Transformations of ethylene dibromide in aqueous systems and groundwater

MISSING IMAGE

Material Information

Title:
Transformations of ethylene dibromide in aqueous systems and groundwater
Physical Description:
viii, 190 leaves : ill. ; 29 cm.
Language:
English
Creator:
Weintraub, Randy Alan, 1958-
Publication Date:

Subjects

Subjects / Keywords:
Dissertations, Academic -- Food Science and Human Nutrition -- UF
Food Science and Human Nutrition thesis Ph. D
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 179-188).
Statement of Responsibility:
by Randy Alan Weintraub.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001563047
oclc - 22711568
notis - AHH6773
sobekcm - AA00004787_00001
System ID:
AA00004787:00001

Full Text











TRANSFORMATIONS OF ETHYLENE DIBROMIDE IN AQUEOUS SYSTEMS
AND GROUNDWATER














By

RANDY ALAN WEINTRAUB


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1989















ACKNOWLEDGMENTS

For valuable advice and guidance on this work, I thank

Dr. H. Anson Moye, Chairman of my Supervisory Committee, and

the other members of the Committee: Dr. Willis B. Wheeler,

Dr. Ross D. Brown, Jr., Dr. Joseph J. Delfino, and Dr. Lori

0. Lim.

I appreciate the knowledgeable assistance from Dr.

Gabriel Bitton, Dr. John P. Toth in collecting mass spectra,

and Kenneth M. Jones in advising me on organic synthesis. I

thank Dr. Barbara Ameer for editorial advice and assistance,

Mrs. Elaine M. Peeples for construction of tables of data,

and Ms. Susan Scherer for professional drawings. Collection

of groundwater samples was made possible through

collaboration with Mr. Dennis Bacon and Mr. Geoffrey Herr,

hydrogeologists with the Water Quality Division of the South

Florida Water Management District. I also thank Dr.

Geoffrey Watts of the Florida Department of Environmental

Regulation for his helpful discussion and interest in the

reactions of ethylene dibromide and hydrogen sulfide.















TABLE OF CONTENTS


page

ACKNOWLEDGMENTS.................. ....................... ii

KEY TO SYMBOLS AND ABBREVIATIONS....................... V

ABSTRACT............................................... vii

INTRODUCTION........................................... 1

LITERATURE REVIEW...................................... 5

Groundwater Contamination by Pesticides............. 5
Ethylene Dibromide................................... 7
Commercial Use...................................... 7
Toxicity............................................. 8
Chemical and Physical Properties................... 11
Ethylene Dibromide in Florida
Groundwater and Soil............................. 11
Sorption to Soil.................................... 17
Degradation in Soil and Water...................... 19
Risk Factors for Groundwater Contamination......... 27

MATERIALS AND METHODS.................................. 29

Materials................................................ 33
Methods............................................... 33
Ethylene Dibromide and Product
Identification and Quantitation
by Gas Chromatography and
Spectrophotometry................................ 33
Liquid Scintillation Counting of
C-ethylene Dibromide........................... 37
Gas Chromatography and Mass Spectrometry
for Sulfur Product Identification................ 38
Oxygen, Redox, and pH Measurements................. 39
Hydrolysis Reactions............................... 40
Kinetics: Temperature and pH Dependence........... 40
Product Identification............................ 43


iii











page


Reactions with Aqueous Hydrogen Sulfide........... 44
Kinetics: Temperature and pH Dependence.......... 44
Product Identification.......................... 48
Ethylene Dibromide in
Sulfate Groundwater............................ 48
Organic Synthesis of a Sulfur Product............ 50

RESULTS............................ ...................... 51

Hydrolysis Reactions................................ 51
Verification of Assay for Hydrolysis
Products.................................... .......... 51
Kinetics: Temperature and pH Dependence.......... 51
Product Identification........................... 67
Reactions with Aqueous Hydrogen Sulfide............ 77
Kinetics: Temperature and pH Dependence.......... 77
Product Identification........................... 108
Ethylene Dibromide in
Sulfate Groundwater............................ 129

DISCUSSION............................................. 138

Analytical Methods... .................... ........... 138
Hydrolysis Reactions................................. ..138
Reactions with Aqueous Hydrogen Sulfide............ 141

SUMMARY AND CONCLUSIONS............................... 150

APPENDIX A ANALYSES OF GROUNDWATER FOR
HYDROLYSIS STUDIES....................... 154

APPENDIX B ANALYSES OF GROUNDWATER FOR HYDROGEN
DEGRADATION SULFIDE STUDIES.............. 156

APPENDIX C MASS SPECTRA OF SULFUR-CONTAINING COMPOUNDS
IN STUDY............ .. ..................... 158

APPENDIX D MICROTOX ASSAY EVALUATION OF PRODUCTS.... 175

REFERENCES.............................. ................ 179

BIOGRAPHICAL SKETCH........... .......................... 189















KEY TO SYMBOLS AND ABBREVIATIONS

ACS American Chemical Society

aq aqueous phase

BDS butyldisulfide

"C degrees Celsius

14C- carbon 14 isotope labeled chemical

ca. approximately

CHT cyclohexylthiol

cm centimeter

cpm counts per minute

d day

dpm disintegrations per minute

DT 1,4-dithiane

DTT 1,3-dithiolane-2-thione

EDB ethylene dibromide

EPA Environmental Protection Agency

ET 1,2-ethanedithiol

eV electron volts

g gram

hr hour

H2S hydrogen sulfide

i.d. inside diameter

"K degrees Kelvin















KEY TO SYMBOLS AND ABBREVIATIONS
(Continued)

L liter

M mole per liter

m meter

mCi microcurie

m/e mass-to-charge ratio

min minute

mL milliliter

mol mole

ND none detected

pg picogram

T temperature

TT 1.3.5-trithiane

TTCD 1,4,7,10-tertathiacyclododecane

TTCH 1,2,5-trithiacycloheptane

TTCO 1,2,5,6-tetrathiacyclooctane

TTCP 1,2,3-trithiacycolpentane

U- uniformly radiolabeled chemical

Ag microgram

AM micromole

yr year















Abstract of Dissertation Presented to the Graduate School
at the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

TRANSFORMATIONS OF ETHYLENE DIBROMIDE IN AQUEOUS SYSTEMS
AND GROUND WATER

By

Randy Alan Weintraub

December 1989

Chairman: Hugh Anson Moye, Ph.D.
Major Department: Food Science and Human Nutrition

Registration of a chemical for use as a pesticide today

requires thorough study of the compound's environmental fate

in water, soil, and air to determine if classification re-

stricting its use is needed. During the 30-year history of

"successful" soil fumigation with ethylene dibromide (EDB,

1,2-dibromoethane), its environmental fate was not closely

studied. This nematicide was not recognized as environ-

mentally persistent or a public health risk until 1983, when

its agricultural use was suspended in the United States.

In this research, the transformations of EDB were

examined both in buffered aqueous solutions and in ground-

water from several locations in Florida. Experiments were

conducted between 25 and 700C to establish temperature-

dependent rate constants and to expedite the reactions.

The ethylene dibromide and products were analyzed by packed


vii








and capillary gas chromatography (GC), liquid scintillation

counting, and a colorimetric assay. The EDB is converted to

ethylene glycol and bromide ion in aqueous solutions and in

abiotic aerobic groundwaters. The rate of degradation of

EDB is dependent upon temperature but independent of pH

between 5 and 9. Extrapolating experimental results to the

average ground- water temperature in Florida of 220C, the

degra- dation half-life of EDB in the tested groundwaters is

1.5 to 2 years.

An alternative transformation of EDB occurs under

anaerobic reducing conditions in groundwater where aqueous

hydrogen sulfide (H2S(aq ) is commonly present. Transfor-

mation by H2S( was not previously recognized as a poten-

tially important reaction with pesticides. The rate of

degradation of EDB under these conditions is increased and

dependent upon temperature, pH, and H2S( ) concentration.

The sulfur-containing products are 1,2-ethanedithiol and

saturated cyclic rings of one, two or four ethylene units

and two, three, or four sulfur atoms. Their structural

identity was elucidated with mass spectrometry (MS) and GC

retention times of products relative to known analytical

standards. One commercially unavailable cyclic disulfide

compound was synthesized and used to positively confirm the

MS and GC findings. Toxicity assessment of sulfur-products,

by Microtoxm screening assay, suggests increased toxicity

relative to parent compound or to its hydrolysis products.


viii














INTRODUCTION

The reported episodes of groundwater contamination by

landfills and septic tanks, hazardous waste sites,

petroleum-based hydrocarbons, halogenated hydrocarbons, and

pesticides have dramatically increased in the United States

in recent years (Pye and Patrick, 1983). These reports were

followed by efforts to address the problem at both the

federal and state levels. An official report from the U.S.

Environmental Protection Agency (EPA) in 1984 stated that

"significant achievements have been made nationally in the

protection and enhancement of water quality" by controlling

the many point sources of pollution (U.S. EPA, 1984a, p. 2).

This report also recognized the contributions of nonpoint

sources to contamination of groundwater. The later term

refers to pollutants which originate principally from

agricultural activities, runoff from urban lands, and mining

activities. In an Association of State and Interstate Water

Pollution Control Administrator's Nonpoint Source Pollution

Survey, 78% of the states indicated that the magnitude of

nonpoint source pollution problems was at least as great as

that of point source problems (U.S. EPA, 1984b). In all

regions studied by the EPA, agricultural activities in

particular were identified as the most widespread nonpoint










source of groundwater contamination, affecting 50% of the

waters in 72% of states (U.S. EPA, 1984b). The severity of

the problem depends to a large extent upon local factors

including the hydrogeology of the region, agricultural

practices, rainfall, terrain, and the particular chemicals

that are dispersed into the environment.

Of all the agricultural pollutants of water, pesticides

are the compounds posing the greatest toxicologic hazard.

Of particular interest recently, and the focus of this

research, is the nematicide fumigant, 1,2-dibromoethane or

ethylene dibromide (EDB). It had been widely used in many

states. By 1984, the chemical had been detected in the

groundwater of at least six states. While EDB was first

registered for agricultural use by the U.S. Department of

Agriculture in 1948, only relatively recently has extensive

information been available on the toxicological properties

of this chemical.

The literature review which follows presents current

knowledge of the persistence, fate, and transport of EDB in

the environment and the evolution of the discovery of this

problem. The state of Florida has had particular interest

in this episode for several reasons that will be discussed.

Close study of the history of EDB usage underscores the

urgency of evaluating and amending current practices that

may lead to groundwater contamination, especially from

unexpected nonpoint sources such as pesticides. Prior to










discoveries of EDB in groundwater, the chemical was thought

to be sufficiently volatile and of sufficiently low water

solubility to not present an environmental hazard. This

belief is reflected in a report prepared for EPA in 1975

which states "ethylene dibromide is not believed to

accumulate in the environment. Limited information suggests

that it degrades at moderate rates in both water and soil"

(Going and Long, 1975, p lb). Review of the literature

suggests that only until relatively recently was the

movement and accumulation of EDB in the subsurface of

interest. Beckman et al. attributed increased bromide

content in various crops to soil pre-treatment with nemato-

cidal compounds, EDB and 1,2-dibromo-3-chloropropane, which

were "readily degradable," liberating inorganic bromide into

the soil (Beckman et al., 1967, p 138).

The detection of EDB in groundwater came about through

increased sensitivity of modern analytical chemistry

instrumentation and a renewal of concern for environmental

contamination by man's daily practices. In the 1960s,

residual EDB was determined by measuring inorganic bromide

in soil and plant tissue (Beckman et al., 1967; Mapes and

Shrader, 1957). These indirect methods involved combustion

or oxidation of EDB followed by assay of bromide by

titration or polarography. Determination of greater than 1

to 10 parts per million (ppm) was possible; however, these

values were highly dependent upon the care taken to










establish the background halide content. In the 1980s, EDB

is detected directly in agricultural commodities (Rains and

Holder, 1981; Morris and Rippon, 1982) and environmental

matrices (Amin and Narang, 1985; Marti et al., 1984) at

concentrations below 10 parts per billion (ppb). These

methods employ purge and trap gas chromatography and mass

spectrometry for achieving lower detection limits and

identity confirmation.

With the benefit of modern analytical chemistry

instrumentation, this study was designed to achieve the

following goals: (1) to describe the kinetics of

degradation of EDB in buffered aqueous systems and in

groundwater, (2) to elucidate the major products of EDB

degradation in groundwater, and (3) to investigate EDB

degradation occurring in buffered aqueous systems and

groundwater in the presence of aqueous hydrogen sulfide

(H2S(aq)), a reactive, natural component in groundwater.














LITERATURE REVIEW

Groundwater Contamination by Pesticides

The Environmental Protection Agency recognized the need

for investigation of groundwater contamination by pesticides

in 1979. The first pesticides discovered in groundwater of

various states were 1,2-dibromo-3-chloropropane (DBCP) and

aldicarb, the later occurring mostly as the sulfone or

sulfoxide (Cohen et al., 1984). Over 50 different

pesticides were found in ground-water basins of California,

one of the largest users of pesticides in the nation. One

of the best known incidences was the discovery of DBCP in

hundreds of wells in the San Joaquin Valley (Litwin et al.,

1983). In 24 other states, a total of 19 pesticides were

detected in groundwater (U.S. EPA, 1987b).

The importance of the safety of groundwater is evident

when faced with the fact that it provides at least half the

potable water to the U.S. population (Pye and Patrick,

1983). The gravity of the situation may be even more acute

in locations such as Florida, where groundwater provides 92%

of the population's potable water supply (Fernald and

Patton, 1984; U.S. EPA, 1987a).

The state of California, in collaboration with the U.S.

EPA in the early 1980s, studied the extent and mechanism of










groundwater contamination by four widely used pesticides:

DBCP, EDB, simazine, and carbofuran. EDB was found at

concentrations between 0.1 and 31.1 ppb in soils at depths

greater than three meters. Numerous wells were also found

to be contaminated (Zalkin et al., 1983). Meanwhile, in the

summer of 1982, irrigation wells in Georgia were found to

contain EDB at levels as high as 100 ppb (Marti et al.,

1984). The greatest extent of contamination, however, was

not uncovered until the state of Florida began monitoring

groundwater wells in the summer of 1983. In Florida,

contamination was found to exist throughout much of the

heavily farmed regions of the state. Concentrations in

private wells as high as 900 ppb were reported1. Florida

responded rapidly by suspending the use of this agrochemical

in July, 1983 (Florida Statutes, 1983). The EPA followed

this action two months later by imposing a nationwide ban on

all agricultural uses of EDB (Fed. Regist., 1983). State

environmental regulatory agencies of South Carolina,

Washington, Arizona, Massachusetts, and Connecticut have

subsequently reported EDB contamination of groundwater. The

range of typical EDB concentrations in these states as well

as California, Georgia, and Florida is 0.05 20 ppb (U.S.

EPA, 1987b).



'Aller, C. Bureau Chief, Groundwater Protection Division,
Florida Department of Environmental Regulation, personal
communication, March 12, 1985.










Ethylene Dibromide

Commercial Use

The EPA estimates that of the 750 to 790 thousand

metric tons (one metric ton equals one thousand Kg) of EDB

formerly produced annually in the United States, only 29 to

33 thousand metric tons were used in agriculture (U.S. EPA,

1983). The major industrial use was as an anti-knock lead-

scavenging additive to leaded gasoline, constituting

approximately 0.03% by weight of leaded gasoline. The major

agricultural use (about 90%) was as a soil fumigant. Other

registered agricultural uses included post-harvest

fumigation, spot fumigation of grain milling machinery, and

several other low volume uses (U.S. EPA, 1983). Its

insecticidal and nematocidal properties were known in the

mid-1920s. Federal pesticide registration for its use as a

soil and grain fumigant was granted in the 1950s. EPA

estimates that approximately one million acres in the U.S.

were treated annually for nematode control with nearly 57

thousand metric tons of EDB. The largest portion of this

was employed in farming of soybean, cotton and peanuts, with

lesser amounts used in farming other crops. The greatest

amounts of EDB have been used in the southeastern United

States, where these crops predominate. EDB application

increased somewhat when a closely related fumigant, DBCP,

was banned in 1978.










Toxicity

The National Cancer Institute in 1973 first reported

that EDB produces stomach cancer in rats and mice with a

very short latent period (Olson et al., 1973). The chemical

was mutagenic at relatively high doses (0.60, 0.092

revertants/nmol) in the Ames Salmonella reversion assay.

The mutagenicity is not enhanced by addition of rat liver

microsomal extract (McCann et al., 1975; Moriya et al.,

1983). EDB induced lethal point mutations in fruit flies

(Buselmaier et al., 1974) and was acutely toxic in all

animals tested, with LD,5 values ranging from 50 mg/Kg in

rabbits to 450 mg/Kg in female mice (National Cancer

Institute, 1978).

The toxic effects of EDB have been postulated to

originate via either of its two known metabolic routes. A

mixed function oxidase (P-450) mediated mechanism through

the intermediate, bromoacetaldehyde, is followed by

conjugation with glutathione (GSH) which is further

metabolized to the water-soluble mercapturic derivative.

The second route involves direct conjugation to GSH, to

produce the half-mustard, S-2-bromoethyl-glutathione. This

reaction is followed by formation of the highly reactive

thiiranium ion, which may also be further metabolized to the

mecapturic acid water-soluble form (van Bladeren et al.,

1980; Shih and Hill, 1981). The later route has been

implicated in producing the most profound toxic effects.










Genotoxic damage involves irreversible binding to DNA (Hill

and Shih, 1978) and more specifically, as a S-[2-(N7-

Guanyl)ethyl]glutathione adduct (Inskeep et al., 1986).

These reactions are dependent on cytosolic components and

reduced glutathione (Ozawa and Guengerich, 1983; Inskeep and

Guengerich, 1984).

Early indications of the toxicity of EDB prompted the

Rebuttable Presumption Against Registration (RPAR)

proceedings by EPA in December, 1977 (Fed. Regist., 1977).

The EPA's final decision to suspend usage of this chemical

was based in part upon the subsequent toxicologic findings

and estimates of cancer risks associated with discoveries of

EDB residues in the diet and drinking water. EPA's Cancer

Assessment Group evaluated the increased cancer risk from

EDB-contaminated drinking water using a one-hit model with

"Weibull" timing (U.S. EPA, 1983, p 79). The resulting

risks are shown in Table 1.

In one of two human epidemiological studies published,

mortality data from workers exposed to the chemical in two

EDB production plants were inconclusive in defining the

relationship between EDB and human cancer (Ott et al.,

1980). One notable conclusion, however, was that far fewer

human neoplasms were observed than what was extrapolated

from animal carcinogenicity studies (Ott et al., 1980).

In the second study, a decrease in fertility of men

occupationally exposed to EDB was significant in one of four

workplaces studied (Wong et al., 1979).











Table 1. U.S. EPA estimates of cancer risk for EDB
groundwater contamination.


CONCENTRATION
of EDB, ppb
(ua/L)

0.02

0.1

5.0

100.0


LIFETIME CANCER RISKa


0.00005

0.00015

0.0075

0.15


Source: U.S. EPA, 1983.

a Assumes consumption of 2 L of water per day, 60 Kg adult,
for 70 years. These probabilities of observation of cancer
are the additional risks estimated for a lifetime exposure.










Evaluating the evidence for EDB carcinogenicity, the

International Agency for Research on Cancer and the American

Medical Association find sufficient evidence from animal

data but inadequate evidence for humans (O'Neill, 1985;

Council on Scientific Affairs, 1988). EPA classifies EDB as

a probable human carcinogen (group B2). Group B2 compounds

are compounds with limited human epidemiologic studies and

sufficient evidence from animal studies.

Chemical and Physical Properties

EDB is the saturated vicinal dibrominated product of

ethylene. It is moderately soluble in water (4300 ppm or

22.9 mM at 30*C), compared to most pesticides and is fairly

volatile (11.6 mm Hg, 1.5 x 10-2 atm at 20C). Having a

Henry's law constant (6.6 x 104 atm-m3/mol) which is

considered mid-range among commonly occurring chemical

contaminants (Lyman, 1982a) combined with a relatively low

affinity for the organic fraction of soil (organic carbon

coefficient, 66 mg/g, as reported by Rogers and Mcfarlane,

1981; 44 mg/g, as cited in Lyman, 1982b), makes the chemical

suspect to move freely through the underground compartment

where it is introduced intentionally as a fumigant or unin-

tentionally from leaking underground gasoline storage tanks.

Ethylene Dibromide in Florida Groundwater and Soil

The locations and levels of EDB in contaminated

groundwater are related to dry and wet time periods and to









12

topography of the land. This pattern presumably reflects the

halting and restarting of dilution of EDB in the aquifer.

For example, a sharp rise in EDB content in the Floridian

aquifer in the eastern part of the Florida panhandle

paralleled the onset of the drought of mid-1986 in the

southeastern United States. In one location studied, the

proximity of sinkholes to areas of EDB application favored

the contamination of underlying limestone aquifers.

As noted earlier, after the initial discoveries of EDB

in groundwater in California, Georgia and Florida, at least

five other states reported finding the compound in their

groundwater. The situation in Florida, however, appeared

the most severe.

Responding to an early indication of the problem, the

state of Florida assembled a multi-departmental task force

headed by the Department of Environmental Regulation. Its

goals were to oversee and implement a state-wide groundwater

surveillance program, to gather available related

information and initiate necessary research, and to offer

assistance to residents of affected areas. Monitoring of

public and private wells since June 1983 has shown that of

the 67 counties in Florida, 22 counties had wells delivering

water contaminated with EDB. Concentrations of EDB detected

have ranged from 0.02 to about 900 ppb (Aller, C., personal

communication, see footnote, p 6). Figure 1 represents the

location of State EDB applications and well water sites









13

known to be contaminated with EDB as of March 1985. The EDB

surveillance program database indicates that by June 1989,

1,703 wells of 13,441 tested (12.7%) were positive for EDB

contamination. Approximately 86% of the positives found

were in the counties of Highlands, Jackson, Polk, Lake, and

Orange. With the exception of Jackson, most of the

contamination (80 to 100%) resulted from the State nematode

eradication program. Ninety percent of the wells tested in

these counties are private wells. Less than 0.1% of wells

tested had EDB concentrations greater than 10.00 ppb. The

majority of the positives were between 0.01 and 1.00 ppb

(Florida Groundwater Protection Task Force Meeting Minutes,

1989).

EDB has been used in Florida by horticultural and

agricultural concerns and in golf course maintenance. Since

EDB was not classified as a restricted use pesticide by the

EPA, users were not required to maintain records. This lack

of record keeping makes it difficult to predict areas at

high risk for contamination and therefore difficult to make

decisions about remediation of this contamination.

Starting in 1960, the Florida Department of Agriculture

documented its use of EDB in their Spreading Decline Control

Program. The disease known as spreading decline is

the result of microbial infection of crops transmitted by

the burrowing nematode. Practices included the

establishment of barriers or buffer zones to prevent the

















C44
0







0






-4
as




u4












C




0
4-C
(0








-4





4-i

*4 4





C
a


















0

















cI
4.)














'O 1
-4H
- )

Cu

C1C


0Q)





F) C
+P-
(0 e
4^ 1













'4
.4
I0
Li


&


S


o Z
y (
*- aeT l
1<^ic










Feldman, 1965). State records indicate that the EDB

application rate had been about 60,000 to 70,000 liters per

year or an estimated total of 660,000 to 770,000 liters

applied since initiation of the program. The application

rate of EDB in this nematode eradication program was 25 to

50 lbs/acre Approximately 9600 cumulative acres were

treated as barriers, which mainly consist of retreated land

which average 417 acres treated per year. EDB was applied

by injection at least eight inches beneath the soil surface

through tubes behind coulter openers or chisels. The

resulting furrows were closed by hilling soil over them, by

press wheels or other means, to prevent loss of the EDB

fumes from the soil (Florida Dept. Agric., 1984). State-

sponsored application of EDB correlates well with cumulative

contamination data, as shown in Figure 1, with the exception

of the northwestern region of the State. The use of EDB in

soil by the State was estimated to be only about 10 percent

of all the EDB applied. Records of non-State-sponsored

applications are scant, making reliable estimates of usage

impossible (Shankland, 1985). Generally, contaminated areas

are consistent with records compiled by inspectors from

information volunteered by agricultural growers for the last

two years of EDB application around the State, with at least



IWritten correspondence dated May 9, 1984 from D.Conner,
Florida Secretary of Agriculture, responding to the
information requested by Florida Senator P. K. Neal,
regarding State EDB soil application.










one notable exception. Records from Dade county in south

Florida indicate that EDB had been applied in the greatest

quantity, yet no contamination has been found. There is no

explanation for this observation; however, more rapid

groundwater flow compared to other areas of the State may

have contributed to faster dilution of the chemical.

Sorption to Soil

Despite the dispersal of significant amounts of EDB

into the environment from pesticide application, emissions

from gasoline-engine combustion and leakage of underground

tanks, little was known about the persistence, fate and

transport of the compound in the natural environment,

especially, in soil and water. Values as divergent as five

to 10 days (Johns, 1976; Leinster et al., 1978) and 14 years

(National Institute of Occupational Safety and Health, 1977)

were cited as half-lives due to hydrolysis in neutral

aqueous solutions at ambient temperature. Early

investigations focused primarily on residual EDB on crops

and on processed or stored agricultural products (Heuser,

1961; Sinclair et al., 1962; Beckman et al., 1967).

The diffusability of EDB through soil is a crucial

factor affecting its ability to reach and destroy soil-borne

pests. An important factor that influences the diffusion is

the adsorptive capacity of the soil for the fumigant vapor.

Early work indicated that there is little penetration into

soil mass above the point of injection of undisturbed soil









18

after treatment. Sorption was found to be primarily due to

colloidal organic matter and therefore affected greatly by

the amount of organic content of the soil (Siegal and

Erickson, 1953). Sorption was rapid with very little, if

any, irreversible sorption. A sharp drop in sorption

followed small increases in moisture content of soils (i.e.,

from 15 to 17% moisture) with sorption leveling at about 25%

soil moisture content. Temperature decreased the amount of

EDB sorbed on soils by about 5% per degree between 10 and

250C (Wade, 1954). The sorption capacity of soil was

closely related to the observed concentration ratio in the

soil-water phase to the vapor phase (Henry's law constant),

with a three-fold decrease in sorption accompanying a

temperature increase from 5 to 250 C (Thomason et al.,

1974). For up to twenty days after simple line injection,

there was very little loss of EDB to the atmosphere through

an undisturbed soil profile. A loss of less than 5% of EDB

was measured after a chisel shank hole injection at a

commonly employed depth of 30.5 cm, during which time a

constant air stream at a rate of 0.80 Km/hr was passing over

the surface of the soil.

In contrast, investigators fit their experimental data

to the Brunauer, Emmett, and Teller (BET) multi-molecular

isotherm adsorption equation. They found that EDB sorption

potentials to soils were essentially restricted to the

external surface of the soil particle and not organic










material (Jurinak and Volman, 1957). The BET adsorption

theory is based primarily on the assumption that Van der

Waal's forces between molecules play the predominant role in

adsorption after the mono-layer is formed. The ability of

EDB to penetrate water structure in close proximity to soil

surfaces is greatly inhibited by increased moisture content

(Call, 1957). The competition between water and EDB for

sites on soil surfaces is also a function of soil type

(Phillips, 1964).

Under field conditions, two or more interacting edaphic

factors in soil were found to determine the overall

diffusion behavior. Penetration in coarse soil was largely

dependent upon moisture content and temperature; medium-

textured soil penetration was dependent upon moisture,

temperature and bulk density; no movement occurred in fine-

textured soil (Townsend et al., 1980). In closely related

studies, sorption of neutral organic compounds was shown to

be essentially a partitioning-controlled process rather than

a physical adsorption process (Choiu and Peters, 1979;

Rogers and McFarlane, 1981).

Degradation in Soil and Water

Evaluation of the ability of microbes to degrade EDB

has been reported by two laboratories. Bacterial batch

cultures of primary sewage effluent incubated under anoxic

denitrifying conditions failed to degrade EDB (Bouwer and

McCarty, 1983). It was found, however, to be rapidly










transformed under anoxic methanogenic conditions in batch

cultures and in a continuous-flow methanogenic fixed-film

column (Bouwer and McCarty, 1985). Anaerobic suspensions

of sewage sludge, activated by percolating oxygen through

the sewage, gave similar findings. The gaseous product in

this case was confirmed to be ethylene (Weintraub et al.,

1986). This laboratory also found that similar suspensions

incubated under aerobic conditions resulted in rapid

degradation to a non-volatile, debrominated and highly

water-soluble product (Jex et al., 1985). Separation of

this product was achieved by high performance liquid and

thin layer chromatography plus chemical derivatization for

gas chromatography and mass spectrometry analysis.

Analytical results suggested a single product which probably

is a condensation or complexation with organic materials or

microbial cells.

The ability of natural microbiota to degrade EDB from

contaminated areas has been reported by two laboratories.

In the first study, soil was obtained from three Florida

farming regions where the chemical was used extensively for

many years. Soil samples from depths of one and three

meters, which were assumed to be anaerobic, were incubated

as such in aqueous suspensions. Only inorganic nutrients

were added and incubated for seven months. The

investigators found that EDB was either not degraded or only

marginally degraded by organisms in Florida soil (Weintraub










et al., 1986). In contrast, the second study found EDB to

be rapidly mineralized to carbon dioxide and bromide ion by

organisms in surface soils under aerobic incubation

conditions. Soil and water were sampled from an EDB-

contaminated groundwater discharge area in Connecticut. The

conversion of EDB to these products was almost complete with

the exception of only small amounts of an unidentified

unextractable product (Pignatello, 1987).

Wilson et al. (1986) found that EDB was rapidly

degraded in aquifer material known to support

methanogenesis. The equilibrium redox potential of the

material used was -272 mV. Concentrations of 200 ppb of EDB

in the aquifer matter-pore water used declined to 27% of the

control concentration by seven weeks of incubation at 170C.

No carbon dioxide was detected as a degradation product.

Although no product was identified, a peak observed by gas

chromatography with flame ionization detection coincided

with the disappearance of EDB; subsequent GC/MS was not

successful in identifying this peak. The report made no

mention of the presence of hydrogen sulfide, commonly

present in the material described and possibly a very

important reactant in the incubation material.

Steinberg, et al. (1987) have reported residual EDB in

soils from Connecticut farmlands highly resistant to

degradation or mobilization. The EDB was released with heat

and organic solvent treatments. Central Florida soils that









22
were treated approximately bi-annually, but had not received

EDB treatment for at least 5 years, had EDB residues up to

60 ng/g of dry soil (Moye and Weintraub, 1988). The EDB in

these soils was released with 750C acetonitrile incubation.

About 1 ngEDB/g dry soil was released from some of the soils

after exposure to 450C dry heat for 8 hr followed by and

aqueous washing. The importance of a fraction of "soil-

bound" EDB in contaminated areas is still being evaluated.

Other modes of degradation of EDB have also been

investigated recently. EDB in aqueous solution is rapidly

photohydrolyzed upon irradiation with a 450-W medium

pressure mercury lamp to yield ethylene glycol and bromide

ion. The rate enhancement for this process was estimated to

be 105 times greater than that of the nonphotolytic pathway

(Castro and Besler, 1985). In another study, EDB was shown

to undergo complete heterogeneous photocatalyzed

transformation to carbon dioxide and hydrogen bromide

(Nguyen and Ollis, 1984). This conversion was catalyzed on

titanium dioxide when illuminated with near-UV radiation.

The titanium dioxide surface is suggested as the site of

formation of hydroxyl radicals which promote the reaction.

Other investigators have found that the superoxide ion

dioxygenates EDB to yield formaldehyde and bromide ion in

aprotic solvents (Calderwood and Sawyer, 1984). The later

three modes of degradation of EDB have been suggested by the

authors as useful ways to dispose of halogenated pesticide










wastes such as EDB. Finally, oxidation of EDB to

formaldehyde in aqueous solution has been reported to occur

at high temperatures (i.e. greater than 800C). The

conversion, however, was not found to be significant under

environmental conditions. In this same laboratory, EDB at

concentrations of 300 and 700 ppb was found to be stable

over a two-week period at 25*C in solutions of aqueous

chlorine which are commonly attained during disinfectant

processes in municipal water facilities. Total concen-

trations of one and 10 ppm of the two species, hypochlorous

acid (HOC1) and hypochlorite ion ('OC1), were varied by

buffering the solutions from pH 6.5 to 9 (Moye et al.,

1986). It is important to note that the above mentioned

studies are laboratory modes of reaction that do not

necessarily reflect reactions in the field.

An additional route of degradation of EDB in the

environment may have been uncovered by recent work by

Schwarzenbach et al. (1985). These investigators detected

numerous dialkyl sulfides and other volatile sulfur-

containing compounds while sampling wells in an area

contaminated seven years earlier by commercially produced

volatile halogenated hydrocarbons. In laboratory studies it

was revealed that at least for a representative test

compound, 1-bromohexane, under highly reducing conditions in

the presence of hydrogen sulfide, nucleophilic substitution

reactions proceed quite readily. A recent report by Watts










and Brown (1986) describing the findings of organic in

contaminated well fields in Palm Beach county, Florida,

includes the sulfur-containing compounds ethyl mercaptan,

triethyl disulfide, and diethyl disulfide, as well as

bromoethane. Although the components of gasoline were also

found in the well water, no EDB was reported. The

possibility of a similar reaction occurring with EDB in

anaerobic groundwater has recently been reported by

Weintraub and Moye (1987).

Vogel and Reinhard (1986) identify the reductive

dehalogenation product of EDB, vinyl bromide, as a major

product in their hydrolysis study of EDB. If, in fact,

vinyl bromide (Moye and Weintraub, 1986), is the major

product, than EDB's degradation in groundwater does not

decrease toxicity (Henschler, 1985).

Summaries of the modes of degradation of EDB discussed

here are presented in Tables 2 and 3. Ongoing and future

work may enable scientists to understand the variations and

relative importance of these and possibly other modes of

degradation in relation to the persistence and hazard

presented by this chemical in the environment.















>1
40
4 (0 U
) Q) 0U
CA C

oM M ic ? 4 ,
, O 'O 3c 0

0 C U-- m -


U)
>1 -4 co



0 *1
u4 4

4J 4
-0 0 O
w in 4o a,
3 co c co
0 Ch -*4 oM X
0 r-4 ) -4 4
S-- s'-


En


o .-
ft I-


H :
^ '0


>4


0

aa

z 0
0 o
E-4
a H
10 o


>1
rl


:1 IQ *
00 00

4-) 0 0

Ir 0 *14
0 C
0
to


*r-


0
O c
0 4
4)
0
0)
( |


I
*









rQ
*






0
l C
0 0
SC




O-1
Hc


(rf

a, 0
C 0
0 at 0


q 0 0
S (0

c O
U

I I U




















r- V 0a
-4 I



(0 i I O

(0 i (0 M 4






a.r a ,o -a a)
u V E-41
04 a4 0 0
1)* 10 a) ,0
(0 4-) 014 1-14


0 0 E
u4< C-u


4-)

0


C --
o
w o
-4 r-4
|Ev-


r-4 -
*- *H
i4C

(a 0
0 C

we
4) 4J
44 Q



















































'0










0(
t35




(0


a) l
ZeC,


>1

rl







(0

r-4

4-)0

3o0
Z 0


44

a)



'0














c9
-H-

-4

















0

No
-0




--4-
3. 0)
4.243
2 -Q


a)

.0 9
Sco
40






Cm
0 (


Q)


0)








4J
0)


14




O-

0'-




+


0
>1


0)




41d
w


0) U


'0
C)

44
.-
(4 (
EAm


41


3 co
H u



SA (


CO

40
4W

r4-

(0 I
I 1r



U 0
U '0


4o
0
c
*04
a

S-



>




+
a)








-H


La
a"M
in
* /
n Za


2
0
H



HE4


UvCI)


z
0
H

IzO
H U










Risk Factors for Groundwater Contamination

Florida's citrus, soybean and peanut farming areas,

where EDB was commonly used, are located on predominantly

sandy and well-drained soils with very low organic matter

content (i.e., less than 1%). The soils are mostly entisol,

spodosol, alfisol and ultisol types. Together with a

relatively high mean annual rainfall and high recharge to

the Floridian aquifer in these regions (Fernald and Patton,

1984), the potential for movement of certain chemicals

through the unsaturated subsurface zone is great. As

previously explained, EDB has this ability to be quite

mobile as well as persistent.

There have been a number of attempts in recent years to

develop comprehensive groundwater protection strategies.

The main goal has been to establish criteria upon which to

base regulatory and monitoring decisions to guard against

groundwater contamination. The EPA began such an effort in

1979 to identify pesticide characteristics, nitrate levels,

and soil pH that would likely result in groundwater

contamination (Cohen et al., 1984). Another approach

currently proposed by the EPA makes use of a numerical

rating scheme for evaluating the potential for groundwater

contamination in a specific site given its geohydrological

setting. The scheme is called DRASTIC, an acronym for (i)

depth to groundwater, (ii) recharge rate, (iii) aquifer

media, (iv) soil media, (v) topography, (vi) impact of the










vadous zone, and (vii) conductivity of the aquifer (Rao et

al., 1985).

More recently, researchers at the University of Florida

have developed a simple scheme for ranking the relative

potentials of different pesticides to intrude into

groundwater (Rao et al., 1985). The index, referred to as

the attenuation factor (AF), is intended for use by

regulatory agencies in a preliminary evaluation of a large

number of pesticides to target chemicals for groundwater

monitoring programs or to initiate site-specific studies.

Of the 40 chemicals evaluated, the top 5 ranking pesticides

were EDB, bromacil, picloram, DBCP, and diuron. Not

surprisingly, EDB ranked as having the greatest potential

for groundwater intrusion in the soils tested.














MATERIALS AND METHODS

Materials
The following is a list of materials used in this study

along with the supplier and stated purity or grade when

provided by the manufacturer. The supplier's location is

listed only the first time it is mentioned.

Analytical Standards

Bromoacetic acid (Aldrich Chemical: Milwaukee, WI; 98%)

Bromoethane (Aldrich Chemical; 99 %)

2-Bromoethanol (Eastman Kodak; Rochester, NY 98%)

1-Butanethiol (Eastman Kodak; 99+%)

Butyldisulfide (Aldrich Chemical: Milwaukee, WI; 98%)

Cyclohexylthiol (Aldrich Chemical; 95+%)

1,2-Dibromoethane (Aldrich Chemical; Gold Label; 99+%)

1,2-Dibromoethane (ethylene dibromide, EDB; EPA Pesti-

cides and Industrial Chemical Repository: Research

Triangle Park, NC; 99+%)

1,4-Dithiane (Aldrich Chemical; 97%)

1,3-Dithiolane-2-thione (Aldrich Chemical; 99%)

1,2-Ethanedithiol (Eastman Kodak; 99.5%)

Ethanethiol (Eastman Kodak; 98%)

Ethyl disulfide (Aldrich Chemical; 99%)

Ethylene glycol (Eastman Kodak; 99+%)










37% Formaldehyde solution (Fisher Scientific: Fair

Lawn, NJ; ACS Certified, 36.6%)

Paraformaldehyde (Fisher Scientific; 95%)

Potassium bromide (Mallinckrodt: Paris, NY; analytical

grade)

1,4,7,10-Tetrathiacyclododecane (Aldrich Chemical: West

Chester; 97%)

1,4,7-Trithiacyclononane (Aldrich Chemical; 98%)

1,3,5-Trithiane (Chem Service: West Chester; 99%)

Vinyl bromide (Aldrich; 98%)

Radiolabeled Chemicals

1C-Formaldehyde (4.4 mCi/mmoL, Pathfinders: St. Louis,

MO; 98%)

U-14C-Toluene (1.8 gCi/mL, New England Nuclear:

Boston, MA)

U- C-Ethylene dibromide (55 mCi/mmoL, Amersham:

Arlington Heights, IL; 95 %)

U-"C-Ethylene dibromide (U-25 mCi/mmoL, Amersham; 95+%)

U-"C-Ethylene glycol (U-4.23 mCi/mmoL, Pathfinders;

98%)

Solvents and Reagents

Aqua-Sol 2 (New England Nuclear)

Buffer solutions, pH 4, 7, 10 (Fisher Scientific;

ACS Certified)

Chloroamine T (Fisher Scientific; ACS Certified)

Chloroform (Fisher Scientific; HPLC-grade)











Chromotropic acid disodium (Fisher Scientific; ACS

Certified)

N,N-Dimethyl-p-phenylenediamine dihydrochloride

(Aldrich Chemical; 99%)

Ferric chloride hexahydrate (Aldrich Chemical; 97%)

Ferric ammonium sulfate (Fisher Scientific, ACS

Certified)

Hexanes (Fisher Scientific; pesticide grade)

Iodine (Fisher Scientific; Certified ACS)

2-Nitrophenylhydrazine (ICN Pharmaceuticals, Plainview,

NY)

Periodic acid (Mallinckrodt; ACS analytical grade)

Potassium nitrate (Fisher Scientific; Certified ACS)

Potassium fluoride (Fisher Scientific; Certified ACS)

Potassium ferrocyanide (Fisher Scientific; Certified)

Potassium ferricyanide (Fisher Scientific; Certified

ACS)

ScintiVerse II (Fisher Scientific)

Sodium sulfide, 9-hydrate (Eastman Kodak; ACS reagent,

98.0-103.0%)

Sodium sulfite (Fisher Scientific; Certified ACS)

Sulfuric acid (Fisher Scientific; Certified ACS)

Triethylamine (Aldrich Chemical; 97%)

Water (Fisher Scientific; HPLC-grade)











Components of Buffers

All of the following chemicals were Fisher Scientific,

ACS Certified grade unless otherwise indicated.

Borax (sodium tetraborate, Sigma Chemical: St. Louis,

MO; 99.9%)

Disodium dihydrogen phosphate hepta hydrate

Disodium borate decahydrate

Disodium carbonate

Glacial acetic acid (reagent ACS)

Hydrochloric acid (reagent ACS)

Monopotassium phosphate

Potassium acetate

Potassium phthalate

Potassium chloride

Sodium monohydrogen phosphate monohydrate

Sodium citrate

Sodium hydroxide

Sodium monohydrogen carbonate

Sodium borate

Sodium carbonate decahydrate

Sodium succinate

Titanium trichloride solution (20%/80% water stabilized

with 3% phosphoric acid)











Gas Chromatographic Materials and Columns

5% Carbowax 20M (Supelco: Bellefonte, PA)

Gas Chrom Q 80/100 mesh (Supelco)

15% OV-17 on Gas-Chrom Q (Alltech Assoc.: Deerfield,

IL)

4% OV-225 on Chromosorb Q (Supelco)

Phenylmethylsilicone crosslinked coated capillary

column (DB-5, J & W Scientific: Folsom, CA)

Polymethylsiloxane crosslinked coated capillary column

(Ultra 1, Hewlett-Packard: Palo Alto, CA)

Polyphenylmethylsiloxane coated wide bore column

(RSL-300, Alltech Assoc.)

Polypropylene glycol (Supelco)

10% Squalane on Chromosorb W (Supelco)



Methods
Ethylene Dibromide and Product Identification and Quantita-
tion by Gas Chromatography and Spectrophotometry

Gas chromatography of EDB and related brominated metabolites

Gas chromatograhpic (GC) conditions were established for

the simultaneous analysis EDB and other closely related

bromominated metabolites (bromoethane, bromoacetic acid and

2-bromoethanol) on a Hewlett-Packard Model 5840A gas

chromatograph (GC) equipped with a "Ni electron capture

detector (EC) and a Hewlett-Packard Model 7671 A

autoinjector. The 2 m x 0.32 cm i.d. glass column used was











packed with 15% polypropylene glycol (PPG) on Gas Chrom Q

80/100 mesh solid support which was coated in this

laboratory. The solid support was coated by slow

rotoevaporation of 50 mL of acetonitrile in which the 3 g of

PPG was dissolved and 17 g of solid support was added. The

resulting stationary phase achieved superior separation of

the analytes compared to others tested. The column packing

which were used perviously EDB and similar analytes and

tested include 5% Carbowax 20M (polyethylene glycol, Dumas

and Bond, 1982), 15% OV-17 (Rains and Holder, 1981), 4% OV-

225, and 10% squalane. The PPG was used for the

determination of ethylene chrorohydrin (2-chloroehtanol) by

Heuser and Scudamore (1967).

The PPG column prepared was used with a 5% methane/95%

Ar carrier gas at a flow of 52 mL/min. Temperatures

maintained during analysis were 100C for the column, 300*C

for the detector and 225C for the injector.

EDB was extracted from aqueous solutions by parti-

tioning into hexane upon mixing. For experiments in which

solutions were incubated in 100 mL serum bottles, the

following procedures were followed. Using a 10 mL syringe a

10 ml aliquot was withdrawn through the Teflon-coated rubber

septa, and slowly delivered to a 10 mL volumetric flask to

assure a precise volume. This volume was transferred to a

15 mL screw-top test tube to which 1.0 mL of hexane was

added and the tube immediately capped. It was inverted 5










more times and allowed to stand for at least 5 min. A

portion of the hexane layer was pipetted into a 1 mL vial

for automated gas chromatographic analysis. In subsequent

experiments where solutions were incubated in 10 mL sealed

ampules or where the partitioning of "C-activity was

analyzed, 2 mL of solutions were transferred with a transfer

pipet to test tubes containing 2.0 mL of hexane. All

extractions of aqueous samples were performed in duplicate.

Extracts were injected in duplicate into the GC. Standard

curves of detector response ranging from 20 to 1500 pg/jAL of

EDB were prepared daily and used for quantitation using

linear regression least squares analysis.

Determination of ethylene glycol

Ethylene glycol concentrations in water were determined

by GC analysis of the 2-nitrophenylhydrazine derivative of

formaldehyde, formyl 2-nitrophenylhydrazone. The method of

Fishkind (1987) for the analysis of formaldehyde in water

was adapted by including a preliminary oxidation step. To 2

or 4 mL of the water sample, a 20 AIL aliquot of 5 x 10-6 M

periodate and the sample was then allowed to stand at room

temperature in the dark for 20 hours. Periodic acid

oxidizes ethylene glycol into two molecules of formaldehyde

(March, 1985). Excess periodic acid was precipitated by the

addition of 50 gL of a saturated solution of potassium

nitrate; the mixture was allowed to stand for 1 hr in an ice

bath (Hillenbrand, 1952).








36

Standard aqueous solutions of ethylene glycol prepared

from analytical grade ethylene glycol were assayed by the

same procedure each day of analysis to construct a standard

curve for quantitation. Aqueous formaldehyde, which was

either formed by the oxidation of ethylene glycol or was

present in standard solutions, was quantitated by the

addition of 640 PL of a 2 M aqueous solution of 2-nitrophen-

ylhydrazine, which had been recrystallized from hexane to

remove background formaldehyde. After the addition of this

derivatizing reagent, the reaction mixture was incubated in

a water bath at 40'C for 1 hr. The derivative 2-

nitrophenylhydrazone was extracted by the addition of 5 mL

of hexane to the tube, which was then mixed on a vortex-

mixer for 1 min. A portion of the hexane layer was pipetted

into a 1 mL vial for automated analysis on a Hewlett-Packard

Model 5890 GC. This instrument was equipped with a Hewlett-

Packard Model 7671 A autosampler, a 63Ni electron capture

detector, and a 2 m x 0.32 column packed with 4% OV-225 on

Chromosorb Q 80/100 mesh. A 5% methane/95% Ar carrier gas

at 30 mL/min, a column temperature of 173*C, a detector

temperature of 3000C, and an injector temperature of 225*C

were employed.

The method was validated with use of 1C-ethylene

glycol (specific activity 4.23 mCi/mmol) and 1C-

formaldehyde (specific activity 4.4 mCi/mmol). Unlabeled

stock standard formaldehyde solutions made from para-










stock standard formaldehyde solutions made from para-

formaldehyde were standardized against accurately made

solutions of bisulfite formaldehyde and 37% formaldehyde

solution using the chromatropic acid spectrophotometric

method of Houle et al. (1970). Other supporting data used

for optimization of these procedures are described by

Fishkind (1987).

Determination of bromide

Bromide ion concentrations in water were analyzed by

Standard Water and Watsewater Methods Procedure No. 405 C

(American Public Health Association, 1980, p 261). Bromide

ion was oxidized by Chloroamine T to bromine which bro-

minates phenol red. The brominated product was measured by

absorbance of the reaction mixture at 590 nm on a Beckman

DU-8 spectrophotometer. Standard aqueous solutions of

bromide prepared from potassium bromide were assayed to

construct a standard curve over a range of 50 to 1000 Ag/L.

Samples were diluted as needed with deionized water to final

concentrations which fell within the linear quanitative

range.

Liquid Scintillation Counting of "C-ethylene Dibromide

Total 1C-activity was measured by liquid scintillation

counting of 1.0 mL of sample pipetted into 15 mL of liquid

scintillation fluid. Either Aqua-Sol 2 or Scinti Verse II

liquid scintillation fluids were used. There was no

difference in results using either of these reagents.









38

Counting was done on a Searle Analytical 92 liquid scintil-

lation counter with a Silent 700 electronic data terminal.

Quenching was evaluated both by comparing additions of

the fortified waters or hexane extracts to additions of

known amounts of "C-toluene (4x106 dpm/mL) and by external

addition of 20 pL of 14C-toluene to fortified waters or

hexane extracts. The dpm/mL in aqueous solutions or hexane

extracts was determined by dividing the measured cpm/min by

the counting efficiency. Efficiency and precision of

extraction of EDB from the waters were evaluated in a

similar fashion.

Gas Chromatographv and Mass Spectrometry
for Sulfur Product Identification

Analyses were performed on a GC with a flame

photometric detector with a 394 nm filter for sulfur-

containing compound selectivity (FPD/S). With this filter,

the predominant chemiluminesence bands of emmision of

sulfur-containing analytes are detected (Farwell and

Barinaga, 1986). Analyses were performed on a 5840A Hewlett

Packard GC equipped with a Hewlett Packard 7671A autosampler

and a 15 M x 0.53 mm i.d. wide bore column with 1.2 pm

polyphenylmethylsiloxane film (RSL-300). The oven

temperature was programmed as follows: it was held at 40C

for 2 min, then increased at a rate of 8*C/min to 90*C, then

250C/min to 2000C. The carrier gas flow of nitrogen at 10

mL/min. FPD/S conditions were as follows: injector











temperature 170*C, detector temperature 225*C, hydrogen 80

mL/min, oxygen 14 mL/min and air 55 mL/min.

Extracts from aqueous incubations, usually a 5:1 ratio

water to hexane, and the sulfur-containing standards in

hexane also were analyzed by GC/MS on a Finnigan 4021 GC/MS.

This instrument was equipped with a 30 M x 0.25 mm i.d.,

0.33 gm phenylmethylsilicone crosslinked film coated

capillary column (DB-5). The GC oven temperature was

programmed to begin at 600C and increase at a rate of

120C/min to 2800C. Mass spectra were obtained in electron

impact (EI) and chemical ionization (CI) modes, both at 70

eV and a source temperature of 150*C. The CI spectra were

collected with isobutane reagent gas at 0.2 torr.

Oxygen. Redox. and pH Measurements

Dissolved oxygen was measured with a membrane oxygen

electrode (model 97-08, Orion Research, Cambridge, MA). The

electrode was calibrated in water-saturated air and zeroed

in a 5% sodium sulfite solution. Measurements were made in

a 100 mL BOD bottle without headspace.

Redox potential (Eh) was measured with a platinum redox

electrode (model 97-80, Orion Research Inc., Cambridge, MA).

This electrode was calibrated using the following solutions:

(1) 0.1 M potassium ferrocyanide with 0.05 M potassium

ferricyanide, (2) 0.01 M potassium ferrocyanide, 0.05 M

potassium ferricyanide, and 0.36 M potassium fluoride, and

(3) a standard ferrous-ferric solution composed of 0.1 M









40
ferrous ammonium sulfate, 0.1 M ferric ammonium sulfate, and

0.1 M sulfuric acid (Light, 1972).

The pH was measured with a glass body combination

electrode (model EA-2, Fisher Scientific, Fair Lawn, NJ) and

calibrated in standard pH 4, 7, and 10 buffer electrode

standardization solutions. All of the above mentioned

electrodes were connected to an Orion model 601A digital

ionanalyzer for operation.



Hydrolysis Reactions
Kinetics: Effects of Temperature and PH

Preparation and incubation of solutions

Groundwater was obtained from shallow wells in three

counties in northcentral and northwestern regions of the

State: Polk, Highlands, and Jackson. Water samples were

characterized by analyses either on-site, at this

laboratory, or at the Extension Soil Testing and Analytical

Research Laboratory, University of Florida (Appendix A).

The water was collected in five-gallon polypropylene

containers and kept on ice during transport to this

laboratory. Water was assayed for residual EDB by the

packed colunm GC analysis.

One hundred ml aliquots of these water samples and of

laboratory deionized water were fortified with EDB to con-

centrations of 10 or 100 ppb with a methanol/water stock

solution of the chemical. Glass serum bottles (100 mL,









41
Wheaton, Millville, NJ) were filled with the solutions, the

headspace purged with nitrogen and the bottles tightly

capped with Teflon-coated rubber septum seal crimp caps

(Supelco, Bellefonte, PA). Before sample preparation, the

serum bottles, seals, and caps were autoclaved (1210C, 15

psi, 20 min) and the waters were either autoclaved (same

conditions) or vacuum-filtered through a 0.20 Am filter

(Millipore, Bedford, MA) to eliminate microbial activity.

Sets of samples prepared in duplicate were incubated in

an inverted position in a darkened water bath. Since EDB

was relatively stable in water at ambient temperatures in

preliminary experiments, incubations were performed at 40,

50, 60, 70 and 80*C 0.50C. Serum bottles were

periodically taken out of the water bath and allowed to

equilibrate to room temperature before sampling. The time

interval between samplings was determined by the observed

rate of disappearance of EDB.

The apparent first-order rate constants for EDB

degradation were obtained by linear regression least squares

analysis of the log percent EDB remaining vs time plot.

Arrhenius temperature dependence was assumed, whereby a plot

of log rate constant vs 1/T "K provides a convenient graphic

method to evaluate the activation energy, Ea, and to

estimate the rate constant of the reaction at any desired

temperature.

At least two samples in each incubation trial were

fortified with a U- C-EDB (25 mCi/mmol) by addition from a










by addition from a 5 tCi/mL methanolic stock solution to a

final concentration of ca. 5000 dpm/mL. These samples were

periodically sampled to check the integrity of the system by

measuring the containment of the 4C-activity. The

radiochemical purity was reported to be 98% (Amersham) and

confirmed in this laboratory to be within 5% of the stated

value as measured by gas chromatography and liquid

scintillation counting. To check the efficiency of hexane

extraction of EDB and its losses during the analysis

procedure, solutions of 14C-EDB were analyzed by identical

procedures.

The degradation of EDB was evaluated as a function of

pH in the three groundwaters and deionized water. A series

of 5 mM buffer systems (Diem, 1962) were used to maintain

groundwater and deionized water fortified with EDB at pH

values ranging from 4.0 to 9.0. The buffers and

concentrations of the components added for the resulting pH

as indicated were as follows: borax/succinic acid: 0.89

mM/4.11 mM pH 4.0, 1.98 mM/3.02 mM pH 5.2; potassium

phthalate/NaOH: 5.00 mM/0.04 mM pH 4.0, 5.00 mM/0.12 mM pH

4.6; borate/KCl/NaOH: 5.00 mM/5.00 mM/0.40 pH 8.0, 5.00

mM/5.00 mM/0.87 mM pH 9.0; borax/monopotassium phosphate:

0.28 mM/4.50 mM pH 8.0, 4.50 mM/0.13 mM pH 9.0; and sodium

carbonate/disodium bicarbonate: 0.40 mM/4.26 mM pH 9.2, 4.60

mM/0.76 mM pH 10.6).

Incubations of the EDB-fortified buffered solutions

were conducted at 62'C. The remaining EDB at approximately











3 day intervals was determined by hexane extraction and GC

analysis on the PPG packed column. Rate constants for EDB

degradation were derived from data collected over 17 days.

Product Identification

Polk, Highlands, and Jackson County groundwater samples

and deionized water were fortified with EDB to a

concentration ca. 3.5 mg/L, including 14C-EDB with a

resulting concentration of ca. 3500 dpm/mL. Incubations

were conducted at 730C until essentially all the EDB had

disappeared. At approximately 2 day intervals, analyses of

EDB, ethylene glycol, and bromide ion were performed. In

addition, samples were analyzed for purgeables by gas

chromatography/mass spectrometry (GC/MS). Within a purge

and trap device (Tekmar model ALS, Cincinatti, OH), helium

gas was bubbled at 8 mL/min through a 25 mL aliquot of a

sample for 22 min at ambient temperature and the effluent

directed onto Tenax adsorbent. The Tenax adsorbent was

heated to direct the purgeables into a GC, with a PPG column

and GC conditions used for EDB analysis. The GC was

interfaced with a Finnigan Model 4021 quadrupole GC/MS for

analysis. Mass spectra, scanning from m/e 50 to m/e 400,

were obtained in electron impact mode with an ionization

energy of 70 eV.

To measure the partitioning of 14C-activity of the

initial "C-EDB and its "C-products, 1 mL aliquots of the

hexane and aqueous phases, after extraction, were pipetted

into 15 mL of scintillation liquid and counted. In









44
addition, 1 mL of the unextracted sample was analyzed in the

same manner.



Reactions with Aqueous Hydrogen Sulfide

Kinetics: Temperature and pH Dependence

Preparation of solutions

Experimental solutions were in 10 mM buffers prepared

with HPLC grade water. Before use, glassware and water were

autoclaved (121 oC, 15 psi, 20 min). The buffers providing

solutions of pH 5, 7 and 9 consisted of the following

components: potassium acetate/acetic acid: 6.40 mM/0.360 mM

pH 5.0, potassium phthalate/sodium hydroxide: 2.36 mM/7.64

mM pH 5.0, sodium monohydrogen carbonate/hydrochloric acid:

8.17 mM/1.83 mM pH 7.0, disodium dihydrogen phosphate

heptahydrate/sodium monohydrogen phosphate monohydrate: 3.81

mM/6.19 mM pH 7.0, disodium borate decahydrate/hydrochloric

acid: 3.65/6.35 mM pH 9.0, disodium carbonate

decahydrate/hydrochloric acid: 0.53 mM/9.47 mM pH 9.0). The

resulting calculated ionic strengths of the solutions were

adjusted to p = 15 mM using 4.8 M KC1. The solutions were

deoxygenated by bubbling a stream of oxygen-free nitrogen

through them for at least 2 hr. The redox buffer, Titanium

(III) citrate of Zehnder and Wahrmann (1976), was used in

the experimental solutions to maintain an oxygen-free water

phase with high electron activity to inhibit oxidation of

sulfides. The redox buffer stock solution was perpared by

adding 3.5 mL of Titanium trichloride solution (20%/80%











water) to a final volume of 50 mL made up in 0.2 M sodium

citrate. One mL of the redox buffer stock solution was

added to each 99 mL of experimental preparation.

All subsequent manipulations of solutions and filling

of sample containers were done in an AtmosBag (Aldrich

Chemical) which contained an nitrogen atmosphere. The

oxygen concentration inside the Atmosbag remained below 0.05

mg/L. Oxygen concentration was monitored with an oxygen

sensor (Bio-Tek Instruments, Inc., Winooski, VT, Model

74223).

Five hundred mL portions of the six buffer solutions

were fortified with methanolic stock solutions of 26.5 mM

EDB and 500 gCi/mL of "C-EDB (specific activity 55

mCi/mmol) to achieve final concentrations of 6.0 AM ca. EDB

and 5000 dpm/mL. Crystals of disodium sulfide nonane

hydrate were washed with deionized water and suction dried

on filter paper before being weighed and dissolved in the

buffered solutions. The final concentrations of total

sulfide were quantitated to be 500 140 and 1000 300 MM

(17 and 34 mg/L HzS(aq,))

Using a 10 mL Repeat-o-syringe pipet (Fisher

Scientific) with a 20 gauge needle, 9 mL of the fortified

buffer solution were slowly transferred to sterile 10 mL

glass ampules (Wheaton, Millville, NJ). Each filled ampule

was purged with a stream of nitrogen, then quickly sealed by

an oxygen and natural gas flame. During this procedure, the

fortified solutions were kept chilled in an ice bath.











Analysis of solutions

Duplicates of each buffer and H2S(a,) concentrations were

analyzed for EDB concentration, '4C-activity, pH, Eh, and

oxygen concentration. Sets of 14 ampules of each buffer and

H2S(aq) concentration were incubated at 25, 40, and 600C
1.00C in the dark. Ampules were analyzed at time intervals

based upon EDB degradation rates determined in preliminary

experiments.

At each time interval, ampules were allowed to come to

room temperature before being opened. Two mL of the sample

was withdrawn for hexane extraction. One mL aliquots of both

the hexane phase and the aqueous phase were transferred to

separate scintillation vials containing 15 mL of liquid

scintillation fluid and counted. The remaining 1.0 mL of

hexane was transferred to a serum bottle for GC analysis of

EDB. The GC analysis was performed on a Hewlett Packard

5890 GC with a model 3392 integrator/plotter and a 7671A

autosampler and an 63Ni electron capture detector. Grob

technique injections were made onto a 25 m x 0.20 mm (film

thickness, 0.33 MM, HP ULtra 1) capillary column temperature

programed with an initial oven temperature of 400C for 2

min, then increased to 900C at 80C/min, and finally

increased at 300C/min to 2000C. A carrier gas flow of

Argon/5% methane was maintained at 1 mL/min. A column purge

flow of 2 mL/min was activated at 1 min and a column make-up

was 30 mL/min. In addition, 1.0 mL of the original










incubated solution was transferred to 15 mL of liquid

scintillation fluid, shaken well, and then analyzed.

Total sulfide concentration was determined by the

procedure of Cline (1969), as modified by Millero (1987).

The method involves the formation of methylene blue from the

sulfide and reagent solution of N,N-dimethyl-p-phenylene-

diamine dihydrochloride and ferric chloride hexahydrate in

acidifided solution. Color was developed in 30 min and

absorbance was read at 670 nm with a Beckman DU8 spectro-

photometer. Calibration solutions of concentrations ranging

from 4 MM to 40 MM H2S(aq were prepared from a sodium sulfide

stock solution for each analysis. Stock solutions of soduim

sulfide were prepared by dissolving ca. 0.5 g of sodium

sulfide crystals that were washed with deionized water and

suctioned dry, in 100 mL of degassed deionized water. The

stock solutions were standardized by the iodometric

titration Standard Water snd Wastewater Method No. 427D

(American Public Health Association, 1980, p 476).

Concentrations of H2S(q, in experimental solutions were

determined by comparison to a standard curve generated by

least-squares regression analysis of the assay response vs.

H2S(aq) concentration.
EDB apparent first-order degradation rate constants

were obtained by linear regression least squares fit of

plots of In percent EDB remaining vs time. The partitioning

of 14 -activity of the initial 14C-EDB and its 1C-products

were plotted versus time.











Product Identification

Hexane extracts from incubated of EDB and H2S(aq) in

buffered solutions from kinetic experiments and the same

buffer systems containing higher concentrations of

reactants, i.e., 60 iM EDB and 1000 or 2000 AM HzS(q, were

subjected to GC and GC/MS analyses. Standard solutions of

11 commercially available sulfur-containing compounds in

hexane were chromatographed and subjected to mass

spectrometry for comparison with products formed in

experimental reaction solutions. Molecular formulae,

structure and suppliers of these standard compounds are

presented in Table 4.

Ethylene Dibromide in Sulfate Groundwater

Groundwater from wells in the Florida Ambient

Groundwater Network, which are known to have relatively

moderate to high concentrations of sulfate, were obtained

from officials of the South Florida Water Management

District. Clean 1 L brown-glass screw-top bottles with

Teflon-lined caps were filled in a manner which avoided

excessive contact with the air and were kept on ice in an

insulated cooler while being transported to the laboratory.

The pH, Eh and oxygen concentration of the samples were

measured. Three mL aliquots were then extracted with 1.0 mL

of hexane and the extracts analyzed by GC to determine if

any detectable compounds were initially present.

Three sets of EDB-fortified solutions were prepared

from each well water sample. Before fortification, one












49








4:r cr 4: 4:
0)


o^ -14?a ~ f .1-l .,. w -4 do O -,I ^P -4 o? dP. -4

t< c ra t on 1: V z ) v a


u


0
-4


4-4
4-4
-4

C
O
4I-





44





a
o

a
0








ri
0







4)






P-q
a




















-4
0














-4
O4
0



*-
*-







-4
-4
0












(0e


C4)
-CO) -
N;]
>-
C U
to N
N ^
N p
u ^u
u N
N

-U

-ou
TN


o3 Wi ) : NO U)N CY 'C N : m m

C4 e a 0S 0
10' 1 N N Cl co o D o
Uko UON u c' 0 V UI U. U U H uH U H UcN


E-1 In
E-4 I a E-ui E-E








0 0 ( -4c r > -

.-4 o S4 V
O 4 1 W o D 4J r 1 4) t0 Y0 )
,. I a% -4 1 u- 1 0 -4t -4 4)n


>4 1I0 >4 a> u

W u 1-4 u U -4 u u W U .04:


uU


U) N~ rn N
: UI
NU N ~ r
10 s I L^ L ~ g" Lg L j
U U: Ci)

U U


U


a

w >
H E-
4 -4
&: z


w 0



E-1

U






0
E















a









50

portion of each groundwater sample was autoclaved for 20 min

(121*C, 15 psi), another portion was degassed by bubbling

purified nitrogen gas through the water for approximately 2

hr, and a third portion of each water was left untreated.

Each portion of groundwater was fortified with a stock

solution of unlabeled EDB to give 1.5 mg/L EDB and also with

a stock solution of 14C-EDB to give a concentration of

approximately 3000 dpm/mL. Sets of 10 mL ampules were

filled with about 8 mL of solution, the air space in the

ampule purged with a stream of purified nitrogen, and then

sealed by flame. The ampules were kept at 25"C in the dark

and periodically analyzed by GC for EDB concentration and

the appearance of extractable products. In addition, the

partitioning of 14C-EDB and C-products after hexane

extraction was determined using the method previously

described. One mL aliquots of the solutions were also

analyzed by liquid scintillation counting to check

containment of the EDB during incubation.

Organic Synthesis of a Sulfur Product

The aliphatic cyclic sulfur-containing compound,

1,2,5,6-tetrathiacyclooctane is not commercially available,

but its synthesis has been described by Goodrow et al.

(1982). The procedure involves slowly reacting a dilute

concentration of 1,2-ethanedithiol in chloroform with iodine

and triethylamine (reagents concentration of approximately

10 AM, Goodrow et al., 1981).















RESULTS

Hydrolysis Reactions

Verification of Assay for Hydrolysis Products

Determination of EDB

Using the conditions previously described, EDB in

hexane extracts was quantitated using GC/EC by comparison of

unknowns to calibration curves consisting of four or five

standard solutions of EDB in hexane. The GC/EC response was

linear and reproducible over a range of 10 to 1200 pg of EDB

injected. Figure 2 shows a sample chromatogram and

calibration curve. The efficiency of extraction of EDB from

aqueous solutions into hexane was 95 to 100% over the

concentrations range using both labeled and unlabeled EDB.

Determination of ethylene glycol

Procedures for quantitation of formaldehyde and

ethylene glycol were validated using standard solutions para

formaldehyde and 37% formaldehyde solution or ethylene

glycol preparated in deionized water. In addition, 14C

labeled analytes were employed (formaldehyde 150 mCi/mg,

ethylene glycol 68 mCi/mg) for verifying the assay.

Since both analytes are very water-soluble and their 2-

nitrophenylhydrazone derivatives formed in the analysis are

extractable into hexane, partitioning of "C-activity was




















0*
C


*c








M



0
0


Cu

rt

1H

0O
Ct
O






































H-
0













0
0



rt
PI













I-J





0


-ct
p,
0,
3e
0l







































u
S-J
O OO





o -
>- o
o 00-
in \ -



c\
tn 'n 3SNOdS38

----------^ :









54
used to test efficiency of the assay. Because 14C-activity

can be measured at extremely low concentrations of labeled

material, a very sensitive assessment of efficiency was used

to test efficiency of the assay. Because "C-activity can

be measured at extremely low concentrations of labeled

material, a very sensitive assessment of efficiency was

possible.

After derivatizing six solutions ranging from 17 to 689

gg/L of 14C-formaldhyde, 86 to 98% was recovered in the

hexane extract. Performing the ethylene glycol procedure on

four solutions containing 288 to 2878 Ag/L of the compound

resulted in the recovery of 89% to 100% of the labeled

compound in the hexane extract.

Calibration curves for ethylene glycol were typically

constructed over a concentration range of 60 to 1500 Ag/L.

Resulting correlations of determination (r2) of linear

regression least squares analyses were larger than 0.99.

Figure 3 shows chromatograms of blanks and results of the

assay of ethylene glycol. Background concentrations of

formaldehyde between 10 and 50 pg/L are common and must be

used to correct the results derived from the analysis of

unknown samples.

Kinetics: Temperature and pH Dependence

The disappearance of EDB in fortified groundwaters and

deionized water was followed over 22 days for incubation at

40, 50 and 600C, 17 days for incubation at 700C, and 6 days












44
0


U,



0
C -H



4-)
N

W



40
-.-I













41
N










4C-
411


C








-4 C





4J
0








o *o


*H 0
1C












-C








C 0
rl







OX


C













.l "56




-
4
w c 4
,- w < / z w



-i F ix
E w
-0 C
CI

I








cJ










,x







'ZJ











0
OD












0
Ic -



Z0 -> -0P
0



3SNOdS38 0:33










for incubations at 800C. These changes correspond to

decreases greater than 90% of initial concentrations of EDB

in the 800C incubations and approximately 20% decreases of

EDB at 400C. At intermediate temperatures, the degree of

degradation fell within this range. Linearity in plots of

the log % EDB concentration remaining vs. time demonstrated

apparent first-order degradation rates. The least-squares

regression coefficients of determination (r2) were greater

than 0.99 at higher temperatures and greater than 0.90 at

lower temperatures. The rate data from the Polk County

groundwater sample initially fortified to a concentration of

100 Jg/L EDB are shown in Figure 4. If EDB degradation in

solution was an acid- or base catalyzed reaction, non-linear

semi-logarithmic plots of observed rate constants vs pH

should result. Figure 5 shows the observed rate constants

for the degradation of EDB in the groundwaters and deionized

water as a function of pH while incubated at 630C. Table 5

shows the observed rate constants for the laboratory EDB-

fortified waters. The rate constants differ only slightly

between the waters and indicate neither acid- nor base-

catalyzed degradation. The larger rate constants observed

at pH 5 and pH 8 as compared to those of pH 7 are not con-

sistent with a pH dependent degradation process. The boric

acid buffer at pH 7.8 to 8.3 may have participated in EDB

degradation, thereby increasing the observed rate constant.

Buffers can catalyze hydrolysis even when present at low mM















0)
4-)




0


C-)
cJ
0














41
4,

0





-4











N
4-4
0
a



























0-4






0
C4

0N -4
r-4








-4q
*>


























0 4
M *q
Qo
(0












Qr 0r









o 59
U 0

-0
LO
1
) *

4-0
I 0

) r o
00


L



o
a 0
>O



00
L -

0 0 -



O
0 1u




0

0
0 -0
I / / I




0)

4,


w
0) 0 LO

U u!uDwaH 8G3 % 50-

0 LO ---
0 in (M Blu 8 8C3 %


















4-4





0
r.






0
.-I
4-1
c4




o








0 r

4,















0
C











-H








Wo
m-'4





























cLn
4-2J
0 .



4g)
H 0


rro .
SJ7













C-




ro


a_
34*



0 '


r-
a-)

41


(0
NI


0
-rI

aC


Un
.0 L
0
(11C
i


C-


CU



En

-)
r














The effect of pH on the EDB degradation rate at 63*C


and initial EDB concentration of 100 gg/L.


pH Buffera 103k (hr1') (tb 2, hr) 10 3kSD


Polk Highlands Jackson Deion.
4.0-4.4 a 2.31 2.31 2.34 2.48 2.360.13
(301) (302) (298) (280)
4.0-4.9 b 2.94 2.61 2.45 2.54 2.630.29
(236) (266) (283) (273)
4.7-5.3 b 2.84 2.56 3.46 3.10 3.230.45
(208) (225) (200) (225)
5.3-5.6 a 2.41 2.44 3.47 3.10 2.400.10
(288) (288) (287) (295)
6.1-6.5 c 2.45 2.98 2.64 2.83 2.730.23
(283) (233) (263) (285)
7.6-7.7 d 2.68 2.68 2.52 2.63 2.9510.62
(259) (259) (275) (263)
7.8-7.9 e 4.39 3.96 3.83 4.17 4.170.38
(158) (175) (181) (166)
8.0-8.3 f 3.76 2.81 3.56 3.62 3.440.44
(266) (247) (195) (191)
8.6-9.0 g 3.07 2.61 2.41 2.95 2.950.48
(224) (267) (288) (235)


a0.005 M buffers:


a = borax/succinic acid
b = potassium phthalate
c = borax/succinic acid
d = borax/phosphate
e,f = boric acid
g = carbonate


aApparent First-order rate constant for degradation of EDB
Average of replicate kinetic runs with 6 data points each
Half-life, t1/2 = 0.693/k

cAll samples in pH range, n=8


Table 5.










concentrations (Perdue and Wolfe, 1983). The observed

first-order EDB degradation rates were used to construct

Arrhenius plots like the one shown in Figure 6. The data

shown are from EDB degradation in Polk County groundwater.

The least-squares regression coefficients of determination

(r2) for these plots ranged from 0.97 to 0.99 for all the

EDB-fortified waters tested. The resulting Arrhenius

kinetic parameters were used to make estimates of EDB

degradation rates at ambient groundwater temperatures. In

Table 6, the extrapolated kinetic parameters derived from

these data are shown for 220C, a common groundwater

temperature in Florida. The kinetic parameters do not

significantly vary with the source of the water or with the

initial concentration of EDB. At 220C in the waters tested,

the predicted half-life (t1/2) ranges from 259 to 772 days.

The activation energy ranges from 19.1 to 24.3 kcal mol1.

The variation among the data is fairly large, with standard

deviations generally ranging from 12 to 20% of their

respective means, except for J10 where the standard

deviation is 58% of the mean. Since the extrapolated values

for rate constants at ambient temperatures are derived from

relatively few data points a fair degree of error is

expected. Sources of error in rate constant may arise from

small deviations in temperature of experimental incubations

and random error (Harris, 1982; Maybey and Mill, 1978).

















4-




0
c1


44


0
0








-4












4
rJ








*4
u











0
































-4
ft
C




















0
















*H
-1












-4 7















0 0c
r-l


























0 0
w0 O
9+-
0
o0


0
o U O
L o

0 0

I /


G.



L 0
3 O
L

0 0 -
L. N /


(J cL' CA C0

(JM-)sqoI 6o|











Table 6. EDB degradation kinetic parameters in
solutions studied as predicted by Arrhenius kinetics plots.
Extrapolated values for 220C are shown.


10 kb
220C. (d-)

2.76 0.41

1.61 0.19

0.64 0.38

2.33 0.57

1.52 0.65

1.93 0.36

2.15 0.25

2.15 0.25


cl -1
22-C. (d ')

259 35

434 47

772 94

309 67

547 317

369 69

326 38

326 38


Ead (kcal
mol ')

19.6 0.2

20.9 0.4

24.2 0.8

19.1 0.6

21.1 2.2

19.7 0.6

19.6 0.6

19.6 0.6


Log Ae
(d-1

12.2 0.5

12.7 0.2

14.9 0.6

11.6 0.4

12.1 0.3

11.9 0.4

11.8 1.7

11.7 1.7


aP=Polk county groundwater
H=Highlands county groundwater
J=Jackson county groundwater
DW = deionized water
10 = 10 Lg/L, concentration of EDB fortification
100= 100 Cg/L, concentration of EDB fortification

b Mean of four replicate runs, except for two replicates
of DW samples;
mean of pseudo-first order rate constant standard
deviation, except DW is mean range of replicates

C Mean standard deviation of half-lives of EDB in days,
except DW is mean range of replicates

d Mean standard deviation of Arrhenius activation energy,
except DW is mean range of replicates

e Mean standard deviation of frequency factor, except DW
is the mean range of replicates


Sample8

P10

P100

H10

H100

J10

J100

DW10

DW100










Product Identification

In experiments using 1C-EDB conducted at elevated

temperatures, the decline in 14 C-activity in the hexane

phase, determined after extraction of incubated solutions,

closely paralleled the degradation of EDB, as quantitated by

GC/EC (Figure 7). These observations are highly

reproducible, as shown in the results from simultaneous

quantitation and partitioning of 14C-EDB and C-products

during incubation at 800C (Table 7). At the completion of

the experiment, the total 14-activity in the solutions

ranged from 92 to 99 percent of the initial activity. This

finding indicates that no EDB or product was lost during the

experiment and that no products volatilized. The next

experiments employed solutions containing higher initial

concentrations of EDB (ca. 3.5 mg/L) than in the prior

experiments for the purpose of quantitating the water-

soluble products formed, i.e., bromide ion and ethylene

glycol (Figures 8a, 8b and 8c). Total C-activity was also

quantitated along with the hexane/aqueous partitioning of

14C-activity during incubation of the EDB-fortified

groundwater and deionized water. During the 17 day

incubations conducted at 730C, the amounts of bromide ion

and ethylene glycol together accounted for 60 to 70% of the

initial EDB present. The extent of EDB degradation in these

incubation was approximately 8 or 9 half-lives. As in

previous experiments, the partitioning of 1C-activity into
































Figure 7. Partitioning of 1C-activity ( 14C-EDB and 1C-
products by liquid scintillation counting) during EDB
degradation at 800C.









69










80 -
80 -






cc 40 -
I-I

0


20 -
z

c o 0
(T0 I--------------------
W 0 20 40 60 80

TIME (hours)




EDB CONCENTRATION 1

EDB CONCENTRATION 2
---0---
TOTAL 14C-ACTIVITY 1

TOTAL 14C-ACTIVITY 2

HEXANE 14C-ACTIVITY 1
--*--
HEXANE 14C-ACTIVITY 2

AQUEOUS 14C-AQUEOUS1

AQUEOUS14C-AQUEOUS2
-A-













Table 7. Simultaneous quantitation of EDB (GC/EC) and partitioning
of 1C-activity ( C-EDB and "C-products by liquid scintillation
counting) in dilute solution during degradation at 800C.


Time/Samplea


0 Hour
Al
A2
B1
B2

18 Hour
Al
A2
B1
B2

44 Hour
Al
A2
Bl
B2

80 Hour
Al
A2
Bl
B2


[EDB]
fug/L)


56.4
59.8
28.9
29.1


41.5(74)b
41.9(70)
21.0(73)
20.1(69)


29.5(52)
29.5(49)
14.5(50)
13.8(47)


16.3(29)
16.5(28)
8.1(28)
7.8(26)


14C
in Hexane
(dpm/mL)


150774(85)
150740(84)
77950(86)
78221(88)


114456(64)
115752(64)
53918(60)
58349(65)


83216(48)
82673(47)
42196(48)
41330(47)


51285(30)
48424(28)
23402(27)
23942(29)


14C
in Aqueous
(dpm/mL)


19815(11)
23425(13)
10535(12)
10495(12)


55464(31)
55860(31)
30415(34)
28965(32)


83315(48)
84900(48)
43405(49)
42765(49)


117290(68)
117025(67)
59705(69)
56695(69)


14C
Total
(dpm/mL)


177950
178550
90700
88970


178230(100)
181680(102)
89940(99)
89270(100)


175190(98)
176560(99)
88590(98)
87360(96)


172890(97)
175880(99)
87110(96)
82050(92)


EDB Degradation Kinetic Data (first-order plot)


Regression Equation


-0.00667x +
-0.00689x +
-0.00679x +
-0.00699x +


2.00
1.99
1.99
1.99


r2= -.9990
r2= -.9989
r2= -.9994
r2 -.9983


Apparent rate
constant (hr ')


1.54 x
1.59 x
1.56 x
1.61 x


10-2
10 2
10-2
10- 2


a10 mL aliquots of 100 mL samples are taken for each analysis
interval; samples 1 and 2 are duplicates
Percent of 0 hour measurement in parenthesis










Figure 8. Partitioning of 1C-EDB and 1C-products and
quantitation of EDB, ethylene glycol, and bromide during
incubation in Florida groundwater and deionized water.
Each point is an average of replicate measurements.

8a. EDB-fortifed Polk and Highlands groundwater incubated
at 730C.







EDB


TOTAL 14C
----i---

14C-HEXANE


14C-WATER
--^--

ETHYLENE GLYCOL


BROMIDE ION
-.A-.~L--















POLK CO. GROUNDWATER


3,o00t\------------ ---- --- --------------


2,000 -


C -
o 1,000 -----

A-


0 100 200
E TIME (hours)
CD

z
0
HIGHLANDS CO. GROUNDWA


S3,---000 ----------
O


2,000



1,000 -



0.
0 100 200

TIME (hours)


300 400


WATER


I~




)---U

3-0----0 400
^^ *
300 40


n--











Figure 8 continued.

8b. EDB-fortified Jackson groundwater and deionized water
incubated at 730C.











EDB
-e--

TOTAL 14C
---Q---

14C-HEXANE


14C-WATER
--A--

ETHYLENE GLYCOL


BROMIDE ION
-.A-)-

















JACKSON CO. GROUNDWATER


t rfn


1,000 A
-J-









0 100 200
TIME (hours)


z
0
I DEIONIZED WATER


z 3," ----- -
w U -------------------c


O
2,000 *




1,0ooo V. -5---


TIME (hours)


--A
A


r


- -












































0
0
o





0
CO






) a4

*4








N
u















-4
*4-i
0






Z 0
0
U 'M
CO O





PL) CO











76



w
(,
cr S0 M

aa
w _-
0 z 4 0
Z I
Sx I

z w -
0 0











0



o O
a: u u CM












o -v





I.









I-









0
w
C0







o 0 0


(-CL/Wid3) A11IAI13V 0L '(qdd) 'ONOD
oy











the aqueous phase after extraction closely paralleled the

decline in EDB concentration. Similar patterns of trans-

formation of EDB were seen in all three fortified ground-

waters and in deionized water. Figure 8c shows the pattern

of EDB transformation during an incubation at a higher

temperature (800C) which was observed for a longer time

period (27 days) in deionized water. In this experiment,

the extent of conversion of EDB to bromide ion and ethylene

glycol ranged between 70 and 90%. Further examination of

extractions of aliquots from solutions in this experiment

did not reveal either additional brominated products of EDB

or intermediates in the conversion of EDB to bromide ion and

ethylene glycol. Purge-and-trap GC/MS analyses of aqueous

aliquots detected only trace quantities of 3- and 5-member

hydrocarbon compounds and remaining EDB.

Vinyl bromide could be formed from EDB if EDB were

reductively dehalogenated. This compound was found by Vogel

and Reinhard (1986) in pentane extracts of incubated aqueous

EDB solutions. Although it could not be detected in any of

the solutions or extracted by GC/EC or by purge-and-trap

GC/MS in this study. Under the GC/EC conditions used in

this study, the detection limit is at least 1000 times less

sensitive for vinyl bromide compared to EDB. If vinyl

bromide were formed in the present study, however, 1C-vinyl

bromide would have partitioned into hexaxe upon extraction.

There was no indication that this occurred. Injections on










GC/EC were made of the hexane phase and aqueous phases of

three replicate extractions of deionized water containing

1000 mg/L vinyl bromide. Vinyl bromide was only found in

the hexane phase and not in the aqueous phase.

Extraction of 2-bromoethanol from dilute aqueous

solutions was 30 to 40% efficient. Furthermore, GC/EC

detection of this compound was also much less sensitive than

EDB; at least 100 times less than EDB.



Reactions with Aqueous Hydrogen Sulfide

Kinetics: Temperature and pH Dependence

During incubations at 25, 40 and 600C, the

disappearance of EDB from pH and redox buffered solutions

was examined along with the partitioning of 14C-EDB and 14C-

products after hexane extraction (Figures 9a, 9b and 9c).

These control solutions, which did not contain H2S( q)I were

studied over 237 days at 250C, 194 days at 400C and for 105

days at 600C. At 250C, neither the concentration of EDB nor

the partitioning of the 14C-EDB between the hexane and

aqueous phases changed noticeably at pH 5, 7, or 9. At 40

and 600C, the disappearance of EDB was paralleled by the

changes in partitioning of the C-activity. This trans-

formation of EDB to water-soluble products in pH and redox

buffered solutions was also observed in the groundwater and

deionized water experiments discussed previously.









Figure 9. Degradation of EDB and partitioning of 14C-
products in buffered reducing solutions as a function of pH
and temperature. Each point is the average of duplicate
analyses.

9a. Solutions buffered at pH 5.





PHTHALATE BUFFER
EDB

ACETATE BUFFER
EDB

PHTHALATE BUFFER
HEXANE C-14
..... .. .....
ACETATE BUFFER
HEXANE C-14

PHTHALATE BUFFER
AQUEOUS C-14

ACETATE BUFFER
AQUEOUS C-14
B-e













6 uM EDB

pH 5


1 0
Z

I -
zu z you
00






cr

UJ 20
aC


SO 100 150 200


250C
40
-U










-o



400C


50 100 150 200


0 20 40 60 80 too


TIME (DAYS)









Figure 9 continued.

9b. Solutions buffered at pH 7.








CARBONATE BUFFER
EDB
........ .......
PHOSPHATE BUFFER
EDB

CARBONATE BUFFER
HEXANE C-14
........A .......
PHOSPHATE BUFFER
HEXANE C-14

CARBONATE BUFFER
AQUEOUS C-14

PHOSPHATE BUFFER
AQUEOUS C-14













6 uM EDB
pH 7


so
S.. 250C


80


60

40


0 50 100 150 200 250

400C













0 50 100 150 200

60 C


TIME (DAYS)


0 20 40 00 80 100








Figure 9 continued.


9c. Solutions buffered at pH 9.







CARBONATE BUFFER
EDB

BORATE BUFFER
EDB

CARBONATE BUFFER
HEXANE C-14

BORATE BUFFER
HEXANE C-14

CARBONATE BUFFER
AQUEOUS C-14

BORATE BUFFER
AQUEOUS C-14












6 uM EDB
pH 9


250C







o



---- *-----


50 100 150 200


150

.u. .


50 100


250

400C




I







200

600 C


0
z
z
W
uj
C13
a
uj ar.
LU 0


I"
wo
0
I-
z
w
C.
wU


0 20 40 60 00 100

TIME (DAYS)









85
The type of pH buffer system (phthalate versus acetate;

carbonate versus phosphate) had little effect on the

results. One exception to this trend was the pH 9 buffer

system (borate versus carbonate). A rate enhancement-effect

of borate (but not carbonate) on the degradation of EDB was

also observed in the previous study of the effect of pH on

EDB hydrolysis.

Where EDB was monitored through more than one half-

life, its degradation followed apparent first-order decay.

EDB, initially present in a concentration of 6 gM, was no

longer detectable at 100 days in each of the three pH

buffering systems held at 600C. In solutions also

containing ca. 500 or 1000 gM H2S( ), very different patterns

of EDB's transformation emerged (Figures 10a 10f) and are

described below. These solutions contained the same

composition of pH and redox buffers as the controls

described above. These solutions were held at 250C and

monitored for 254 days at pH 5 and 7 and for 150 days at pH

9. At 400C, solutions of pH 5 and 7 were monitored for 93

days and for 82 days at pH 9. At 600C, the solutions at pH

5 and 7 were watched for 62 days and for 19 days at pH 9.

At the start of the incubation of these solutions, the

dissolved oxygen was less than 0.07 ppm. The initial redox

potentials for the solutions buffered at pH 5, 7 and 9

ranged from -240 to -258 mV, -238 to -291 mV and -284 to -

345 mV, respectively. The first notable difference produced

by the addition of H2S (q) was that EDB was transformed in








Figure 10. Degradation of EDB and partitioning of 1C-
products in buffered reducing solutions containing H2S(aq)
as a function of pH and temperature. Each point is the
average of duplicate analyses with relative differences less
than 3%.

10a. Solutions buffered at pH 5 with an H2S(aq) concentration
of ca. 500gM.



PHTHALATE BUFFER
EDB

ACETATE BUFFER
EDB

PHTHALATE BUFFER
HEXANE C-14
.....A ... ...
ACETATE BUFFER
HEXANE C-14

PHTHALATE BUFFER
AQUEOUS C-14

ACETATE BUFFER
AQUEOUS C-14
B--












EDB, 500 uM H23


250C
-a """**


200 250

400C












80 100

60 C


0 10 20 30 40 50 60


TIME (DAYS)


6 uM

pH 5


0 20








Figure 10 continued.


10b. Solutions buffered at pH 7 with an H2S (q) concentration
of ca. 500gM.







CARBONATE BUFFER
EDB

PHOSPHATE BUFFER
EDB

CARBONATE BUFFER
HEXANE C-14

PHOSPHATE BUFFER
HEXANE C-14

CARBONATE BUFFER
AQUEOUS C-14
........a .......
PHOSPHATE BUFFER
AQUEOUS C-14













6 uM EDB, 500 uM HS

pH 7


100 250C




sO
80




20
Q...... ....
0



F-0
Z 0 -
Z 0 50 100 150 200 250

0 ". ~40 C
c Z











0
0 20 40 60 80




OC oot 0 O
'"I 60 C


TIME (DAYS)








Figure 10 continued.


10c. Solutions buffered at pH 9 with an H2S(aq) concentration
of ca. 500gM.







CARBONATE BUFFER
EDB

BORATE BUFFER
EDB

CARBONATE BUFFER
HEXANE C-14
........A .......
BORATE BUFFER
HEXANE C-14

CARBONATE BUFFER
AQUEOUS C-14

BORATE BUFFER
AQUEOUS C-14
-6-













6 uM EDB, 500 uM H2s

pH 9

250C

so*

80 .-


40 -







2 0 l0oo) 40 C
LU Z

m 0 80 1


g ^ so C
< 0 0

Z 0


w 20 -


0 20 40 60 80

'" o600C

60 6
A.






20 -


0
0 2 4 6 8 10 12 14


TIME (DAYS)








Figure 10 continued.


10d. Solutions buffered at pH 5 with an H2S( ) concentration
of ca. 10OO0M.







PHTHALATE BUFFER
EDB

ACETATE BUFFER
EDB

PHTHALATE BUFFER
HEXANE C-14

ACETATE BUFFER
HEXANE C-14

PHTHALATE BUFFER
AQUEOUS C-14

ACETATE BUFFER
AQUEOUS C-14
p9