Ministry of the Environment. The data collected cover the
period from the mid-1970*s to early 1985. A research team, appointed by the
Association, reviewed data on the quality of water at 31 representative Canadian
and United States communities and 24 offshore sites to evaluate the human
health implications.
For each of the 31 communities, data consisted of: 1) background infor-
mation on the community; 2) treatment plant schematics and associated .treatment
process information; and 3) water quality data. Water sample types included
raw water (treatment plant intake), distribution water (treated water), and
tap water. Water quality data collected included general parameters (e.g.,
alkalinity, turbidity), microbiological and radiological parameters, inorganic
parameters, and organic parameters (including volatiles, base/neutrals,
pesticides and PCBs, and phenols and acids). For each parameter, the water
type, time period, concentration (mean, range), number of samples, and detection
limit are presented.
For most of the volatile organics, including toxaphene, the available data
indicate that there were very low levels of these contaminants in the raw,
treated, or tap water. Most of the values found were "not detected" or near
the detection limit (Canadian Public Health Association 1986).
-------
Unspecified Sources — National Study
In an EPA survey (USEPA 1977b) of pesticide contamination in commercial
drinking water sampled during 1975 and 1976, 27 of 58 samples analyzed were
found to be positive for toxaphene. The levels of toxaphene detected were at
or below 0.05 ug/1. The detection limit was not reported.
Unspecified Sources — Regional Study
In a report on source identification of pollutants entering a sewage
treatment plant, Levins et al. (1979a) tested two drinking water sources in a
drainage basin in Georgia. Although the detection limits were not reported,
toxaphene was not detected in either of the samples tested.
13.2.1.2 Occurrence in Water Other than Drinking Water
One national study and several regional studies were performed to monitor
levels of toxaphene in water other than drinking water. All of the studies
reported on levels of toxaphene in surface water; they are discussed below.
National Study
The National Surface Water Monitoring Program (NWMP 1982) presented data
on levels of toxaphene in surface water samples collected throughout the U.S.
during the period 1975-1979. During this time, 2,479 samples were collected
and analyzed for toxaphene. Although no detection limit was given for toxa-
phene, 11 positive samples (or 0.4%) were found during testing. The range of
positive values was 0.0 to 1.65 ug/1.
Regional Studies
Mattraw (1975, as cited in USEPA 1980b) did not detect toxaphene in any
surface water samples collected and analyzed during an organochlorine residue
survey in Florida. The number of samples analyzed and the detection limit were
not reported.
-------
Toxaphene was detected in 11 (or 55%) of 20 Mississippi Delta lakes sampled
by Herring and Cotton (1970/ as cited in USPEA 1980b). The maximum reported
concentration was 1.92 ug/1 (the detection limit was not reported).
13.2.2 Ambient Air
The highest toxaphene concentrations in air are found where toxaphene has
been used for agricultural application, particularly in the southern United
States and during the growing season. Arthur et al. (1976) collected and
analyzed 156 air samples from the Mississippi Delta, site of intensive cotton
production, between 1972 and 1974. Average monthly toxaphene concentrations in
air were 0.258 ug/m3 in 1972, 0.082 ug/m3 in 1973, and 0.160 ug/m3 in 1974,
with a maximum detected concentration of 1.747 ug/m3. No detection limit was
reported.
Bidleman et al. (1976, as cited in USEPA 1980b) reported 21 air samples
from Rhode Island and Georgia in 1975, and in Arizona and Kansas in 1974, with
toxaphene concentrations of 0.00004 to 0.007 ug/m3. The number of positive
samples and the detection limit were not reported.
Bidleman (1981-, as cited in Reinert et al. 1982) reported an average
toxaphene concentration of 0.011 ug/m3 in air samples collected in South
Carolina in 1978. The number of samples collected, the number of positive
samples identified, and the detection limit were not reported.
The highest reported toxaphene level was 8.7 ug/m3 in a city in Arkansas
in 1970, with toxaphene detected in 3.5 percent of the 2,479 samples collected
from locations that were selected for their potentially high concentrations of
pesticides in: ambient air. The mean of the positive samples was 5.2 ug/m3
(Kutz et al. 1976). The detection limit and range were not reported.
13.2.3 Soil/Sediments
Three studies were identified that involved the examination of the occur
rence of toxaphene in soil and/or sediments; two were soil studies and one was
a sediment study.
-------
Carey et al. (1978, 1979) presented data obtained during the 1971 and 1972
National Soils Monitoring Program. During the late summer and fall of 1971,
composite soil samples were obtained from 1,486 10-acre sites in 37 states.
Data were collected from 1,483 sites in the same states in the summer and fall
of 1972. The minimum detection limits for toxaphene ranged from 50 to 100
ug/kg. During the 1971 sampling period, 92 (or 6.2%) of 1,486 soil samples
collected were found to contain residues of toxaphene in the range of 180 to
36,330 ug/kg (Carey et al. 1978). During the 1972 sampling period, 76 (or
5.1%) of 1,483 soil samples analyzed were found to contain residues of toxaphene
in the range of 220 to 46,580 ug/kg (Carey et al. 1979).
Patty (1981, 1982, as cited in TXNT 1986) reported that 45 percent toxaphene
remained in soils 20 years after (soil) treatment. Between 90 and 95 percent
of the toxaphene residues were detected in the 30 cm layer. The range of
contamination was not presented.
Itie National Surface Water Monitoring Program (NWMP 1982) presented data
on residues of toxaphene in sediment samples collected between 1975 and 1979.
Of the 937 sediment samples analyzed, 11 samples were found to be contaminated
with detectable levels of toxaphene. The maximum reported value of toxaphene
was 814.49 ug/kg.
13.2.4 Food
13.2.4.1 Dietary Intake
The Food and Drug Administration (FDA) conducts Total Diet Studies (also
known as Market Basket Surveys) to evaluate the occurrence of various substances,
including toxaphene, in food consumed by adults, toddlers, and infants. According
to the resultivof a recently published survey for FY 81/82, toxaphene was
detected in several food categories including meat, fish, and poultry; grain
and cereal products; leafy vegetables; root vegetables; garden fruits; and oils
and fats. In the adult diet study, the highest concentration of toxaphene was
detected in root vegetables (0.0059 ppm) (Gartrell et al. 1986a). Toxaphene
was only detected in the oils and fats category of infant and toddler diets.
The level of toxaphene in the infant diet was 0.323 ppm, while in the toddler
-------
diet toxaphene was present in oils and fats at a level of 0.0408 ppm {Gartrell
et al. 1986b).
The grain and cereal products category accounted for about 62 percent of total
dietary intake of toxaphene for the adult male. Meat, fish, and poultry and
root vegetables also contributed significantly—12 and 11.3 percent, respectively.
(Gartrell et al. 1986a.\
Total daily intake of toxaphene by infants and toddlers was provided by
only one source—oils and fats. For infants this totaled 0.658 ug/day and for
toddlers intake was 0.493 ug/day (Gartrell et al. 1986b).
Hie FDA provided mean daily intakes of toxaphene reflecting detections of
toxaphene in 12 total diet studies conducted from April 1982 to April 1985
(FDA 1986). For the 6- to 11-month old infant, daily intake of toxaphene was
0.065 ug/day. For ttie 2-year old toddler, toxaphene intake was 0.310 ug/day.
Toxaphene intakes for adult males and females are presented in Table 13-1. For
the adult male, toxaphene intakes ranged between 0.576 and 0.603 ug/day. Daily
toxaphene intakes for adult females ranged between 0.367 and 0.492 ug/day.
Intakes were highest for the 60- to 65-year old age group for both sexes.
Table 13-1. Summary of FDA Total Diet Study Estimates for Toxaphene Intakes
for Adult Males and Females
Sex/Age Group Intake (ug/day)
14-16
year
old
female
0.367
14-16
year
old
male
0.584
25-30
year
old
female
0.369
25-30
year
old
male
0.576
60-65
year
old
female
0.492
60-65
year
old
male
0.603
Sources FDA 1986.
-------
13.3 EXPOSURE SUMMARY
Several studies have been conducted that provide useful data on the extent
of occurrence of toxaphene in drinking water, air, and food. According to the
monitoring data collected during the late 1970's, most of the reported toxaphene
levels in drinking water were below 0.1 ug/1. Current levels in drinking water
are unknown because of a lack of data, but are expected to be even lower than
in the 1970's due to the restrictions placed on the major uses of the compound
in 1982. Also with the cancellation of the remaining minor uses of toxaphene
in 1986, coupled with the volatilization and biodegradation of the remaining
toxaphene in the environment, future levels in drinking water are not expected
to be highter than those levels reported in the past. According to available
data from the late 1970's and 1980, ambient air levels in agricultural areas
where toxaphene was used were reported to reach as high as 8.7 ug/m3. However,
typical levels ranged from less than 0.01 to 1.75 ug/m^. While current data on
toxaphene concentrations in ambient air are unavailable, levels today, as well
as those in the future, are expected to be negligible. Recent dietary studies
report that the daily intake of toxaphene for 25-30 year old males ranged from
0.576 to 0.603 ug/day, with a mean daily intake of 0.576 ug/day. However,
future intake levels are expected to decrease. Exposure estimates for toxaphene
are presented in Table 13-2.
Table 13-2. Exposure Estimates for Toxaphene
Reported Exposure
Estimated Adult
Source
Levels (low-high)
Intake
Drinking Water
0 - <0.1 ug/1
0 - <0.2 ug/day
Diet*
—
0.6 ug/day
Air
Negligible
Negligible
* Note: 25-10 year old adult male.
The current available information on occurrence of toxaphene is insufficient
to determine the national distribution of intake by any of the three routes.
However, EPA believes that intakes from diet will generally be less than 1 ug/day
and that air exposure to toxaphene is expected to be negligible. If toxaphene
does occur in drinking water at levels of more than a few tenths of a ug/1, it
is likely to be the major source of intake. However, for the majority of
individuals in the United States diet is the major source.
-------
references
Achari, R.G., S.S. Sandhu, and W.J. Warren. 1975. Chlorinated hydrocarbon
residues in groundwater. Bull. Environ. Contam. Toxicol. 13(1):94-96.
Akermark, B. 1978. Photochemical reactions of phenoxy acids and dioxins.
In: Ramel, C. (ed.). Chlorinated phenoxy acids and their derivatives.
Stockholm: Swedish Natural Research Council* pp. 75-81.
Altom, J.D., and J.P. Stritzke. 1973. Degradation of dicamba, picloram, and
four phenoxy herbicides in soils. Weed Sci. 21(6):556-560.
Arthulr, R.D., j.d. Cain, and B.F. Barrentine. 1976. Atmospheric levels of
pesticides in the Mississippi Delta. Bull. Environ. Contain. Ibxicol.
15(2):129-134.
Baker# D.B. 1983. Herbicide Contamination in municipal water supplies of
northwestern Ohio. Draft final report. Tiffin, Olio: Water Quality
Laboratory, Heidelberg College.
Baker, D.B., K.A. Krieger, and J.V. Setzler. 1981. The concentrations and
transport of pesticides in northwestern Ohio rivers. Prepared by Water
Quality Laboratory, Heidelberg College, Tiffin, Ohio, for Lake Erie
Wastewater Management Study, U.S. Army Corps of Engineers, Buffalo, New
York.
Baker, J.L., H.P. Johnson, M.A. Borcherding, and W.R. Payne. 1979. Nutrient
and pesticide movement from field to stream: A field study. In: Best
Management Practices for Agriculture and Silviculture. Proc. of the 1978
Cornell Agriculture Waste Management Conference, R.C. Roehr, D.A. Haith,
M.F. Walker and C.S. Marten (eds.) Ann Arbor Sci. Publ. Inc., Ann Arbor,
Michigan 16:213-245. Cited in USEPA 1984d.
Baker, J.L., and T.A. Austin, 1983. Impact of agricultural drainage wells on
groundwater quality. Completion Report 1981-1983. Departments of
Agricultural Engineering and Civil Engineering, Iowa State University.
EPA Grant No. G007228010.
Baker, M.D.# C.I. Mayfield, and W.E. Inniss. 1980. Degradation of
chlorophenols in soil, sediment, and water at low temperature. Water
Res. 14»1765-1771. Cited in DSEPA 1986.
Banerjee, S., P.H. Howard, A.M. Rosenberg, A.E. Dombrowski, H. Sikka, and D.L.
Tulles. 1984. Development of a general kinetic model for biodegradation
and its application to chlorophenols and related compounds. Environ.
Sci. and Technol. 18:416-422. Cited in DSEPA 1986.
Barks, J.H. 1978. Water quality in the Ozark National Scenic Rivervays,
Missouri. Washington, DC: D.S* Geological Survey, U.S. Department of
Interior. Geological Survey Water-Supply Paper 2048.
Barnett, A.P., et al. 1967. Loss of 2,4-D in washoff from cultivated fallow
land. Weeds 15:133-137. Cited in NRCC 1978 and in Bovey and Young 1980.
-------
Barnett, R.W., A.J. D'Ercole, J.D. Cain, and R.D. Arthur. 1979. Organo
chlorine pesticide residues in human milk samples from women living in
Northwest and Northeast Mississippi, 1973-75. Pestic. Monit. J.
13(2):47-51.
Beestman, G.B., and J.M. Deming. 1974. Dissipation of acetanilide herbicides
from soils. Agron. J. 66:308-311. Cited in USEPA 1984d.
Benvenue, A., J.N. Ogata, and J.W. Hylin. 1972b. Organochlorine pesticides
in rainwater, Oahu, Hawaii, 1971-1972. Bull. Environ. Contain. Toxicol.
8(4):238-241. Cited in Cirelli 1978.
Benvenue, A., J.W. Hylin, Y. Xawano, and T.W. Kelley. 1972a. Organochlorine
pesticide residues in water, sediment, algae, and fish, Hawaii,
1970-1971. Pestic. Monit. J. 6(1):56-64.
Berg, G.L. (ed.). 1986. Farm Chemicals Handbook. Willoughby, Ohio: Meister
Publishing Co.
Bidleman, T.F. 1981. Atmos Environ. 15:619. Cited in Reinert et al. 1982.
Bidleman, T.F., et al. 1976. High molecular weight chlorinated hydrocarbons
in the air and sea: Rates and mechanisms of air/sea transfer. In:
Windom H.L., and R.E. Dace (eds.). Marine pollutant transfer. DC Heath
and Co. Cited in USEPA 1980b.
Bigler, W. 1983. Personal communication to S. Oohen. Epidemiology Program,
Department of Health and Rehabilitative Services, Jacksonville, Florida.
Cited in USEPA 1983c.
Boland, P.A. 1981. National screening program for organics in drinking
water. Part II. Data. Prepared by SRI International, Menlo Park,
California; for Office of Drinking Water, U.S. Environmental Protection
Agency, Washington, DC. EPA Contract No. 68-01-4666.
Bovey, R.W., and A.L. Young. 1980. Ihe Science of 2,4,5-T and Associated
Herbicides. New York: John Wiley & Sons.
Bradshaw, J.S., E.L. Loveridge, K.P. Rippee, J.L. Peterson, D.A. White, J.R.
Barton, and D*K. Fuhriman. 1972. Seasonal variations in residues of
chlorinated hydrocarbon pesticides in the water of the Utah Lake drainage
system - 1970-1971. Pestic. Monit. J. 6(3)>166-170.
Breidenbach, A.W. et al. 1967. Chlorinated hydrocarbon pesticides in major
river basins, 1957-65. Pub. Health Rep. 82:139. Cited in USEPA 1980f.
Britton, L.J., K.E. Goddard, and J.C. Briggs• 1983. Quality of rivers of the
United states, 1976 water year — Based on the National Stream Quality
Accounting Network (NASQAN). Washington, DC: U.S. Geological Survey,
U.S. Department of Interior. Open-file report 80-594.
Brockway, D.L., J. Hill, J.R. Maudsley, and R.R. Lassiter. 1979.
Development, replica'bility, and modeling of naturally derived microcosms.
Int. J. Environ. Stud. 13(2):149-158. Cited in USEPA 1984g.
-------
—i.odtnann, N.V. 1976. Continuous analysis of chlorinated hydrocarbon
pesticides in the lower Mississippi River. Bull. Environ. Contain.
Toxicol. 15(1):33-39. Cited in USEPA 1984g.
Brodzinsky, R., and H.B. Singh. 1982. Volatile organic chemicals in the
atmosphere: An assessment of available data. Prepared by SRI
International, Menlo Park, California, for Environmental Sciences
Research Laboratory, Office of Research and Development, U.S.
Environmental Protection Agency, Research Triangle Park, North Carolina.
Contract No. 68-02-3452.
Bromilow, R.H., R.J. Baker, M.A.H. Freeman, and K. Gorog. 1980. Pestic. Sci.
11:371-378. Cited in Cohen et al. 1984.
Brookhart, G., and L.D. Johnson. 1977. Analysis of fish tissue for kepone,
mirex, atrazine, linuron, and alachlor. Prepared by Anal. Biochem. Lab.,
Inc. for U.S. Environmental Protection Agency. NTIS PB=285865. Cited in
USEPA 1984d.
Buhler, D.R., M.E. Rasmusson, and H.S. Nekaue. 1973. Occurrence of hexa
chlorophene and pentachlorophenol in sewage and water. Environ. Sci. and
Technol. 7(10):929-934. Cited in Cirelli 1978.
Burlinson, N.E., L.A. Lee, and D.H. Rosenblatt. 1982. Kinetics and products
of hydrolysis of 1,2-dibrooo-3-chloroprane. Environ. Sci. and Technol.
16(9):627-632.
California. 1982. Pesticide use report. Annual 1982. Sacramento,
California: Department of Food and Agriculture, State of California.
Callahan, M.A., et al. 1979. Water-related environmental fate of 129
priority pollutants. Vol. I. Prepared by Versar and SRI International
for Monitoring and Data Support Division, Office of Water Planning and
Standards, U.S. Environmental Protection Agency, Washington, DC.
EPA-440/4-79-029.
Canadian Public Health Association. 1986. Comprehensive survey of the status
of Great Lakes drinking water. Prepared in cooperation with Health and
Welfare Canada, and the Ministry of the Environment. Published by
Canadian Public Health Association. Ottawa, Canada, August 1986.
Carey, A.E., and F.W. Kutz. 1983. Trends in ambient concentrations of
agrochemicals in humans and the environment of the United States.
Presented at the International Conference on Environmental Hazards of
Agrochemicals in Developing Countries, Alexandria, Egypt, November 8-12,
1983.
Carey, A.E., J.A. Gowen, H. Tai, w.G. Mitchell, and G.B. Weirsma. 1978.
Pesticide residue levels in soils and crops, 1971 -- National Soils
Monitoring Program (III). Pesticide Monit. J. 12(3):117-136.
Carey, A.E., J.A. Gowen, H. Tai, W.G. Mitchell, and G.B. Wiersma. 1979.
Pesticide residue lavels in soils and crops from 37 states, 1972 —
National Soils Monitoring Program (IV). Pesticide Monit. J.
12(4):209-229.
-------
Carter, G. and M.B. Riley. 1981. Pestic. Monit. J. 15:139-142. Cited in
Cohen et al. 1984.
Castro, C.E., and N.O. Belser. 1968. Biviron. Sci. and Techno1. 2:779-783.
Cited in Cohen et al. 1984.
Castro, T.F., and T. Yoshida. 1971. Degradation of organo-chlorine
insecticides in flooded soils in the Fhillipines. J. Agxic. Food Chem.
19(6):1168-1170. Cited in Callahan et al. 1979.
Cautreels, W., K. Vancoumenberge, and L.A. Guzman. 1977. Comparison between
the organic fraction of suspended water at a background and an urban
station. Sci. Total Environ. 8:79-88.
Chemical Regulation Reporter. 1978. Heptachlor. EPA administrator approves
agreement for five-year phase-out of crop uses. Chem. Reg. Rep.
1(50):1904-1905.
Chemical Regulation Reporter. 1984. _ Ethylene dibromide. EDB found in
Connecticut groundwater wells at unacceptable levels, state officials
say. Chem. Reg. Rep. 8(4):100-101.
Cirelli, O.P. 1978. Pentachlorophenol: Position document 1. Washington,
DC: Special Pesticide Review Division, U.S. Environmental Protection
Agency. EPA/SPRD-80/85.
Cochrane, W.P., R. Greenhalgh, and N.E. Looney. 1976. Residues in apples
sprayed with fenoprop. Canadian J. Plant Sci. 56:207-210. Cited in
USEPA 1979a.
Cohen, D.B., and G.W. Bowes. 1984. Hater quality and pesticides: A
California risk assessment program (Volume 1). State Hater Resources
Control Board, Toxic Substances Control Program, Sacramento, California.
Cohen, S.Z. 1981. Summary report — DBCP in groundwater in the Southeast.
Washington, DC: Hazard Evaluation Division, U.S. Environmental
Protection Agency. Cited in Cohen et al. 1984.
Cohen, S.Z*, C* Eiden, and N.N. Lorder* 1986. Evaluation of pesticides in
groundwater* W*Y* Garner, R.C. Honeycutt, H.N. Nigg (eds.)* American
Chemical Society, Washington, DC. ACS Symposium Series 315.
Cohen, S.Z*# 6.M. Creeger, R.F. Carsel, and C.G* Enfield. 1984. Pesticide
contamination of groundwater from agricultural uses. American Chemical
Society Symposium Series Treatment and Disposal of Pesticide Hastes*
Cong?ton, B., P.P. Bazydlo, and G. Zweig. 1972* Field evaluation of methods
of collection and analysis of airborne pesticides* Vol. 1. Field
evaluations and analysis. U.S. NT IS PB 214 008. Cited in Grover et al.
1976*
CPCR. 1986. Crop protection chemicals reference. Chemical and
Pharmaceutical Press, New York, New York.
-------
Crosby, D.G. 1976. Nmbiological degradation of herbicides in the soil. Ins
Audus, L.J. (ed.). Herbicide::: Physiology, Biochemistry, Ecology.
Vol. 2. New Yorfcc Academic Press, pp. 65-97.
Crosby, D.G., K.I. Beynon, P.A. Greve, F. Korte, G.G. Still, and J.W. VonJc.
1981. Environmental chemistry of pentachlorophenol• Pure Appl. Chem.
53-1051-1080. Cited in USEPA 1986.
Dappen, G. 1974. Pesticide analysis from urban storm runoff, frepared by
University of Nebraska, Lincoln, Nebraska, for Office of Water Research
and Technology, Nebraska Wesleyan University. Project No. A-025-NEB(2).
Datta, P.R. No date. Memorandum: Review of six documents regarding
monitoring of pesticides in Northwestern Ohio rivers. Sent to: D.J.
Severn, Chief - Exposure Assessment Branch, HED. Office of Pesticides
and Toxic Substances, U.S. Environmental Protection Agency, Washington,
DC.
De Filippi, R.P., V.J. Kyukonis, R.J. Robey, and M. Modell. 1980.
Supercritical fluid regeneration of activated carbon for adsorption of
pesticides. EPA-600/2-80-054. Research Triangle Park, North Carolina:
U.S. Environmental Protection Agency. Cited in USEPA 1985a.
Deeley, G.M. 1986. Estimated hydrolysis half-lives of DBCP in California
groundwater. Shell Development Company. Interoffice memorandum from
G.N. Deeley to J.D. Colthart. January 14, 1986.
Demayo A. 1972. Gas chromatographic determination of the rate constant for
hydrolysis of heptachlor. Bull. Environ. Oontamin. and Toxicol.
8(4):234-237. Cited in Callahan et al. 1979.
Dudley/ D.R., and J.R. Karr. 1980. Pesticides and PCB residues in the Black
Creek watershed, Allen County, Indiana, 1977-1978. Pestic. Monit. J.
13(4):155-157. Cited in USEPA 1984d.
Eichelberger, J.W., and J.J. Lichtenberg. 1971. Persistence of pesticides in
river water. Environ. Sci. and Technol. 5(6):541-544. Cited in Callahan
et al. 1979.
Eichers, T.R., P.A. Andrilenas, and T.W. Anderson. 1978. Farmers' use of
pesticides in 1976. Washington, DC: National Economic Analysis
Division, Economics/ Statistics, and Cooperative Service, U.S. Department
of Agriculture. Agricultural Economic Report No* 418.
Elliott, W.M. 1979. Portland's public water supply -- pure and simple.
Proceedings of the National Conference on Environmental Engineering. New
York: ASCE.
Englande, A. J., J.K. Sknith, and J.N. English. 1978. Potable water quality of
advanced wastewater treatment plant effluents. Prog. Wat. Tech.
10(1/2):17-39.
Estabrooks, F. No date. Memorandum of pesticide information in New York
surface water. From: Frank Estabrooks, U.S. Department of Environmental
Conservation, Albany, New York.
-------
Fang, C.H. 1977. Studies on the photodecomposition of herbicide alachlor.
15(1-2):14-22. Cited in USEPA 1984d.
FDA. 1980a. (Food and Drug Administration.) Compliance program report of
findings. FY75 pesticide program (7320.07) and FY76 pesticides and
metals program (7320.55). Washington, DC: Food and Drug Administration,
D.S. Department of Health, Education, and Welfare.
FDA. 1980b. (Food and Drug Administration.) Conypliance program report of
findings. FY77 total diet studies — Adult (7320.73). Washington, DC:
Food and Drug Administration, U.S. Department of Health, Education, and
Welfare.
FDA. 1980c. (Food and Drug Administration.) Compliance program report of
findings. FY77 total diet studies — Infants and toddlers (7320.74).
Washington, DC: Food and Drug Administration, D.S. Department of Health,
Education, and Welfare.
FDA. 1981a. (Food and Drug Administration.) Compliance program report of
findings. FY77 pesticides and metals program (7320.79). Washington, DC:
Food and Drug Administration, U.S. Department of Health and Human
Services. FDA/BF-82/87.
FDA. 1981b. (Food and Drug Administration.) The FDA surveillance index.
Washington, DC: Bureau of Foods, Food and Drug Administration.
1982a. (Food and Drug Administration.) Compliance program report of
findings. FY79 total diet studies — Adult (7305.002). Washington, DC:
Food and Drug Administration, U.S. Department of Health and Human
Services. FDA/BF-82/98.
1982b. (Food and Drug Administration.) Compliance program report of
findings. FY79 total diet studies — Infants and toddlers (7305.002).
Washington, DC: Food and Drug Administration, U.S. Department of Health
and Human Services. FDA/BF-82/97.
1982c. (Food and Drug Administration.) Compliance program report of
findings. FY 79 pesticides and metals in fish program (7305.007).
Washington, DC: Food and Drug Administration, U.S. Department of Health
and Human Services. FDA/BF-82/88.
1983* (Food tad Drug Administration.) Data on EDB in grain from Dr.
Sanford Miller, delivered to John A. Moore, DVM, November 2, 1983. Cited
in USEPA 1984b.
1986. Food and Drug Administration. Memorandum from E. Gunderson,
Division of Contaminants Chemistry, Center for Food Safety and Applied
Nutrition, Washington, DC, to Dr. Paul S. Price, Office of Drinking
Water, U.S. Environmental Protection Agency, Washington, DC. November 6,
1986.
Florida Department of Agriculture. 1984. Transmittal letter to Dr. Richard
D. Schmitt from Doyle Conner, January 19, 1984. Cited in USEPA 1984b.
FDA.
FDA.
FDA.
FDA.
FDA.
-------
Fogel, S., R.L. Lancione, and A.E. Sewall. '*982. Enhanced biodegradation of
methoxychlor in soil under sequential environmantal conditions. Appl.
Environ. Microb. 44(1):t13-120. Cited in USEPA 1984g.
FRDS. 1984* (Federal Reporting Data System.) Computer printout; dated April
4/ 1984, containing data on organic chemical MCL violations, FY 1979-
1983. U.S. Environmental Protection Agency, Washington, DC.
Freeden, F.J.H., J.G. Saha, and M.H. Balba. 1975. Residues of methoxychlor
and other chlorinated hydrocarbons in water, sand, and selected fauna
following injections of methoxychlor blackfly larvicide into the
Saskatchewan River. Festic. Monit. J. 8(4):241-246. Cited in USEPA
1984g.
Gartrell, M.J., J.C. Craun, D.S. Podrebarac, and E.L. Gunderson. 1986a.
Pesticides, selected elements, and other chemicals in adult total diet
samples, October 1980-March 1982. J. Assoc. Anal Chem. 69:(1)146-161.
Gartrell, M.J., J.C. Craun, D.S. Podrebarac, and E.L. Gunderson. 1986b.
Pesticides, selected elements, and other chemicals in infant and toddler
total diet samples, October 1980-March 1982. J. Assoc. Anal. Chem.
69:(1):123-145.
Georgia Department of Agriculture. 1984. Letter from J.R. Conley to Dr.
Richard D. Schmitt, January 23, 1984 and supplemental submission of
January 27, 1984. Cited in USEPA 1984b.
Gianessi, L.P. 1986. A national pesticide usage data base. Prepared by
Resources for the Future for U.S. Environmental Protection Agency, Office
of Standards and Regulations, Washington, DC. Cooperative Agreement CR
811858-01-0.
Gilliom, R.J., R.B. Alexander, and R.A. Smith. 1985. Pesticides in the
Nation's rivers, 1975-1980, and implications for future monitoring. U.S.
Geological Survey Water-Supply Paper 2271. U.S. Government Printing
Office, U.S. Department of Interior.
Glaze, M.H. 1982. Preliminary quantitative usage analysis of carbofuran.
Washington, DC: Office of Pesticide Programs, U.S. Environmental
Protection Agency.
Going, J.E., and J.L. Spigarelli. 1976. Sampling and analysis of selected
toxic substances. Task IV -- Ethylene dibromide. Prepared by Midwest
Research Institute, Kansas City, Missouri for U.S. Environmental
Protection Agency. EPA-560/6-76-021. Cited in Brodzinsky and Singh
1982.
Greenberg, M., R. Anderson, J. Keene, A. Kennedy, G.W. Page, and S.
Schowburow. 1982. Empirical test of the association between gross
contamination of wells with toxic substances and surrounding land use.
Environ. Sci. Technol. 16(1):14-19. Cited in USEPA 1984g.
Grocery Manufacturers of America. 1984a. Transmittal letter of January 20,
1984 from Sherwin Gardner to Edwin L. Johnson. Cited in USEPA 1984b.
-------
Grocery Manufacturers of America. 1984b* Supplemental, submission of January
28, 1984. Cited in USEPA 1984b.
Grover, R., L.A. Kerr, K. Wallace, et al. 1976. Residues of 2,4-D in air
samples from Saskatchewan 1966-1975* J. Environ. Sci. Health (Part B).
B11:4 331-347.
Hamaker, J.W. 1975. Interpretation of soil leaching experiments. In:
Hague, R. and V.H. Freed (eds.). Environmental Dynamics of Pesticides.
Plenum Press.
Hansen, J.L., and M.H. Spiegel. 1983. Environ. Toxicol. Chen. 2:147-153.
Cited in Cohen et al. 1984.
Hanthorn, M., C. Osteen, R. McDowell, and L. Roberson. 1982. NRE staff
report. 1980 pesticide use on soybeans in the major producing states.
Washington, DC: Natural Resource Economics Division, Economic Research
Service, U.S. Department of Agriculture. ERS Staff Report No.
AGES820106.
Hargrove and Merkle. 1971. (No reference provided in USEPA 1984.) Cited in
USEPA 1984d.
Helling, C.J., and B.C. Turner. 1968. Pesticide mobility: Determination by
soil thin layer chromatography. Sci. 162:562.
Herring, J., and D. Cotton. 1970. Pesticide residues of 20 Mississippi delta
lakes. Proc. 4th Annu. Conf. S.E. Assoc. Game Fish Comm. 482. Cited in
USEPA 1980b.
Hiatt, V. 1976. Personal communication. Cited in NAS 1977.
Holden, P.W. 1986. Pesticides and Groundwater Quality. Issues and Problems
in Four States. Prepared for the Board of Agriculture, National Research
Council. National Academy Kress. Washington, DC. 1986.
Holtorf, R.C. 1982. Preliminary quantitative usage analysis of aldicarb as a
pesticide. Washington, DC: Office of Pesticide Programs, U.S.
Environmental Protection Agency.
Hunt et al. 1980. (Complete reference not provided; missing page from USEPA
1984). Cited in USEPA 1984d.
Hunt, L.B., and R.J. Sachs. 1969. Response of roluns to DDT and
methoxychlor. J. Wildlife Mgmt. 33(2):336-345. Cited in USEPA 1984g.
Idaho Department of Health and Welfare* 1984. Letter to Charles Berry,
Office of Drinking Water, U.S. Environmental Protection Agency,
summarizing data on groundwater contamination incidents. Sent by: "
Charles D. Brokopp, State Epidemiologist, Division of Health, Idaho
Department of Health and Welfare.
Ide, A., Y. Niki, F. Sakamoto, and I. Watanalie. 1972. Decomposition of
pentachlorophenol in paddy soil. Agric. Biol. Chea. 36:1937-1944. Cited
in USEPA 1986.
-------
Irwin/ G.A., and H.G. Healy. 1978. Chemical and physical quality of selected
public water supplies in Florida, August-September 1976. Tallahassee,
Florida: Water Resources Division, U.S. Geological Survey. USGS/WRI
78-21.
Johns, R. 1976. Air pollution assessment of ethylene dibromide. NTIS
PB-256736. p. 44. Cited in USEPA 1984f.
Johnson, R.D., and D.D. Manske. 1976. Pesticide residues in total diet
samples (IX). Pestic. Monit. J. 9(4):157-169.
Johnson, R.D., and D.D. Manske. 1977. Pesticide and other chemical residues
in total diet samples (XI). Pestic. Monit. J. 11(3)t116-131.
Johnson, R.D., D.D. Manske, D.H. Mew, and D.S. Podrebarac. 1979. Pesticides
and other chemical residues in infant and toddler total diet samples (I),
August 1974-July 1975. Pestic. (tonit. J. 13(3):87-98.
Johnson, R.D., D.D. Manske, D.H. New, and D.S. Podrebarac. 1981a. Pesticide,
heavy metal, and other chemical residues in infant and toddler total diet
samples (II), August 1975-July 1976. Pestic. Monit. J. 15(1).
Johnson, R.D., D.D. Manske, and D.S. Podrebarac. 1981b. Pesticide, metal,
and other chemical residues in adult total diet samples (XII), August
1975-July 1976. Pestic. Monit. J. 15(1):54-69.
Jonsson, V., G.J.K. Liu, J. Armbruster, L.L. Kettlehut, and B. Drucker. 1977.
Chlorohydrocarbon pesticide residues in human milk in Greater St. Louis,
Missouri, 1977. Amer. J. Clin. Nutr. 30(7):1106-1109.
Jovanovich A.P., and S.Z. Cohen. In review. Monitoring groundwater in
Georgia for ethylene dibromide (EDB). A preliminary reconnaissance in
Seminole County, Georgia. Hazard Evaluation Division, U.S. Environmental
Protection Agency. Cited in USEPA 1983c.
Joyce, J.C., and H.C* Sikka. 1977. Residual 2,4-D levels in the St. Johns
River, Florida. J. Aquat. Plant Manag. 15:76-82.
Jury, W.A., W.F. Spencer, and W.J. Farmer. 1983. Behavior assessment model
for trace organics in soil: I. Model description. J. Environ. &ial.
12(4)*558-564.
Kadoum and Mock. 1978. Herbicide and insecticide residues in tailwater pits:
Water and pit bottoms from irrigated corn and sorghum fields. J. Agric.
Food Chem. 26(10:45-50. Cited in USEPA 1984e.
Keith, L.H.# H.W. Garrison, F.R. Allen, et al. 1976. Identification of
organic compounds in drinking water from 13 U.S. cities. In:
Identification and Analysis of Organic Pollutants in Water. Ann Arbor
Sciences Publishers, Inc., Ann Arbor, Michigan, p. 329-373. Cited in
USEPA 1984d.
Kelley, R., and Wnuk, M. . 1986. Little synthetic organic compound municipal
well sampling survey. Iowa Department of Water, Air, and Waste
Management, Des Moines, Iowa.
-------
Kenaga, E.E., and C.A.I. Goring. 1978. Relationship between water
solubility, soil sorption, octanol water partition coefficient and
concentrations of* chemicals in biota. In: Eaton, J.G., P.R. Parrish,
and A.C. Hendricks (eds.). Aquatic Toxicology. Proc. 3rd Ann. Symp. on
Aquatic Toxicology, Am. Soc. for Testing and Materials. New Orleans,
Louisiana, October 17-18, 1978.
Kirk-Othmer. 1979. Cyclopentadiene and dicyclopentadiene. In: Kirk-Othmer
Encyclopedia of Chemical Technology. Vol. 7, Third ed.
New York: John Wiley and Sons. pp. 425-426.
Kirk-Othmer. 1981. Insect control technology. In: Kirk-Othmer Encyclopedia
of Chemical Technology, Vol. 13, 3rd Edition. New York: John Wiley and
Sons. pp. 413-485.
Konasevich, D., W. Traversy and H. Zar. 1978. Status Report on Organic and
Heavy Metal Contaminants in the Lakes Erie, Michigan, Huron and Superior
Basins. Great Lakes Water Quality Board, p. 37. Cited in USEPA 1984d.
Kuch, P. 1986. Letter to Frederic A. Zafran, October 28, 1986, SAIC, McLean,
Virginia.
Kuch, P.J. 1980. Preliminary quantitative usage analysis of 2,4-D.
Washington, DC: Office of Pesticide Programs, U.S. Environmental
Protection Agency.
Kutz, F.W., A.R. Yobs, and H.S.C. Yang. 1976. National pesticide monitoring
programs. In: R.E. Lee (ed.). Air pollution from pesticides and
agriculture processes. Cleveland, Ohio: CRC Press, pp. 95-136.
lAskowski, D.A., C.A.I. Goring, P.J. McCall, and R.L. Swann. 1982.
Terrestrial Environment. Chapter 6. In: Conway, R.A. (ed.).
Environmental Risk Analyses for Chemicals. Van Nostrand Reinhold Co.
Lawson, E.R. 1976. 2,4,5-T residues in storm runoff from small watersheds.
J. Soil Water Conserv. 31:217-219.
Leidy, R.B., M.D. Jackson, W.A. Skroch, and T.J. Sheets. 1975. Residue
studies with silvex in apples. Bull. Environ. Contam. Toxicol.
13(3):338-341. CLted in OSEPA 1979a.
Leigh, G.M. 1969. Degradation of selected chlorinated hydrocarbon
insecticides. J. Water Poll. Oontr. Fed. 41(11)(Pt. 2):R450-R460. Cited,
in Callahan et al. 1979.
Lemley, A.T., and W.Z. Siong. 1984. Hydrolysis of aldicarb, aldicarb
sulfoxide, and aldicarb sulfone at parts per billion levels in aqueous
media. J. Agric. and Food Chem. 32(4):714-719.
Leung, SY.Th, R.V. Bulkley and J.J. Richard. 1982. Pesticide accumulation in
a new impoundment in Iowa. Water Res. Bull. 18(3)>485-493. Cited in
USEPA 1984d.
-------
Levins# P., J« Adams, P. Brenner, S. Coons, X. Thrun, and J. Varone. 1979a.
Sources of toxic pollutants found in influents to sewage treatment
plants* IV. R.R*. Clayton Drainage Basin, Atlanta
report. Rrepared-by Arthur D. Little, Inc., for Office of Hater Planning
and Standards, U.S. Environmental Protection Agency, Washington, DC. EPA
Contract No. 68-01-3857.
Levins, P., J. Adams, K» Sirun, G. Harris, and A. Wechsler. 1979b. Sources
of toxic pollutants found in influents to sewage treatment plants. II.
Muddy Creek Drainage Basin, Cincinnati, Ohio. Final Report. Prepared by
Arthur D. Little, Inc., for Office of Hater Planning and Standards, U.S.
Environmental Protection Agency, Hashingon, DC. EPA Contract No.
68-01-3857.
Lipsey, R.L. 1981. Florida statewide pesticide use survey. Gainesville,
Florida: Institute of Food and Agricultural Sciences, university of
Florida.
Lu, P.Y., Metcalf, R.L., Hirwe, A.S., and Williams, J.H. 1975. Evaluation of
environmental distribution and fate of heptachlorocyclopentadiene,
chlordene, heptachlor, and heptachlor epoxide in a laboratory model
ecosystem. J. Agric, Food Chem. 23(5):967-973. Cited in Callahan et al.
1979.
Luczak, J. 1969. Stability of methoxychlor in natural waters.
Rocz. Panstiv. Zakl. Hig. 20(2):147-154. Cited in USEPA 1984g.
Lutz, J.F., G.E. Byers, and T.J. Sheets. 1973. The persistence and movement
of picloram and 2,4,5-T in soils. J. Environ. Quality. 2(4):485-488.
Lyman, W.J., H.F. Reehl, and D.H. Rosenblatt (eds.). 1982. Handbook of
Chemical Property Estimation Methods. Environmental Behavior of Organic
Compounds. McGraw-Hill Book Company.
Mabey, W.R., et al. 1981. Aquatic fate process data for organic priority
pollutants. Produced by SRI International for Monitoring Data Support
Division, Office of Hater Regulations and Standards, U.S. Environmental
Protection Agency. EPA 440/4-81-014.
Maddy, K.T., H.F. Fong, J.A. Lowe, D.H. Oonrad, and A.S. Fredrickson. 1982.
A study of wel£./ water in selected California USA communities for residues
of 1,3-dichluropropene chloroallyl alcohol and 49 organo phosphate or
chlorinated hydrocarbon pesticides. Bull. Environ. Oontam. Technol.
29(3)t354-359, Cited in USEPA 1984g.
Manske, D.D., and P.E. Corneliussen. 1974. Pesticide residues in total diet
samples (VII). Pestic. Monit. J. 8(2)x110-124.
Manske, D.D., and R.D* Johnson. 1975. Pesticide residues in total diet
samples (VIII). Pestic. Monit. J. 9(2):94-105.
Manske, D.D., and R.D. Johnson. 1977. Pesticide residues in total diet
samples (X). In: Savage, E.P. (ed.). Environmental chemicals. Human
and animal health. Proceedings of the 5th Annual Conference, Colorado
State University, Fort Collins, Colorado, pp. 131-166.
-------
Mattraw, H.C. 1975. Occurrence of chlorinated hydrocarbon insecticides,
southern Florida, 1968-1972. Pestic. Monit. J. 9:106. Cited in USEPA
1980b.
McKenry, M.V., and I.J. Ihomason.
1,2-Dibromoethane compounds:
conditions in several soils.
1984f.
1974. 1,3-Dichloropropane and
I. Movement and fate as affected by various
Hilgardia 42:393-438. Cited in USEPA
Midwest Research Institute (MRI). 1984. Sampling and analysis for ethylene
dibromide. Preliminary data submitted January 20, 1984. EPA Contract
Ho. 68-02-3938. Cited in USEPA 1984b.
Miles, J.R.W., Tu, C.M., and HarriB, C.R. 1969. Metabolism of heptachlor and
its degradation products by soil microorganisms. J. Econ. Entoool.
62(6):1334-1338. Cited in Callahan et al. 1979.
Miles, J.R.W., Tu, C.M., and Harris, C.R. 1971. Degradation of heptachlor
epoxide and heptachlor by a mixed culture of soil microorganisms. J.
Econ. Bitomol. 64(4):839-841. Cited in Callahan et al. 1979.
Mink, J.F. 1981. DBCP and EDB in soil and water at Ruhia, Oahu, Hawaii.
Prepared for Del Monte Corp., Honolulu, Hawaii. Cited in Cohen et al.
1984.
Monsanto Agricultural Cong>any. 1986. Information to support the registration
of Lassot herbicides. EAP Reg. Nos. 524-285, 524-296, 524-314, 524-329,
524-341, 524-344. Alachlor in raw and finished drinking water derived
from surface sources from 24 community water systems located in regions
of extensive Lassot use. Compiled by S.R. Muench, Monsanto Agricultural
Company, St. Louis, Missouri.
Moss, L.P., E.J. Kirsh, R.F. Wukasch, and C.P.L. Grady. 1983.
Pentachlorophenol biodegradation. I. Aerobic Hater Res. 17:1575-1584.
Cited in USEPA 1986.
MRI. 1975. (Midwest Research Institute.) Substitute chemical program.
Initial scientific and minieconomic review of aldicarb. Prepared by
Midwest Research Institute, Kansas City, Missouri, and Office of
Pesticide Programs, U.S. Environmental Protection Agency, for Office of
Pesticide Programs, U.S. Environmental Protection Agency, Washington, DC.
EPA-540/1-75-01£.
MSU. 1984. (Mississippi State university.) Cooperative Agreement Progress
Report: Mississippi Pesticide Hazard Assessment Project. Pesticide
residue monitoring of drinking water from shallow wells in the
Mississippi Delta. Prepared by: Larry Lane, Principal Investigator,
Mississippi State Chemical Laboratory, Mississippi State University.
NAS. 1977. National Academy of Sciences. Drinking water and health. Vol.
1. Washington, DC: National Academy of Sciences.
Nelson, S., M. Iskander, M. Volz, S. Khalifa, and R. Haberman. 1981. Sci.
Total. Biviron. 21:35-40. Cited in Cohen et al* 1984.
-------
Newsome, W.H. et al. 1977. Residues of dibromochloropropane in root crops
grown in fumigated soil* J. Agric. Food Chem. (25) Cited in USOA 1978.
Norris, L.A. 1966. Degradation of 2,4-0 and 2,4,5-T in forest litter. J.
Forestry. 64:475-476.
North Carolina Department of Agriculture. 1984. Letter from Leonard p.
Blanton to Richard J. Johnson, January 10, 1984 and supplemental
submission of January 27, 19B4. Cited in USEPA 1934b.
NRCC. 1978. (National Research Council of Canada). Associate Committee on
Scientific Criteria for Environmental Quality.
NWMP. 1982. (National Surface Water Monitoring Program.) Conputer data file
printout. March 1982.
Offutt, C.K. 1984. Memorandum: Aldicarb in Rhode Island Groundwater. From:
Carolyn K. Offutt, Chief - Environmental Processes and Guidelines
Section, Exposure Assessment Branch, HED. Sent to: Paul Lapsley, Chief
- Special Review Branch, Registration Division. Office of Pesticides and
Toxic Substances, U.S. Environmental Protection Agency, Washington, DC.
Paasivinta, J., K. Heinola, T. Humppi, et al. 1985. Polychlorinated phenols,
guaiacols, and catechols in the environment. Chemaphene. 14(5):469-491.
Parejko, R., R. Johnston, and R. Keller. 1975. Chlorohydrocarbons in Lake
Superior lake trout (Savelinus namaycuBh). Bull. Environ. Gontam.
Toxicol. 14(4):480-488.
Parks, J.R. 1983. Pesticide use on fall potatoes in the United States. U.S.
Department of Agriculture. ERS Staff Report No. AGES830113. 33.pp.
Parr, J.F., and S. Skoith. 1976. Degradation of toxaphene in selected
anaerobic soil environments. Soil Sci. 121(1):52-57. Cited in Callahan
et al. 1979.
Patty. 1981. (No further information provided.) Cited in TXNT 1986.
Patty. 1982. (No further information provided.) Cited in TXNT 1986.
Pellizzari and Bunch. 1979. Ambient air carcinogenic vapors — Improved
sampling and analytical techniques and field studies. Prepared by
Research Triangle Institute, Research Triangle Park, North Carolina, for
U.S. Environmental Protection Agency. EPA-600/2-79-081.
Pellizzari, E.D., M.D. Erickson, and R.A. Zweidinger. 1979. Formulation of a
preliminary assessment of halogenated organic compounds in man and
environmental media. Prepared by Research Triangle Institute, Research
Triangle Park, North Carolina, for Office of Toxic Substances, U.S.
Environmental Protection Agency, Washington, DC. EPA-560/13-79-006.
-------
Pellizzari E.D., T. Bartwe11, H. Zelon, C. Leninger, M. Erickson, and C.
Sparacino* 1982. Total exposure assessment methodology (TEAM):
Prepilot study -- Northern New Jersey. Prepared by Research Triangle
Institute, Research Triangle Park, North Carolina, for office of Research
and Development, U.S. Environmental Protection Agency, Washington, DC.
Contract No. 68-02-3849.
Peoples, S.A., K.T. Maddy, W. Cusick, T. Jackson, C. Cooper, and A.S.
Frederickson. 1980. Bull. Environ. Contain. Toxicol. 24:611-618. Cited
in Cohen et al. 1984.
Pignatello, J.J., M.M. Martinson, J.G. Steiert, R.E. Carlson, and R.L.
Crawford. 1983. Biodegradation and photolysis of pentachlorophenol in
artifical freshwater streams. Appl. Environ. Microbiol. 46:1024-1031.
Cited in USEPA 1986.
Pinto, E. 1980. Report of groundwater contamination study in Wicomico,
Maryland. Wicomico County Health Department, State of Maryland (plus
1981-1982 addenda). Cited in Cohen et al. 1984.
Ramlit Associates, Inc. 1983. Groundwater Contamination by Pesticides: A
California Assessment. Submitted to: State Water Resources Control
Board, Sacramento, California. Submitted by: Ramlitt Associates, Inc.,
Berkeley, California. Publication No. 83-4SP.
Rao, P.S.C., and J.M. Davidson. 1980. Estimates of pesticide retention and
transformation parameters required in non-point source pollution models.
Zn: Overcash, M.R., and J.M. Davidson (eds.). Environmental Impact of
Non-Point Source Pollution. Ann Arbor Science Publishers, Inc.
Ray et al. 1983. (No further information provided). Cited in TXNT 1986.
Reinert, J.C., D.J. Severn, W.S. Rabert, E.C. Saito, M.L. Jones, and G.A.
Fristrom. 1982. The atmospheric transport of toxaphene. Washington,
DC: Office of Pesticide Programs, U.S. Environmental Protection Agency.
SAIC. 1981a. Locations where volatile organic chemicals have been detected
in water supplies. Prepared by JRB Associates/SAIC, McLean, Virginia,
for Office of Drinking Water, U.S. Environmental Protection Agency,
Washington, DC.
SAIC. 1981b* [Science Applications International Corporation (JRB)]. Review
of literature on herbicides, including phenoxy herbicides and associated
dioxins. Prepared for the Veterans Administration, Washington, DC, NTIS.
Sandhu, S.S., W.J. Warren, and P. Nelson. 1978. Pesticidal residue in rural
potable water. J. Am. Mater works Assoc. 70(1):41-45.
Savage, E.P. 1976. National study to determine levels of chlorinated
hydrocarbon insecticides in human milk: 1975-1976, and supplementary
report to the national human milk study: 1975-1976. Final report.
Washington, DC: Human Effects Monitoring Branch, U.S. Environmental
Protection Agency. kPA-540/9-78-005.
-------
Sch.-.-ht, R.A. 1974. Pesticides in the Illinois waters of Lake Michigan.
Prepared by Illinois Environmental Protection Agency, Chicago, Illinois,
for Office of Research and Development, U.S. Environmental Protection
Agency, Washington, DC. EPA-660/3-74-002.
Schafer, M.L. et al. 1969. Pesticides in drinking water. Environ. Sci. and
Technol. 3:1261.
Schepers, J.S., E.J. Vavricka, and G.E. Schuman. 1980. Agricultural runoff
during a drought period. J. WPCF 52(4). Cited in Baker et al. 1981.
Schmitt, C.J., M.A. Ribick, J.L. Ludke, and T.W. May. 1981. Organochlorine
residues in freshwater fish, 1976-1979: National Pesticide Monitoring
Program. Columbia, Missouri: Fish and Wildlife Service, U.S. Department
of the Interior.
Schultz, D.P., and P.D. Herman. 1971. Residues of 2,4-D in pond waters, mud,
and fish. Pestle. Monit. J. 8(3):173-179.
Seiber, J.N., M.P. Catahan, and C.R. Barril. 1978. Loss of carbofuran from
rice paddy water: Chemical and physical factors. J. Environ. Sci.
Health. B. B13(2):131-148. Cited in USEPA 19846.
Senn, L.H. 1983. Personal communication, College of Agricultural Sciences,
Clemson University. Cited in Cohen et al. 1984.
Sittig, M. (ed.). 1980. Pesticide Manufacturing and Toxic Materials Control
Encyclopedia. Park Ridge, New Jersey. Noyes Data Corp*
Spalding, R.F., G.A. Junk, and J.J. Richard. 1980. Pestic. Monit. J.
14(2):70-73. Cited in Cohen et al. 1984.
SRI International. 1983. 1983 Directory of Chemical Producers in the United
States of America. SRI International.
Stanley, C.W., J.B. Barney,, M.R. Helton, and A.R. Yobs. 1971. Environ. Sci.
and Technol. 5:430. Cited in Grover et al. 1976.
State of Maryland. 1983. Summary Report: Results of a Maryland ground water
herbicide survey* Fall 1983. Office of Environmental Programs.
Department of Health and Mental Hygiene.
Strachan, et al* 1980. Health and environmental effects profile of
methoxychlor. U.S. Environmental Protection Agency, Environmental
Criteria and Assessment Office, Cincinnati, Ohio. Cited in USEPA 1984g.
Strassmann, S.C., and F.W. Rutz. 1977. Insecticide residues in human milk
from Arkansas and Mississippi, 1973-74. Pestic. Monit J. 10(4):130-133.
Stumm, W., and J.J. Morgan. 1970. Aquatic Chemistry. New York: John Wiley
& Sons. p. 69.
.Swan, R.L., P.J. NcCall, and S.M. Unger. 1979. Volatility of pesticides from
soil surfaces. Unpublished manuscript. .Cited in Lyman et al. 1982.
-------
Swann, R.L.f. P McCall, and S.M. Unger. 1979. Am. Qiem. Soc. Nat. Mtg.
Pestic. Chem. Div- Abscracts. 1979. Cited in Cohen et al. 1984.
Tanaka, F.S., R.G. Wren, and E.R. Mansager. 1981. Survey for surfactant
effects on the photodegradation of herbicides in aqueous media. J.
Agric. Pood Chem. 29(2):227-330. Cited in USEPA 1984d.
Tinsley, I.J.. 1979. Chemical Concepts in Pollutant Behavior. Mew York:
Wiley-Interscience. p. 265.
Truhlar, J.F., and L.A. Reed. 1976. Occurrence of pesticide residues in four
streams draining different land-use areas in Pennsylvania, 1969-1971.
Pestic. Monit. J. 10(3):101-110.
Tucker, R.K., and T.A. Burke. 1978. A second preliminary report on the
findings of the state groundwater monitoring project. New Jersey:
Department of Environmental Protection.
TXNT (TOXNET Data Base Printout). 1986. Accessed and printed out by SA2C
(Science Applications International Corporation). October 1986.
USDA. 1978. (U.S. Department of Agriculture). The biologic and economic
assessment of DBCP. Washington, DC: U.S. Department of Agriculture.
Technical Bulletin No. 1666.
USDA. 1980a. (U.S. Department of Agriculture.) The biologic and economic
assessment of lindane. Washington, DC: U.S. Department of Agriculture.
Technical Bulletin No. 1647.
USDA. 1980b. (U.S. Department of Agriculture.) The biologic and economic
assessment of pentachlorophenol, inorganic arsenicals, creosote. Volume
II: Non-wood-preservatives. Washington, DC: U.S. Department of
Agriculture. Technical Bulletin No. 1658-11.
USDA. 1983. (U.S. Department of Agriculture). Inputs. Outlook and
situation. Washington, DC: Economic Research Service, U.S. Department
of Agriculture. IOS-2.
USDA. 1984a. (U.S. Department of Agriculture.) Letter of January 23, 1984
from John w. Marshall to Dr. John A. Moore, DVM. Cited in USEPA 1984b.
USDA. 1984b. (U.S. Department of Agriculture.) Letter of January 31, 1984
from Kenneth A. Giles to Dr. R.D. Schmitt. Cited in USEPA 1984c.
USDA. 1984c. (U.S. Department of Agriculture.) Monitoring phase biological
residue reports for calendar years 1980-1983. Computer printout from
Food Safety and Inspection Service, U.S. Department of Agriculture,
Washington, DC.
USDA. 1984c. (U.S. Department of Agriculture.) Monitoring phase:
biological residue reports for calendar years 1980-1983. Congputer
printout from Food Safety and Inspection Service, U.S. Department of
Agriculture, Washington, DC.
-------
USEPA. 1975a* (U.S. Environmental Protection Agency.) Analytical report:
New Orleans water supply study. Region VI, U.S. Environmental Protection
Agency. EPA-906/9-75-003. Cited in Pellizzari et al. 1979.
USEPA. 1975b. (U.S. Environmental Protection Agency.) Preliminary
assessment of suspected carcinogens in drinking water. Washington, DC:
Office of Toxic Substances, U.S. Environmental Protection Agency.
USEPA. 1975c. (U.S. Bivironmental Protection Agency..) Survey of organic,
metal, and other inorganic parameter concentrations in selected Region V
drinking water supplies. Preliminary report. Chicago, Illinois: Water
Division and Surveillance and Analyses Division, U.S. Environmental
Protection Agency.
USEPA. 1976. (U.S. Environmental Protection Agency.) EPA actions to cancel
and suspend uses of chlordane and heptachlor as pesticides—Economic and
social implications. Washington, DC: Office of Pesticide Programs, U.S.
Environmental Protection Agency. EPA-540/4-76-004.
USEPA. 1977a. (U.S. Environmental Protection Agency.) Lindane: Position
document 1. Washington, DC: Special Pesticide Review Division, U.S.
Environmental Protection Agency.
USEPA. 1977b. (U.S. Environmental Protection Agency.) Toxaphene: Position
document 1. Washington, DC: Special Pesticide Iteview Division, U.S.
Environmental Protection Agency. EPA/SPRD-80/55.
USEPA. 1979a. (U.S. Environmental Protection Agency.) Silvex: Position
document 1/2/3. Preliminary determination concerning a rebuttable
presunqption against registration of pesticide products containing
2-(2,4,5-trichlorophenoxy)propionic acid (silvex). Washington, DC:
Special Pesticide Review Division, U.S. Environmental Protection Agency.
EPA/SPRD-80/52.
USEPA. 1979b. (U.S. Environmental Protection Agency.) Notice of intent to
cancel certain registrations of pesticide products containing silvex;
suspension order for silvex. Washington, DC: S&ecial Pesticide Review
Division, U.S. Environmental Protection Agency. EPA/SPRD-80/102.
USEPA. 1980a. (U.S. Environmental Protection Agency.) Ambient water quality
criteria fax,chlordane. Washington, DC: Office of Water Regulations and
Standards. EPA-440/5-80-027.
USEPA. 1980b. (U.S. Bivironmental Protection Agency.) Ambient water quality
criteria for toxaphene* Washington, DC: Office of Water Regulations and
Standards# U.S. Bivironmental Protection Agency. KPA-440/5-80-076.
USEPA. 1980c. (U.S. Environmental Protection Agency.) Fate and transport of
hazardous constituents. RCRA Subtitle C background document. Appendix
B. Washington, DC: Office of Solid Waste.
USEPA. 1980d. (U.S. Bivironmental Protection Agency.) Interim report:
Monitoring methods development in the Beaumont-Lake Charles area.
Washington, DC: Office of Research and Development, U.S. Environmental
Protection Agency. EPA-600/4-80-046.
-------
USEPA. 1980e. (U.S. fiivironmental cterrMon Agency.) Lindane position
document 2/3. Washington, DC: Office of Pesticide Programs, U.S.
Environmental Protection Agency.
USEPA. 1980f• (U.S. environmental Protection Agency.) Ambient water quality
criteria for heptachlor. Washington, D.C.: Office of Water Regulations
and Standards, U.S. Bivironmental Protection Agency. EPA-440/4-80-052.
USEPA. 1962a. (U.S. Environmental Protection Agency.) EPA index to
pesticide chemicals. Alachlor. Washington, DC: Office of Pesticide
Programs, U.S. Environmental Protection Agency.
USEPA. 1982b. (U.S. environmental Protection Agency.) EPA index to
pesticide chemicals. Aldicarb. Washington, DC: Office of Pesticide
Programs, U.S. Environmental Protection Agency.
USEPA. 1982c. (U.S. Environmental Protection Agency.) Toxaphene: Decision
document and Federal Register notice. Washington, DC: Office of
Pesticide Programs, U.S. Environmental Protection Agency.
EPA-540/9-82-027.
USEPA. 1982d. (U.S. Environmental Protection Agency.) 2,4-D fact sheet.
Washington, DC: U.S. Environmental Protection Agency.
USEPA. 1983a. (U.S. Environmental Protection Agency). Aldicarb (Temik) in
groundwater/citrus. Issue paper dated February 10, 1983, obtained from
Office of Pesticide Programs, U.S. Environmental Protection Agency,
Washington, DC.
USEPA. 1983b. (U.S. environmental Protection Agency.) Analysis of the risks
and benefits of seven chemicals used for subterranean termite control.
Washington, DC: Office of Pesticide Programs, U.S. environmental
Protection Agency. EPA-540/9-83-005.
USEPA. 1983c. (U.S. Environmental Protection Agency.) Ethylene dibromide
(EDB) position document 4. Washington, DC: Office of Pesticide
Programs, U.S. Environmental Protection Agency.
USEPA. 1983d. (U.S. environmental Protection Agency.) Lindane position
document 4. Washington, DC: Office of Pesticide Programs, U.S.
Environmental Protection Agency.
USEPA. 1984a* (U.S. Environmental Protection Agency.) EDB facts. EPA
decision 2/3/84. Washington, DC: Office of Public Affairs, U.S.
Environmental Protection Agency.
USEPA. 1984b. (U.S. environmental Protection Agency.) Ethylene dibromide
(EDB). Scientific support and decision document for grain and grain
milling fumigation uses. Washington, DC: Office of Pesticide Programs,
U.S. Environmental Protection Agency.
USEPA. 1984c. (U.S. environmental Protection Agency.) Ethylene dibrcmide
(EDB). Technical support document for quarantine fumigation uses.
Washington, DC: Office of Pesticide Programs, U.S. environmental
Protection Agency.
-------
USEPA. 1984d. (U.S. Environmental Protection ^ency.) Health and
environmental effects profile for alachlor. Research and Development.
Prepared for Office of Solid Waste and Emergency Response, by
Environmental Criteria and Assessment Office, U.S. Environmental
Protection Agency. June 1984.
USEPA 1984e. (U.S. Environmental Protection Agency.) Health and
environmental effects profile for carbofuran. Prepared for Office of
Solid Waste and finergency Response, Washington, DC, by Experimental
Criteria and Assessment Office, Cincinnati, Ohio.
USEPA. 1984f• (U.S. Environmental Protection Agency.) Health and
environmental effects profile for 1,2-dibrcmoethane. Environmental
Criteria and Assessment Office. Research and Development. August 1984.
USEPA. 1984g. (U.S. Bivironmental Protection Agency.) Health and
environmental effects profile for methoxychlor. Office of Research and
Development. Prepared by Environmental Criteria and Assessment Office,
for Office of Solid Waste and Baergency Response, U.S. Environmental
Protection Agency. September 1984.
USEPA. 1984h. (U.S. Environmental Protection Agency.) Proposal for comment
on intent to cancel use of dibromochloropropane in Hawaii.
46 FR 19594. Washington, DC: U.S. Environmental Protection Agency.
USEPA. 1984i« (U.S. Bivironmental Protection Agency.) Rural water survey.
Computer data provided by Department of Sociology, Cornell University,
Ithaca, New York.
USEPA. 1985a. (U.S. Bivironmental Protection Agency.) Alachlor Health
Advisory. Draft. Washington, DC: Office of Drinking Water, U.S.
Environmental Protection Agency.
USEPA 1985b. (U.S. Environmental Protection Agency.) Carbofuran Health
Advisory. Draft. Office of Drinking Water. U.S. Environmental
Protection Agency.
USEPA. 1985c. (U.S. Bivironmental Protection Agency.) Lindane Health
Advisory. Draft. Office of Drinking Water, U.S. Environmental
Protection Agency;
USEPA. 1985d. (U.S. Environmental Protection Agency.) Toxaphene Health
Advisory.- Draft. Office of Drinking Water, U.S. Environmental
Protection Agency.
USEPA. 1985e. (U.S. Bivironmental Protection Agency.) Methoxychlor Health
Advisory. Draft. Office of Drinking Water* U.S. Environmental
Protection Agency.
USEPA. 1986. (U.S. Environmental Protection Agency.) Health and
environmental effects profile for pentachlorophenol. Office of Research
and Development. Prepared for Office of Solid Waste and Emergency
Response by Office of Health and Environmental Assessments, Cincinnati,
Ohio.
-------
OSGS. 1983. (U.S. Geological Survey.) Quality of rivers ->f the United
States, 1976 water year - Based on the National Stream Quality Accounting
Network (NASQAN). Prepared by L.J. Britton, E. Goddard, and J.C. Briggs.
U.S. Geological Survey Report 80-594.
Velsicol Chemical Company. No date. Heptachlor 2EC. Product profile.
Chicago, Illinois.
Verschueren, K. 1983. Handbook of Environmental Data on Organic Chemicals.
Second Edition. V&n Nostrand Reinhold Company.
Waggoner, P.E. 1985. Memorandum from the Connecticut Agricultural Experiment
Station on Pesticides in Connecticut Groundwater. P. Gough, (ed.) News
of Science, New Haven, Connecticut.
Wall, G.T., L.P. Wilding, and N.E. Skueck. 1978. Physical, chemical, and
mineralogical properties of fluvial consolidated bottom sediments in
Northwestern Ohio. J. Environ. Qual. 7(3):319-325.
Wallace, L.A., E. Pellizzari, T. Hartwell, M. Rozenzweig, N. Erickson,
C. Sparacino, and H. Zelon. 1984. Personal exposure to volatile organic
compounds. Biviron. Research. 35:293-319.
Warry and Chan. 1981. (No further information provided.) Cited in TXNT
1986.
Washington State Department of Social and Health Services. 1985. results and
implications of the investigation of ethylene dibromide in groundwater in
western Washington. February 1985. Water Supply and Waste Section,
Washington State Department of Social and Health Services.
Watanabe, I. 1973. Decoag>osition of pesticides by soil microorganisms -
Special emphasis on flooded soil condition. JARQ. 7(1):15-18.
Wauchope, R.D. 1978. The pesticide content of surface water draining from
agricultural fields. A review. J. Environ. Quality. 7(4):459-472.
Wedberg, J.L., S. Moore, F.J. Amore, and H. McAvoy. 1978. Organochlorine
insecticide residues in bovine milk and manufactured milk products in
Illinois, 1971-1976.' Pestic. Monit. J. 11(4):161-164.
Wiersma, G.B«, H. Tai, and P.F. Sand. 1972. Pesticide residue levels in
soils# fiscal year 1969, National Soils Monitoring Program. Pestic.
Monit. J. 6(3)(194-228. Cited in USEPA 1984g.
Windholz, M. (ed.)* 1976. the Merck Index. Ninth Edition. Rahway, New
Jersey: Merck and Company, Inc.
Wolfe, N.L., R.G. Zepp, G.L. Baughman, R.C. Fincher, and J.A. Gordon. 1976.
Chemical and photochemical transformation of selected pesticides in
aquatic systems. Office of Research and Development, U.S. Environmental
Protection Agency. EPA-600/3-76-067. Cited in Callahan et al. 1979.
-------
Wolfe, N.L., R.G. Zepp, G.L. Baughman, and R.C. Hollis. 1977. Methoxychlor
and DDT degradation in water: rates and products. Environ. Sci. and
Technol. 11:1077-1081. CLted in USEPA 1984g.
Wong, L., P. Byre, J. Mink, A. Hepton, J. Hylin, K. Chenchin, T. Arizumi, and
others contributing to a project analyzing soil and groundwater in Oahu
for EDB during the summer of 1983* Department of Health and Agriculture,
U.S. Geological Survey, and Dole, Del Monte, and Maui Pineapple
companies. Cited in USEPA 1983c.
Woodham, D.W., M.C. Ganyard, C.A. Bond, and R.G. Reeves. 1975. Monitoring of
agricultural pesticides in a cooperative pest management project in North
Carolina, 1971, first year of study. In: Savage, E.P. (ed.).
Environmental chemicals. Human and animal health. Proceedings of 4th
Annual Conference, Oolorado State University, Fort Collins, Colorado.
Zaki, M.H., D. Mor&n, and D. Harris. 1982. Am. J. Public Health
72:1391-1395. Cited in Cohen et al. 1984.
Zalkin, F., M. Wilkerson, and R.J. Oshima. 1983. Pesticide movement to
groundwater, Volume II. Pesticide contamination in the soil profile at
DBCP, EDB, simazine, and carbofur an application sites. Sacramento,
California: California Department of Food and Agriculture. Cited in
Cohen et al. 1984.
Zalkin, F., R.J. Oshima, R. Sava, and M. Wilkerson. In progress. Pesticide
movement to groundwater. Sacramento, California: California Department
of Food and Agriculture. Cited in USEPA 1983c.
Zepp, R.G., N.L. Wblfe, J.A. Gordon, and B.C. Fincher. 1976. Light-induced
transformations of methoxychlor in aquatic systems. J. Agric. Food Chem.
24(4):727-733. Cited in USEPA 1984g.
Zepp, R.G., N.L. Wolfe, J.A. Gordon, and G.L. Baughman. 1975. Dynamics of
2,4-D esters in surface waters. Hydrolysis, photolysis, and
vaporization. Biviron. Sci. and Technol. 9(13):1144-50.
-------
APPENDIX A. METHODOLOGIES EMPLOYED FOR CALCULATING
ENVIRONMENTAL FATE PARAMETERS
Halt-life Computations
The following equation was used to calculate the half-life for removal of
a chemical from a given medium due to a single first order or pseudo first
order rate process i:
tl/21 = In 2/Ki (A-l)
where
tx/Jl = half-life in units of time due to removal by process i
= first order or pseudo first order rate constant for the removal of
the chemical from the given medium by process i.
The following equation was used to calculate the overall half-life for removal
of a chemical from a given medium due to all first order and pseudo first
order rate processes for which a rate constant is known or estimated (Mabey et
al. 1981):
t1/2 = In 2/pc. (A-2)
Volatilization From Surface Waters
Methods for estimating volatilization half-lives for chemicals in surface
water are reviewed by Lyman et al. (1982).
If the concentration of a chemical in the water column is approximately
uniform and nuch greater than the concentration.of the chemical in air at the
air/water interface, the rate of volatilization will approximately follow
first order kinetics so that equation (A-l) can be used to estimate volatil-
ization half-lives (Lyman et al. 1982). The use of equation (A-l) to estimate
volatilization half-lives requires the computation of a volatilization rate
constant (Kv).
A-l
-------
The following equation was used to estimate Ky values for a given
chemical (Lyman et al. 1982).
(A—3)
where
Kv = volatilization rate constant in hr"1
H' = dimensionless Henry's constant
kq 3 gas phase exchange coefficient in cm/hr
k1 = liquid phase exchange coefficient in cm/hr
Z' = mean water depth in cm
The following equations were used to estimate kg, kx for wind velocities
<1.9 m/s and kL for wind velocities between 1.9 m/s and 5.0 m/s, respectively.
(Southworth 1979 in Lyman et al. 1982):
k = 1137.5
<3
wind curr
+ V
(A-4).
k. = 23.51 V0,969/Z °"613 J32/M
1 curr
curr
(A-5)
kx = 23.51 v0^969/z0 . 673 J32m e Q 526 (vwind-1.9)
(A-6)
where
kg s gas phase exchange coefficient in cm/hr
kx 3 liquid phase exchange coefficient in cm/hr
Vuind > mean near surface wind velocity in m/s
Vcurr o mean water current in m/s
Z t> mean water depth in m
Equations (A-4) through (A-6) can be used to compute k? and k; values in
lakes as well as rivers and streams since lakes also have currents due to
-------
stream inflows, temperature gradients and wind (Reid and Vood 1976). Hovever,
lake currents are generally not determined as often as river or stream
currents and are therefore frequently not available for incorporation into
equations (A-4) through (A-6). Therefore, the following equations derived by
Liss and Slater (1974) for the air/sea interface were used in place of
equations (A-4) through (A-6) to estimate kg and kx for turbulent lakes:
kg a gas phase exchange coefficient in cra/hr
kx » liquid phase exchange coefficient in cm/hr
M s molecular weight
However, since the air/sea interface is generally more turbulent than most
air/lake interfaces,, the values of k? and k1 computed from equations (A-7) and
(A-8) may be too high, and the corresponding estimated volatilization half-
lives too low for most lakes (Lyman et al. 1982).
Width and depth averaged river and stream currents can range from
<0.1 m/s under stagnant conditions to >10 m/s under storm runoff conditions.
However, under non-runoff conditions, most mean river and stream currents fall
within the range of 0.2 m/s to 2 m/s (Reid and Vood 1978) which is the range
SAIC assumed for computational purposes. Wind speeds can range from <0.1 m/s
under stagnant conditions to >30 m/s in storms. Hovever, under non-storm con-
ditions, mean near surface wind velocities typically fall within the range of
0.5 m/s to 5 at/s (Perkins 1974) which is the range SAIC assumed for compu-
tational purposes. The mean depths of the Great Lakes range from 17 m for
Lake Erie to 148 m for Lake Superior and may exceed 10 m in some sections of a
few large rivers (Reid and Vood 1978). However, the mean depths of most
rivers, streams, and lakes appear to range from'1 m to 10 m (Reid and Vood
1978) which is the range SAIC assumed for computational purposes.
(A-7)
kl = 20 J44/M
(A-8)
where
A-3
-------
Volatilization half-lives of chemicals in surface waters decrease with
decreasing depth, increasing water current and increasing wind speed. To
calculate a theoretical range of typical volatilization half-lives for a given
chemical in rivers, streams and turbulent lakes, SA1C did the following:
o Haximum k and kx values for the volatilization of a given chemical
from rivers and streams,, were estimated by substituting a water depth
of 1 m, a current of 2 m/s for river and streams or 0.2 m/s for lakes,
a wind speed of 5 m/s and the molecular weight into equations
-------
where
t. 3 volatilization half-life of the chemical on the soil surface in
' days
Koc = organic carbon normalized soil/water equilibrium partition
coefficient for the chemical and soil of interest
S 3 aqueous solubility of the chemical in mg/1
VP = vapor pressure of the chemical in torr
Equation (A-9) is only applicable for chemicals applied to and present at the
soil surface. In addition, since equation (A-9) vas empirically derived from
data on chemicals with vapor pressures much less than 1 torr, it vas not used
to predict the volatilization half-lives at the soil surface of chemicals vith
vapor pressures exceeding 1 torr.
Sorption and Biomagnification
The equilibrium binding isotherms for the adsorption of organics in
aqueous solutions at lov concentrations to sediments or soil are frequently
linear vith near zero intercepts and can be represented by the following
equation (Lyman et al. 1982):
x/n> - Ks/w C., (A-10)
where
x/m = ug of chemical adsorbed per g of sediments or soil
K = sediment or soil/vater equilibrium partition coefficient
S / W
C#9 - aqueous equilibrium concentration of the chemical in ug/ml (mg/1)
Although the units of K /(J are ml of aqueous solution lg of sediment or soil,
8 / w
K. essentially can be assumed to be unitless since the density of an aqueous
9 /
solution is generally approximately 1 g/ml.
A-5
-------
SAIC used the following equation to predict sediment or soil/vater
equilibrium partition coefficients for a given chemical and sediments or soils
with organic carbon fractions ranging from 0.01 to 0.08 (Lyman et al. 1982):
* K.c E.c <*-">
where
Ks/w = sediment or soil/water equilibrium partition coefficient
Koc = organic carbon normalized sediment or soil/vater equilibrium
partition coefficient
foc = organic carbon fraction of sediments or soil
Koc values were estimated from the octanol/vater partition coefficient and
aqueous solubility in Section 1). Equation (A-ll) may not be suitable for
estimating Ks/U values for chemicals adsorbed to sediments or soils with foc
<0.01 or £oc >0.08 (as cited in Lyman et al. ).
The sediment or soil/vater equilibrium partition coefficient (K can
9 / V
be used to qualitatively predict the behavior of a chemical with respect to
adsorption to bottom and suspended sediment, and leaching from soil. However,
values of K should be used with caution. As long as the aqueous equil-
S / w
ibrium concentration of a chemical is low enough so that its binding isotherm
can be represented by equation (A-10), the value of Kg/W for the chemical and
a given sediment or soil will give the ratio at equilibrium of the ug of
chemical adsorbed per g of sediment or soil to the aqueous equilibrium
concentration of the chemical in wg/ml (mg/1). However, at higher aqueous
equilibrium concentrations and/or in the presence of high concentrations of
other organics which compete with the chemical of interest for binding sites,
the equilibrium binding isotherm for the chemical may become non-linear with a
decreasing slope and may even decrease to a zero slope if the binding capacity
of the sediment or soil is exceeded. In such regions, the value of will
over-estimate the ratio at equilibrium of the chemical concentration in the of
sediment or soil to the aqueous equilibrium concentration of the chemical*. In
addition, if equilibrium has not been obtained, the value of Kj/w will be
A-6
-------
greater than the actual ratio in a non-equilibrium system in vhich the con-
centration of the chemical in solution is still decreasing due to adsorption
and vill be less than the ratio in a non-equilibrium system in vhich the
concentration of the chemical in solution is still increasing due to
deadsorption (leaching).
It should also be pointed out that at equilibrium, the ratio of the mass
of chemical adsorbed to suspended and bottom sediment to the mass of chemical
dissolved in the water column is given by:
R a ratio of the mass of chemical adsorbed to suspended and bottom
sediment to the mass of chemical dissolved in the vater column
K 3 sediment/vater equilibrium partition coefficient
S / V
Mb = mass of suspended sediment and exposed bottom sediment
M = mass of water
W
In most aquatic systems, Mb/Mw appears to be <10"3 (USGS 1983). Therefore, in
such systems, K„/w would have to be >102 for Rg/w to be >10_1.
As previously mentioned, the soil/water equilibrium partition coefficient
can be used to qualitatively predict the susceptibility of a compound to
leaching from the soil. The following equation can be used to estimate Rf
values for the soil thin layer chromatography (TLC) of organics (Hamaker in
Hague and Freed 1975):
R , = K . (M /M )
s/v »/wvs n '
(A-12)
where
(A-13)
-------
where
Rf a estimated soil thin layer chromatography Rf value
K = organic carbon normalized soil/water equilibrium partition
oc coefficient
f = organic carbon fraction of soil
OC
dg = density of soil in g/cm3
6 s soil pore fraction.
Based on observed correlations between soil TLC Rf values and the
susceptibility of compounds to leaching from soil, Hilling and Turner (1968)
as cited by Hamaker in Hague and Freed (1975) defined the following five
mobility classes for the leaching of organics from a soil with f = 0.014,
p = 2.5 g/cm3 and 0= 0.5 based upon the magnitude of the Rf value:
Hobility Class Rf Range
1. (Very immobile) 0 - 0.99
2. (Modertely immobile 0.1 - 0.34
3. (Intermediate mobility) 0.35 - 0.64
4. (Moderately mobile) 0.65 - 0.89
5. (Very mobile) 0.90 - 1.00
An experimental bioconcentration factor (BCF) for a given chemical and
aquatic organism is equal to the steady ratio of the concentration of the
chemical in the organism in tig chemical per g of tissue to the chemical
concentration in aqueous solution in ug/ml (ntg/1). Bioconcentration factors
generally increase with increasing lipid content for similar aquatic species.
The bioconcentration factors that are estimated from the octanol/water
partition coefficient and aqueous solubility are computed from equations
derived froo'data on a number of different aquatic organisms with different
lipid contents and therefore do not represent any specific organism or lipid
content. However, they can be used to qualitatively predict the bioconcen-
tration potential of aquatic organisms in general and the general potential
for biomagnification. in aquatic ecosystems. Stern and Valker (1978) in ASTH
(1978), indicate that compounds with a BCF value below 100, between 100 and
1,000 and above 1,000 probably have low, moderate, and high bioconcentration
A-8
-------
potentials, respectively. Since the potential for biomagnification is
dependent upon the b-ioconcentration potential, the ranges of BCF values listed
above can also be-"Used to predict biomagnification potential.
Abiotic Degradation: Hydrolysis, Photooxidation, Photolysis
SAIC used the following equation to estimate pseudo first order
hydrolysis rate constants for chemicals in water as a function of pH (Lyman et
al. 1982)
K, . k, 10"'" » . kB 10'"
where
kH = pseudo first order hydrolysis rate constant in time'1
kA = second order acid catalyzed rate constant
kR o first order rate consatant for neutral hydrolysis
kB a second order base catalyzed rate constant
pH = log (H+ J
1^, = [H+j {OH" J
Values of kA, kRt and kB were taken from the literature. The pH of most
natural waters is within the range of 5 to 9 (Bouwer 1978; Stumm and Morgan).
Therefore, for any chemical with known or estimated values of kA, kn, and kB,
pseudo first order hydrolysis rate constants at pHs of 5, 6, 7, 8 and 9 were
estimated from equation (A-14).
SAIC used the following equation to estimate pseudo first order photo-
oxidation rate constants for chemicals in shallow sunlit surface waters (Hill
1980 in Mabey et al. 1981):
» kio2 <10~iaM> + (10~9m)
where
Koj| = pseudo first order photooxidation rate constant (hr-1)
K = second order rate constant for oxidation by sunlit oxygen
r02 (M hr )
A-9
-------
» assumed concentration of sunlit oxygen in shallow sunlit
surface vaters
- second order rate constant for oxidation by peroxy radicals
(M hr )
- assumed concentration of peroxy radicals in shallov sunlit
surfce vaters.
Hethods for estimating direct photolysis rate constants for chemicals in
aqueous solution are reviewed by Lyman et al. (1982). The magnitude of the
aqueous photolysis rate constant for a given chemical depends upon the
intensity of solar radiation, the attenuation of solar radiation by the water
body, and the magnitudes of the absorption coefficients for the chemical at
wavelengths above 290 nm (Lyman et al. 1982). Due to time restrictions and a
lack, of sufficient data on the magnitudes of aborption coefficients at wave-
lengths above 290 no, SAIC did not attempt to estimate direct photolysis rate
constants.
(10"12M)
Ko
(10_,M)
Biodegradation
SAIC used the following equation to estimate pseudo first order rate
constants for the biodegradation of organics in natural vaters (Nabey et al.
1981):
K„ o kBiol [bacteria cells] (A-16)
where
K a pseudo first order biodegradation rate constant in
water (hr" )
kBlet • second order biodegradation rate constant in vater
(ml cell" hr" )
[bacteria calls] » concentration of bacteria cells (cells/ml).
Mabey et al. (1981) divided numerous organics into five groups according
to their known or estimated susceptibility to biodegradation in vater and
assigned an estimated aqueous second order biodegradation rate constant kBi((1
to each of the five groups. SAIC used#these estimated kBiol values to compute
aqueous pseudo first order rate constants from equation (A-16).
A-10
-------
Baughman and Lassiter (1978) in Lyoan et al. (1982) suggest the use of
the following active bacteria concentrations (in the absence of data) to
compute aqueous first order biodegradation rate constants: 10 cell/ml for
oligotrophic lakes, 104 cells/ml for streams« and 10* cells/ml for ponds and
eutrophic lakes, representing a range of 10/cell to 109/cell for natural
surface vaters. McCabe et al. (1970) in Bouver (1978) reported 17 percent of
the groundwater samples from 621 veils had aerobic bacteria plate counts of
0/ml, 60 percent had counts betveen 1/ml and 100/ml and 7 percent had counts
exceeding 1,000/nl, representing a non-zero range of 1 cell/ml to > 1,000
cells/ml. SAIC assumed the range of active bacteria concentrations in surface
vaters and groundvaters was 10 cell/ml to 10s cell/ml, and 1 cell/ml to 103
cell/ml, respectively for computational purposes.
A-ll
------- |