Aldicarb studies in groundwaters from citrus groves in Indian River County, Florida

MISSING IMAGE

Material Information

Title:
Aldicarb studies in groundwaters from citrus groves in Indian River County, Florida
Series Title:
Florida Water Resources Research Center Publication Number 76
Physical Description:
Book
Creator:
Dierberg, F. E.
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Notes

Abstract:
The disappearance of aldicarb 2-methyl-2(methylthio)propionaldehyde O-(methylcarbamoyl)oxime and its two toxic degradation products, aldicarb sulfoxide and aldicarb sulfone, were measured in laboratory studies using groundwaters and subsoils collected from citrus groves in Indian River County, Florida, and incubated under controlled conditions which best represented the in situ environment. The presence of aquifer material on the decomposition of total toxic residues from aldicarb addition in anaerobic groundwaters was pronounced, suggesting bacteria were important in decomposing aldicarb to nontoxic residues. However, aquifer material had only a minor effect on the rate of total toxic residue disappearance when aldicarb sulfoxide or aldicarb sulfone was the primary toxic aldicarb residue, suggesting that chemical hydrolysis in solution was more important in degrading aldicarb sulfoxide and aldicarb sulfone. Based on hydrolysis experiments in sterile pH-buffered distilled water for aldicarb, hydrolysis rate constants became second-order (kOH = 1.94 x 10^3 L mole-1 day-1 at 200 C) at pH 8 and above; acid-catalyzed hydrolysis occurred at pH 4, but not to the same extent as base-catalyzed hydrolysis. Oximes did not interfere with the analysis of total toxic residues under the conditions of the procedures used in this study; nitriles interfered in a positive fashion, but only when toxic residue concentrations were < 10% of the initial concentration. The half-life times for total toxic residue disappearance of aldicarb and its two sulfur-oxidized derivatives in groundwater-saturated subsoils ranged from 10-26 days, suggesting a resumption to the reported faster aerobic degradation rates in the upper soil layers after having undergone slow degradation. In unsaturated subsoils. Based on the degradation rates found in this study, hydrologic parameters obtained for Indian River County subsoils, and amounts of total toxic residue reported entering Florida groundwaters, it was estimated that toxic residues in aldicarb-contaminated groundwaters in Indian River County would migrate only short distances (1-17 ft) before conversion of toxic residues to non-toxic residues was completed. Thus, the exclusion zone of 300 ft from the nearest drinking water well for applying aldicarb is a reasonable restriction for protecting the groundwater resources in Indian River County.

Record Information

Source Institution:
University of Florida Institutional Repository
Holding Location:
University of Florida
Rights Management:
All rights reserved by the source institution and holding location.
System ID:
AA00001574:00001


This item is only available as the following downloads:


Full Text











a-..


PFp~%i~ ~k~C~d~7 Lb~ r
.:. i~ st~
'' ii :.' ~ J
': r
'L
C :~
i G; ,
b
-~; .?---: ~~p- ;f.~ r
,.
,i
-li, i L


Publication No. 76



ALDICARB STUDIES IN GROUNDWATERS FROM CITRUS GROVES
IN INDIAN RIVER COUNTY, FLORIDA



By


F. E. Dierberg



Department of Environmental Science & Engineering
Florida Institute of Technology
Melbourne


-;

~ ~ .. 'i'[ -" "i I ";'
.~ ~ ~ ~- _;-',. -, .
.... "":
:,-. ~ ,, :.- .
D r:'', .. '' T .w
>...,. ~ ~ ~ ~ ~ ~ ~ ` --,., ,."., ., ...,
-..,l. T: "--- -: w, ",= .'.
.'7.=_: ,. -, ..,.' 1,,. ,,, ,: .;

...~~~4 :':, .. "- .:' .-:-: .;'
..o ) '-- .. : :. .. .. : ,-_ .. ,:. ., -,. .; . -,,
t:,~~~~~~ .'._-: o ,-"..:t ; :-,' .:.'' -. .(.-.:-'.,, ., ." -.. .- .;'- :. :' .. .,.o ,i ., ;






r.~-: ,-, -1 -
-~-".,-.,..




'- -:: ... ..:.. -
,. .. r
-,~~~- ':; --",. ".',,i" -
"" :-:-p- : i --

')C~l .. .c Li','... .. ,- .. .-.
', :.L. ._ ... .-., -.o ',


"r~s~gp~i~gaS_ ~r:-

"~? i; -i' ~
%:i
, '1. ":


..

1.,


C P~P~S~
.;
..,- I;~.
, i J
:i i ~i-~'
-----.r r
IL 1~

r-- ~aop~
1 i` ,
.r
i 'I 'j-
;i
i-


h; ..79

^ ,.'


'i-
;;

4
r.


f~9~-~BB~ q~a~e~Ji
'
d
:i
~.


F~8ai~ 1

.'-'

-


-

r
-
I sl$""";p: ;i












TABLE OF CONTENTS


Abstract . . . . . .


Acknowledgements . . . . .


Introduction . . . . .


Objectives . . . . . .


Site Description . . . . .


Sampling Procedure . . . . .


Analytical Methods . . . . .


Field Measurements . . . .


Laboratory Measurements . . .


Laboratory Incubations of Aldicarb, Aldicarb
Sulfoxide, and Aldicarb Sulfone Degradation .


Aqueous Hydrolysis of Aldicarb . . .


Migration of TTR in Groundwaters in Indian River


Quality Assurance . . . . .


Results . . . . . .


Interferences From Non-toxic Aldicarb Residues:


Page


Siv


Svi


1


6


S 7

. 10


. 11


. 11


11


County

. .


. .


Nitriles and Oximes


Hydrolysis Rates in Sterile pH-buffered Distilled Water


Groundwater Characteristics . . . .


Degradation of TTR From Aldicarb, Aldicarb Sulfoxide,
and Aldicarb Sulfone Amended to Groundwaters From
Private Water Supply Wells . . . .


Degradation of TTR From Aldicarb, Aldicarb Sulfoxide,
and Aldicarb Sulfone Amended to Groundwaters and
Saturated Aquifer Material . . . .


Discussion . . . . . . .


Hydrolysis Rates in pH-buffered Distilled Water .


Disappearance of Toxic Residues of Aldicarb in the
Saturated Zone . . . . .


. . . . .


I

. .












Page


Comparison of Degradation Rates of TTR
in the Unsaturated and Saturated Zones


Field Monitoring Studies and Models For
TTR Intrusion into Florida Groundwaters


Lateral Transport of TTR in Shallow
Groundwaters in Indian River County .


Summary and Recommendations . . .


Bibliography . . . . .


Appendix I . . . . .


Appendix II . . . . .


Appendix III . . . . .


. . . 49



. . . 50



. . . 52


. . . 55


. . . 61


. . . 66


. . . 68


. . . 72











ABSTRACT


The disappearance of aldicarb [2-methyl-2(methylthio)propionaldehyde

O-(methylcarbamoyl)oxime] and its two toxic degradation products, aldicarb

sulfoxide and aldicarb sulfone, were measured in laboratory studies using

groundwaters and subsoils collected from citrus groves in Indian River County,

Florida, and incubated under controlled conditions which best represented the

in situ environment. The presence of aquifer material on the decomposition of

total toxic residues from aldicarb addition in anaerobic groundwaters was pro-

nounced, suggesting bacteria were important in decomposing aldicarb to non-

toxic residues. However, aquifer material had only a minor effect on the rate

of total toxic residue disappearance when aldicarb sulfoxide or aldicarb sulfone

was the primary toxic aldicarb residue, suggesting that chemical hydrolysis in

solution was more important in degrading aldicarb sulfoxide and aldicarb sulfonc.

Based on hydrolysis experiments in sterile pH-buffered distilled water for

aldicarb, hydrolysis rate constants became second-order (koH = 1.94 x 103 L

mole-1 day-1 at 200C) at pH 8 and above; acid-catalyzed hydrolysis occurred at

pH 4, but not to the same extent as base-catalyzed hydrolysis. Oximes did

not interfere with the analysis of total toxic residues under the conditions

of the procedures used in this study; nitriles interfered in a positive fashion,

but only when toxic residue concentrations were <10% of the initial concentrating

The half-life times for total toxic residue disappearance of aldicarb and

its two sulfur-oxidized derivatives in groundwater-saturated subsoils ranged

from 10-26 days, suggesting a resumption to the reported faster aerobic degrad,

tion rates in the upper soil layers after having undergone slow degradation in

unsaturated subsoils. Based on the degradation rates found in this study,

hydrologic parameters obtained for Indian River County subsoils, and amounts











of total toxic residue reported entering Florida groundwaters, it was estimated

that toxic residues in aldicarb-contaminated groundwaters in Indian River

County would migrate only short distances (1-17 ft) before conversion of toxic

residues to non-toxic residues was completed. Thus, the exclusion zone of

300 ft from the nearest drinking water well for applying aldicarb is a reason-

able restriction for protecting the groundwater resources in Indian River County.









ACKNOWLEDGEMENTS


This study was supported in part under an annual allotment grant from

the Florida Water Resources Research Center to Forrest E. Dierberg, Associate

Professor of Environmental Science and Engineering, Florida Institute of

Technology. Chris Given served as research assistant and completed his M.S.

thesis on this project. The assistance of Brian Combs, Indian River Agri-

cultural Extension Agent, in locating accessible sampling sites and in helping

with sample collection is gratefully acknowledged. Larry Pollack, a graduate

student in the department, also contributed to the research findings found

in this report.

The interest and assistance of Carl Miles, Department of Environmental

Engineering Sciences, University of Florida, helped to make this report more com-

plete. The co-operation of Willis Wheeler and David Atherton, of the Pesticide

Research Lab, Institute of Food and Agricultural Sciences, tUniversity of

Florida, Gainesville, and Robert Patton of the Florida Department of Environ-

mental Regulation, Tallahassee, in the inter-laboratory exchange of samples

is appreciated. Victor Carlisle of the Soil Science Department, University of

Florida, and Carol Wettstein of the U.S. Soil Conservation Service provided

the information on the soils in Indian River County, Finally, the U.S.

Environmental Protection Agency, Pesticides and Industrial Chemicals Repository,

Research Triangle Park, North Carolina, and Union Carbide kindly provided

standards for aldicarb and its decomposition products.









INTRODUCTION


Aldicarb (TemikR) is an effective but non-selective systemic insecticide,

miticide, and nematicide; its acute mammalian toxicity (rat acute oral LD50,

1 mg/kg; rabbit dermal LD50, 5 mg/kg) (ICET 1983) makes it one of the most toxic

of all currently-registered insecticides. Aldicarb and its toxic oxidized

metabolites sulfoxidee and sulfone), like other carbamate esters, exert their

insecticidal activity through reversible inhibition of the enzyme acetyl-

cholinesterase. There is no evidence that either aldicarb or its metabolites

are associated with any adverse mutagenic, carcinogenic, or teratogenic effects.

As it stands now, the conditions of exposure can be viewed as a series of low-

level exposures rather than continuous chronic exposure. Detailed reviews on

the toxicology of aldicarb and its sulfoxide and sulfone metabolites can be

found in reports issued by the Institute for Comparative and Environmental

Toxicology (ICET 1983) and aldicarb's manufacturer, Union Carbide (1983).

Because of its high mammalian toxicity, aldicarb is available only in granular

formulations (5-20% active ingredient) for soil incorporation.

Its primary use in Florida is to protect citrus groves Cin the central and

southern parts of the state) and potatoes (in north Florida) from aphids, mites

and nematodes (Jones and Back 1984). Concern for the potential of aldicarb or

its toxic metabolites to contaminate underground aquifers used for drinking

water in Florida heightened after reports that aldicarb had been detected

for the first time in groundwater in Suffolk County, New York, in August

1979 (Zaki et al. 1982). Since the sandy soils in Suffolk County were not

too dissimilar from Florida soils and because of the high water solubility

(=6,000 ppm) and non-volatility (vapor pressure = 1 x 10 mm Hg at 250C) of









the pesticide, both of which would preclude significant adsorption onto soil

particles, extensive monitoring, field, and laboratory studies were sub-

sequently undertaken in the state (IFAS 1983). Other states such as Wisconsin

(Chesters et al. 1982) also initiated monitoring networks after detecting

aldicarb residues in groundwaters.

The regulatory history of aldicarb in Florida can be briefly summarized

below:


August 1982






January 28, 1983


October 1983




January 1, 1984


Media reported Temik-contaminated water in

other states; Florida began testing for and

finding traces in underground water supplies.


State-wide ban went into effect.


Nation-wide ban on ethylene dibromide (EDB)

increased the potential use of Temik.


Ban lifted, but with the following restrictions:

Sno more than 5 Ibs active ingredient (a.i.) can be

used per acre (formerly 15 lbs a.i. per acre).


applied only once a year between January 1

and April 30 (for citrus).


. cannot be applied within 300 ft of any drinking

water well.


. use suspended in an area if drinking water

>10 ppb.

. notification of impending treatment must be posted

prominently on property where it is to be applied.











The degradation of aldicarb in plants, animals and soil is dominated by

two processes: oxidation of sulfur to obtain sulfoxide and sulfone analogs,

and cleavage of the carbamate ester bond (Fig. 1). Oxidation is the major

pathway for aldicarb metabolism in most systems (ICET 1983). Which degradation

pathway predominates is important since hydrolysis of the ester linkage

detoxifies aldicarb and its sulfur-oxidized derivatives, while oxidation of

sulfur yields metabolites that retain-the toxicity of the parent compound.

Aldicarb oxidizes readily to sulfoxide in plants (Maitlen et al. 1968) and

soils (Smelt et al. 1978c: Bromilow et al. 1980). It is the sulfoxide which is

the most potent cholinesterase inhibitor of the group (Fig. 1), and responsible

for the high systemic activity and long-term persistence of insecticidal

activities.

Much of the research to date on aldicarb and its sulfur-oxidized derivatives

in Florida have been fate studies in soils and crops (IFAS 1983). Little is

known regarding the chemical behavior of aldicarb and its oxidized

metabolites in shallow groundwaters in Florida where drinking water is obtained

from many private wells. Funding for the project began in August 1983; field

sampling, laboratory studies, and chemical analyses continued through June 1984.

Continued research on aldicarb in groundwater is being conducted beyond the funding

period of this report. Preliminary data contained in this report were presented at

the Florida Academy of Sciences meeting on March 30, 1984, in Boca Raton, Florida.

The practical aspects of the study focused on the degradation of aldicarb

and its oxidized metabolites in groundwaters from 3 shallow wells (a 20 m deep)

located near citrus groves in Indian River County. Two other wells (% 1.4 m deep)

inside citrus groves were augered to study the effects of aquifer material on the

rate of degradation. In one case, groundwaters and aquifer material were
















DEHYDRATION


CH3 0
CH3SCCH=NOCNHCH3
CH3

aldicarb
(0.84 mg/kg)


OCH 0
111 3 11
CH3SCCH=NOCNHCH3
CH3

aldicarb sulfoxide
(0.49-1.13 mg/kg)


OCH3 0
CH3SCCH=NOCNHCH3
OCH3

aldicarb sulfone
(20-45 mg/kg)


CH
I 3
CH3SCCH=NOH
CH3

aldicarb oxime
(2380 mg/kg)


OCH3
11I 3
CH3SCCH=NOH >
CH3

sulfoxide oxime
(8060 mg/kg)


I
OCH
CH3SCCH=NOH --
OCH3

sulfone oxime
(1590 mg/kg)


CH3

3
CH3CCN
CH3


aldicarb nitrile
(570 mg/kg)


OCH
ni 3
CH3SCC-=N
CH3

sulfoxide nitrile
(4000 mg/kg)


OCH
CH SCC=N
3,111
OCH3

sulfone nitrile
(350 mg/kg)


Source: The Institute for Comparative and Environmental Toxicology 1983


Degradative Pathways of Aldicarb. Large Arrows Indicate Major
Pathways. Values in Parentheses are Acute Oral LD50's for Rats.


Figure 1.


HYDROLYSIS









sterilized while replicate subsamples were left unsterilized so as to determine

what role, if any, the groundwater microflora play in the degradation process.

Finally, carefully conducted hydrolysis experiments in pH buffered distilled

water of 4, 6, 7, 8, 9 and 10 were carried out at room temperature to further

understand the significance of acid or base catalysis and to obtain reliable

half-life values for the pH range found for natural waters.










OBJECTIVES


The objectives of this study were as follows:

1. To determine the extent that non-toxic oximes and nitriles interfered

in the analysis of total toxic aldicarb residues by gas chromatography;

2. To perform a carefully controlled hydrolysis experiment for understanding

what effect pH has on hydrolyzing aldicarb under sterile conditions;

3. To examine shallow groundwaters with and without aquifer material collected

in citrus groves for their capacity to degrade aldicarb and its two

toxic derivatives, sufoxide and sulfone, to non-toxic residues in labora-

tory experiments; and

4. To integrate the kinetic expressions of these experiments with a simple

groundwater transport model and field data published in the literature

in arriving at an estimate of the distances groundwater contaminated with

toxic residues of aldicarb would migrate in Indian River County.











SITE DESCRIPTION


Five well sites bordering or within citrus groves in Indian River County

(Figure 2) provided groundwaters and aquifer material for aldicarb, aldicarb

sulfoxide, and aldicarb sulfone additions in the laboratory degradation studies.

Three of the wells Ryall, Luther, and Sexton were private water supply

wells with attached pumping systems. Two of the wells directly supplied the

Luther and Sexton residences; the Ryall well was used for irrigation, but

water from it was sometimes consumed by grove workers. Well construction

reports were not available to verify their depths, but owners believed that

their wells were relatively shallow, ranging between 16 and 20 m, closely

approximating the depths which are commonly used in providing drinking water

to private homes. Only groundwaters were sampled from the 2-inch diameter

piping of these wells. The remaining two wells (BBC and Lindsey Wabasso)

were drilled by using a bucket auger, and their very shallow depths (1.3-1.5 m)

coincided to the top of the water table in the unconfined aquifer. Any

transport of toxic residues to the deeper groundwaters which are used for

domestic water supply would have to first pass through this layer.

The soils in Indian River County are currently being mapped by the SCS.

The soils in the field sites which were hand-augered (BBC and Lindsey Wabasso

groves) are classified as Pineda sand (Arenic Glossaqualfs), Riviera sand

(Arenic Glossaqualfs), and Wabasso sand (Alfic Haplaquods). They are all

nearly level (slopes less than 2 percent), poorly drained, slowly to very

slowly permeable soils. Permeability is rapid (15-51 cm/hr) in the sandy A

horizons and slow to very slow in the sandy clay loam B horizons ( <0.5 cm/hr).

Clay content averages 12 to 30% in the B horizon where subsoils representing

aquifer material were sampled. The water table is within 10 inches of the













ATLANTIC
OCEAN


Location of the Five Study Wells in Indian
River County, Florida. Li = Lindsey Wabasso
grove; R = Ryalls; Lu = Luther; S = Sexton;
BBC = BBC grove.


Figure 2.










surface for 1 to 6 months in most years and 10 to 30 inches deep most of the

rest of the year. Some areas are flooded for periods ranging from a few

days to about 3 months.









SAMPLING PROCEDURE


Groundwater samples from the three private water supply wells (Ryall,

Luther, Sexton) (Fig. 2) within 30 m of citrus groves were collected on

December 9, 1983, and March 26, 1984 after pumping an amount of water greater

than three volumes of the water standing in the pipe. Water samples were

taken from the closest accessible point to the well head, usually from the pump

itself or just after the pump. Care was exercised in avoiding entrapment of

air and degassing in the sample containers by flushing several volumes of

groundwater before caping each container, Samples to be used in laboratory

incubations were put into sterilized 4-L amber-colored glass containers after

at least one volume had been allowed to overflow and placed in a cooler filled

with ice for transport to the laboratory. Several BOD and Nalgene bottles

were filled in the same manner with sample groundwaters for field and laboratory

measurements.

In addition to sampling only groundwaters from private water supply wells,

two shallow wells each at the BBC and Lindsey Wabasso sites (Fig. 2) were

hand-augered on February 22, 1984 and January 30/April 23, 1984, respectively,

using a 3-inch (ID) x 14-inch (length of containment area) soil auger. The

wells were located in a furrow separating two rows of citrus trees. Groundwater

was sampled through 1-inch (OD) Flex PVC heavy wall tubing connected to a

12 gpm maximum flow "Guzzler" hand pump; aquifer material from saturated sub-

soils was withdrawn from these self-constructed shallow wells (1,3-1.5 m)

using the soil auger, The same field and laboratory analyses were done on

these augered wells as already described for the private water supply wells.








ANALYTICAL METHODS


Field 'Measurements

Redox potential was measured with a platinum electrode and a saturated

calomel electrode (SCE) as the reference electrode. Calibration of electrodes

and potentiometer (Fisher Accumet 640 Mini-Meter) was made against a Fe +2/Fe+3
standard solution (Light 1972). Adjustments were made in the measured poten-

tials to a standard hydrogen reference electrode (Light 1972) at pH 7 (Patrick

and Mahapatra 1968). In addition to redox potential, dissolved oxygen (Leeds

and Northrup 7932 portable dissolved oxygen meter), specific conductance (YSI

Model 33 S-C-T meter), and pH (Fisher Accumet 640 Mini-Meter) were also measured

potentiometrically using the appropriate sensors. Temperature was recorded

using a mercury thermometer.

Laboratory Measurements

General Water Chemistry Analyses

The following chemical constituents were determined titrimetrically

according to the procedures given in Standard Methods For the Examination of

Water and Wastewater (APHA 1976): total alkalinity (0.02 N H2SO4), total hard-

ness C0.02 M EDTAY, and sulfide (iodine), Total iron was measured by atomic

absorption spectroscopy with a Perkin-Elmer Model 460 atomic absorption

spectrometer after preserving sample by acidifying to pH<2 in the field.

Standard plate counts were conducted anaerobically by the Vacuum and

Gas Displacement Method (Benson 1973) on the groundwaters and saturated sub-

soils from the private water supply wells and augered shallow wells before

(within 24 hours of field sampling) and at the end of each incubation.










Aldicarb Residue Analyses

After adding 0.10 mg/L of aldicarb or one of its S-oxides (aldicarb

sulfoxide and aldicarb sulfone), pH-buffered distilled water, groundwater,

and subsoil were analyzed for the sum of aldicarb and its S-oxides at various

time intervals according to the methodology developed by Union Carbide (1980).

This metholdlogy purports to determine the sum total of toxic aldicarb

residues (Fig. 1) Ci.e. aldicarb + aldicarb sulfoxide + aldicarb sulfone),

which has been referred to as the "total toxic residue" (TTR). It wasn't

learned until later into the research period of the potential "positive"

interference from non-toxic nitriles. An assessment of the extent that the

nitrile may have interfered in those earlier TTR analyses where it hadn't

been removed is presented in the RESULTS section,

At appropriate intervals, fifty or one-hundred mL aliquots were removed

from each duplicate incubation container (BOD bottle or Mason jar) of pH-buffered

distilled water solution or groundwater and placed into a 125-mL or 250-mL

separatory funnel. Oxidation of aldicarb and aldicarb sulfoxide to aldicarb

sulfone was accomplished by adding 2 mL of peracetic acid. Conversion of

aldicarb and aldicarb sulfoxide to aldicarb sulfone is necessary since

chromatographic peaks of the former are indistinguishable from the solvent peak

using gas chromatograph (GC) (Galoux et al. 1979). The contents in the

separatory funnel were mixed and allowed to stand for 30 minutes with occasional

mixing, After 30 minutes, 15 mL of 10% NaHCO3 were added, mixed, and allowed

to stand with occasional mixing for 15 minutes. Fifty mL of methylene chloride

were added with frequent venting to release evolved CO2. After the layers

separated, the lower methylene chloride layer was drained through approximately

80 g of prewet sodium sulfate in a 4-inch funnel with a glass wool plug. The









extraction was repeated with another 50 mL of methylene chloride and the

extracts combined. After rinsing the sodium sulfate bed with an additional

20 mL of methylene chloride, the combined extracts and rinse were collected

and concentrated in a 450C water bath by evaporating just to dryness with a

stream of dry N2 gas* The residue was then dissolved in acetone to 1-2 mL

and stored in 3-mL septum-capped vials at -50C until analyzed by GC,

Groundwater-saturated subsoil samples (50 g dry wtY were extracted

using the method of Galoux et al. (1979), The subsoils C50 g dry wt) were mixed

with 50 mL of acetone-water (40:60), shaken for 30 min and centrifuged for

5 min at 3000 rpm. The extract was transferred to a 250-mL separation funnel.

A second extraction was performed with 40 mL of methanol-water (50:50), and the

two extracts were combined. The acetone extract was next extracted 3 times with

50 mL methylene chloride in a 500-mL separatory funnel. After draining the

methylene chloride fraction through sodium sulfate beds, combining, and con-

centrating to approximately 2-mL in a 450C water bath, 10 mg of m-chloroper-

benzoic acid was added to oxidize the aldicarb and aldicarb sulfoxide, if

present, to aldicarb sulfone (Smelt et al. 19786). The residue was dissolved

in 1-2 mL acetone for GC analysis after taking to dryness under N2 in a 450C

water bath.

As previously stated, if non-toxic nitrile derivatives of aldicarb and its

S-oxides are present in the sample, they would serve as a positive interferent

(i.e. included with the toxic residues). The high injection port temperature

(2600C) pyrolyzes the toxic aldicarb and its S-oxides to nitriles

(Knaak et al. 1966; Trehy et al. 1984), which are then detected by the flame

photometric detector in the GC. To remove the nitrile interference, 0.5-cm(ID)

x 60-cm glass columns were filled to a depth of 14 cm with 5 g of PR grade

Florisil. After pre-washing the columns with 25 mL of methylene chloride but










before it reached the top of the Florisil, the eluted pre-wash was discarded

and the sample immediately added to the top of the column. When the sample

reached the top of the column, 100 mL of 5% acetone in ethyl ether were

added to the column, and the eluate discarded when the solvent reached the

top of the Florisil. A second solvent fraction consisting of 50 mL of 50%

acetone in ethyl ether was then added to the column, and the eluate collected

in a 125-mL Erlenmeyer flask. The second fraction was evaporated under an N2

gas stream in a 450C water bath to less than 2 mL and then transferred to a

3 mL septum cap vial. The flask was rinsed with small amounts of acetone and

the rinse added to the vial. The volume in the vial was reduced under N2 gas

to a final volume of 1 mL.

Gas chromatography was performed on a Perkin-Elmer Sigma 300 GC equipped

with a flame photometric detector with a 394 nm filter to quantify the sulfone.

A coiled glass column, Im x 2mm (ID), packed with 5% SP-1000 on Supelcoport

(100/120 mesh) was used for separation. Normal operating conditions were 2600C

injector temperature, 1750C column temperature, 2500C detector temperature,

and helium, hydrogen, and air flow rates of 35, 20, and 26 mL/min, respectively.

The minimum detectable concentration was 1,8 ng absolute or 300 ng/mL in a

6 uL injection, which corresponded to total aldicarb residues extracted from

water solutions originally containing 3 to 12 ppb, depending on the volume

of water and the final volume of the extract. Chromatograms were reported

on a Hewlett-Packard 3390A Integrator in the linearized mode and a Varian 9176

recorder in the non-linearized mode.

Laboratory Incubations of Aldicarb, Aldicarb
Sulfoxide. and Aldicarb Sulfone Degradation

Incubation containers consisting of BOD bottles for groundwaters and

946-mL Mason jars for groundwater-saturated subsoils were initially filled to









completely occupy the entire volume of the container, The BOD Bottles were

stoppered and a small amount of silicone stopcock grease was placed around the

outside edge of the stopper and Bottle to help prevent any exchange of gases

from occurring. All containers were incubated in the dark at the ambient

temperatures of the groundwaters measured in the field (22 + 2 to 26 + 2 C).

All pH values were checked before and after each incubation period,

To insure that anaerobic conditions were maintained during those times

when aliquots of groundwater and groundwater-saturated subsoil were withdrawn

from incubation containers, disposable polyethylene gLove bags with a stream of

N2 constantly purging the bag were used. A Leeds and Northrup dissolved

oxygen meter equipped with a BOD probe set on "air calibrate" recorded zero

oxygen in the bags and headspace of the containers after sampling; also, the

water of those samples incubated in Mason jars were devoid of oxygen as recorded

by the dissolved oxygen meter.

Sterile and Non-sterile Degradation of Aldicarb

Duplicate sterile and non-sterile groundwaters from Ryall, Sexton, and

Luther wells sampled on December 9, 1983, were spiked with aldicarb in BOD

bottles to yield a final concentration of 0,10 mg/L, and incubated at measured

ambient field temperature of 24 + 20C. Groundwaters (500 mL) with and without

saturated subsoils (1800 g dry wt. from 1.1 to 1.5 m depths) obtained from the

Lindsey Wabasso grove on January 30, 1984, were also sterilized and left non-

sterile at 22 + 2 C so as to determine what effects the presence of aquifer

material had on aldicarb decomposition. A second site (BBC grove) was cored to

retrieve groundwaters with and without saturated subsoil from a depth of 0.9 to

1.2 m on February 22, 1984, but all incubations using water (500 mL) and aquifer_

material (1200 g dry wt) from this site remained non-sterile. Sterile











and non-sterile controls without added aldicarb were run for each well.

Sterilization was accomplished by adding sodium azide (NaN3) to give

a final concentration of 0.1% in the groundwater-only incubations (Sharom

et al. 1980) and 0.4% in the groundwater-saturated subsoils. To deter-

-mine whether sterility was maintained throughout the incubations,


thioglycollate broth media was employed on the first and last days to test for

contamination in the sterilized samples and was negative in only the groundwater

containers.

Fifty-mL aliquots of groundwater from the private water supply wells

incubated in the BOD bottles were withdrawn at times 0, 5, 11, 34 and 90 days

and analyzed for TTR. In the case of the groundwater with and without aquifer

material from the Lindsey Wabasso and BBC sites, subsampling from the 946-mL

Mason jars serving as incubation containers at time 0 occurred when the

sediment had settled (approximately 1 hour) after being shaken; subsampling

occurred again after 30 (Lindsey Wabasso) and 23 (BBC) days had elapsed.

Saturated subsoil material was frozen until extraction.


Non-sterile Degradation of Sulfoxide and Sulfone

Duplicate BOD bottles containing non-sterile- groundwaters from the private

water supply wells were spiked with sulfoxide or sulfone to yield a final concen-

tration of 0.10 mg/L. Shallow groundwaters and saturated subsoils from 0.9- 1.2 m

(BBC) and 1.2-1.5 m (Lindsey Wabasso) depths were obtained on February 22

and April 23, 1984, from BBC and Lindsey Wabasso groves, respectively. As

before, sulfoxide or sulfone was amended to the Mason jar incubation containers

to produce an initial concentration of 0.10 mg/L. Sterile samples were not run

because results from the sterile vs. non-sterile aldicarb experiments indicated








sterile conditions were unattainable for incubations where the aquifer material

was included. Controls without amended sulfoxide or sulfone were set up in

duplicate for all groundwaters from all wells with and without added subsoil. An

incubation temperature for the deeper groundwaters from the private water supply

wells was 24 + 20 C; groundwaters and groundwater-saturated subsoils from the

augered wells were incubated at 22 + 20C CBBC) and 26 + 20C (Lindsey Wabasso).

Either 50 or 100-mL aliquots were removed at 0, 5, 20 and 40 days from the BOD

bottles used to incubate the groundwaters sampled from the private water supply

wells. Water and saturated subsoil were withdrawn from the Mason jars used

as incubation containers for the augered wells at BBC and Lindsey Wabasso

groves at time Q and after 23 or 25 days, and extracted as previously described.

Aqueous Hydrolysis of Aldicarb

Buffered reaction solutions for the aldicarb hydrolysis experiment were

prepared by adding 3 mL of a 30 mg/l aldicarb solution in water to a solution of
_2
297 mL of distilled-deionized-carbon-filtered water and 1.67 x 10-2 M of pH
-3
buffer solution C4.1 x 10-3 M buffer for pH 8.85) in duplicate BOD bottles to

give a 0.10 mg/L aldicarb concentration. The sterile buffer systems and their

final pH values after incubating 15-89 days were potassium hydrogen phthalate-

hydrochloric, acid CpH 3,95), potassium dihydrogen phosphate-sodium hydroxide

(pH 6.02, 7.06, and 7,96), sodium tetraborate-hydrochloric acid (pH 8.85), and

sodium bicarbonate-sodium hydroxide (pH 9.85). Buffer solutions and glassware

were autoclaved before use. The BOD bottles were plugged with sterilized foam

plugs and placed in an incubator in the dark at 200 + 2 C. Controls without

aldicarb were included. All pH values were checked before and after each

kinetic run.












Pseudo-first-order rate constants, k, were obtained from the slope of

the line C-2,30 X regression coefficient) obtained by a linear least-squares

analysis of the data for those experiments where samples were collected and

analyzed over multiple time intervals (i.e., aldicarb hydrolysis in pH-buffered

distilled water; Ryall, Sexton, and Luther well-water). For the shallow wells

at BBC and Lindsey Wabasso where aquifer material and groundwaters were sampled

and analyzed for only one time interval, the integrated first-order rate equation

was used
-k = In C/Co C)
t




where Co is the initial concentration of aldicarb or one of its primary oxida-

tion products and C is the remaining concentration at time t, the corresponding

half-life was from

t = 0.693/k (2)


The second-order reaction rate constants, kOH and k,, were determined from

the slope between pH 8 and pH 10 and between pH 4 and pH 6 in a plot of log k

vs. pH according to the relationships


kH = and (3)
OH OH


kk (4)
[H-4










Migration of TTR in Groundwaters in Indian River County

Lateral movements of theoretical TTR plumes were calculated using Darcy's

Law

dh
v = -K-/ a
dl (5)

where v = velocity Cm day- ); K = hydraulic conductivity (m day-1);

dh
d- = hydraulic gradient (m/m); and a = porosity,

Values for the parameter K were obtained from U.S. Soil Conservation Service

Soil Interpretation Records for the soil series (Pineda sand, Riviera Sand, and

Wabasso sand) surveyed recently to be present at the BBC and Lindsey Wabasso

sites. Water table data from observation wells in Indian River County provided

by the Soil Survey of the U.S. Soil Conservation Service (C. Wettstein, pers. comm.,

May 30, 1984) were converted to heights above sea'level after adjusting-for,

elevation differences in ground level between well locations. The average

of the differences between water table surfaces based on 12 measurements

during an 18.month period CNovember 1982 May 1984) divided by the distance between

two wells generated realistic hydraulic gradients (dh/dl) for the county. A

total of five observation wells were used to compute three different hydraulic

gradients. Porosity was assumed to be 33%.










QUALITY ASSURANCE


In order to insure reliability in accuracy and precision of analyses, the

following quality control measures were carried out:

1. All sample incubations were done in duplicate. Expressed as percentages of

their averages, the differences for all aldicarb samples were 11% (0-30%);

for aldicarb sulfoxide samples 9% (0-33%); and for the aldicarb sulfone

samples 22% (3-62%).

2. Blanks were analyzed for each well water, aquifer material, and aldicarb

substrate combination, and were always below the limit of detection (i.e.,

43 to <12 ppb, depending on the volume of water extracted and final

volume of the extract).

3. A total of 22 spikes were performed (16 with aldicarb spiked into water;

4 with aldicarb sulfone spiked into water; and 2 with aldicarb spiked into

saturated subsoils). Recoveries ranged from: 93 to 109% (ave. = 103%)

for the aldicarb-spiked water samples, 104% for all the aldicarb sulfone-

spiked waters, and 88 to 92% (ave. = 90%) for the aldicarb-spiked subsoils.

Results were not corrected for the recovery percentage.

4. Inter-laboratory comparisons with two independent laboratories: 1) Florida

Dept. of Environmental Regulation; and 2) Pesticide Research Lab of the

Institute of Food and Agricultural Sciences, University of Florida,

Gainesville. A total of 4 samples were exchanged (2 samples to each lab),

and their results were within 20%, except for one sample which differed

by 29%. The samples, concentrations, and gas/liquid chromatographic

systems are presented in Appendix I.

5. Twenty-two groundwater samples underwent separation by liquid chroma-

tography using Florisil columns after oxidation by peracetic acid. Two










eluants were used: one to retrieve the non-toxic aldicarb residues

.(fraction 1 = 5% acetone in ethyl ether) and the other to elute toxic aldi-

carb residues (fraction 2 = 50% acetone in ethyl ether). The data from

which errors were calculated for recovery of all residues after Florisil

separation as well as for not removing nitriles by Florisil separation

are provided in Tables 1 and 2 of the RESULTS section. A discussion of the

importance of the computed errors is also presented in the RESULTS section.

6. Standards of aldicarb sulfone and aldicarb sulfone nitrile were added to

water and separated by Florisil into two fractions: one containing the

nitrile and the other containing the sulfone. Recoveries were approxi-

mately 111% for the nitrile and 93% for the sulfone. The data are pro-

vided in Table 2 of the RESULTS section.

7. Readings of pH for the laboratory experiments were taken at each time a

subsample was withdrawn for TTR analysis and pH was found to increase

approximately 1 pH unit within 10 days of the initial time in the private

water supply wells (Ryall, Sexton, and Luther). The pH hydrolysis experi-

ment for aldicarb and the hand-augered wells showed pH variations within

0.2 pH units during the incubation period, except for the Lindsey

Wabasso samples taken on January 30, 1984, which had an increase of

Nv0.9 pH unit.

8. A check as to whether positive interference from oximes would result for

those samples not passed through a Florisil column revealed that they would

not interfere at the concentrations of toxic aldicarb residues used in this

investigation. Details of the check-out procedure are given in the

RESULTS section.

9. Anoxic conditions were maintained throughout the laboratory experimental

studies since dissolved oxygen could not be detected at the end of the










incubation period in the groundwaters or saturated subsoils from the augered

wells which were contained in Mason jars; oxygen was also not detected in

the atmospheres in the BOD bottles used for experiments on aldicarb

hydrolysis and degradation of TTR in waters from private wells for

aldicarb, aldicarb sulfoxide, and aldicarb sulfone amendments.

10. An evaluation of the buffer catalysis contribution to the hydrolysis of

aldicarb for each pH-buffered solution indicated that there was minimal

contribution, if any, from the nature and concentration of the acid-base

system used to buffer the pH. A full account of the evaluation is provided

in Appendix II.

11. Tests for enumerating bacterial densities at the end of each incubation

period were always conducted to evaluate whether sterilization had been

maintained throughout the duration of the experimental period, or, in

the case of non-sterile samples, that incubation conditions had closely

approximated the in situ environment such that bacterial densities

after the incubation period did not differ significantly for densities

measured at the beginning of the incubation period. In all cases,

bacteria densities remained close to their initial values during the

incubation period. Raw data for bacterial counts can be found in Appendix

III.










RESULTS


Interferences From Non-toxic Aldicarb Residues: Nitriles and Oximes

Care must be exercised in labeling as total toxic residues any sample

extractions which have not been passed through a Florisil column to remove

nitriles and oximes (Romine 1974) prior to GC analysis. If a significant

proportion of aldicarb residues is comprised of the non-toxic oximes or

nitriles, then they would serve as a positive interferent by chromatographing

with the toxic aldicarb and its S-oxides; neither nitriles nor oximes should

be included in any analytical procedure designed for measuring TTR.

Since the method followed at the beginning of this study was an unpublished

one (Union Carbide 1980) which made no mention of removing non-toxic aldicarb

residues from the toxic aldicarb residues by Florisil separation, twenty-two

groundwater samples collected toward the end of the period of investigation

(when the possibility of a nitrile interference became known) were eluted

through Florisil columns and fractions .1 and 2 analyzed for aldicarb residues

(Table 1). Ten of the 22 samples were split before Florisil separation and

one aliquot put through a Florisil column and the other aliquot left without

Florisil separation (Table 2).

Large ranges in percentage recoveries (57-129%) were found for the ground-

water samples passed through Florisil-columns when compared to the aliquots

left without separation by liquid chromatography, with an average of 99% for

the 10 samples (Table 2). Interferences from nitriles, which would result in

an over-estimation of TTR, became quantitatively important only near the end

of the incubation period in those samples that had >90% disappearance of the

initial TTR (Table 1). Although the absolute error was large for those par-

ticular samples because of the low concentrations of TTR remaining ( <10 ppb),











Table 1. Amount of Non-toxic Nitriles and Toxic Aldicarb Residues Remaining

in Duplicate Groundwaters and Subsoils After Florisil Separation.


Compound Added

aldicarb sulfoxide
aldicarb sulfoxide

aldicarb sulfone
aldicarb sulfone

aldicarb sulfoxide
aldicarb sulfoxide

aldicarb sulfone
aldicarb sulfone

aldicarb sulfoxide
aldicarb sulfoxide

aldicarb sulfone
aldicarb sulfone


Amount Remaining, ppb

Time 0 Time 20 Days Time 42 Days

TR T TTRa Nitrileb TTRa Nitrileb

97 54 ND -
105 4 10

81 43 3 ND 5
97 28 ND -

97 34 ND 4 5
98 35 4 -

80 12 ND ND -
92 37 ND 13

95 43 6 9 9
92 40 13 6

90 20 5 ND ND
94 30 ND ND


Sample

Wab. water-A
Wab. water-B

Wab. subsoil-A
Wab. subsoil-B

Wab. water-A
Wab. water-B

Wab. subsoil-A
Wab. subsoil-B


Compound Added

aldicarb sulfoxide
aldicarb sulfoxide

aldicarb sulfoxide
aldicarb sulfoxide

aldicarb sulfone
aldicarb sulfone

aldicarb sulfone
aldicarb sulfone


Amount Remaining, ppb

Time 0 Time 25 Days

TTR TTRa Nitrileb

79 54 ND
80 53 3

51 12 ND
62 10 5

90 30 3
90 29 11

64 10 5
78 16 ND


a TTR = eluate from fraction 2 (50% acetone in ethyl ether) of

b Nitrile = eluate from fraction 1 (5% acetone in ethyl ether)

ND -non-detectable (<3 ppb)


Florisil column

of Florisil column


Sample

Ryall-A
Ryall-B

Ryall-A
Ryall-B

Sexton-A
Sexton-A

Sexton-A
Sexton-B

Luther-A
Luther-B

Luther-A
Luther-B


Lindsey
Lindsey

Lindsey
Lindsey

Lindsey
Lindsey

Lindsey
Lindsey











Table 2. Recovery of Aldicarb Sulfone and Aldicarb Sulfone Nitrile From Water


and Sediments After Eluting Through Florisil Columns.


Measured, ppb
No Florisila Florisilb
Amount Added Florisil Fraction Fraction
To Water Separation 1 2


No
Florisil
Separation


Recovery,
Florisila
Fraction


1 2


35 ppb Sn
62 ppb Sn nitrile
35 ppb Sn
+
62 ppb Sn nitrile

Amount Measured in
Sample Water or Subsoil


Lind. Wab. Water
Sx added; 25 days

Lind. Wab. Water
Sn added; 25 days

Lind. Wab. Subsoil
Sn added; 25 days

Lind. Wab. Subsoil
Sx added; 25 days


Ryall
Sn added;

Sexton
Sn added;

Luther
Sn added;

Luther
Sx added;

Luther
Sx added;

Luther
Sx added;


20 days



20 days



20 days



20 days



42 days



42 days


14



46



21



26



65



14


Average % Recovery For Samples


Florisil"
Fraction


107



100



57



96



71



129


119


S 99%


a Florisil fraction 1 = 5% acetone in ethyl ether and contains aldicarb Sn nitrile.
b Florisil fraction 2 = 50% acetone in ethyl ether and contains aldicarb Sn
ND = non-detectable (<3 ppb)
Sx = Sulfoxide; Sn = Sulfone











the relative error based on the initial TTR concentrations was small (average

of 6% with a range of 0-14% for the 8 groundwater samples from the private

water supply wells spiked with aldicarb sulfone or aldicarb sulfoxide).

Considering that this is less than the individual percentage errors associated

with recoveries of aldicarb sulfone nitrile and aldicarb sulfone from Florisil

separation of water spiked with standards (111% and 93%, respectively) (Table 2),

recoveries of total toxic plus non-toxic residuals remaining in groundwater sam-

ples after Florisil separation (57-129%) (Table 2), inter-laboratory comparisons

of split samples (-20%), and the average error between duplicate samples (9

and 22% for aldicarb sulfoxide and aldicarb sulfone, respectively), the extra

time and effort involved in the Florisil removal of nitriles may not be

warranted. Hansen and Spiegel (1983) also observed in their hydrolysis studies

that presence of nitriles was important only after 98 to 99% of the aldicarb

sulfoxide or sulfone had been degraded. Furthermore, Union Carbide (Romine,

pers. comm.; May 1, 1984) has not found nitriles of aldicarb or its sulfoxide

or sulfone as residues in potable groundwater from their monitoring of many ground-

water sources in several states, including Florida. Neglecting to remove

nitriles from some of the earlier samples apparently was not a significant

source of error in analyzing for TTR.

Apparently aldicarb oxime does not pose as an interferent since it is not

thermally decomposed to nitrile at an injection port temperature of 3500C

(Knaak et al. 1966). However, the aldicarb sulfoxide oxime and aldicarb

sulfone oxime decompose to nitriles at 3500C (Knaak et al. 1966), and thus

could also serve as a positive interferent, especially since any aldicarb

oxime and aldicarb sulfoxide oxime present would be oxidized to aldicarb

sulfone oxime by the peracetic acid oxidation step. Studies done in our

lab using either aldicarb sulfoxide oxime-or aldicarb sulfone oxime at final










concentrations of 5 ug/mL in the acetone extract, and without a Florisil

separation step, did not elicit a response on the GC under the operating

conditions set for the instrument. This was true regardless of whether or not

the oxime had undergone oxidation by peracetic acid. Larger amounts of the

oxime (100 ug/mL) in acetone did produce a full-scale response by the detector.

Clearly, oximes would not have been detected at the low levels of residues

(<5-10 ug/mL) that were analyzed in our samples. This could have been due to

poor detector sensitivity to oximes; and/or low recovery rates of oximes

from extracts (Maitlen et al. 1968); and/or removal of oximes quantitatively

by conversion to their respective aldehydes (which do not interfere with the

analysis) by acid hydrolysis in the peracetic acid oxidation step (Beckman

et al. 1969).

Hydrolysis Rates in Sterile pH-buffered Distilled Water

Plots of log percent remaining vs. time are shown in Figure 3 in an effort

to determine the contribution of H+, OH-, and H20 to the rate of degradation.

The actual data at each duplicated pH are presented in Table A-2 in the Appen-

dix III. At pH = 6, 7, and 8 aldicarb hydrolyzes slowly, but increases at

higher and lower pH levels (Figure 3). Plotting log k vs. pH in Figure 4

demonstrates that there are only slight changes in the pseudo-first-order

rate constant in the pH 6 to 8 range. The least squares estimates of all the

pseudo-first-order hydrolysis rate constants, half-lives, and coefficients

of determination are given in Table 3. Only the tests conducted at pH =

4, 9, and 10 showed degradation sufficient for estimating a rate constant.

Therefore, the rates and resulting half-life values for pH 6-8 are only

estimates since the slopes of the log percent remaining vs. time regres-

sion lines in Figure 3 were probably not significantly different from

zero. To accurately estimate the half-life for these pH conditions,







I-

pH=4



24


pH=6


I I


pH=7






pH=8
S I I I


2


pH=9
111111111 I


I I 1N


10 20 30 40 50
DAYS


pH=10
I I I I


60 70 80 90


Figure 3. Hydrolysis at 200C of Aldicarb in Sterile pH-buffered Distilled Water
Solutions. Each Data Point is the Mean of Duplicate Samples.
28


k--





















0
_0

-0
O
o


-2-


-3-


-4-


pH


Log k vs. pH for Aldicarb Hydrolysis at 200C in Sterile pH-buffered Distilled Water Solutions.


Figure 4.












Table 3. Pseudo-first-order Rate Constants (d), Half-life Values (th),

and Coefficient of Determination of the Regression Line (r2)

For Aldicarb Hydrolysis at 200C in pH buffered Distilled Water.


Period (days)



89

89

89

89

89

15


k (day-1)



5.3 x 10-3

1.2 x 10-3

8.1 x 10-4

2.1 x 10-3

1.3 x 10-2

1.2 x 10-1


pH


3.95

6.02

7.06

7.96

8.85

9.85


t (days)



131

559

861

324

55

6


r2



0.86

0.90

0.21

0.62

0.98

1.00









experiments lasting longer than 89 days would have to be performed. The

data obtained at pH = 7 showed a slight increase in aldicarb and its oxida-

tion products over time, resulting in a low coefficient of determination

(r2 = 0.21) and a positive slope. Above pH 8 the pseudo-first-order rate

constant increases with increasing pH and the slope of the line is approxi-

mately +1 (Figure 4), indicating aldicarb hydrolysis is sensitive to hydroxyl

ions in aqueous solutions. At pH <6, the rate of hydrolysis appears to be

acid catalyzed, but not to the extent as for base catalysis since the slope

is less than 1. The nonlinearity of the plot between pH 6 and 8 is inter-

preted as resulting from competing reactions of aldicarb with-water, hydrogen,

and hydroxide.

The second-order reaction rate constant for base hydrolysis, kOH, was

first-order with respect to hydroxide because the plot of log k vs. pH

(Figure 4) yielded a +1 slope at pH > 8. The kOH value calculated from Equa-

tion 3, using the data obtained at pH 7.96, 8.85, and 9.85, was 1.94 x 103 +

3.54 x 102 L mole-1 day-1. The acid hydrolysis constant, kH, which was not

first-order with respect to hydrogen, had a computed value (based only on

the data acquired at pH 3.95) of 4.72 x 101 L mole-1 day-1.

Groundwater Characteristics

Private Water Supply Wells

The groundwaters from the three private water supply wells contained

similar heat values and concentrations for the following constituents:

temperature, specific conductance, pH, alkalinity, and hardness (Table 4).

Sexton well, however, exhibited higher levels of total Fe (0.12 mg/L) and

sulfide (6.6-8.9 mg/L) while having stronger reducing conditions (-145 to

-163 mV). Luther well yielded approximately an order-of-magnitude higher

bacteria cell concentration (55-77 cells/mL) than tne other two wells, but











Table 4. Physical, Chemical, and Biological Characteristics of Groundwaters

Sampled From Private Water Supply Wells.


Date Sampled

Depth (m)

Temp (oC)

Sp. Cond. (uS/cm)

D. 0. (mg/L)

Eh7 (mV)

pH

Alkalinity (mg CaCO3/L)

Hardness (mgCaCO 3/L)

Total Fe (mg/L)

Sulfide (mg/L)

Bacteria (cells/mL)a

Before

After


Ryall

12/9/83 3/26/84

20 20

24 24

1126 1363

0.0 0.0

+264 +274

6.8 7.8

292 292

363 440

0.01 0.04

0.25 0.38


WELL SITE

Sexton

1/9/83 3/26/84

17 17

24 25

1073 1049

0.0 0.0

-163 -145

7.2 6.8

355 353

298 300

0.11 0.12

8.88 6.58


Luther

12/9/83 3/26/84

? ?

24 24

789 839

0.1 0.1

+326 +360

7.3 7.2

219 289

178 272

0.05 0.20

0.38 0.57


plate counts before and after the incubation period.


aAverage of two anaerobic










still was low when compared to cell densities from subsoils (Table 5).

The low densities of bacteria were maintained throughout the incubation

period, attesting to the sterile techniques and precautions taken not

to alter the original temperature and redox potential of the groundwaters

during the laboratory incubations. Generally, the groundwaters can be character-

ized as being hard, anoxic, and reduced, with high alkalinities, low iron,

and neutral pH.

Augered Wells

The shallow groundwaters from the augered wells at BBC and Lindsey Wabasso

groves differed from the deeper water supply wells by containing more dissolved

solids and bacteria cells (Table 5). Oxidation-reduction potential and pH

were consistent with those values reported for the deeper groundwaters at

Ryall, Sexton, and Luther wells. Lindsey Wabasso groundwater had a higher

alkalinity and hardness than of any sampled groundwater.

Degradation of TTR From Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone
Amended to Groundwaters From Private Water Supply Wells

The least-squares estimates of the pseudo-first-order rate constants,

half-life values, and coefficients of determination of the regression lines

when the percent TTR remaining is plotted logarithmically for 0.10 mg/L

additions of aldicarb, aldicarb sulfoxide, and aldicarb sulfone are presented

in Table 6. The linear coefficients for the straight lines are high (r2 =

0.88-1.00), indicating that some confidence can be placed on the pseudo-

first-order rate constants and half-life values which were determined from

the data. The raw data for each duplicate incubation container and at each

time that a sample was withdrawn and measured for TTR are presented in

Appendix III (Tables A-3 and A-4).

Notwithstanding a high variability in the data during the earlier stages

of incubation, Figure 5 depicts that approximately 60% of the initial 0.10 mg/L










Table 5. Physical, Chemical, and Biological Characteristics of Groundwaters

Sampled From Angered Shallow Wells.


Well Site


Date Sampled

Depth Cm)

Temp (C)

Sp. Cond. (jS/cm)

D. 0. (mg/L)

E (mV)
h7
pH

Alkalinity C(g CaCO3/L)

Hardness (ng CaCO3/L)

Total Fe (mg/L)

Bacteria (cells/mL)a

Before

After

Bacteria Ccells/g dry wt)a

Before

After


Lindsey Wabasso
1/30/84 4/23/84

1.5 1.5

22 26

2750 2462

0.0 0.1


+111

7.1

358

836





7.5 x 102

2.2 x 102



1.1 x 105

1.5 x 104


+83

7.3

442.

860

0,16



1.3 x 10

9,0 x 102



6.0 x 104
8.0 x 103


BBC
2/22/84

1.3

22



0.4

+321

7.4

106

202

0.12



3.6 x 104





4.6 x 104
9.0 x 103


aAverage of two anaerobic plate


counts before and after the incubation period.










Table 6. Pseudo-first-order Rate Constants (k), Half-life Values (t) and

Coefficients of Determination of the Regression Lines (r2) at

240C For the Disappearance of TTR in Sterile and Unsterile Ground-

waters Amended With 0.10 mg/L of Aldicarb, Aldicarb Sulfoxide, or

Aldicarb Sulfone. Each Value Is a Mean of Duplicate Samples.


Period (days)


90

90

90

90

90

90


k (day-1)


5.77 x 10-3

6.74 x 10-3

6.90 x 10-3

6.88 x 10-3

7.80 x 10-3

6.74 x 10-3


2.95

7.71

5.23



4.51

6.65

6.46


x 10-2

x 10-2

x 10-2



x 10-2

x 10-2

x 10-2


*Nitriles removed from sample


Compound


Aldicarb


Well


Ryall


Sterile


Yes

No

Yes

No

Yes

No


Sexton




Luther


t (days)


120

103

100

101

89

103


Aldicarb
Sulfoxide






Aldicarb
Sulfone


Ryall*

Sexton*

Luther*



Ryall*

Sexton*

Luther*


r2


0.96

0.92

0.90

0.98

0.88

0.92



1.00

0.98

0.99



1.00

0.99

1.00











100



90


80



70


Sterile Unsterile
o Luther -
---- Ryal Is
-- Sexton ---


5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90


DAYS


Percent Remaining of TTR From an Initial 0.100 mg/L Aldicarb Inoculum in Sterile and Non-sterile
Groundwaters From Three Water Supply Wells. Each Data Point is the Mean of Duplicate Samples.


Figure 5.










aldicarb inoculum remained as a toxic residue in all three groundwaters after

90 days at 240C, regardless of whether the water was sterile or non-sterile.

The pseudo-first-order rate constants for the TTR remaining after aldicarb

was added to the groundwaters fell within a narrow range of 5.77 x 10-3 to

7.80 x 10-3/day, which equalled 89 and 120 days when transformed into their

respective half-life times. TTR concentrations disappeared at faster rates

(half-life of 9-24 days) when either of the S-oxides was added (Table 6);

there were no major differences in the degradation of TTR between sulfoxide

and sulfone or among wells. The differences in the decomposition of TTR

observed for amendments of aldicarb and its S-oxides cannot be from the removal

of nitriles in only the S-oxide samples (and not the aldicarb samples) for

reasons previously described. It should be pointed out that these measured

rates overestimate the true rates of TTR degradation since the pH of the

water supply wells increased approximately 1 pH unit in each well water during

the incubation period. Still, they represent rates under pH conditions which

can exist at different times of the year. For instance, the Ryall well under-

went a pH change of 1 (6.8 vs. 7.8) between the two sampling periods in

December and March (Table 4).

Degradation of TTR From Aldicarb, Aldicarb Sulfoxide, and
Aldicarb Sulfone Amended to Groundwaters and Saturated Aquifer Material

As was found for the private water supply well waters, aldicarb added

to groundwaters without the subsoil aquifer material decomposed to non-toxic

residuals at very low rates (Table 7): half-life values of 112 days at the BBC

and 178 days at Lindsey Wabasso groves. Agreement among the replicate samples

was high without any differences being measured (cf. Appendix III). When

aldicarb was added to aquifer material consisting of saturated subsoil from

the top of the water table, the rate of TTR disappearance was increased by

an order-of-magnitude. For example, the half-life times decreased from 112










Table 7. Percent Remaining, Pseudo-first-order Rate Constants (k), and Half-

life Values (t) For the Disappearance of TTR in Groundwaters and

Aquifer Material From Lindsey Wabasso (LW) and BBC Groves Amended

With Aldicarb, Aldicarb Sulfoxide, or Aldicarb Sulfone. Each Value

Is a Mean of Duplicate Samples.


Aldicarb

Groundwater

Aquifer Material

Groundwater

Aquifer Material



Aldicarb Sulfoxide

Groundwater

Aquifer Material

Groundwater

Aquifer Material



Aldicarb Sulfone

Groundwater

Aquifer Material

Groundwater

Aquifer Material


Site

BBC

BBC

LW

LW


BBC

BBC

LW

LW





BBC

BBC

LW

LW


Temp
(oC)

22

22

22

22


Period
(days)

23

23

30

30


% TTR
Remaining



87

27

89

<3


k
(day-1)

6.21 x 10-3

5.76 x 10-2

3.88 x 10-3


2.09

2.64

1.60

6.44





2.64

3.77

4.43

6.86


10-2

10-2

10-2

10-2


x 10-2

x 10-2

x 10-2

x 10-2


*Nitriles removed from sample


t
(days)

112

12

178









days to 12 days when aquifer material was present from the BBC site. For the

Lindsey Wabasso grove, the conversion of TTR to non-toxic residues was so

fast that no detectable TTR was found after 30 days when aquifer subsoil

was present. This should be compared to a half-life of 178 days without

aquifer material being present. Adsorption of aldicarb onto the clays in

the subsoil did not occur since extraction of the sediments consistently

produced non-detectable levels ( < 6 ng/g (dry wt.)).

A similar trend was also noticed for the disappearance of TTR from sul-

foxide and sulfone additions to groundwaters with and without aquifer material

(Table 7), but not to the extent that was recorded for aldicarb. Instead of

the 9-fold difference in the degradation rates with and without aquifer

material that was found in the aldicarb amendments, sulfoxide or sulfone

amended to saturated aquifer subsoils increased the conversion of TTR to

non-toxic products by only 1.3 to 4.0 times. Thus, the presence of aquifer

material resulted in only slightly greater rates of degradation of TTR from

either sulfone-or sulfoxide-amended groundwaters: half-life times of 16-43

days without aquifer material and 10-26 days in the presence of aquifer

material. The removal of nitriles from some of the S-oxide amended samples

but not from the aldicarb-amended samples-could not have accounted for the

differences in the rates of TTR disappearance measured for aldicarb amended

and S-oxide amended samples since the percentage of the initial TTR comprised

by nitriles was small (Table 1). The small differences between the rate

constants from the two groves for sulfoxide and sulfone amendments in Table 7

were probably due more to the higher incubation temperature (260C) used for

the Lindsey Wabasso samples than from the removal of nitriles from the Lindsey

Wabasso samples. Adsorption onto clays was negligible: no aldicarb sulfoxide

or aldicarb sulfone was found above the limits of detection (6 ng/g (dry wt.)).











DISCUSSION


Hydrolysis Rates in pH-buffered Distilled Water

Laboratory hydrolysis studies of any xenobiotic using sterile pH-buffered

distilled water can only be interpreted as representing a "worst case" situa-

tion since all the environmental factors such as volatilization, adsorption,

plant uptake, leaching, and microbial degradation present under field condi-

tions have been omitted. Moreover, laboratory studies of hydrolysis reaction

rates are not only a function of pH, but also of the nature and concentration

of the acid-base system used to buffer the pH, which is called buffer catalysis

(Perdue and Wolfe 1983). A detailed account showing the effect of buffer

catalysis was negligible under the experimental conditions used in this inves-

tigation (Appendix II). The extent of error in not considering the environ-

mental conditions in the field and not using actual well waters and aquifer

material when interpreting hydrolysis data will be discussed later in this

report. Still, as Hansen and Spiegel (1983) point out, hydrolysis rates

obtained from laboratory studies can be used to establish upper bounds for

the half-lives of aldicarb in groundwater.

Comparisons of hydrolytic half-life values reported by other investiga-

tors for aldicarb in sterile, pH-buffered distilled water are presented in

Table 8. For those cases when the raw data were available, the rate constants

derived for temperatures other than 200C were adjusted to a temperature of

200C by constructing Arrhenius plots. Not only is there a scarcity of pub-

lished literature on aldicarb hydrolysis, but only a few of the published

studies included the range of pH values which bracket the pH of natural waters.

Carbamates such as aldicarb typically are quite resistant to hydrolysis

at neutral pH values, but are relatively unstable under alkaline conditions of









Table 8. A Comparison of Hydrolytic Half-lives for the Disappearance of TTR

in Sterile pH-buffered Distilled Water Amended With Aldicarb.


Half-life(days)

131

559

861

324

55

6


This

This

This

This

This

This


Reference

study

study

study

study

study

study


Lemley and

Lemley and


Zhong 1983

Zhong 1983


Hansen and Spiegel 1983


Trehy et al. 1984


3.95

6.02

7.06

7.96

8.85

9.85


Temp (oC)

20

20

20

20

20

20


12.90

13.39


4.0 min

1.3 min


8.5


8.2


None given









pH (Faust and Gomaa 1972), yielding aldicarb oxime (which is stable in basic

medium), methylamine, and carbonate from the cleavage of the -0 bond

(Lemley and Zhong 1983). Trehy et al. (1984) found from GC/MS analysis that

the degradation product for aldicarb in sterile anaerobic water was also

aldicarb oxime. Oximes can undergo a dehydration to become another non-toxic

aldicarb residue: nitriles. Hansen and Spiegel (1983) felt that sulfoxide

nitrile and sulfone nitrile became important only after 98 or 99% of the

aldicarb sulfoxide or aldicarb sulfone had been hydrolyzed at pH 8.5; however,

they never measured the nitriles. Presumably, aldicarb nitrile would also

become a dominant degradation product under similar circumstances.

The Hansen and Spiegel (1983) work is the closest data set comparable

to the conditions of our experiment. Their data (adjusted to 200C) for pH 8.5

yield a half-life of 69 days, which is almost twice as fast as the pseudo-

first-order rate constant extrapolated from Figure 4 at pH 8.5 (k = 5.8 x 10-3/day),

corresponding to a half-life of 120 days. Little confidence can be placed

on the rate constants obtained at pH 7.5 in either study since the slope of

the regression line, which is equal to k, was not significantly different from

zero in the Hansen and Spiegel study and also was probably not so in this study;

having only two replicates precluded statistically testing the hypothesis of

whether the slope was significantly different than zero.

To the authors' knowledge, the only published values for second-order

rate constants of aldicarb hydrolysis is from Lemley and Zhong (1983). Based

on using high hydroxide and aldicarb concentrations, and a different method

(i.e., titrimetric) to measure the progress of hydrolysis, they found koH for

aldicarb to be 1.35 x 103 1 0.03 x 103 L mole-i day-1 at 15 OC. After adjust-

ing to a temperature of 200C (assuming the activation energy of the aldicarb

is the same as the activation energy measured by Lemley and Zhong for aldicarb









sulfoxide (= 15.2 + 0.1 kcal/mol), the kOH becomes 2.12 x 103 L mole- day1,

which compares favorably to the 1.94 x 103 L mole-1 day-1 rate measured by us.

There is some confusion in the literature as to whether acid-catalyzed

hydrolysis of aldicarb and its S-oxides occur. Lemley and Zhong (1983) have

measured proton-catalysis for aldicarb sulfone, but the reaction rates were

slow and unmeasurably low when the acid concentrations were below 2M. Hansen

and Spiegel (1983), based on their communications with L. Tobler (1980),

reported no acid-catalysis of aldicarb at 770C down to a pH of 2, a finding

which is inconsistent with the results of this study.

Disappearance of Toxic Residues of Aldicarb in the Saturated Zone

Although hydrolysis experiments using pH-buffered sterile distilled

water are easier to perform than experiments using groundwaters and their

aquifer material from specific sites, their results are not as meaningful

because microbiological and aquifer catalytic effects are not taken into

consideration. Neither do field studies lend themselves to the accurate

degradation rate of TTR in groundwaters since the residues are subject to

dispersion, dilution, and recharge, all of which are beyond the control of

the investigator. Therefore, TTR degradation rates are best measured by

laboratory studies using actual groundwaters and saturated subsoils collected

in the field and incubated under controlled conditions which best represent

the in situ environment.

The half-life values for groundwaters and aquifer material obtained from

the well sites are compared to the hydrolytic degradation half-life times in

Table 9 after appropriate adjustments for pH and temperature differences. In

those samples which experienced an upward shift in the pH during the incuba-

tion period, an average pH was used for the basis of comparison to the hydrolysis

studies in sterile, pH-buffered distilled water. The pH shift to more alkaline










Table 9. Comparison of Degradation Rates For TTR Estimated by Hydrolysis

With Degradation Rates Measured in Groundwaters in the Absence and

Presence of Aquifer Material.


Site


Ryall
Sexton
Luther
BBC
Lind. Wab.
BBCb
Ryallc
Sextonc
Luther0

Ryall
Sexton
Luther
BBC
Lind. Wab.
BBCb
Lind. Wab.b

Ryall
Sexton
Luther
BBC
Lind. Wab.
BBCb
Lind. Wab.b


Residue
Added


aldicarb
aldicarb
aldicarb
aldicarb
aldicarb
aldicarb
aldicarb
aldicarb
aldicarb

sulfoxide
sulfoxide
sulfoxide
sulfoxide
sulfoxide
sulfoxide
sulfoxide

sulfone
sulfone
sulfone
sulfone
sulfone
sulfone
sulfone


pH
Initial Ave.


6.8
7.2
7.3
7.4
7.1
7.4
6.8
7.2
7.3

7.8
7.2
7.4
7.4
7.4
7.4
7.4

7.8
7.2
7.4
7.4
7.4
7.4
7.4


Study
Temp
(C)


Measured
Half-life
(days)


Estimated
Half-life
Based on
Distilled
Water
Hydrolysis
(days)a


211
238
211
648
456
515
211
238
211


7.9
7.9
7.6
7.3
7.4
7.4
7.3

7.9
7.9
7.6
7.3
7.4
7.4
7.3


aAldicarb half-life times


based on data presented in Fig.


4 and Table 3;


sulfoxide and sulfone half-life times based on data presented by Porter et al.
(1984).

bAquifer material present

CSterilized groundwater


Estimated
Measured









values by as much as one pH unit was unexpected. It indicated the waters

were not well buffered at their initial neutral pH values, the high alkalini-

ties notwithstanding. It is assumed the pH shift occurred because of losses

of CO2 during sequential subsamplings for TTR analysis. Apparently the change

occurred within the first 10 days of the incubation period. If a buffer

had been added to maintain the pH at its initial value, even more of an

error may have been introduced because of: 1) changing the ionic strength

of the waters (which may affect rate constants); 2) exerting a buffer catalysis

effect; and 3) serving as a nutrient or energy source to the microbiota.

For the locations listed, the half-life times measured in the groundwater

only and groundwater + aquifer samples are a factor of 1 to 43 shorter than

half-life estimates based on distilled water hydrolysis. Metabolic activities

of the microbiota cannot be invoked in explaining the faster degradation of

aldicarb in well waters since there were no differences between the rates of

aldicarb disappearance between sterile and non-sterile groundwaters (estimated

t /measured t] was 2 in all cases). Even though the anaerobic plate counts

probably underestimated bacterial cell densities because of the bias toward

gram-negative bacteria when a high proportion of the total bacteria in the

unconsolidated sediments of the saturated zone has been reported to be gram-

positive (White et al. 1983), the low bacteria densities found for the three

private water supply wells (Table 4) indicate a negligible microbiological

influence on TTR degradation rates. Therefore, the well waters contained

dissolved constituents which, when incubated under conditions closely resembling

those found in the field, increased the rate of degradation of aldicarb and

its toxic derivatives.

Based solely on the degradation rates for aldicarb-amended groundwaters, a

substantial risk of contaminating groundwaters would exist if aldicarb should









leach into the saturated zone. However, in the presence of aquifer material,

which is a more realistic condition, aldicarb and its toxic residues disappeared

9 times faster at the BBC site where the estimated t_/measured t_ ratio was

increased to 43 (Table 9). The higher densities of bacteria in the groundwater-

saturated subsoils (Table 5) probably accounted for the faster decay of TTR,

since aldicarb was not significantly sorbed on the sandy clay loam subsoil

(<6 ng/g (dry wt)) which served as the water table aquifer at this site,

thereby reducing the likelihood that surface catalyzed degradation was respon-

sible for the faster subsoil degradation rate. Sterile conditions (through

NaN3 additions) were not achieved in the subsurface soils from Lindsey Wabasso

(see Appendix III) probably because of the protection afforded to the micro-

biota by extracellular polysaccharide polymers secreted under the conditions

of unbalanced growth (i.e. nutrient deprivation) usually found in underground

waters (Uhlinger and White 1983).

It was not the objective of this investigation to determine the individual

degradation products, although such a study on these groundwaters would be

useful. Since the incubations were anaerobic, oxidation of aldicarb to

sulfoxide by bacteria would be less likely than bacterially-mediated hydrolysis

to aldicarb oxime. Oxidation processes cannot be entirely ruled out, but an

oxidizing agent other than oxygen would have to serve as the terminal electron

acceptor.

TTR concentrations in aldicarb sulfone- or aldicarb sulfoxide-enriched

well waters decreased at rates which were more consistent with expected values

published for hydrolysis in sterile, pH-buffered distilled water (Table 9).

By using a detailed study on the effects of pH and temperature on the hydrolysis

of aldicarb sulfoxide and aldicarb sulfone (Porter et al. 1984), we inter-

polated the estimated half-life times for Ryall, Sexton, and Luther wells to









be 30-50 days for aldicarb sulfoxide and 12-23 days for aldicarb sulfone.

These should be compared to the nearly equal measured half-life times of 9-24

days and 10-15 days for aldicarb sulfoxide and aldicarb sulfone, respectively

(Table 9). Since these samples were not sterilized, the role which microbiota

in the groundwaters had in the degradation was unknown. However, the low

bacteria densities (6-63 cells/mL) measured for the March 26, 1984 well water

samples (Table 4) indicated only a negligible contribution from the microbiota

could have occurred. It is therefore believed that most of the increase in

degradation rates was due to non-biological effects, with chemical hydrolysis

accounting for much of it. This is consistent with the findings of Delfino

and co-workers (1984), who found aldicarb sulfoxide to have a half-life of

13-15 days in autoclave aerobic groundwaters.

When aldicarb sulfoxide and aldicarb sulfone were separately added to

saturated subsoils obtained from Lindsey Wabasso and BBC groves (containing

5-100 times more bacteria than without aquifer material), only slightly

higher rates of TTR disappearance (<-50% higher for 3 of the 4 site-substrate

combinations) than what had been observed for well waters alone were recorded

(Tables 7 and 9). This differs sharply from the 9-fold increase in TTR

rate of degradation previously described for aldicarb-amended saturated

subsoils. The exception was for sulfoxide at the Lindsey Wabasso site, where

the decomposition rate for TTR was increased 4 times by the presence of

aquifer material. These data indicate that using only groundwaters may

suffice in testing the potential of sulfoxide and sulfone to degrade in

aquifers. The 50% overestimation of half-life values may be an acceptable

error (especially since it is a conservative one) considering the time and

expense expended in obtaining aquifer material from deep wells and that contam-

inated groundwaters have been found to contain 50% each of aldicarb sulfoxide










and aldicarb sulfone (Porter et al. 1984). However, more studies are necessary

before this procedure could be routinely practiced because the nature of the

aquifer material can be important. Miles (pers. comm., April 4, 1984) found

limestone decreased the hydrolysis rate of aldicarb sulfoxide and aldicarb

sulfone five-fold. This may be true only for deep limestone aquifers such as

the Floridan, and not for the shallower unconfined aquifers that are composed

of sands and clays.

In summary, the effect of the presence of aquifer material on the decompo-

sition of TTR from aldicarb addition was pronounced; however, aquifer material

had only small effects on rates of TTR disappearance for sulfone and sulfoxide.

Because aldicarb, aldicarb sulfoxide, and aldicarb sulfone sorb weakly in saudy-

and clay soils with low organic matter, surface catalyzed degradation is less

likely to be more important than microbial or solution processes in causing

faster aldicarb degradation in saturated subsoils. If bacteria are essential

in the anaerobic degradation of aldicarb to non-toxic oximes in groundwaters,

they are relatively unimportant in degrading sulfone and sulfoxide, a finding

consistent with studies conducted on Long Island (Porter et al. 1984). Also,

for sulfoxide and sulfone, chemical hydrolysis in solution proceeds fast

enough to be a major degradation pathway by itself. Since all the experiments

were conducted under anaerobic conditions, oxidation reactions would probably

be unimportant relative to hydrolysis reactions, which produce oximes (Figure 1).

It therefore appears that without bacteria, aldicarb hydrolysis is slow com-

pared to sulfoxide and sulfone hydrolysis in anaerobic groundwaters, which

conforms to the findings reported by Hansen and Spiegel (1983), Delfino (1984),

Porter et al. (1984), and this investigation. Bacteria are necessary for

catalyzing the hydrolysis of only aldicarb. When this occurs, the rate of

TTR disappearance equals that for chemical hydrolysis of aldicarb sulfoxide

and aldicarb sulfone.











Comparison of Degradation Rates of TTR
in the Unsaturated and Saturated Zones

To understand the significance of the findings in this report, the

broader picture of the likelihood of aldicarb or one of its toxic residues

reaching the water table from the vadose zone should be presented. The sus-

ceptibility of groundwaters to aldicarb contamination is a combination of the

potentials for TTR to reach the water table and to persist after it has entered

the groundwater system. Numerous studies of aldicarb behavior in soils in

the unsaturated zone have demonstrated rapid conversion to sulfoxide, which

in turn is more slowly biodegraded to sulfone (Bromilow et al. 1980; Smelt et al.

1978 b, c). However, one study (Coppedge et al. 1967) reported aldicarb to

decompose slowly in a fine sand soil: 27% of the applied aldicarb remained

after 4 weeks. Smelt et al. (1978c) calculated 91-100% of aldicarb was con-

verted to its sulfoxide, which was higher than the 60-80% values given by

Coppedge et al. (1967) and Bull et al. (1970), and the 67-92% conversion

reported by Bromilow et al. (1980). For two studies which investigated aldi-

carb sulfone degradation in soils, both found rates to be slower than what

had been reported for aldicarb sulfoxide (Smelt et al. 1978a) and aldicarb

(Hornsby et al. 1984)

Generally, soil scientists have found slower rates in the degradation of

aldicarb, aldicarb sulfoxide, and aldicarb sulfone in deeper soil layers than

in corresponding top layers of the soil profile (Smelt et al. 1978a,b; Hornsby

et al. 1984), presumably because of the lower microbial activity in the subsoil.

Typical half-life values reported in the literature for laboratory studies

of aldicarb, aldicarb sulfoxide, and aldicarb sul.fone losses in the upper soil

layers were 1-23, 13-14, and 24-158 days, respectively. For deeper soil layers

(70-180 cm) in the unsaturated zone, aldicarb sulfoxide had a reported half-

life of 53-475 days (Smelt et al. 1978b), while half-life times of 46-o0











(Smelt et al. 1978a) and 54-296 days for aldicarb sulfone (Hornsby et al. 1984)

have been published. These data imply that once TTR penetrate to the deeper

layers of the aerated soil zone, little further degradation can be expected.

When compared to the half-life times found in the upper soil layers of the

unsaturated zone, the half-life range of 10-26 days for TTR disappearance

measured for saturated subsoils in this study suggest a resumption to the

faster degradation rates recorded for the upper soil layers can be expected

for TTR entering the shallow water table.

Field Monitoring Studies and Models For
TTR Intrusion into Florida Groundwaters

Based on the laboratory degradation studies for deep soils in the un-

saturated zone and field investigations in Florida (Jones and Back 1984;

Hornsby et al 1984) and other states (Rothschild et al. 1982; Porter et al.

1984), the question no longer is whether toxic aldicarb residues are reaching

water tables, but at what concentrations are they entering water tables, how

fast is the TTR decomposing to non-toxic residues in groundwaters, and how

far would TTR travel in groundwaters before it is degraded to non-toxic

products.

Jones and Back (1984) reported degradation rates of aldicarb residues

in Florida soils decreased as they moved down through the soil column due to

fewer soil biota available to metabolize the residues. After several months,

when most of the remaining residues are 60 to 120 cm below the soil surface,

the disappearance of the residues became immeasurably low. Their data, which

encompassed six citrus grove sites throughout Florida, clearly indicated

significant percentages of TTR (i.e., aldicarb + aldicarb sulfoxide + aldicarb

sulfone) can remain in the lower 1.2-2.2 m of unsaturated soil. Contamination

of groundwaters from wells near the six citrus groves during a 1.5 year period

was slight: out of 67 groundwater samples taken from a total of 21 wells in








the 6 locations, waters from only two wells (at the Hillsborough site) had

intermittent trace amounts (1 ppb) of aldicarb residues.

Although the field data were too variable to make predictions of the

amount of aldicarb residues which will leach into the saturated zone, Jones

and Back (1984) estimated that, based on a computer model (PESTAN), <1% of

the aldicarb applied to citrus groves will leach more than three feet below

the soil surface. However, many assumptions were made in reaching this con-

clusion. These included using average soil and climatic properties, apply-

ing soil characteristics to Florida soils from sandy soil on Long Island, and

adopting the faster degradation rates found for the upper soil horizons.

Considering that the selection of input parameters for the modeling produced

a "best case" simulation, the conclusion that < 1% of the applied aldicarb

would leach below three feet of the soil surface would have to be viewed with a

large degree of uncertainty. Furthermore, the field data provided by Jones

and Back (1984) show that from 0.5 to 5.7% of the applied aldicarb remains as

TTR in the unsaturated zone from 166 to 344 days after application. Most of

the remaining residues would be located at the lower depths of the unsaturated

zone, where degradation is slow and migration to the nearby saturated zone is

likely to occur. The importance of looking at site-specific contamination,

rather than relying on generalized computer simulations or averages of field

data, can be readily recognized by noting that the highest fraction of aldi-

carb residues remaining (5.3% in Hillsborough County) at any of the six field

sites, was associated with the longest time after application (344 days).

Another example of the variability of different soils to degrade aldicarb

and its toxic residues is given by Hornsby et al. 1984, where 4-8 percent of

the applied aldicarb residues reached groundwater in a citrus grove on ridge

soils in Seminole County, Florida, while no TTR were detected in the upper

portion of the saturated zone at a flatwoods soil site in Polk County, Florida.










In further modeling efforts, Jones et al. (1984) recognized that the model

PESTAN, although simpler and easier to use than other models, is generally

not appropriate for predicting the leaching of aldicarb residues in Florida

citrus groves. By comparing three existing simulation models (PESTAN, PISTON,

and PRZM) with each other and with field monitoring and laboratory studies,

they found using site-specific factors such as soil hydraulic properties, soil

organic matter, pesticide degradation rates, and daily rainfall data were

important in determining the extent of pesticide leaching.

Lateral Transport of TTR
in Shallow Groundwaters in Indian River County

The results of this investigation showed measured TTR half-life times of

10-26 days in the presence of aquifer material, regardless of whether the

initial toxic residue had been aldicarb or one of its S-oxides. Using the

number of half-lives (7) required to reduce the highest TTR concentration

(1.26 mg/L) reported for Florida groundwaters (Hornsby et al. 1984) to levels

less than the state standard of 10 ppb, and our measured half-life times,

70-182 days would have to elapse before TTR concentrations in that water

(without dilution or dispersion) could decrease to acceptable TTR levels.

This means that if Temik was applied only once a year, and toxic residues of

this magnitude moved into the saturated zone as a pulse over a short time

interval, TTR would be below the state drinking water standard (10 ppb)

after one-half year. How far would such a plume travel in that period of time?

Considering the calculated hydraulic gradients of 0.00019-0.00078 m/m

derived from observation wells in Indian River County and reported hydraulic

conductivities of 3.6-12.2 m/day for the overlying sandy soils located

at the BBC and Lindsey Wabasso groves, and assuming a porosity of 33%,

the plume of contaminated water would travel 0.0021-0.029 m/day from the

boundaries of application. Thus, a distance of 0.38-5.3 m (1.3-17 ft) would










be traveled by a contaminated plume during the one-half year required for

the TTR to decrease to concentrations <10 ppb. These calculated distances

of lateral plume migration are well within the 300 ft exclusion zone

adopted by the state for applying aldicarb near drinking water wells.

This simple calculation assumes a homogeneous and isotropic shallow ground-

water aquifer with no contribution from dispersion or dilution toward reducing

TTR concentrations. It therefore represents the travel distance of TTR in

the most extreme case, especially since the highest reported concentration

of TTR in Florida groundwaters was used.

Another approach to evaluating the extent of migration of contaminated

groundwater in Indian River County is by calculating how much of the 5 lb

a.i./acre per year application rate as required under the new restrictions may

enter into the groundwater and be transmitted laterally. If 8% of the applied

aldicarb were to enter into the groundwater as reported by Hornsby et al.

(1984), then 352 ppb of TTR would be concentrated in the surface to 0.1 m

layer of groundwater. Rothschild et al. (1982) concluded that most TTR

leached to groundwaters under aldicarb-treated fields was near the water table.

If an average half-life of 16 days measured in this study for a TTR comprised

of 50% each of aldicarb sulfoxide and aldicarb sulfone (Porter et al. 1984)

is assumed, then it would require 80 days (or five half-lives) for the TTR

to degrade to a concentration that is in compliance with the 10 ppb standard

adopted by Florida for potable waters. Using the same hydraulic conductivi-

ties and gradients and porosity as before, then only 0.17-2.3 m (0.6-8 ft)

distance would be traveled by the contaminated plume. Considering the slow

rate of groundwater movement in Florida soils, coupled with any further reduc-

tion in concentration within the contaminated plume from dilution and disper-

sion, there should be little chance that wells used for potable water would










be contaminated above the 10 ppb standard if the present restrictions are

observed. Even if a larger hydraulic gradient exists because of downward

gradients artificially created during well pumping (Rothschild et al. 1982),

the effect of a 10-fold increase in the hydraulic gradient increases the dis-

tance traveled by the TTR plume to 6-80 ft, still well within the 300 ft

exclusion zone. Apparently the restriction that aldicarb cannot be applied

within 300 ft of any drinking water well seems reasonable. It should be

emphasized that the estimated small distances of plume migration apply only

to Indian River County, which in turn are further limited to those few

sandy soils used to derive the TTR degradation rates.









SUMMARY AND RECOMMENDATIONS


Laboratory degradation experiments were conducted for the purpose of

measuring the degradative activity of a small concentration (0.10 mg/L) of

aldicarb and its toxic oxidation products sulfoxidee and sulfone) in ground-

waters from five wells located near citrus groves in Indian River County.

The experiments were designed to evaluate what effect microbiota and

aquifer material would have in influencing the rate of detoxific-ation of toxic

residues. Attention was given to mimick as closely as possible the actual

in situ conditions of the shallow groundwaters during the incubations in

the lab so that realistic reaction rates could be projected. Reliable

predictions of the duration of aldicarb contamination in groundwater could then

be made. Once having arrived at the reaction rates, converting to half-lives

facilitated comparisons to other degradation rates reported in the literature

for the saturated and unsaturated zones.

Our measured rates for total toxic residue disappearance were also combined

with data from the literature which stated the amounts and concentrations of

toxic residues entering the water table in another Florida county to yield

the time required to degrade the toxic residues to less than 10 ppb. Using

field measurements of water table fluctuations and hydraulic conductivities

reported for Indian River County, the daily distance of groundwater flow could

be calculated which, when multiplied by the amount of time necessary to reduce

toxic residues to less than 10 ppb, gave the total distance moved by the con-

taminated plume.

Two companion studies sought to: 1) adequately describe the hydrolysis

of aldicarb in various pH-buffered solutions; and 2) evaluate the suspected

interference of non-toxic nitriles and oximes in the methodology used for the










analysis of toxic residues.

To best summarize the results of this investigation, four recommendations

are given. Following each recommendation the pertinent conclusions are

provided upon which the recommendation was based.

Recommendation 1

The positive interference from non-toxic oximes and nitriles in being

included with the total toxic residue (TTR) are minor, suggesting that the

liquid chromatography step used in the analytical procedure to remove them

may be omitted under some experimental circumstances.


Oximes did not interfere in the analysis of TTR under the conditions used

in this investigation; nitriles did interfere, but only when >90% of the TTR

had disappeared. The percentage error of the initial TTR contributed by non-

toxic nitriles rarely exceeded 10%, and was on the average, less than the

percentage errors associated with either: i) the Florisil separation method;

ii) agreement between duplicate samples; iii) inter-laboratory analyses of

split-samples; or iv) recoveries of known standards without Florisil separation.

Recommendation 2

The best method for determining the degradation rate of toxic pesticides

in groundwaters and gaining an understanding of the mechanisms is from labora-

tory experiments using groundwater-saturated aquifer material incubated under

in situ environmental conditions. However, in cases where obtaining aquifer

material is difficult or impossible, incubating only groundwater may be sufficient

in producing realistic TTR degradation rates for aldicarb sulfoxide and aldicarb

sulfone. This does not apply to aldicarb, where incubating in the absence of

aquifer material would cause a large underestimation of the rate of degradation

in anaerobic groundwaters.









TTR disappearance in anaerobic groundwaters containing low densities of

bacteria and without aquifer material was slow (half-life times = 101-178

days) for aldicarb and fast (half-life times = 9-43 days) for aldicarb sul-

foxide and aldicarb sulfone, indicating that chemical hydrolysis in the slightly

alkaline groundwaters was responsible for converting the sulfoxide and sulfone

into non-toxic derivatives but was not operating to the same extent in aldicarb

deactivation. The persistence of aldicarb in neutral to slightly alkaline

(pH 8) waters was supported by an hydrolysis experiment in sterile pH-buffered

distilled waters. At pH 7, the rate of TTR disappearance from the parent

compound was immeasurable; a half-life of 324 days was measured at pH 8,

with rates of TTR disappearance increasing as the pH increased. The second-

order reaction rate (koH) for base-catalyzed hydrolysis was 1.94 x 103 L mole-1

day-.

When aquifer material was added, rates of TTR disappearance increased

9-fold to 5.76 x 10-2 day-1 for aldicarb in one shallow well, indicating

microbes were essential in degrading the parent compound to non-toxic residues.

However, aldicarb sulfoxide and aldicarb sulfone degraded only slightly

more in the presence of aquifer material (half-life times = 10-26 days) than

they did in the absence of aquifer material, suggesting chemical hydrolysis

is more prominent in reducing TTR associated with these S-oxides than any

microbially-mediated pathway.

Recommendation 3

Restricting the application rate of Temik to 5 lb a.i./acre and not apply-

ing it within 300 ft of a potable water supply well should be continued.


Shallow groundwaters of Indian River County apparently possess the capability

to rapidly degrade TTR from aldicarb and its S-oxides. Half-life times of

10-26 days stand out in contrast to the 2-3 year half-life projected by Porter









et al. (1984) for TTR disappearance in groundwater on Long Island. The reasons

for the faster degradation rates of TTR in Florida are probably the more

favorable temperatures and pH. The neutral to slightly alkaline pH of Florida's

groundwaters (6.8-7.8) serves to hydrolyze the sulfoxide and sulfone, whereas

the acid pH of Long Island groundwater (4.2-5.9) inhibits hydrolysis.

A combination of slow groundwater movement and fast degradation rates

limits the migration distances of aldicarb-contaminated plumes in Indian River

County. Estimates of lateral distances traversed in the saturated zone of the

soils studied in this investigation by a plume of groundwater contaminated

with TTR were only 1-17 ft since the TTR was converted to non-toxic residues

within one-half year. The 300 ft exclusion zone and application rate of

5 lb a.i./acre insure adequate protection of the groundwater resources in

Indian River County.

Recommendation 4

Studies on surface and ground water contamination by aldicarb and its

oxidized toxic products should be continued.



Although these results point to rapid degradation and slow transport of

TTR in groundwaters, care should be exercised in extrapolating this conclusion

to other sites and soils in Florida. The danger of generalizing results

obtained at one site to other sites has been discussed. Even for the specific

soils used in the degradation experiments, the limited number of replicates

and field samplings performed in the short study period warrant a cautious

interpretation of the fast degradation rates.

Monitoring the groundwaters of some citrus fields receiving annual aldi-

carb applications should be performed for a period of several years to provide

a long-term data base which includes the extremes in meteorological conditions

affecting residue transport and deactivation.








Future experimental studies should be directed toward evaluating the

effect of salt content on the hydrolysis rate and the chemical or biological

reductive conversion of sulfone to sulfoxide and sulfoxide to aldicarb. There

is some evidence that both could have an effect on aldicarb. Fukuto et al. (1967)

reported that the observed first-order rate constants for the hydrolysis of

p-nitro-N-methylcarbamate increased upon decreasing the ionic strength of the

phosphate buffer. He suggested that a lowering of the rate constant was

from decreasing activities of hydroxide ion and/or the carbamate because of

increasing ionic strength, and not from catalysis of the phosphate ions present

in the buffer. The environmental implication is that the hydrolysis rate of

toxic aldicarb residues may be inhibited in estuarine environments.

Reductive conversion of a sulfoxide to its parent compound has been

reported for phorate (Walter-Echols and Lichtenstein 1977) in sediments. The

prevailing evidence of this occurring in reduced environments for sulfoxide

or sulfone is to the contrary. Studies such as this one have shown rapid

disappearance of TTR under anaerobic conditions and no one has published any

data implying that this mechanism is operating in the chemistry of sulfoxide

or sulfone conversions. Still, there is an abundance of reduced compounds in

groundwaters (e.g. sulfides, ferrous iron) which could serve as a reductant,

and with the possibility of marketing sulfone as the active ingredient (instead

of aldicarb) in commercial insecticides, more research should be done on

evaluating whether reductive conversion could occur. Even if a reductive

pathway was not found, the information that would be generated on the conver-

sion rates to other toxic and non-toxic compounds for each toxic residue

species in an anaerobic aquifer system would be beneficial.

Given the shallow distances between land surface and the water table,

the geographical closeness of residential and agricultural land uses,










permeability of sandy soils, low adsorption potential, and field studies in

Florida showing 4-8% of the applied aldicarb available to enter the ground-

water, a continuing scientific effort directed toward monitoring, modeling,

and reaching best management practices is prudent.









BIBLIOGRAPHY


APHA CAmerican Public Health Association) 1976. Standard methods for the

examination of Water and Wastewater. Washington, D,C,

Beckman, H., B.Y. Giang, and J, Qualia. 1969, Production and detection of

derivatives of Temik and its metabolites as residues. J, Agr. Food

Chem. 17:70-74.

Benson, H.T. 1973. Microbiological applications; A laboratory manual in

general biology, 2nd ed. W.MC. Brown Publishing Co., Dubuque, Iowa,

Bromilow, R.H., R.J. Baker, M.A.H. Freeman, and K. Gorog, 1980. The degrad-

ation of aldicarb and oxamyl in soil, Pest, Sci, 11;371-378,

Bull, D. L., R.A. Stokes, J.R. Coppedge, and R. L, Ridgway, 1970, Further

studies of the fate of aldicarb in soil, J, Econ. Entomol, 63:1283.

Chesters, G., M.P. Anderson, B. Shaw, J.M, Harkin, M, Meyer, E, Rothschild,

and R. Manser. 1982. Aldicarb in groundwater. Water Resources Research

Center, University of Wisconsin, Madison. 38 pp.

Coppedge, J.R., D. A. Lindquist, D.L. Bull, and H,W, Dorough, 1967, Fate of

2-methyl 2-(methylthio) propionaldehyde 0-mOethylcarbamoyl) oxime (Temik)

in cotton plants and soil, J. Agr. Food Chem. 15;902-910,

Coppedge, J.R., D. L. Bull, and R.L. Ridgway. 1977. Movement and persistence of

aldicarb in certain soils, Archiv. of Environ, Contamination and Toxicol. 5:129-141

Delfino, J. J. (ed). 1984. The fate of industrial organic compounds in drinking

water aquifers. Dept. of Environmental Engineering Sciences, University of

Florida, Gainesville, 64 p.

Faust, S.D., and H.M. Gomaa. 1972. Chemical hydrolysis of some organic phos-

phorus and carbamate pesticides in aquatic environments. Environmental Letters

3:171-206.









Fukuto, T.R., 1M.A,H. Fahmy, and R,L, Metcalf, 1967, Alkaline hydrolysis, anticholine-

sterase, and insecticidal properties of some nitro-substituted phenyl car-

bamates. J. Agric. Food Chem. 15:273-281.

Galoux, M., J.-C, van Damme, A. Bernes, and J, Potvin. 1979. Gas-liquid chroma-

tographic determination of aldicarb, aldicarb sulfoxide, and aldicarb sulfone

in soils and water using a Hall electrolytic conductivity detector.

J. Chromategr. 177:245-253,

Hansen, J.L., and M.H, Spiegel, 1983, Hydrolysis studies of aldicarb, aldicarb

sulfoxide and aldicarb sulfone. Environ, Toxicol, Chem. 2:147-153.

Hornsby, A.G., P.S.C. Rao, W,B. Wheeler, P. Nkedi-Kizza, and R.L, Jones. 1984.

Fate of aldicarb in Florida citrus soils, I, Field and laboratory studies,

In: D.M. Nielsen (ed), Proceedings of a Conference on Characterization and

Monitoring of the Vadose (Unsaturated) Zone, December 8-10, 1983, Las Vegas,

Nevada, National Water Well Association, Worthington, Ohio.

ICET (Institute for Comparative and Environmental Toxicology). 1983. A toxicological

evaluation of aldicarb and its metabolites in relation to the potential human

health impace of aldicarb residues in Long Island ground water, Cornell

University, Ithaca, N.Y. 90 pp.

IFAS (Institute of Food and Agricultural Sciences). 1983, Aldicarb research task

force report. Gainesville, Florida. 58 pp.

Jones, R.L,, and R.C. Back. 1984. Monitoring aldicarb residues in Florida soil

and water. g nviron. Toxicol. Chem, 3.,;-20

Jones, R. L., P.S.C. Rao, and A.G, Hornsby. 1984, Fate of aldicarb in Florida

citrus soil, 2. Model evaluation. In; D,M. Nielsen (ed), Proceedings of a

Conference on Characterization and Monitoring of the Vadose (Unsaturated)

Zone, December 8-10, 1983, Las Vegas, Nevada. National Water Well Association,

Worthington, Ohio,









Knaak, J.E., M.J. Tallant, and L.J. Sullivan. 1966. The metabolism of 2-

methyl-2-(methylthio)propionaldehyde O-(methylcarbamoyl) oxime in the rat.

J. Agric. Food Chem. 14:573-578.

Lemley, A.T,, and W.Z. Zhong. 1983. Kinetics of aqueous base and acid hydrolysis
of aldicarb, aldicarb. sulfoxide, and aldicarb sulfone. J, Environ,,Sci, Health

B18:189-206.

Light, T.S. 1972. Standard solution for redox potential measurements.

Anal, Chem. 44:1038-1039,

Maitlen, J.C,, L.M, McDonough, and M. Beroza, 1968, Determination of residues

of 2-Methyl-2-(methylthio) propionaldehyde O-n(ethylcarbamoyl) oxime

CUC-21149, Temik), its sulfoxide, and its sulfone by gas chromatography.

J. Agr. Food Chem. 16,549-553.

Patrick, W.H., Jr., and I.C, Mahapatra. 1968, Transformations and availability

to rice of nitrogen and phosphorus in water-logged soils. Advances in

Agronomy 20:323-359.

Perdue, E.M., and N.L. Wolfe. 1983. Prediction of buffer catalysis in field

and laboratory studies of pollutant hydrolysis reactions. Environ. Sci.

Technol. 17 635-642.

Porter, K.S., A,T, Lemley, H.B, Hughes, and R,L, Jones, 1984. Developing

information on aldicarb levels in Long Island groundwater, In;

Proceedings on the Second International Conference on Groundwater Quality

Research, March 26-29, 1984, Tulsa, Oklahoma, U.S. Environmental

Protection Agency. (in press),

Rothschild, E.R, R.J,Manser, and M,P, Anderson, 1982, Investigation of aldicarb

in ground water in selected areas of the Central Sand Plain of Wisconsin.

Ground Water 20,437-445,









Sharom, N.S,, J.RW, Miles, CR, Harris, and F,L, McEwin. 1980. Persistence

of 12 insecticides in water, Water Res, 14.1089-1093.

Smelt, J.H., M, Leistra, W,H. Houx, and A, Dekker. 1978a. Conversion rates

of aldicarb and its oxidation products in soils. I. Aldicarb sulphone,

Pestic. Sci. 9;279-285.

1978b, Conversion rates of

aldicarb and its oxidation products in soils. II, Aldicarb sulphoxide.

Pestic, Sci, 9_286-292,

1978c, Conversion rates of

aldicarb and its oxidation products in soils, III, Aldicarb, Pestic,

Sci. 9:293-300,

Trehy, M.L,, R.A. Yost, and J.J. McCreary, 1984. Determination of aldicarb,

aldicarb oxime and aldicarb nitrile in water by GC/MS, Anal Chem, 55

(in press)

Uhlinger, D, J., and D.C. White, 1983. Relationship between the physiological

status and the formation of extracellular polysaccharide glycocalyx in

Pseudomonas atlantica, Appl. Environ, Microbiol, 45:64-70.

Union Carbide Corporation, Agricultural Products Company, Inc. 1980, A method

for the determination of aldicarb residues in water, Unpublished,

1983, Temik

aldicarb pesticide; A scientific assessment. Research Triangle Park,

N.C. 71 pp.

Walter-Echols, G., and E.P. Lichtenstein. 1977. Microbial reduction of phorate

sulfoxide to phorate in a soil-lake mud-water microcosm, J, Econ, Entom,

70:505-509,










White, D,C., J.S. Nickels, J.H, Parker, R,H. Findlay, M,J, Gehron, G,A. Smith,

and R.F. Martz, 1984. Biochemical measures of the biomass community

structure and metabolic activity-of the ground water microbiota.

Chapt, XX. In: C,H. Ward (ed), Proc, First Intern. Conf. Ground Water

Quality Research. Wiley Interscience, N.Y.


Zaki, -M.H., D. Moran, and D. Harris, 1982. Pesticides in groundwater: The

aldicarb story in Suffolk County, N,Y, Am, J, Public Health 72:1391-1395.

































APPENDIX I


SUMMARY OF COMPARISONS BETWEEN LABORATORIES

IN THE SPLIT SAMPLING PROGRAM












Sample

Ryal well-B
Aldicarb amended
Non-sterile
T = 0 days

Aldicarb hydrolysis
pH = 10
Time = 5 days

Aldicarb hydrolysis-A
pH = 10
Time = 15 days

Luther well-B
Aldicarb amended
Non-sterile
Time = 90 days


FIT





101 ppb




49 ppb


FDERa
Tallahassee


IFAS
Pesticide Research Labb
Gainesville


120 ppb




60 ppb


18 ppb


16 ppb


51 ppb


36 ppb


a TTR determination using a Hewlitt-Packard model 5700 GC with a nitrogen-

phosphorus detector and a 1% SP1000 on Carbopack B 60/80 mesh column;

injector, column and detector temperatures were 250, 200, and 3000C,

respectively.

b TTR determination using a Perkin-Elmer Series 4 high pressure liquid chromato-

graph with a Gilson 121 fluorometer (excitation: 305-395 nm; emission:

430-470 nm) and a Zorbax C-8 (15 cm x,4.6 mm) stationary phase. The mobile

phase consisted of a 10 min. linear gradient with an initial composition of

4% CH3CN, 16% CH3OH, 80% H20 and ending with a final composition of 14%

CH3CN, 56% CH30H, 30% H20. A post-column derivitization step included

0.5 mL/min of 0.05 M NaOH at 950C followed by 0.5 mL/min of OPA at ambient

temperature.















APPENDIX II


EFFECT OF BUFFER CATALYSIS ON THE HYDROLYSIS

OF ALDICARB IN STERILE, pH-BUFFERED SOLUTIONS







EFFECT OF BUFFER CATALYSIS ON THE HYDROLYSIS OF
ALDICARB IN'STERILE, pH-BUFFERED SOLUTIONS


In laboratory studies of hydrolysis of the aldicarb, conditions of con-

stant pH are desired to simplify kinetic interpretations, Pseudo-first-order

kinetics are usually observed only at a constant pH for acid-base-catalyzed

hydrolysis of a pollutant, P:

d[Pj = -k IP] (A-1)
dt obsed

where Kobsed = [H20]+ k 3 H30 + 4_C + ( [HIB] + k fBi]) (A-2)

and HBi and B. are the ith Bronsted acid-base pair in solution. Eq. A-2

states that hydrolysis reaction rates are a function not only of pH (the first

three terms) but also of the nature and concentration of the acid-base system

used to buffer the pH, called buffer catalysis, The first three terms of

Eq. A-2 can contribute to kobsed in all aqueous solutions, their contribution

being predictable if the second-order rate constants (k 20, k3 and k H-)

and pH are known. At constant pH, the combined contributions of H20, H30 ,

and OH to.kobsed are constant and can be represented by a pseudo-first-order

rate constant, k :
w

02 + k2304H30 + k330] WIOH (A-2)

Thus, the observed pseudo-first-order rate constant (kobsed) equals the

pseudo-first-order rate constant for catalysis by solvent species (k ) plus

the buffer catalysis contribution (the last two terms of Eq. A-2).

Perdue and Wolfe (1983) developed a theoretic basis for assessing a

maximum contribution of buffer catalysis for hydrolysis reactions in aqueous

systems. They derived a buffer catalysis factor (BCF) for the buffers commonly

employed to buffer solutions to a constant pH. The potential significance of








buffer ctalysis in aqueous solutions can he expressed;

k obed/kw = 1 + CBCBCF CA-4)
obsed w B

where CB is the concentration of the buffer catalyst. When kobsed /k l, then

the contribution from buffer catalysis to the rate of hydrolysis is negligible.

Perdue and Wolfe suggested that when kobsed/kw >1.10 i.e., buffer catalysis

contribution that equals or exceeds 10% of the combined kinetic contribution

of H20, H3C., and OH-), then a 10% or greater increase in kobsed results from

buffer catalysis, and should be viewed as having potential significance.

Substituting the molar concentrations and published BCFs for the buffers

used in the aldicarb hydrolysis investigation yielded k osed/kw ratios ranging

from 1.23 to 2.67 (Table A-i). Even though these represent a potential for

significant buffer catalysis of 23 to 167% in kobsed, the pH buffers which

exhibited the highest ratios (i.e., pH = 6.02, 7.06, and 7.96) corresponded to

the slowest kobsed, while those.buffers which had faster kobsed i,e,

pH = 3,95, 8.85, 9,85) were associated with lower kobsed/k ratios, indicating

that the potential for buffer catalysis was not realized. It is important

to recognize that the relative contribution of buffer catalysis to kobsed is

a function of pH, catalyst, and substrate, but that the preceding calculations

predict the maximum contribution of buffer catalysis for a particular catalyst

and pH only (ignoring the type of substrate). Buffer catalysis would be some-

what less important for any real substrate, such as aldicarb:, and our data

indicate a negligible contribution from the buffers at the strength used in

our hydrolysis experiment. Fukuto et al.(1967) found varying concentrations of

phosphate buffer anions did not participate in the hydrolytic reaction of

p-nitrophenyl N-methycarbamate, which supports our conclusion that buffer

catalysis is probably unimportant in hydrolyzing carbamates,









Table A-i.


Potential for Maximum Contribution of Buffer Catalysis to the

Observed Pseudo-first-order Rate Constants (kobsed) in the

Hydrolysis of Aldicarb Using Various pH Buffers in Distilled Water.


Buffer Concentration
(CB)


Buffer Catalysis Factora
(BCF)


phthalate

phthalate

phosphate

phosphate

borate

carbonate


50

100

80


kobsed/kw

(=1 + CB(BCF))


1.50

1.84

2.67

2.34

1.23

1.42


aperdue and Wolfe (1983)


pH


3.95

6.02

7.06

7.96

8.85

9.85


1.67

1.67

1.67

1.67

4.10

1.67


10-2

10-2

10-2
io2


10-2

10-3

10-2















APPENDIX III


RAW DATA









Hydrolysis Data for Aldicarb
Water at 200 + 20 C.


in Sterile pH-buffered Distilled


Aldicarb Remaining, ppb


pH Day 0


3.95-A 111
3.95-B 117


6.02-A
6.02-B


133


7.06-A 108
7.06-B 117


7.96-A
7.96-B


8.85-A 88
8.85-B 120


9.85-A 108
9.85-B 108


Table A-2


Day 5


121
118


133
139


121
130


133
127


Day 15


112
128


122
150


128
137



150


Day 37


120
106


127
130


130
127


127
134


Day 89


72
76


120
124


132
124


100
120











Raw Data For Aldicarb Degradation in Duplicate Sterile and

Non-sterile Groundwaters at 24 + 20C.


Well

Ryall -A

Ryall. -B

Ryall -A

Ryall -B


Sterile

Yes

Yes

No

No


Day 0

94

99

99

101


Day 5

92

90

95

107


Yes

Yes

No

No


Sexton

Sexton

Sexton

Sexton



Luther

Luther

Luther

Luther


85

118

89

104



98

118

118

124


Yes

Yes

No

No


TTR Remaining, ppb

Day 11

84

81

91

98



78

78

88

84



95

91

95

98


Table A-3


Day 34

73

70

57

80


Day 90

54

56

58

54


II I__










Table A-4. Raw Data For Aldicarb Sulfoxide and Aldicarb Sulfone Degradation in
Duplicate Non-Sterile Groundwaters at 24 + 20C.



TTR Remaining, ppb


Aldicarb-
Well g Ide__ Day 0 Day 5 Day 20 Day 42

Ryall A Sulfoxide 97 85 54* --

Ryall B Sulfoxide 105 92 -- 4*

Ryall A Sulfone 81 68 43* <3*

Ryall B Sulfone 97 79 28* <3*



Sexton -A Sulfoxide 97 83 34* 4*

Sexton -B Sulfoxide 98 84 35* 4*

Sexton -A Sulfone 80 63 12* <3*

Sexton -B Sulfone 92 81 37* <3*



Luther -A Sulfoxide 95 92 43* 9*

Luther -B Sulfoxide 92 82 40* 13*

Luther -A Sulfone 90 65 20* <3*

Luther -B Sulfone 94 60 30* <3*



* Nitrile removed from sample.











Table A-5. Raw Data For Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone


Degradation in Duplicate Non-sterile Groundwaters With and

Without Subsoil From the BBC and Lindsey Wabasso Groves.


TTR Remaining, ppb


Substrate


Aldicarb
Aldicarb
Aldicarb
Aldicarb

Aldicarb
Aldicarb
Aldicarb
Aldicarb

Aldicarb
Aldicarb
Aldicarb
Aldicarb


Sulfoxide
Sulfoxide
Sulfoxide
Sulfoxide

Sulfone
Sulfone
Sulfone
Sulfone


Subsoil


Absent
Absent
Present
Present

Absent
Absent
Present
Present

Absent
Absent
Present
Present


Day 0


Day 23


Day 0


Aldicarb
Aldicarb
Aldicarb
Aldicarb


Absent
Absent
Present
Present


Day 30


100
100
62
65


Day 0


Aldicarb
Aldicarb
Aldicarb
Aldicarb

Aldicarb
Aldicarb
Aldicarb
Aldicarb


Sulfoxide
Sulfoxide
Sulfoxide
Sulfoxide

Sulfone
Sulfone
Sulfone
Sulfone


Day 25


Absent
Absent
Present
Present

Absent
Absent
Present
Present


30*
29*
10*
16*


*Nitrile removed from sample.


Well


BBC-A
BBC-B
BBC-A
BBC-B

BBC-A
BBC-B
BBC-A
BBC-B

BBC-A
BBC-B
BBC-A
BBC-B


Lind.
Lind.
Lind.
Lind.


Wab.-A
Wab.-B
Wab.-A
Wab. -B


Lind.
Lind.
Lind.
Lind.

Lind.
Lind.
Lind.
Lind.


Wab. -A
Wab.-B
Wab.-A
Wab.-B

Wab.-A
Wab.-B
Wab.-A
Wab.-B


~










Table A-6. Bacterial Counts on Groundwater Samples.


Sampling Date Well


9 December 1983 Ryall




Sexton




Luther


26 March 1984


Ryall


Sexton




Luther


30 January 1984


Lind. Wab. (sed.)


Lind. Wab.


23 April 1984


Lind. Wab.


Lind. Wab.


22 February 1984


(water)




(sed.)




(water)


BBC (sed.)


BBC (sater)


Pour Plate
(cells/mL for water and
cells/g (dry wt) for sediments)


start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:

start:

finish:


6; 5; 4

10; 12

2; 3; 2

7; 8

58; 51; 55

42; 33

3; 5

6; 8

1; 4

6; 7

75; 79

63; 64

1.2 x 105; 1.0 x 105

1.4 x 104; 1.6 x 105

800; 700

228; 219

6.3 x 104; 5.7 x 104

8.0 x 103; 8.0 x 103

1500; 1200

800; 1100

3.9 x 104; 5.4 x 104

9.0 x 103; 9.0 x 103

31,000; 42,000

1800; 2300


x=5

x=ll

x=2

x=7

x=55

x=37

x=4

x=7

x=3

x=6

x=77

x=63

x=l.1 x .105

x=1.5 x 104

x=750

x=223

x=6.0 x 104

x=8.0 x 103

x=1350

x=950

x=4.6 x 104

x=9.0 x 103

R=36,000

R=2,050




Full Text

PAGE 1

j r -t, r WATER IiRESOURCES researc center Publication No. 76 ALDICARB STUDIES IN GROUNDWATERS FROM CITRUS GROVES IN INDIAN RIVER COUNTY, FLORIDA By F. E. Dierberg Department of Environmental Science & Engineering Florida Institute of Technology Melbourne UNIVERSITY OF FLORIDA

PAGE 2

TABLE OF CONTENTS Abstract Acknowledgements Introduction Objectives Site Description Sampling Procedure Methods Field Measurements Laboratory Measurements Laboratory Incubations of Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone Degradation Aqueous Hydrolysis of Aldicarb Migration of TTR in Groundwaters in Indian River'County Quality Assurance Results Interferences From Non-toxic Aldicarb Residues: Nitriles and Oximes ... Hydrolysis Rates in Sterile pH-buffered Distilled Water Groundwater Characteristics Degradation of TTR From Aldicarb, Sulfoxide, and Aldicarb Sulfone Amended to Groundwaters From Private Water Supply Wells Degradation of TTR From Aldicarb, Aldicarb Sulfoxide,' and Aldicarb Sulfone Amended to Groundwaters and Saturated Aquifer Material Dis cussion Hydrolysis Rates in pH-buffered Distilled Water Disappearance of Toxic Residues of Aldicarb in the Saturated Zone ii Page iv vi 1 6 7 10 11 11 11 14 17 19 20 23 23 27 31 33 37 40 40 43

PAGE 3

Comparison of Degradation Rates, of TTR in the unsaturated and Saturated Zones Field Monitoring Studies and Models For TTR Intrusion into Florida Groundwaters Lateral Transport of TTR in Shallow Groundwaters in Indian River County Summary and Recommendations Bibliography Appendix I Appendix II Appendix III iii Page 49 50 52 55 61 66 68 72

PAGE 4

ABSTRACT The disappearance of aldicarb [2-methyl-2(methylthio)propionaldehyde O-(methylcarbamoyl)oxime] and its two toxic degradation products, aldicarb sulfoxide and aldicarb sulfone, were measured in laboratory studies using groundwaters and subsoils collected from citrus groves in Indian River County, Florida, and incubated under controlled conditions which best represented the in situ environment. The presence of aquifer material on the decomposition of total toxic residues from aldicarb addition in anaerobic groundwaters was pro-nounced, suggesting bacteria were important in decomposing aldicarb to non-toxic residues. However, aquifer material had only a minor effect on the rate of total toxic residue disappearance when aldicarb sulfoxide or aldicarb sulfone was the primary toxic aldicarb residue, suggesting that chemical hydrolysis in solution was more important in degrading aldicarb sulfoxide and aldicarb sulfone. Based on hydrolysis experiments in sterile pH-buffered distilled water for aldicarb, hydrolysis rate constants became second-order (kOH = 1.94 x 103 L mole-l day-l at 200C) at pH 8 and above; acid-catalyzed hydrolysis occurred at pH 4, but not to the same extent as base-catalyzed hydrolysis. Oximes did not interfere with the analysis of total toxic residues under the conditions of the procedures used in this study; nitriles interfered in a positive fashion, but only when toxic residue concentrations were <: 10% of the initial concentrd t l' The half-life times for total toxic residue disappearance of aldicarb and its two sulfur-oxidized derivatives in groundwater-saturated subsoils ranged from 10-26 days, suggesting a resumption to the reported faster aerobic degracL' tion rates in the upper soil layers after having undergone slow degradation In unsaturated subsoils. Based on the degradation rates found in this study, hydrologic parameters obtained for Indian River County subsoils, and amounts iv

PAGE 5

of total toxic residue reported entering Florida groundwaters, it was estimated that toxic residues in aldicarb-contaminated groundwaters in Indian River County would migrate only short distances (1-17 ft) before conversion of toxic residues to non-toxic residues was completed. Thus, the exclusion zone of 300 ft from the nearest drinking water well for applying aldicarb is a reasonable restriction for protecting the groundwater resources in Indian River County. v

PAGE 6

ACKNOWLEDGEMENTS This study was supported in part under an annual allotment grant from the Florida Water Resources Researcn Center to Forrest E. Dieroerg. Associate Professor of Environmental Science and Engineering, Florida Institute of Technology. Chris Given served as researcn assistant and completed his M.S. thesis on this project, The assistance of Brian Combs, Indian River Agricultural Extension Agent, in locating accessiBle sampling sites and in helping with sample collection is gratefully acknowledged. Larry Pollack, a graduate student in the department, also contributed to the research findings found in this report. The interest and assistance of Carl Miles, Department of Environmental Engineering Sciences, University of Florida, helped to make this report more complete. The co-operation of Willis Wheeler and David Atherton, of the Pesticide Research Lab, Institute of Food and Agricultural Sciences, University of Florida, Gainesville, and Robert Patton of the Florida Department of Environmental Regulation, Tallahassee, in the inter-laboratory exchange of samples is appreciated. Victor Carlisle of the SoU Science Department, University of Florida, and-Carol Wettstein of the U.S. Conservation Service provided the information on the soUs in Indian County, Finally, the U.S. Environmental Protection Agency, Pesticides and Industrial Chemicals Repository, Research Triangle Park, North Carolina, and Union Carbide kindly provided standards for aldicarb and its decomposition products. vi

PAGE 7

INTRODUCTION Aldicarb (Temik R ) is an effective but non-selective systemic insecticide, miticide, and nematicide; its acute mammalian toxicity (rat acute oral LDSO' 1 mg/kg; rabbit dermal LDSO' S mg/kg) (ICET 1983) makes it one of the most toxic of all currently-registered insecticides. Aldicarb and its toxic oxidized metabolites (sulfoxide and sulfone), like other carbamate esters, exert their insecticidal activity through reversible inhibition of the enzyme acetyl-cholinesterase. There is no evidence that either aldicarb or its metabolites are associated with any adverse mutagenic, carcinogenic, or teratogenic effects. As it stands now, the conditions of exposure can be viewed as a series of low-level exposures rather than continuous chronic exposure. Detailed reviews on the toxicology of aldicarb and its sulfoxide and sulfone metabolites can be found in reports issued by the Institute for Comparative and Environmental Toxicology (ICET 1983) and aldicarb's manufacturer, Union Carbide (1983). Because of its high mammalian toxicity, aldicarb is available only in granular formulations (5-20% active ingredient) for soil incorporation. Its primary use in Florida is to protect citrus groves (in the central and southern parts of the state) and potatoes (in north Florida) from aphids, mites and nematodes (Jones and Back 1984). Concern for the potential of aldicarb or its toxic metabolites to contaminate underground aquifers used for drinking water in Florida heightened after reports that a1dicarb had been detected for the first time in groundwater in Suffolk County, New York, in August 1979, (Zaki et al. 1982). Since the sandy soils in Suffolk County were not too dissimilar from Florida soils and because of the high water solubility -4 0 (=6,000 ppm) and non-volatility (vapor pressure = 1 x 10 mm Hg at 25 C) of 1

PAGE 8

the pesticide, both of which would preclude significant adsorption onto soil particles, extensive monitoring, field, and laboratory studies were sub sequently undertaken in the state (IFAS 1983). Other states such as Wisconsin (Chesters et ale 1982) also initiated monitoring networks after detecting aldicarb residues in groundwaters. The regulatory history of aldicarb in Florida can be briefly summarized below: August 1982 January 1983 October 1983 January 1, 1984 Media reported Temik-contaminated water in other states; Florida began testing for -and finding -traces in underground water supplies. State-wide ban went into effect. Nation-wide ban on ethylene dibromide (EDB) increased the potential use of Temik. Ban lifted, but with the following restrictions: no more than 5 lbs active ingredient (a.i.) can be used per acre (formerly 15 lbs a.i. per acre). applied only once a year between January 1 and April 30 (for citrus), cannot be applied within 300 ft of any drinking water well. use suspended in an area if drinking water >10 ppb. notification of impending treatment must be posted prominently on property where it is to be applied. 2

PAGE 9

The degradation of aldicarb in plants, animals and soil is dominated by two processes: oxidation of sulfur to obtain sulfoxide and sulfone analogs, and cleavage of the carbamate ester bond (Fig. 1). Oxidation is the major pathway for aldicarb metabolism in most systems (IeET 1983). Which degradation pathway predominates is important since hydrolysis of the ester linkage detoxifies aldicarb and its sulfur-oxidized derivatives, while oxidation of sulfur yields metabolites that retain the toxicity of the parent compound. Aldicarb oxidizes readily to sulfoxide in plants (Maitlen et al. 1968, and soils (Smelt et al. 1978c: Bromilow et al. 1980). It is the sulfoxide which is the most potent cholinesterase inhibitor of the group (Fig. 1), and responsible for the high systemic activity and long-term persistence of insecticidal activities. Much of the research to date on aldicarb and its sulfur-oxidized derivatives in Florida have been fate studies in soils and crops (IFAS 1983). Little is known the chemical behavior of aldicarb and its oxidized metabolites in shallow groundwaters in Florida where drinking water is obtained from many private wells. Funding for the project began in August 1983; field sampling, laboratory studies, and chemical analyses continued through June 1984. Continued research on aldicarb in groundwater is being conducted beyond the funding period of this report. Preliminary data contained in this report were presented at the Florida Academy of Sciences meeting on March 30, 1984, in Boca Raton, Florida. The practical aspects of the study focused on the degradation of aldicarb. and its oxidized metabolites in groundwaters from 3 shallow wells 20 m deep) located near citrus groves in Indian River County. Two other wells 1.4 m deep) inside citrus groves were /iu.&e+,ed to the effects of aquifer -material on the rate of degradation. tn one case, groundwaters and aquifer lll8.terial were 3

PAGE 10

o X I D A T I o N CH3 0 I CH3 aldicarb (0.84 mg/kg) '" HYDROLYSIS ) .. .. :---=..:...-CH3 aldicarb sulfoxide (0.49-1.13 mg/kg) .1 OCH3 0 II J II CH3SCCH=NOCNHCH3 III OCH3 aldicarb sulfone (20-45 mg/kg) DEHYDRATION yH 3 ) CH3 CH3 aldicarb oxime aldicarb nitrile (2380 mg/kg) (570 mg/kg) 1 OCH3 III OCHg n I CH3 SyCH=NOH )-CH3 CH3 sulfoxide oxime sulfoxide nitrile (8060 mg/kg) (4000 mg/kg) 1 OCH3 III OCH3 III CH3SCCH=NOH ) CH3SCC=N III III OCH3 'OCH 3 sul fone oxime sul fone nitri 1 e (1590 mg/kg) (350 mg/kg) Source: The Institute for Comparative and Environmental Toxicology' 1983 Figure 1. Degradative Pathways of Aldicarb. Large Arrows Indicate Maj or Pathways. Values in Parentheses are Acute Oral LDSO's for Rats. 4

PAGE 11

sterilized while replicate sub samples were left unsterilized so as to determine what role, if any, the groundwater microflora play in the degradation process. Finally, carefully conducted hydrolysis experiments in pH buffered distilled water of 4, 6, 7, 8, 9 and 10 were carried out at room temperature to further understand the significance of acid or base catalysis and to obtain reliable half-life values for the pH range found for natural waters. 5

PAGE 12

OBJECTIVES The objectives of this study were as follows: 1. To determine the extent that non-toxic oximes and nitriles interfered in the analysis of total toxic aldicarb residues by gas chromatography; 2. To perform a carefully controlled hydrolysis experiment for understanding what effect pH has on hydrolyzing aldicarb under sterile conditions; 3. To examine shallow groundwaters with and without aquifer material collected in citrus groves for their capacity to degrade aldicarb and its two toxic derivatives, sufoxide and sulfone, to non-toxic residues in laboratory experiments; and 4. To integrate the kinetic expressions of these experiments with a simple groundwater transport model and field data published in the literature in arriving at an estimate of the distances groundwater contaminated with toxic residues of aldicarb would migrate in Indian River County. 6

PAGE 13

SITE DESCRIPTION Five well sites bordering or within citrus groves in Indian River County (Figure 2) provided groundwaters and aquifer material for aldicarb, aldicarb sulfoxide, and aldicarb sulfone additions in the laboratory degradation studies. Three of the wells -Ryall, Luther, and Sexton -were private water supply wells with attached pumping systems. TWo of the wells directly supplied the Luther and Sexton residences; the Ryall well was used for irrigation, but water from it was sometimes consumed by grove workers. Well construction reports were not available to verify their depths, but owners believed that their wells were relatively shallow, ranging between 16 and 20 m, closely approximating the depths which are commonly used in providing drinking water to private homes. Only groundwaters were sampled from the 2-inch diameter piping of these wells. The remaining two wells (BBC and Lindsey Wabasso) were drilled by using a bucket auger, and their very depths (1.3-1.5 m) coincided to the top of the water table in the unconfined aquifer. Any transport of toxic residues to the deeper groundwaters which are used for domestic water supply would have to first pass through this layer. The soils in Indian River County are currently being mapped by the SCS. The soils in the field sites which were hand-augered (BBC and Lindsey Wabasso groves) are classified as Pineda sand (Arenic Glossaqualfs), Riviera sand (Arenic Glossaqualfs), and Wabasso sand (Alfic Haplaquods). They are all nearly level (slopes less than 2 percent) poorly drained, slowly to very slowly permeable soils. Permeability is rapid (15-51 cm/hr) in the sandy A horizons and slow to very slow in the sandy clay loam B horizons (
PAGE 14

I NDIAN RiVER COUNTY D Slue Cypress Lake o I Km 29 I f:::. SSC Figure 2. Location of the Five Study Wells in Indian River County, Florida. Li =0 Lindsey Wabasso grove; R = Ryalls; Lu = Luther; S = Sexton; BBC = BBC grove. 8 f:::. Lu ATLANTIC OCEAN

PAGE 15

surface for 1 to 6 months in most years and 10 to 30 inches deep most of the rest of the year. Some areas are flooded for periods ranging from a few days to about 3 months. 9

PAGE 16

SAMPLING PROCEDURE Groundwater samples from the three private water supply wells (Ryall, Luther, Sexton) (Fig. 2) within 30 m of citrus groves were collected on December 9, 1983, and March 26, 1984 after pumping an amount of water greater than three volumes of the water standing in the pipe. Water samples were taken from the closest accessible point to the well head, usually from the pump itself or just after the pumP'. Care was exercised in avoiding entrapment of air and degassing in the sample containers by flushing several volumes of groundwater before caping each container. Samples to be used in 1aooratory incubations were put into sterilized 4-L amber-co10red glass containers after at least one volume had been allowed to 'overflow and placed in a cooler filled with ice for transport to the laboratory. Several BOD and Na1gene oott1es were filled in the same manner with sample groundwaters for field and 1aooratory measurements. In addition to sampling only groundwaters from private water supply wells, two shallow wells each at the BBe and Lindsey Waoasso sites {Fig. 2) were hand-augered on February 22, 1984 and January 30/April 23, 1984, respectively, using a 3-inch (ID) x 14-inch (length of containment area) soil auger. The wells were located in a furrow separating two rows of citrus trees. Groundwater was sampled through l-inch (ODt Flex PVC heavy wall tubing connected to a 12 lIpm maximum flow "Guzzler" hand pump; aquifer material from saturated subsoils was withdrawn from these self-constructed shallow wells (1.3-1.5 m) using the soil auger. The same field and laboratory analyses were done on these augered as already described for the private water supply wells. lO

PAGE 17

ANALYTICAL .METHODS Field Redox potential was measured with a platinum electrode and a saturated calomel electrode (SCE) as the reference electrode. Calibration of electrodes +2 +3 and potentiometer (Fisher Accumet 640 Mini-Meter) was made against a Fe /Fe standard solution (Light 1972). Adjustments were made in the measured poten-tials to standard hydrogen reference electrode (Light 1972) at pH 7 (Patrick and Mahapatra 1968). In addition to redox potential, dissolved oxygen (Leeds and N:orthrup 7932 portable dissolved oxygen meter), specific conductance (YSI Model 33 S-C-T meter), and pH (Fisher Accumet 640 Mini-Meter) were also measured potentiometrically using the appropriate sensors. Temperature was recorded using a mercury thermometer. Laboratory Measurements General Water Chemistry Analyses The following chemical constituents were determined titrimetrically according to the procedures given in Standard For the Examination of Water and Wastewater CAPRA 1976): total (0.02 N H 2 S0 4), total hardness (0.02 M EDTA1, and sulfide (iodine), Total iron was measured by atomic absorption spectroscopy with a Perkin-Elmer Model 460 atomic absorption spectrometer after preserving by acidifying to pH<2 in the field. Standard plate counts were conducted anaerobically by the Vacuum and Gas Displacement Method (Benson 1973) on the groundwaters and saturated sub-soils from the private water supply wells and augered shallow wells before (within 24 hours of field sampling) and at the end of each incubation. 11

PAGE 18

Aldicaro Residue Analyses After adding 0.10 mg/L of aldicarb or one of its S-oxides (aldicarb sulfoxide and aldicarb sulfone), pH-buffered distilled water, groundwater, and subsoil were analyzed for the sum of aldicarb and its S-oxides at various time :intervals according to the methodology by Union Carbide (1980). This meb.hqldJJogy purports to determine the sum total of toxic aldicarb residues (Fig. 1) (i.e. aldicarb + sulfoxide + aldicarb sulfone), which has been referred to as the "total toxic r.esidue" (TTR). It wasn It learned until later into the research period of the potential "positive" interference from non-toxic nitriles. 1m assessment of the extent that the nitrile may have interfered in those earlier TTR analyses where it hadn't been removed is presented in the RESULTS sect ion. At appropriate intervals, fifty or one-hundred mL aliquots were removed from each duplicate in cub at ion container (BOD bottle or Mason jar) of pH-buffered distilled water solution or groundwater and placed into a l25-,mL separatory funnel. Oxidation of aldicarb and aldicarb sulfoxide to aldicarb sulfone was accomplished by adding 2 mL of peracetic acid. Conversion of aldicarb and aldicarb sulfoxide to: aldicarb sulfone is necessary since chromatographic peaks of the former are indistinguishable from the solvent peak using gas chromatograph (GC) al. 1979). The contents in the separatory funnel were mixed and allowed to stand for 30 minutes with occasional mixing. After 30 minutes, 15 mL of 10% NaHC03 were added 1 mixed, and allowed to stand with occasional mixing for 15 minutes. Fifty mL of methylene chloride were added with frequent vent ing to release evolv-ed CO2 After the layers separated, the lower methylene chloride layer was drained through approximately 80 g of prewet sodium sulfate in a 4-inch funnel with a glass wool plug, The 12

PAGE 19

extraction was repeated with another 50 mL of methylene chloride and the extracts combined. After rinsing the sodium sulfate bed with an additional 20 mL of methylene chloride, the combined extracts and rinse were collected and concentrated in a 450C water bath by evaporating just to dryness with a stream of dry N2 The residue was then dissolved in acetone to 1-2 mL and stored in 3-mL septum-capped vials at -SoC until analyzed by GC, Groundwater-saturated subsoil samples (SO g dry wtt were extracted using the method of Ga10ux et a1.. (19]9.1. The subsoils (50 g ,dry wt 1 were mixed with 50 mL of acetone-water (40:60), shaken for 30 min and centrifuged for S min at 3000 rpm. The extract was transferred to a 250-mL separation funnel. A second extraction was performed with 40 mL of methanol-water (50:S0), and the two extracts were combined. The acetone extract was next extracted 3 times with 50 mL methylene chloride in a SOO-mL separatory funnel. After draining the methylene chloride fraction through sodium sulfate beds, combining, and con-o centrating to approximately 2-mL in a 45 C water bath, 10 mg of m-chloroper-benzoic acid was added to oxidize the a1dicarb and a1dicarb sulfoxide, if present, to aldicarb sulfone (Smelt a1. 191786).. The residue was dissolved in 1-2 mL acetone for GC analysis after taking to dryness under N2 in a 4Soc water bath. As previously stated, if nitrile derivatives of aldicarb and its S-oxides are present in the they would serve as a positive interferent (i.e. included with the toxic residues). The high injection port temperature (260C) pyrolyzes the toxic aldicarb and its S-oxides to nitriles (Knaak a1. 1966; Trehy et a1. 1984), which are then detected by the flame photometric detector in the GC. To remove the nitrile interference, O. S-cmCID) x 60-cm glass columns were filled to a depth of 14 em with S g of PR grade F1orisil. After pre-washing the columns with 25 mL of :methylene chloride but 13

PAGE 20

before it reached the top of the Florisil, the eluted pre-wash was discarded and the sample added to the top of the column. When the sample reached the top of the column, 100 1UL of 5% acetone in ethyl ether were added to the column, and the eluate discarded when the solvent reached the top of the Florisi!. A solvent fraction consisting of 50 mL of 50% acetone in ethyl ether was then added to the column, and the eluate collected in a l25-mL Erlenmeyer flask. The second fraction was evaporated under an N2 gas stream in a 450C water bath to less than 2 mL and then transferred to a 3 mL septum cap vial. The flask was rinsed with small amounts of acetone and the rinse added to the vial. The volume in the vial was reduced under N2 gas to a final volume of 1 mL. Gas chromatography was performed on a Perkin-Elmer Sigma 300 GC equipped with a flame photometric detector with a 394 nm filter to quantify the sulfone. A coiled glass colUlllIl, 1m x 2mm (ID), packed with 5% SP-lOOO on Supelcoport (100/120 mesh) was used for separation. o Normal operating conditions were 260 C injector temperature, l750C column temperature, 2500C detector temperature, and helium, hydrogen, and air flow rates of 35, 20, and 26 mL/min, respectively. The minimum detectable concentration was 1.8 ng absolute or 300 ng/mL in a 6 injection, which corresponded to total aldicarb residues extracted from water solutions originally containing 3 to 12 ppb, depending on the volume of water and the final volume of the extract. Chromatograms were reported on a Hewlett-Packard 3390A Integrator in the linearized mode and a Varian 9176 recorder in thenon'-linearized mode. Laboratory Incuoations of Aldicarli, Aldicarb Sulfoxide. and Aldicarb Sulfone Degradation Incubation' containers consisting of BOD bottles for groundwaters and 946-mL Mason jars for groundwater-saturated subsoils were initially filled to 14

PAGE 21

completely occupy the entire volume of the container, The BOD Dottles were stoppered and a small amount of silicone stopcock grease was placed around the outside edge of the stopper and Dottle to help prevent any exchange of gases from occuring. All containers were incubated in the dark at the ambient o temperatures of the groundwaters measured in the field (22 + 2 to 26 + 2 C). All pH values were checked before and after each incubation period, To insure that anaerobic conditions were maintained during those times when aliquots of groundwater and groundwater-saturated subsoil were withdrawn from incubation containers, disposable polyethylene bags with a stream of N2 constantly purging the bag were used. A Leeds and Northrup dissolved oxygen meter equipped with a BOD probe set on "air calibrate" recorded zero oxygen in the Bags and beadspace of the containers after sampling; also, the water of those samples incubated in Mason jars were devoid of oxygen as by the dissolved oxygen meter. Sterile and Non-sterile Degradation of Ald;i.carb Duplicate sterile and non-sterile groundwaters from Ryall, Sexton, and Luther well:s sampled on December 9, 1983, were spiked with aldicarb in BOD bottles to yield a final concentration of 0,10 mg/L, and incubated measured o ambient field temperature of 24 + 2 C. Groundwaters (500 mL) with and without saturated subsoils (1800 g dry wt. from 1.1 to 1.5 m depths) obtained from the Lindsey Wabasso grove on January 30, 1984, were also sterilized and left nonsterile at 22 + 20C so as to determine what effects the presence of aquifer material had on aldicarb decomposition. A second site (BBC grove) was cO.red to retrieve groundwaters with and without saturated subsoil from a depth of 0.9 to 1.2 m on February 22, 1984, but all incubations using water (500 mL) and aquifer_ material (1200 g dry wt) from this site remained non-sterile. Sterile 15

PAGE 22

and non-sterile controls without added aldicarb were run for each well Sterilization was accomplished by adding sodium azide (NaN3 ) to give a final concentration of 0.1% in the groundwater-only incubations (Sharom et ale 1980) and 0.4% in the groundwater-saturated subsoils. To deter lIline whether sterility was maintained throughout the incubations, thioglycollate broth media was employed on the first and last days to test for contamination in the sterilized samples and was negative in only the groundwater containers. aliquots of groundwater from the private water supply wells incubated in the BOD bottles were withdrawn at times 0, 5, 11, 34 and 90 days and analyzed for TTR. In the case of the groundwater with and without aquifer material from the Lindsey Wabasso and B13C sites, subsampling from the 946-mL Mason jars serving as incubation containers at time 0 occurred when the sediment had settled (approximately 1 hour) after being shaken; subsampling occurred again after 30 (Lindsey Wabasso) and 23 (BBC) days had elapsed. Saturated subsoil material was frozen until extraction. Non-sterile Degradation of Sulfoxide and Sulfone. Duplicate BOD bottles containing non-sterile, from the private water supply wells were spiked with sulfoxide or sulfone to yield a final concentration of 0.10 mg/L. Shallow groundwaters and saturated -subsoihfrom 0.91.2 m (BBC) and 1.2-1.5 m (Lindsey Wabasso) depths were obtained on February 22 and April 23, 1984, from BBC and Lindsey Wabasso groves, respectively. As before, sulfoxide or sulfone was amended to the Mason jar incubation containers to produce an initial concentration of 0.10 mg/L. Sterile samples were not run because results from the sterile vs. non-sterile aldicarb experiments indicated 16

PAGE 23

sterile conditions were unattainable for incubations where the aquifer Eateria1 was included. Controls without amended sulfoxide or sulfone were set up in duplicate for all groundwaters from all wells with and without added subsoil. An incubation temperature for the deeper groundwaters from the private water supply o wells was 24 + 2 groundwaters and groundwater-saturated subsoils from the o 0 augered wells were incubated at 22 + 2 C (BBC) and 26 + 2 C (Lindsey Wabasso). Either 50 or 100-mL a1iquots were removed at 0, 5, 20 and 40 days from the BOD bottles used to incubate the groundwaters sampled from the private water supply wells. Water and saturated subsoil were withdrawn from the jars used as incubation containers for the augered wells at BBC and Lindsey Wabasso groves at time and after 23 or 25 days, and extracted as previously described. Aqueous Hydrolysis of Aldicarb Buffered reaction solutions for the a1dicarb hydrolysis experiment were prepared by adding 3 mL of a 30 mg11 a1dicarb solution in water to a solution of 297 mL of distil1ed-deionized-carbon-filtered water and 1.67 x 10-2 M of pH -3 buffer solution (4.1 x 10 M buffer for pH 8.85) in duplicate BOD bottles to give a 0.10 mg/L a1dicarb concentration. The sterile buffer systems and their final pH after incubating 15-89 days were potassium hydrogen phtha1atehydrochloric' acid (PH 3,95), potassium dihydrogen phosphate-sodium hydroxide (PH 6.02,7.06, and 7.96), sodium tetraborate-hydrochloric acid (pH 8.85), and sodium bicarbonate-sodium hydroxide (PH 9. 85}. Buffer solutions and glassware were autoc1aved before use. The BOD bottles were plugged with sterilized foam plugs and placed in an incubator in the dark at 200 + 20C. Contro1s without a1dicarb were included. All pH values were checked before and after each kinetic run. 17

PAGE 24

Pseudo-first-order rate constantsJ kJ were obtained from tlie slope of the line (-2.30 X regression coefficient} obtained by a linear least-squares analysis of the data for those experiments where samples were collected and analyzed over mUltiple time intervals (i.e., aldicarb in pH-buffered distilled water; Ryall, Sexton, and Luther well-waterL For the shallow wells at BBC and Lindsey Wabasso where aquifer material and groundwaters were sampled and analyzed for only one time interval, the integrated first-order rate equation was used U) where Co is the initial concentration of aldicarb or one of its primary oxidation products and C is the remaining concentration at time t, the corresponding half-life was from tl = 0.693/k (2) The second-order reaction rate -constants, k and k d in d f OH were eterm e rom the slope between-pH 8 and pH 10 and between pH 4 and pH 6 in a plot of log k vs. pH according to the relationships kOH = _k_ IOHJ and (3) = k IH+] (4) 18

PAGE 25

Migration of in Groundwaters in Indian River County Lateral movements of theoretical TTR plumes were calculated using Darcy's Law dh v = -K-/a d1 (5) where v = velocity (m day); K = hydraulic conductivity (m day); dh d1 = hydraulic gradient CJn/m); and a = porosity. Values for the parameter K were obtained from U.S. Soil Conservation Service Soil Interpretation Records for the soil series (Pineda sand, Riviera Sand, and Wabasso sand) surveyed recently to be present at the BBC and Lindsey Wabasso sites. Water table data from observation wells in Indian River County provided by the Soil Survey of the U.S. Soil Conservation Service (C. Wettstein, pers. 'corom., May 30, 1984) were converted to heights above sea" level. after adj1,lstiri.g for elevation in ground level between well locations. The average of the differences between water table surfaces based on 12 measurements during B?18.month period (November 1982 -May 1984) divided by the distance between two wells Eenerated realistic hydraulic gradients (dh/d1) for the county. A total of five obser.vation wells were used to compute three different hydraulic gradients. Porosity was assumed to be 33% 19

PAGE 26

QUALITY ASSURANCE In order to insure reliability in accuracy and precision of analyses, the following quality control measures were carried out: 1. All sample incubations were done in duplicate. Expressed as percentages of their averages, the differences for all aldicarb samples were ll% (0-30%); for aldicarb sulfoxide samples % (0-33%); and for the aldicarb sulfone samples % (3-62%). 2. Blanks were analyzed for each well water, aquifer material, and aldicarb substrate combination, and were always below the limit of detection (i.e., <3 to ppb, depending on the volume of water extracted and final volume of the extract) 3. A total of 22 spikes were performed (16 with aldicarb spiked into water; 4 with aldicarb sulfone spiked into water; and 2 aldicarb spiked into saturated subsoils). Recoveries ranged from: 93 to 109% (ave. = 103%) for the aldicarb-spiked water samples, 104% or all the aldicarb sulfonespiked and 88 to 92% (ave. = 90%) for the aldicarb-spiked subsoils. Results were not corrected for the recovery percentage. 4. Inter-laboratory comparisons with two independent laboratories: 1) Florida Dept. of Environmental Regulation; and 2) Pesticide Research Lab of the Institute of Food and Agricultural Sciences, University of Florida, Gainesville. A total of 4 samples were exchanged (2 samples to each lab) and their results were within.%, except for one sample which differed by 29%. The samples, concentrations, and gas/liquid chromatographic systems are presented in Appendix I. S. Twenty-two groundwater samples underwent separation by liquid chromatography using Florisil columns after oxidation by peracetic acid. Two 20

PAGE 27

eluants were used: one to retrieve the non-toxic aldicarb residues ,(fraction 1 = 5% acetone in ethyl ether) and the other to elute toxic aldicarb residues (fraction 2 = 50% acetone in ethyl ether). The data from which errors were calculated for recovery of all residues after Florisil separation as well as for not removing nitriles by Florisil separation are provided in Tables 1 and 2 of the RESULTS section. A discussion of the importance of the computed errors is also presented in the RESULTS section. 6. Standards of aldicarb sulfone and aldicarb sulfone nitrile were added to water and separated by Florisil into two fractions: one containing the nitrile and the other containing the sulfone. Recoveries were approximately 111% for the nitrile and 93% for the sulfone. The data are provided in Table 2 of the RESULTS section. 7. Readings of pH for the laboratory experiments were taken at each time a subsample was withdrawn for TTR analysis and pH was found to increase approximately'l pH unit within 10 days of the initial time in the private water supply wells (Ryall, Sexton, and Luther). The pH hydrolysis experiment for aldicarb and the hand-augered wells showed pH variations within .2 pH units during the incubation period, except for the Lindsey Wabasso samples taken on January 30, 1984, which had an increase of IV 0 9 pH uni t. 8. A check as to whether positive interferences from oximes would result for those samples not passed through a Florisil column revealed that they would not interfere at the concentrations of toxic aldicarb residues used in this investigation. Details of the check-out procedure are given in the RESULTS section. 9. Anoxic conditions were maintained throughout the laboratory experimental studies since dissolved oxygen could not be detected at the end of the 21

PAGE 28

incubation period in the groundwaters or saturated subsoils from the augered wells which were contained in Mason jars; oxygen was also not detected in the atmospheres in the BOD bottles used for experiments on aldicarb hydrolysis and degradation of TTR in waters from private wells for aldicarb, aldicarb sulfoxide, and aldicarb sulfone amendments. 10. An evaluation of the buffer catalysis contribution to the hydrolysis of aldicarb for each pH-buffered solution indicated that there was minimal contribution, if any, from the nature and concentration of the acid-base system used to buffer the pH. A full account of the evaluation is provided in Appendix II. 11. Tests for enumerating bacterial densities at the end of each incubation period were always conducted to evaluate whether sterilization had been maintained throughout the duration of the experimental period, or, in the case of non-sterile samples, that incubation conditions had closely approximated the in situ environment such that bacterial densities after the incubation period did not differ significantly for densities measured at the beginning of the incubation period. In all cases, bacteria densities remained close to their initial values during the incubation period. Raw data for baqterial counts can be found in Appendix III. 22

PAGE 29

RESULTS Interferences From Non-toxic Aldicarb Residues: Nitriles and Oximes Care must be exercised in labeling as total toxic residues any S'ample extractions which have not been passed through a Florisil column to remove nitriles and oximes (Romine 1974) prior to GC analysis. If a significant proportion of aldicarb residues is comprised of the non-toxic oximes or nitriles, then they would serve as a positive interferent by chromatographing with the toxic aldicarb and its S-oxides; neither nitriles nor oximes should be included in any analytical procedure designed for measuring TTR. Since the method followed at the beginning of this study was an unpublished one (Union Carbide 1980) which made no mention of removing non-toxic aldicarb residues from the toxic aldicarb residues by Florisil separation, twenty-two groundwater samples collected toward the end of the period of investigation (when the possibility of a nitrile interference became known) were eluted through Florisil columns and fractions .1 and 2 analyzed for aldicarb residues (Table 1). Ten of the 22 samples were split before Florisil separation and one aliquot put through a Florisil column and the other aliquot left without Florisil separation (Table 2). Large ranges in percentage recoveries (57-129%) were found for the groundwater samples passed through Florisil-columns when compared to the aliquots left without separation by liquid chromatography, with an average of 99% for the 10 samples (Table 2). Interferences from nitriles, which would result in an over-estimation of TTR, became quantitatively important only near the end of the incubation period in those samples that had >90% disappearance of the initial TTR (Table 1). Although the absolute error was large for those particular samples because of the low concentrations of TTR remaining 10 ppb) 23

PAGE 30

Table l. Amount of Non-toxic Nitriles and Toxic Aldicarb Residues Remaining in Duplicate Groundwqters and Subsoils After Florisil Separation. Amount Remaining, ppb Time 0 Time 20 Days Time 42 Days Sample Compound Added TTR TTR a Nitrileb TTR a Nitrileb Ryall-A aldicarb sulfoxide 97 54 ND Ryall-B aldicarb sulfoxide 105 4 10 Ryall-A aldicarb sulfone 81 43 3 ND 5 Ryall-B aldicarb sulfone 97 28 ND Sexton-A aldicarb sulfoxide 97 34 ND 4 5 Sexton-A aldicarb sulfoxide 98 35 4 Sexton-A aldicarb sulfone 80 12 ND ND Sexton-B aldicarb sulfone 92 37 ND 13 Luther-A aldicarb sulfoxide 95 43 6 9 9 Luther-B aldicarb sulfoxide 92 40 13 6 Luther-A aldicarb sulfone 90 20 5 ND ND Luther-B aldicarb sulfone 94 30 ND ND Amount Remaining, ppb Time 0 Time 25 Days Sample Compound Added TTR TTRa Nitrileb Lindsey Wab. water-A aldicarb sulfoxide 79 54 ND Lindsey Wab. water-B aldicarb sulfoxide 80 53 3 Lindsey Wab. subsoil-A aldicarb sulfoxide 51 12 ND Lindsey Wab. subsoil-B aldicarb sulfoxide 62 10 5 Lindsey Wab. water-A aldicarb sulfone 90 30 3 Lindsey Wab. water-B aldicarb sulfone 90 29 11 Lindsey Wab. subsoil-A aldicarb sulfone 64 10 5 Lindsey Wab. subsoil-B aldicarb sulfone 78 16 ND a TTR = eluate from fraction 2 (50% acetone in ethyl ether) of Florisil column b Nitrile eluate from fraction 1 (5% acetone in ethyl ether) of Florisil column ND ;:: -non-detectable 3 ppb) 24

PAGE 31

Table 2. Recovery of Aldicarb Sulfone and Aldicarb Sulfone Nitrile From Water and Sediments After Eluting Through Florisil Columns. Measured, ppb Recovery, % No Florisila Florisilb No Florisila Florisilb Amount Added Florisil Fraction .=T..::o:.-..:.W:-;a::..t::..e::..r:._____ Separ ati on 1 35 ppb Sn 62 ppb Sn nitrile 35 ppb Sn + 62 ppb Sn nitrile Amount Measured in Sample Water or Subsoil Lind. Wab. Water Sx added; 25 days Lind. Wab. Water Sn added; 25 days Lind. Wab. Subsoil Sn added; 25 days Lind. Wab. Subsoil Sx added; 25 days Ryall Sn added; 20 days Sexton Sn added; 20 days Luther Sn added; 20 days Luther Sx added; 20 days Luther Sx added; 42 days Luther Sx added; 42 days 57 35 15 14 46 21 26 65 14 16 NO 70 68 NO 11 5 5 3 NO 5 3 9 13 Fraction Florisil Fraction __ ..;:2:...__ Separation 1 34 NO 31 54 29 10 10 43 12 20 43 9 6 95 114 100 107 100 57 96 71 129 119 NO 113 110 Average % Recovery For Samples 99% Fraction 2 97 NO 89 a Florisil fraction 1 = 5% acetone in ethyl ether and contains aldicarb Sn nitrile. b Florisil fraction 2 = 50% acetone in ethyl ether and contains aldicarb Sn NO non-detectable ppb) Sx = Sulfoxide; Sn = Sulfone 25

PAGE 32

the relative error based on the initial TTR concentrations was small (average of 6% with a range of 0-14% for the 8 groundwater samples from the private water supply wells spiked with aldicarb sulfone or aldicarb sulfoxide). Considering that this is less than the individual percentage errors associated with recoveries of aldicarb sulfone nitrile and aldicarb sulfone from Florisil separation of water spiked with standards (111% and 93%, respectively) (Table 2), recoveries of total toxic plus non-toxic residuals remaining in groundwater samples after Florisil separation (57-129%) (Table 2) inter-laboratory comparisons of split samples and the average error between duplicate samples (9 and 22% for aldicarb sulfoxide and aldicarb sulfone, respectively), the extra time and effort involved in the Florisi1 removal of nitriles may not be warranted. Hansen and Spiegel (1983) also observed in their hydrolysis studies that presence of nitriles was important only after 98 to 99% of the aldicarb sulfoxide or sulfone had been degraded. Furthermore, Union Carbide (Romine, pers. comm.; May 1984) has not found nitriles of aldicarb or its sulfoxide or sulfone as residues in potable groundwater from their monitoring of many groundwater sources in several states, including Florida. Neglecting to remove nitriles from some of the earlier samples apparently was not a significant source of error in analyzing for TTR. Apparently aldicarb oxime does not pose as an interferent since it is not thermally decomposed to nitrile at an injection port temperature_of 3500C (Knaak et ale 1966). However, the aldicarb sulfoxide oxime and aldicarb sulfone oxime decompose to ni triles at 3500C (Knaak et al. 1966), and thus could also serve as a positive interferent, especially since any aldicarb oxime and aldicarb sulfoxide oxime present would be oxidized to aldicarb sulfone oxime by the peracetic acid oxidation step. Studies done in our lab using either aldicarb sulfoxide oximeor aldicarb sulfone oxime at final 26

PAGE 33

concentrations of 5 ug/mL in the acetone extract, and without a Florisil separation step, did not elicit a response on the GC under the .operating conditions set for the instrument. This was true regardless of whether or not the oxime had undergone oxidation by peracetic acid. Larger amounts of the oxime (100 ug/mL) in acetone did produce a full-scale response by the detector. Clearly, oximes would not have been detected at the levels of residues -10 ug/mL) that were analyzed in our samples. This could have been due to poor detector sensitivity to oximes; and/or low recovery rates of oximes from extracts (Maitlen et ale 1968); and/or removal of oximes quantitatively by conversion to their respective aldehydes (which do not with the analysis) by acid hydrolysis in the peracetic acid oxidation step (Beckman et al. 1969). Hydrolysis Rates in Sterile pH-buffered Distilled Water Plots of log percent remaining vs. time are shown in Figure 3 in an effort to determine the contribution of H+, OH-, and H 2 0 to the rate of degradation. The actual data at each duplicated pH are presented in Table A-2 in the Appendix III. At pH = 6, 7, and 8 aldicarb hydrolyzes slowly, but increases at higher and lower pH levels (Figure 3). Plotting log k vS. pH in Figure 4 demonstrates that there are only slight changes in the rate constant in the pH 6 to 8 range. The least squares estimates of all the pseudo-first-order hydrolysis rate constants, half-lives, and coefficients of determination are given in Table 3. Only the tests conducted at pH = 4, 9, and 10 showed degradation sufficient for estimating a rate constant. Therefore, the rates and resulting half-life values for pH 6-8 are only extimates since the slopes of the log percent remaining vs. time regres-sion lines in Figure 3 were probably not significantly different from zero. To accurately estimate the half-life for these pH conditions, 27

PAGE 34

<9 Z -z
PAGE 35

o ; -I ..I >. -2 0 ""0 "'--" ()) 0 N -3 \.0 -4 4 5 6 7 8 9 10 pH Figure 4. Log k vs. pH for Aldicarb Hydrolysis at 20C in Sterile pH-buffered Distilled Water Solutions.

PAGE 36

Table 3. Pseudo-first-order Rate Constants (d), Half-life Values and Coefficient of Determination of the Regression Line (r2) For Aldicarb Hydrolysis at 200C in pH -buffered Distilled Water. pH Period (days) k (day-l) (days) 3.95 89 5.3 x 10-3 131 6.02 89 1.2 x 10-3 559 7.06 89 8.1 x 10-4 861 7.96 89 2.1 x 10-3 324 8.85 89 1.3 x 10-2 55 9.85 15 1.2 x 10-1 6 30 r2 0.86 0.90 0.21 0.62 0.98 1.00

PAGE 37

experiments lasting longer than 89 days would have to be performed. The data obtained,at pH = 7 showed a slight increase in aldicarb and its oxidation products over time, resulting in a low coefficient of determination (r2 = 0.21) and a positive slope. Above pH 8 the pseudo-first-order rate constant increases with increasing pH and the slope of the line is approximately +1 (Figure 4), indicating aldicarb hydrolysis is sensitive to hydroxyl ions in aqueous solutions. 'At pH <6, the rate of hydrolysis appears to be acid catalyzed, but not to the extent as for base catalysis since the slope is less than 1. The nonlinearity of the plot between pH 6 and 8 is interpreted as resulting from competing reactions of aldicarb with'water, hydrogen, and hydroxide. The second-order reaction rate constant for base hydrolysis, kOH' was first-order with respect to hydroxide because the plot of log k vs. pH (Figure 4) yielded a +1 slope at pH > 8. The kOH calculated from Equation 3, using the data obtained at pH 7.96, 8.85, and 9.85, was 1.94 x 103 3.54 x 102 L mole-l day-I. The acid hydrolysis constant, kH' which was not first-order with respect to hydrogen, had a computed value (based only on the data acquired at pH 3.95) of 4.72 x 101 L mole-l day-I. Groundwater Characteristics Private Water Supply Wells The groundwaters from the three private water supply wells contai?ed similar heat values and concentrations for the following constituents: temperature, specific conductance, pH, alkalinity, and hardness (Table 4). Sexton well, however, exhibited higher levels of total Fe (0.12 mg/L) and sulfide (6.6-8.9 mg/L) while having stronger reducing conditions (-145 to -163 mV). Luther well yielded approximately an order-of-magnitude higher bacteria cell concentration (55-77 cells/rnL) than the other two wells, but 31

PAGE 38

Table 4. Physical, Chemical, and Biological Characteristics of Groundwaters Sampled From Private Water Supply Wells. WELL SITE Ryall Sexton Luther Date Sampled 12/9/83 3/26/84 1/9/83 3/26/84 12/9/83 3/26/84 Depth (m) 20 20 17 17 ? ? Temp (oC) 24 24 24 25 24 24 Sp. Condo (uS/em) 1126 1363 1073 1049 789 839 D. 0. (mg/L) 0.0 0.0 0.0 0.0 0.1 0.1 Eh7 (mV) +264 +274 -163 -145 +326 +360 pH 6.8 7.8 7.2 6.8 7.3 7.2 Alkalinity (mg CaC0 3/L) 292 292 355 353 219 289 Hardness (mgcaC03/L) 363 440 298 300 178 272 Total Fe (mg/L) 0.01 0.04 0.11 0.12 0.05 0.20 Sulfide (mg/L) 0.25 0.38 8.88 6.58 0.38 0.57 Bacteria (cells/mL) a Before 5 4 2 3 55 77 After 11 7 7 6 37 63 aAverage of two anaerobic plate counts before and after the incubation period. 32

PAGE 39

still was low when compared to cell densities from subsoils (Table 5). The low densities of bacteria were maintained throughout the incubation period, attesting to the sterile techniques and precautions taken not to alter the original temperature and redox potential of the groundwaters during the laboratory incubations. Generally, the groundwaters can be character-ized as being hard, anoxic, and reduced, with high alkalinities, low iron, and neutral pH. Augered Wells The shallow groundwaters from the augered wells at BBe and Lindsey Wabasso groves differed from the deeper water supply wells by containing more dissolved solids and bacteria cells (Table 5). Oxidation-reduction potential and pH were consistent with those values reported for the deeper groundwaters at Ryall, Sexton, and Luther wells. Lindsey Wabasso groundwater had a higher alkalinity and hardness than of any sampled groundwater.. Degradation of TTR From Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone Amended to Groundwaters From Private Water Supply Wells The least-squares estimates of the pseudo-first-order rate constants, half-life values, and coefficients of of the regression lines when the percent TTR remaining is plotted logarithmically for 0.10 mg/L additions of aldicarb, aldicarb sulfoxide, and aldicarb sulfone are presented in Table 6. The linear coefficients for the straight lines are high (r2 0.88-1.00), indicating.that some confidence can be placed on the pseudo-first-order rate constants and half-life values which were determined from the data. The raw data for each duplicate incubation container and at each time that a sample was withdrawn and measured for TTR are presented in Appendix III (Tables A-3 and A-4) Notwithstanding a high variability in the data during the earlier stages of incubation, Figure 5 depicts that approximately 60% of the initial 0.10 mg/L 33

PAGE 40

Table 5. Physical 1 Chemical, and Biological Characteristics of Groundwaters Sampled From Augered Shallow Wells. Well Site Lindse:x: Wabasso BBC Date Sampled 1/30/84 4/23/84 2/22/84 Depth (m) 1.5 1.5 1.3 Temp (C) 22 26 22 Sp. Condo (pS/em) 2750 2462 D. o. (rng/L) 0.0 0.1 0.4 (mV) +111 +83 +321 pH 7.1 7.3 7.4 Alkalinity (mg CaC03/L) 358 442, 106 Hardness Gng CaC03/L) 836 860 202 Total Fe (mg/L) 0,16 0.12 Bacteria (cells/mL) a Before 7.5 x 102 3 1.3 x 10 3.6 x 10 4' After 2.2 x 102 9,0 x 10 2 3 2.0 x 10 Bacteria (cells/g dry wt) a Before 1.lz c; 10-6.0 x 104 4.6 x 10 4 4 8.0 x 103 10 3 After 1.5 x 10 9.0 x a Average of two anaerobic plate counts before and after the incubation period. 34

PAGE 41

Table 6. Pseudo-first-order Rate Constants (k), Half-life Values and Compound Aldicarb Aldicarb Sulfoxide Aldicarb Sulfone Coefficients of Determination of the Regression Lines (r2) at 240C For the Disappearance of TTR in Sterile and unsterile Ground-waters Amended with 0.10 mg/L of Aldicarb, Aldicarb Sulfoxide, or Aldicarb Sulfone. Each Value Is a Mean of Duplicate Samples. Well Sterile Period (days) k (day-I) (days) Ryall Yes 90 5.77 x 10-3 120 No 90 6.74 x 10-3 103 Sexton Yes 90 6.90 x 10-3 100 No 90 6.88 x 10-3 101 Luther Yes 90 7.80 x 10-3 89 No 90 6.74 x 10-3 103 Ryall* No 42 2.95 x 10-2 24 Sexton* No 42 7.71 x 10-2 9 Luther* No 42 5.23 x 10-2 13 Ryall* No 20 4.51 x 10-2 15 Sexton* No 20 6.65 x 10-2 10 Luther* No 20 6.46 x 10-2 11 *Nitriles removed from sctmple 35 0.96 0.92 0.90 0.98 0.88 0.92 1.00 0.98 0.99 1.00 0.99 1.00

PAGE 42

w 0'1-100 90 80 (9 z -z
PAGE 43

aldicarb inoculum remained as a toxic residue in all three groundwaters after 90 days at 24oC, regardless of whether the water was sterile or non-sterile. The pseudo-first-order rate constants for the TTR remaining after aldicarb was added to the groundwaters fell within a narrow range of 5.77 x 10-3 to 7.80 x lO-3/day, which equalled 89 and 120 days when transformed into their respective half-life times. TTR concentrations disappeared at faster rates (half-life of 9-24 days) when. either of the S-oxides was added (Table 6) ; there were no major differences in the degradation of TTR between sulfoxide and sulfone or among wells. The differences in the decomposition of TTR observed for amendments of aldicarb and its S-oxides cannot be from the removal of nitriles in only the S-oxide samples (and not the aldicarb samples) for reasons previously described. It should be pointed out that these measured rates overestimate the true rates of TTR degradation since the pH of the water supply wells increased approximately 1 pH unit/in each well water during the incubation period. Still, they represent rates under pH conditions which can exist at different times of the year. For instance, the Ryall well under-went a pH change of 1 (6.8 vs. 7.8) between the two sampling periods in December and March (Table 4). Degradation of TTR From Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone Amended to Groundwaters and Saturated Aguifer Material As was found for the private water supply well waters, aldicarb added to groundwaters without the subsoil aquifer material decomposed to non-toxic residuals at very low rates (Table 7) :"half-life values of 112 days at the BBC and 178 days at Lindsey Wabasso groves. Agreement among the replicate samples was high without any differences being measured (cf. Appendix III). When aldicarb was added to aquifer material consisting of saturated subsoil from the top of the water table, the rate of TTR disappearance was increased by an order-of-magnitude. For example, the half-life times decreased from 112 37

PAGE 44

Table 7. Percent Remaining, Pseudo-first-order Bate Constants (k), and Half-life Values For the Disappearance of TTR in Groundwaters and Aquifer Material From Lindsey Wabasso (LW) and BBC Groves Amended With Aldicarb, Aldicarb Sulfoxide, or Aldicarb Sulfone. Each Value Is a Mean of Duplicate Samples. Aldicarb Site Groundwater BBC Aquifer Material BBC Groundwater LW Aquifer Material LW Aldicarb Sulfoxide Groundwater BBC Aquifer Material BBC Groundwater LW Aquifer Material LW Aldicarb Sulfone Groundwater BBC Aquifer Material BBC Groundwater LW Aquifer Material LW *Nitriles removed from sample Temp (OC) 22 22 22 22 22 22 26 26 22 22 26 26 Period (days) 23 23 30 30 23 23 25 25 23 23 25 25 38 % TTR Remaining --=------87 27 89 62 -54 67* 20* 54 42 33* 18* 6.21 x 10-3 5.76 x 10.,..2 3.88 x 10-3 2.09 x 10-2 2.64 x 10-2 1. 60 x 10-2 6.44 x 10-2 2.64 x 10-2 3.77 x 10-2 4.43 x 10-2 _6.86 x 10-2 (days) 112 12 178 33 26 43 11 26 18 16 10

PAGE 45

days to 12 days when aquifer material was present from the BBC site. For the Lindsey Wabasso grove, the conversion of TTR to non-toxic residues was so fast that no detectable TTR was found after 30 days when aquifer subsoil was present. This should be compared to a half-life of 178 days without aquifer material being present. Adsorption of aldicarb onto the clays in the subsoil did not occur since extraction of the sediments consistently produced non-detectable levels ( 6 ng/g (dry wt.. A similar trend waS also noticed for the disappearance of TTR from sulfoxide and sulfone additions to groundwaters with and without aquifer material (Table 7) but not to the extent that was recorded for aldicarb. Instead of the 9-fold difference in the degradation rates with and without aquifer material that was found in the aldicarb amendments, sulfoxide or sulfone amended to saturated aquifer subsoils increased the conversion of TTR to non-toxic products by only 1.3 to 4.0 times. Thus, the presence of aquifer material resulted in only slightly greater rates of degradation of TTR from either sulfone-or sulfoxide-amended groundwaters: half-life times of 16-43 days without aquifer material and 10-26 days in the presence of aquifer material. The removal of nitriles from some of the S-oxide amended samples but not from the aldicarb-amended samplescould not have accounted for the differences in the rates of TTR disappearance measured for aldicarb amended and S-oxide amended samples since the percentage of the initial TTR comprised by nitriles was small (Table 1). The small differences between the rate constants from the two groves for sulfoxide and sulfone amendments in Table 7 were probably due more to the higher incubation temperature (260C) used for. the Lindsey Wabasso samples than from the removal of nitriles from the Lindsey Wabasso samples. Adsorption onto clays was negligible: no aldicarb sulfoxide or aldicarb sulfone was found above the limits of detection (6 ng/g (dry wt.. 39

PAGE 46

DISCUSSION Hydrolysis Rates in pH-buffered Distilled Water Laboratory hydrolysis studies of any xenobiotic using sterile pH-buffered distilled water can only be interpreted as representing a "worst case" situation since all the environmental factors such as volatilization, adsorption, plant uptake, leaching, and microbial degradation present under field conditions have been omitted. Moreover, laboratory studies of hydrolysis reaction rates are not only a function of pH, but also of the nature and concentration of the acid-base system used to buffer the pH, which is called buffer catalysis (Perdue and Wolfe 1983). A detailed account showing the effect of buffer catalysis was negligible under the experimental conditions used in this investigation (Appendix II). The extent of error in not considering the environmental conditions in the field and not using actual well waters and aquifer material when interpreting hydrolysis data will be discussed later in this report. Still, as Hansen and Spiegel (1983) point out, hydrolysis rates obtained from laboratory studies can be used to establish upper bounds for the half-lives of aldicarb in groundwater. Comparisons of hydrolytic half-life values reported by other investigators for aldicarb in sterile, pH-buffered distilled water are presented in Table 8. For those cases when the raw data were available, the rate constant;:; derived for temperatures other than 20C were adjusted to a temperature of 200C by constructing Arrhenius plots. Not only is there a scarcity of published literature on aldicarb hydrolysis, but only a few of the published studies in'cluded the range of pH values which bracket the pH of natural waters. Carbamates such as aldicarb typically are quite resistant to hydrolysis at neutral pH values, but are relatively unstable under alkaline conditions of 40

PAGE 47

Table 8. A Comparison of Hydrolytic Half-lives for the Disappearance of TTR in sterile pH-buffered Distilled water Amended with Aldicarb. pH 3.95 6.02 7.06 7.96 8.85 9.85 12.90 13.39 8.5 8.2 Temp (OC) 20 20 20 20 20 20 15 15 20 None given Half-life (days) 131 559 861 324 55 6 4.0 min 1. 3 min 69 43 41 Reference This study This study This study This study This study This study Lemley and Zhong 1983 Lemley and Zhong 1983 Hansen and Spiegel 1983 Trehy et al. 1984

PAGE 48

pH (Faust and Gomaa 1972) yielding aldicarb oxime (which is stable in basic medium), methylamine, and carbonate from the cleavage of the bond (Lemley and Zhong 1983). Trehy et ale (1984) found from GC/MS analysis that the degradation product for aldicarb in sterile anaerobic water was also aldicarb oxime. Oximes can undergo a dehydration to become another non-toxic aldicarb residue: nitriles. Hansen and Spiegel (l983) felt that sulfoxide nitrile and sulfone nitrile became important only after 98 or 99% of the aldicarb sulfoxide or aldicarb sulfone had been hydrolyzed at pH 8.5; however, they never measured the nitriles. Presumably, aldicarb nitrile would also become a dominant degradation product under similar circumstances. The Hansen and Spiegel (l983) work is the closest data set comparable to the conditions of our experiment. Their data (adjusted to 200e) for pH 8.5 yield a half-life of 69 days, which is almost twice as fast as the pseudo-first-order rate constant extrapolated from Figure 4 pH 8.5 (k = 5.8 x 10-3/day) I corresponding to a'half-life of 120 days. Little confidence can be placed on the rate constants obtained at pH 7.5 in either study since the slope of the regression line, which is equal to k, was not significantly different from zero in the Hansen and Spiegel study and also was probably not so in this study; having only two replicates precluded testing the hypothesis of whether the slope was significantly different than zero. To the authors' knowledge, the only published values for secorid-order rate constants of aldicarb hydrolysis is from Lemley and Zhong (l983). Based on using high hydroxide and aldicarb concentrations, and a different method (i.e., titrimetric) to measure the progress of hydrolysis, they found kOH for aldicarb to be 1.35 x 103 0.03 x L mole-l'day-l at 15 e. After adjust-ing to a temperature of 200e (assuming the activation energy of the aldicarb is the same as the activation. energy measured by Lemley and Zhong for aldicarb 42

PAGE 49

sulfoxide (= 15.2 0.1 kcal/mol), the kOH becomes 2.12 x 10 3 L mole-l day-I, which compares favorably to the 1.94 x 103 L mole-l day-l rate measured by us. There is some confusion in the literature as to whether acid-catalyzed hydrolysis of aldicarb and its S-oxides occur. Lemley and Zhong (1983) have measured proton-catalysis for aldicarb sulfone, but the reaction rates were s'low and unmeasurably low when the acid concentrations were below 2M. Hansen and Spiegel (1983), based on their communications with L. Tobler (1980), reported no acid-catalysis of aldicarb at 770C down to a pH of 2, a finding which is inconsistent with the results of this study. Disappearance of Toxic Residues of Aldicarb in the Saturated Zone Although hydrolysis experiments using pH-buffered sterile distilled water are easier to perform than experiments using groundwaters and their aquifer material from specific sites, their results are not as meaningful / because microbiological and aquifer catalytic effects are not taken into consideration. Neither do field studies lend themselves to the accurate degradation rate of TTR in groundwaters since the residues are subject to dispersion, dilution, and recharge, all of which are beyond the control of the investigator. Therefore, TTR degradation rates are best by laboratory studies using actual groundwaters and saturated subsoils collected in the field and incubated under controlled conditions which best represent the in situ environment. The half-life values for groundwaters and aquifer material obtained from the well sites are compared to the hydrolytic degradation half-life times in Table 9 after appropriate adjustments for pH and temperature differences. In those samples which experienced an upward shift in the pH during the incuba-tion period, an average pH was used for the basis of comparison to the hydrolysis studiesin sterile, pH-buffered distilled .water. The pH shift to more alkaline 43

PAGE 50

Table 9. Comparison of Degradation Rates For TTR Estimated by Hydrolysis With Degradation Rates Measured in Grol.mdwaters in the Absence and Presence of Aquifer Material. Estimated Half-life Based on Distilled Study Measured Water Residue ,EH Temp Half-life Hydrolysis Estimated Site Added Initial Ave. (OC) (days) (days) a Measured Ryall aldicarb 6.8 8.0 24 103 211 2 Sexton aldicarb 7.2 7.9 24 101 238 2 Luther aldicarb 7.3 8.0 24 103 211 2 BBC aldicarb 7.4 7.3 22 112 648 6 Lind. Wab. aldicarb 7.1 7.5 22 178 456 3 BBcb aldicarb 7.4 7.4 22 12 515 43 RyallC aldicarb 6.8 8.0 24 120 211 2 SextonC aldicarb 7.2 7.9 24 100 238 2 Lutherc aldicarb 7.3 8.0 24 89 211 2 Ryall sulfoxide 7.8 7.9 24 24 30 1 Sexton sulfoxide 7.2 7.9 24 9 30 3 Luther sulfoxide 7.4 7.6 24 13 50 4 BBC sulfoxide 7.4 7.3 22 33 140 4 Lind. Wah. sulfoxide 7.4 7.4 26 43 70 2 BBcb sulfoxide 7.4 7.4 22 26 110 4 Lind. Wah. b sulfoxide 7.4 7.3 26 11 140 11 Ryall sulfone 7.8 7.9 24 15 12 1 Sexton sulfone 7.2 7.9 24 10 12 1 Luther sulfone 7.4 7.6 24 11 23 2 BBC sulfone 7.4 7.3 22 26 70 3 Lind. Wah. sulfone 7.4 7.4 26 16 38 2 BBCb sulfone 7.4 7.4 22 18 62 3 Lind. Wah. b sulfone 7.4 7.3 26 10 44 4 aAldicarb half-life times based on data presented in Fig. 4 and Table 3; sulfoxide and sulfone half-life times based on data presented by Porter et al. (1984) bAquifer material present CSterilized groundwater 44

PAGE 51

values by as much as one pH unit was unexpected. It indicated the waters were not well buffered at their initial neutral pH values, the high alkalinities notwithstanding. It is assumed the pH shift occurred because of losses of C02 during sequential subsamplings for TTR analysis. Apparently the change occurred within the first 10 days of the incubation period. If a buffer had been added to maintain the pH at its initial value, even more of an error may have been introduced because of: 1) changing the ionic strength of the waters (which may affect rate constants); 2) exerting a buffer catalysis effect; and 3) serving as a nutrient or energy source to the microbiota. For the locations listed, the half-life times measured in the groundwater only and groundwater + aquifer samples are a factor of 1 to 43 shorter than half-life estimates based on distilled water hydrolysis. Metabolic activities of the microbiota cannot be invoked in explaining the faster degradation of aldicarb in well waters since there were no differences between the rates of aldicarb disappearance between sterile and non-sterile groundwaters (estimated was 2 in all cases). Even though the anaerobic plate counts probably underestimated bacterial cell densities because of.the bias toward gram-negative bacteria when a high proportion of the total bacteria in the unconsolidated sediments of the saturated zone has been reported to be grampositive (White et 1983), the low bacteria densities found for the three private water supply wells (Table 4) indicate a negligible microbiological influence on TTR degradation rates. Therefore, the well waters contained dissolved constituents which, when incubated under conditions closely resembling those found in the field, increased the rate of degradation of aldicarb and its toxic derivatives. Based solely on the degradation rates for aldicarb-amended groundwaters, a substantial risk of contaminating groundwaters would exist if aldicarb should 45

PAGE 52

leach into the saturated zone. However, in the presence of aquifer material, which is a more realistic aldicarb and its toxic residues disappeared 9 times faster at the BBe site where the estimated ratio was increased to 43 (Table 9). The higher densities of bacteria in the groundwatersaturated subsoils (Table 5) probably accounted for the faster decay of TTR, since aldicarb was not significantly sorbed on the sandy clay loam subsoil ng/g (dry wt which served as the water table aquifer at this site, thereby reducing the likelihood that catalyzed degradation was responsible for the faster subsoil degradation rate. Sterile conditions (through NaN 3 additions) were not achieved in the subsurface soils from Lindsey Wabasso (see Appendix III) probably because of the protection afforded to the microbiota by extracellular polysaccharide polymers secreted under the conditions of unbalanced growth (i.e. nutrient deprivation) usually found in underground waters (Uhlinger and White 1983). It was not the" objective of this investigation to determine the individual degradation products, although such a study on these groundwaters would be useful. Since the incubations were anaerobic, oxidation of aldicarb to sulfoxide by bacteria would be less likely than bacterially-mediated hydrolysis to aldicarb oxime. Oxidation processes cannot be entirely ruled out, but an oxidizing agent other than oxygen would have to serve as the terminal electron acceptor. TTR concentrations in aldicarb sulfone-or aldicarb sulfoxide-enriched well waters decreased at rates which were more consistent with expected values published for hydrolysis in sterile, pH-buffered distilled water (Table 9). By using a detailed study on the effects of pH and temperature on the hydrolysis of aldicarb sulfoxide and aldicarb sulfone (Porter et ale 1984), we interpolated the estimated half-life times for Ryall, Sexton, and Luther wells to 46

PAGE 53

be 30-50 days for aldicarb sulfoxide and 12-23 days for aldicarb sulfone. These should be compared to the nearly equal measured half-life times of 9-24 days and 10-15 days for aldicarb sulfoxide and aldicarb sulfone, respectively (Table 9). Since these samples were not sterilized, the role which microbiota in the groundwaters had in the degradation was unknown. However, the low bacteria densities (6-63 cells/mL) measured for the March 26, 1984 well water samples (Table 4) indicated only a negligible contribution from the microbiota could have occurred. It is therefore believed that most of the increase in degradation rates was due to non-biological effects, with chemical hydrolysis accounting for much of it. This is consistent with the findings of Delfino and co-workers (1984), who found aldicarb sulfoxide to have a half-life of 13-15 days in autoclave aerobic groundwaters. When aldicarb sulfoxide and aldicarb sulfone were separately added to saturated subsoils obtained from Lindsey Wabasso and BBe groves (containing 5-100 times more bacteria than without aquifer material), only slightly higher rates of TTR disappearance higher for 3 of the 4 site-substrate combinations) than what had been observed for well waters alone were recorded (Tables 7 and 9). This differs sharply from the 9-fold increase in TTR rate of degradation previously described for aldicarb-amended saturated subsoils. The exception was for sulfoxide at the Lindsey Wabasso site, where the decomposjtion rate for TTR was increased 4 times by the presence of aquifer material. These data indicate that using only groundwaters may suffice testing the potential of sulfoxide and sulfone to degrade in aquifers. The 50% overestimation of half-life values may be an acceptable error (especially since it is a conservative one} considering the time and expense expended in obtaining aquifer material from deep wells and that contaminated groundwaters have been found to contain 50% each of aldicarb sulfoxide 47

PAGE 54

and aldicarb sulfone (Porter et al. 1984). However, more studies are necessary before this procedure could be routinely practiced because the nature of the aquifer material can be important. Miles (pers. comm., April 4, 1984) found limestone decreased the hydrolysis rate of aldicarb sulfoxide and aldicarb sulfone five-fold. This may be true only for deep limestone aquifers such as the Floridan, and not for the shallower unconfined aquifers that are composed of sands and clays. In summary, the effect of the presence of aquifer material on the decompo-sition of TTR from aldicarb addition was pronounced; however, aquifer material had only small effects on rates of TTR disappearance for sulfone and sulfoxide. Because aldicarb, aldicarb sulfoxide, and aldicarb sulfone sorb weakly in saudy and clay soils with low organic matter, surface catalyzed degradation is less likely to be more important than microbial or solution processes in causing faster aldicarb degradation in saturated subsoils. If bacteria are essential in the anaerobic degradation of aldicarb to non-toxic oximes in groundwaters, they are relatively unimportant in degrading sulfone and sulfoxide, a finding consistent with studies conducted on Long Island (Porter et ala 1984). Also, for sulfoxide and sulfone, chemical hydrolysis in solution proceeds fast enough to be a major degradation pathway by itself. Since all the experiments were conducted under anaerobic conditions, oxidation reactions would probably be unimportant relative to hydrolysis reactions, which produce oximes (Figure 1). It therefore appears that without b.:lcteria, aldical.-b hydrolysis is slow com-pared to sulfoxide and sulfone hydrolysis in anaerobic groundwaters, which conforms to the findings reported by Hansen and Spiegel (1983), Delfino (1984), Porter et ala (1984) and this investiga'tion. Bacteria are necessary for catalyzing the hydrolysis of only aldicarb. \'1hen this occurs, the rate of TTR disappearance equals that fl.. ... r chemical hydrolysis of aldicarb sulfoxide and aldicarb sulfone. 48

PAGE 55

Comparison of Degradation Rates of TTR in the Unsaturated and Saturated Zones To understand the significance of the findings 'in this report, the broader picture of the likelihood of aldicarb or one of its toxic residues reaching the water table from the vadose zone should be presented. The sus-ceptibility of groundwaters to aldicarb contamination is a combination of the potentials for TTR to reach the water table and to persist after it has entered the groundwater system. Numerous studies of aldicarb behavior in soils in the unsaturated zone have demonstrated rapid conversion to sulfoxide, which in turn is more slowly biodegraded to sulfone (Bromilow et ale 1980; Smelt et ale 1978 b, c). However, one study (Coppedge et ale 1967) reported aldicarb to decompose slowly in a fine sand soil: 27% of the applied aldicarb remained after 4 weeks. Smelt et ale (1978c) calculated 91-100% of aldicarb was con-verted to its sulfoxide, which was higher than the 60-80% values given by Coppedge et ale (1967) and Bull et ale (1970), and the 67-92% conversion reported by Bromilow et al. (1980) I For two studies which investigated aldi-carb sulfone degradation in soils, both found rates to be slower than what had been reported for aldicarb sulfoxide (Smelt et ale 1978a) and aldicarb Generally, soil scientists have found slower rates in the degradation of aldicarb, aldicarb sulfoxide, and aldicarb sulfone in deeper soil layers than in corresponding top layers of the soil profile (Smelt et ale 1978a,b; Hornsby I et ale 1984), presumably because of the lower microbial activity in the subsoil. Typical half-life values reported in the literature for laboratory studies of aldicarb, aldicarb sulfoxide, and aldicarb sul.fone losses in the upper soil layers were 1-23, 13-14, and 24-158 days, respectively. For deeper soil layers (70-180 em) in the unsaturated zone, aldicarb sulfoxide had a reported half-life of 53-475 days (Smelt et ale 1978b) while half-life times of 46-00 49

PAGE 56

(Smelt et al. 1978a) and 54-296 days for aldicarb sulfone (Hornsby_ et al. 1984) have been published. These data imply that once TTR penetrate to the deeper layers of the aerated soil zone, little further degradation can be expected. When compared to the half-life times found in the upper soil layers of the unsaturated zone, the half-life range of 10-26 days for TTR disappearance measured for saturated subsoils in this study suggest a resumption to the faster degradation rates recorded for the upper soil layers can be expected for TTR entering the shallow water table. Field Monitoring Studies and Models For TTR Intrusion into Florida Groundwaters Based on the laboratory degradation studies for deep soils in the un-saturated zone and field investigations in Florida (Jones and Back 1984; Hornsby al 1984) and other states (Rothschild et al. 1982; Porter et al. 1984), the question no longer is whether toxic aldicarb residues are reaching water tables, but at what concentrations are they entering water tables, how fast is the TTR decomposing to non-toxic residues in groundwaters, and how far would TTR travel in groundwaters before it is degraded to non-toxic products. Jones and Back (1984) reported degradation rates of aldicarb residues in Florida soils decreased as they moved down through the soil column due to fewer soil biota available to metabolize the residues. After several months, when most of the remaining residues are 60 to 120 em below the soil surface, the disappearance of the residues became immeasurably low. Their data, which encompassed six citrus grove sites throughout clearly indicated significant percentages of TTR (i.e., aldicarb + aldicarb sulfoxide + aldicarb sulfone) can remain in the lower 1.2-2.2'm of unsaturated soil. Contamination of groundwaters from wells near the six citrus groves during a 1.5 year period was slight: out of 67 groundwater samples taken from a total OI 21 wells in 50

PAGE 57

the 6 locations, waters from only two wells (at the Hillsborough site) had intermittent trace amounts (1 ppb) of aldicarb residues. Although the field data were too variable to make predictions of the amount of aldicarb residues which will leach into the saturated zone, Jones and Back (1984) estimated that, based on a computer model (PESTAN), < 1% of the aldicarb applied to citrus groves will leach more than three feet below the soil surface. However, many assumptions were made in reaching this conclusion. These included using average soil and climatic properties, applying soil characteristics to Florida soils from sandy soil on Long Island, and adopting the faster degradation rates found for the upper soil horizons. Considering that the selection of input parameters for the modeling produced a "best case" simulation, the conclusion that < 1% of the applied aldicarb would leach below three feet of the soil surface \vould have to be viewed with a large degree of uncertainty. Furthermore, the field data provided by Jones and Back (1984) show that from 0.5 to 5.7% of the applied aldicarb remains as TTR in the unsaturated zone from 166 to 344 days after application. Most of the remaining residues would be located at the lower depths of the unsaturated zone, where degradation is slow and migration to the nearby saturated zone is likely to occur. The importance of looking at site-specific contamination, rather than !elying on generalized computer simulations or averages of field data, can be readily recognized by noting that the highest fraction of aldicarb residues remaining (5.3% in Hillsborough County) at any of the six field sites, was associated with the longest time after application (344 days). Another example of the variability of different soilsto degrade aldicarb and its toxic residues is given by Hornsby et al. 1984, where 4-8 percent of the applied aldicarb residues reached groundwater in a citrus grove on ridge soils in Seminole County, Florida, while no TTR were detected in the upper portion of the saturated zone at a flatwoods soil site in Polk County, Florida. 51

PAGE 58

In further modeling efforts, Jones et ale (1984) recognized that the model PESTAN, although simpler and easier to use than other models, is generally not appropriate for predicting the leaching of aldicarb residues in Florida citrus groves. By comparing three existing simulation models (PESTAN, PISTON, and PRZM) with each other and with field monitoring and laboratory studies, they found using site-specific factors such as soil hydraulic properties, soil organic matter, pesticide degradation rates, and daily rainfall data were important in determining the extent of pesticide leaching. Lateral Transport of TTR in Shallow Groundwaters in Indian River County The results of this investigation showed measured TTR half-life times of 10-26 days in the presence of aquifer material, regardless of whether the initial toxic residue had been aldicarb or one of its S-oxides. Using the number of half-lives (7) required to reduce the highest TTR concentration (1.26 mg/L) reported for Florida groundwaters (Hornsby et ale 1984) to levels less than the state standard of 10 ppb, and our measured half-life times, 70-182 days would have to elapse before TTR concentrations in that water (without dilution or dispersion) could decrease to acceptable TTR levels. This means that if Temik was applied only once a year, and toxic residues of this magnitude moved into the saturated zone as a pulse over a short time interval, TTR would be below the state drinking water standard (10 ppb) after one-half year. How far would such a plume travel in that period of time? Considering the calculated hydraulic gradients of 0.00019-0.00078 m/m derived from observation wells in Indian River County and reported hydraulic conductivities of 3.6-12.2 m/day for the overlying sandy soils located at the BBC and Lindsey Wabasso groves, and assuming a porosity of 33%, the plume of contaminated water would travel 0.0021-0.029 m/day from the boundaries of application. Thus" a distance of 0.38-5.3 m (1. 3-17 ft) would 52

PAGE 59

be traveled by a contaminated plume during the one-half year required for the TTR to decrease to concentrations <10 ppb. These calculated distances of lateral plume migration are well within the 300 ft exclusion zone adopted by the state for applying aldicarb near drinking water wells. simple calculation assumes a homogeneous and isotropic shallow groundwater aquifer with no contribution from dispersion or dilution toward reducing TTR concentrations. It therefore represents the travel distance of TTR in the most extreme case, especially since the highest reported concentration of TTR in Florida groundwaters was used. Another approach to evaluating the extent of migration of contaminated groundwater in Indian River County is by calculating how much of the 5 lb a.i./acre per year application rate as required under the new restrictions may enter into the groundwater and be transmitted laterally. If 8% of the applied aldicarb were to enter into the groundwater as reported by Hornsby et ale (1984), then 352 ppb of TTR would be concentrated in the surface to 0.1 m layer of groundwater. Rothschild et al. (1982) concluded that most TTR leached to groundwaters under aldicarb-treated fields was near the water table. If an average of 16 days measured in this study for a TTR comprised of 50% each of aldicarb sulfoxide and aldicarb sulfone (Porter et ale 1984) is assumed, then it would require 80 days (or five half-lives) for the TTR to degrade to a concentration that is in compliance with the ppb standard adopted by Florida for potable waters. Using the same hydraulic conductivities and gradients and porosity as before, then only 0.17-2.3 m (0.6-8 ft) distance would be traveled by contaminated plume. Considering the slow rate of groundwater movement in Florida soils, coupled with any further reduction in concentration within the contaminated plume from dilution and dispersion, there should be little chance that .wells used for potable wate'r would 53

PAGE 60

be contaminated above the 10 ppb standard if the present restrictions are observed. Even if a larger hydraulic gradient exists because of downward gradients artificially created during well pumping (Rothschild et ale 1982), the effect of a la-fold increase in the hydraulic gradient increases the distance traveled by the TTR plume to 6-80 ft, still well within the 300 ft exclusion zone. Apparently the restriction that aldicarb cannot be applied within 300 ft of any drinking water well seems reasonable .. It should be emphasized that the estimated small distances of plume migration apply only to Indian River County, which in turn are further limited to those few sandy soils used to derive the TTR degradation rates. '54

PAGE 61

SUMMARY AND RECOMMENDATIONS Laboratory degradation experiments were conducted for the purpose of measuring the degradative activity of a small concentration (0.10 mg/L) of aldicarb and its toxic oxidation products (sulfoxide and sulfone) in ground-waters from five wells located near citrus groves in Indian River County. The experiments were designed to evaluate what effect microbiota and aquifer material would have in influencing the rate of detoxific'ation of toxic residues. Attention was given to mimick as closely as possible the actual in situ conditions of the shallow groundwaters during the incubations in the lab so that realistic reaction rates could be projected. predictions of the duration of aldicarb contamination in groundwater could then be made. Once having arrived at the reaction rates, converting to half-lives facilitated comparisons to other degradation rates reported in literature for the saturated and unsaturated zones. Our measured rates for total toxic residue disappearance were also combined with data from the literature which stated the amounts and concentrations of toxic residues entering the water table in another Florida county to yield the time required to degrade the toxic residues to less than 10 ppb. Using field measurements of water table fluctuations and hydraulic conductivities reported for Indian River County, the daily distance of groundw.lter flow could be calculated which, when multiplied by the amount of time necessary to reduce toxic residues to less than 10 ppb, gave the total distance moved by the con-taminated plume. Tw'o companion studies sought to: 1) adequately describe the hydrolysis of aldicarb in various pH-buffered solutions; and 2) evalua tc t.ll..:suspected interferences of non-toxic nitrilcs
PAGE 62

analysis of toxic residues. To best summarize the results of this investigation, four recommendations are given. Following each recommendation the pertinent conclusions are provided upon which the recommendation was based. Recommendation 1 The positive interferences from non-toxic oximes and nitriles in being included with the total toxic residue (TTR) are minor, suggesting that the liquid chromatography step used in the analytical procedure to remove them may be omitted under some experimental circumstances. oximes did not interfere in the analysis of TTR under the conditions used in this investigation; nitriles did interfere, but only when ;>90% of the TTR had disappeared. The percentage error of the initial TTR contributed by non-toxic nitriles rarely exceeded 10%, and was on the average, less than the percentage errors associated with either: i) the Florisil separation method; ii) agreement between duplicate samples; iii) inter-laboratory analyses of split-samples; or iv) recoveries of known standards without Florisil separation. Recommendation 2 The best method f0r determining the degradation rate of toxic pesticides in groundwaters and gaining an understanding of the mechanisms is from labora-tory experiments using groundwater-saturated aquifer material incubated under in situ environmental conditions. However, in cases where obtaining aquifer material is difficult or impossible, incubating only groundwater may be sufficient in producing realistic TTR degradation rates for aldicarb sulfoxide and aldicarb sulfone. This does not apply to aldicarb, where incubating in the absence of aquifer material would cause a large underestimation of the rate of degradation in anaerobic groundwaters. 56

PAGE 63

TTR disappearance in anaerobic groundwaters containing low densities of bacteria and without aquifer material was slow (half-life times = 101-178 days) for aldicarb and fast (half-life times = 9-43 days) for aldicarb sulfoxide and aldicarb sulfone, indicating that chemical hydrolysis in the slightly alkaline groundwaters was responsible for converting the sulfoxide and sulfone into non-toxic derivatives but was not operating to the same extent in aldicarb deactivation. The persistence of aldicarb in neutral to slightly alkaline (pH 8) waters was supported by an hydrolysis experiment in sterile pH-buffered distilled waters. At pH 7, the rate of disappearance from the parent compounq was immeasurable; a half-life of 324 days was measured at pH 8, with rates of TTR disappearance increasing as the pH increased. The secondorder reaction rate (kOH) for base-catalyzed hydrolysis was 1.94 x 103 L mole-l day-l. When aquifer material was added, rates of TTR increased 9-fold to 5.76 x 10-2 day-l for aldicarb in one shallow well, indicating microbes were essential in degrading the parent compound to non-toxic residues. However, aldicarb sulfoxide and aldicarb sulfone degraded only slightly more in the presence of aquifer material (half-life times = 10-26 days) than they did in the absence of aquifer suggesting chemical hydrolysis is more prominent in reducing TTR associated with these S-oxides than any microbially-mediated pathway. Recommendation 3 Restricting the application rate of Temik to 5 Ib a.i./acre and not applying it within 300 ft of a potable water supply well should be continued. Shallow groundwaters of Indian River County apparently possess the capability to rapidly degrade TTR from aldicarb and its S-oxides. Half-life times of 10-26 days stand out in contrast to the 2-3 year half-life projected by Porter 57

PAGE 64

et ale (1984) for TTR disappearance in groundwater on Long Island. The reasons for the faster degradation rates of TTR in Florida are probably the more favorable temperatures and pH. The neutral to slightly alkaline pH of Florida's groundwaters (6.8-7.8) serves to hydrolyze the sulfoxide and sulfone, whereas the pH of Long Island groundwater (4.:2-5.9) inhibits hydrolysis. A combination of slow groundwater movement and fast degradation rates limits tile migration distances of aldicarb-contaminated plumes in Indian River county. Estimates of lateral distances traversed in ele saturated zone of the soils studied in this investigation by a plume of groundwater contaminated with TTR were only 1-17 ft since the TTR was converted to non-toxic residues within one-half year. The 300 ft exclusion zone and application rate of 5 Ib a.i./acre insure adequate protection of the groundwater resources in Indian River County. Recommendation 4 Studies on surface and ground water contamination by aldicarb and its oxidized toxic products should be continued. Although these results point to rapid degradation and slow transport of TTR in groundwaters, care should be exercised in extrapolating this conclusion to other sites and soils in Florida. The danger of generalizing results obtained at one site to other sites has been discussed. Even for the specific soils used in the degradation experiments, the limited number of replicates and field samplings performed in the short study pel.-iod warrant a cautious interpretation of the fast degradation rates. Monitoring the groundwaters of some citrus fields receiving annual aldicarb applications should be performed for a period of several years to provide a long-term data base which includes the extremes in meteorological conditions affecting residue transport and deactiv;1tion. 58

PAGE 65

Future experimental studies should be directed toward evaluating the of salt content on the hydrolysis rate and the chemical or biological reductive conversion of sulfone to sulfoxide and sulfoxide to aldicarb. There is some evidence that both could have an effect on aldicarb. Fukuto et ale (1967) reported that the observed first-order rate constants for the hydrolysis of p-nitro-N-methylcarbarnate increased upon decreasing ti1e ionic strength of the phosphate buffer. He suggested that a lowering of the rate constant was from decreasing activities of hydroxide ion and/or the because of increasing ionic strength, and not from catalysis of the phosphate ions present in the buffer. The environmental implication is that the hydrolysis rate of toxic aldicarb residues may be inhibited in estuarine environments. Reductive conversion of a sulfoxide to its parent compound has been reported for phorate (';valter-Echols and Lichtenstein 1977) in sediments. The prevailing evidence of this occurring in reduced environments for sulfoxide or sulfone is to the contrary. Studies such as this one have shown rapid disappearance ot' 'ITR under anaerobic conditions and no one has published any data implying that this mechanism is operating in the chemistry of sulfoxide or sulfone conversions. Still, there is an abundance of reduced compounds in groundwaters (e.g. sulfides, ferrous iron) whicl1 could serve as a reductant, and with the possibility of marketing sulfone as the active in<:.rredient (instead of aldicarb) in commercial insecticides, more research should be done on evaluating whether reductive conversion could occur. Even if a reductive pathway was not found, the information that would be generated on the conver-sion rates to other toxic and non-toxic compounds for each toxic residue species in an anaerobic aquifer system would be beneficial. Given the shallow distances between land surface and the \"ater table I the geographical closeness of residential and agricultural land uses, 59

PAGE 66

permeability of sandy soils, low adsorption potential, and field studies in Florida showing 4-8% of the applied aldicarb available to enter the groundwater, a continuing scientific effort directed toward monitoring, modeling, and reaching best management practices is prudent. 60

PAGE 67

BIBLIOGMPHY APHA (American Puolic Realto AssociationL 1976. Standard methods for the examination of Water and Wastewater. Washington, D,C, Beckman, H., B.Y. Giang, and J. 1969. Production and detection of derivatives of Temik and its metabolites as residues. J, Agr. Food Chem. 17 ;70-74. Benson, H.T. 1973. Microbiological applications; A laboratory manual in general biology, 2nd ed. W.M,C. Brown Publishing Co., Dubuque, Iowa, Bromilow, R.H., R.J. Baker, M.A.H. Freeman, and K. 1980. The degrad-ation of aldicarb and oxamyl in soil. Pest. Sci, 11;3]1-3]8, Bull, D. L., R.A. Stokes, J.R. Coppedge, and R. L, Ridgway, 19]0, Further studies of the fate of aldicarb in soil. J. Econ. Entomol. 63 :1283. Chesters, G., M.P. Anderson, B. Shaw, J.M, Harkin, N, Meyer, E. Rothschild, and R. Manser. 1982. Aldicarb in groundwater. Water Resources Research Center, University of Wisconsin, Madison. 38 pp. Coppedge, J.R., D. A. Lindquist, D.L. Bull, and H.W. Dorough. 1967, Fate of 2-methyl 2-Cmethylthio) propionaldehyde O-Cmethylcarbamoyl) ox:1:me (Temik) in cotton plants and soil. J. Agr. Food Chern. Coppedge, J.R., D. L. Bull, and R.L. Ridgway. 1977. Movement and persistence of aldicarb in certain soils. of Environ. and roxicol. 5:129-141 Delfino, J. J. (ed). 1984. The fate of industrial organic compounds in drinking water aquifers. Dept. of Environmenta1.Engineering Sciences, University of Florida, 64 p. Faust, S.D., and H.M. Gomaa. 1972. Chemical hydrolysis of some organic poosphorus and carbamate pesticides in aquatic environments. Environmental Letters 3:171-206. 61

PAGE 68

Fukuto, T.R., N.A.H. :Fahmy, and R,L, -Metcalf, 1967, Alkaline hydrolysis, anticholinesterase, and insecticidal properties of some nitro-substituted phenyl car bamates. J. Agric. Food Chern. 15:2]3-281. Galoux, M., J,-C. van Damme, A. Bernes, and J. Potvin. 1979. Gas-liquid chroma .... tographic determination of a1dicarb, aldicarb sulfoxide, and aldicarb sulfone in soils and water using a Hall electrolytic conductivity detector. J. Chromategr. 177;245-253. Hansen, J.L., and M.H. Spiegel, 1983. sulfoxide and aldicarb sulfone. Hydrolysis studies of aldicaro, aldicarfi Environ. Toxicol. Chern. Hornsby, A.G., P.S,C. Rao, W.B. Wheeler, P. Nkedi-Kizza, and R.L. Jones. 1984. Fate of a1dicarb in Florida citrus soils. I. Field and studies. In: D.N. Nielsen (ed), Proceedings of a Conference on Characterization and Monitoring of the Vadose (Unsaturated) Zone, December 8-10, 1983, Las Vegas, Nevada. National Water Well Association, Worthington,' Ohio. lCET (Institute for Comparative and Environmental Toxicology). 1983. A toxicological evaluation of aldicarb and its metabolites in relation to the potential human health impace of aldicarb residues in Long Island ground water. Cornell TIniversity, Ithaca, N.Y, 90 pp. IFAS (Institute of Food and Agricultural Sciences). 1983. Aldicarb research task force report. Gainesville, Florida. 58 pp. Jones, R. L., and R. C. Back. 19_84. .Monitoring aldicarb residues in Florida soil and water. p: nviron. Toxicol. Chern. 3,;9-..,.20-, Jones, R. L., P.S,C. Rao, and A.G, Hornsby. 1984. Fate of aldicarb in Florida citrus soil. 2. Model evaluation. In; D.M. Nielsen (ed), Proceedings of a Conference on Characterization and .Monitoring of the Vadose (Unsaturated) Zone, December 8-10, 1983, Las Vegas, Nevada. National Water Well Association, Worthington, Ohio. 62

PAGE 69

Knaak, J.E., M.J.Ta11ant, and L.J. Sullivan. 1966. The metabolism of 2-methyl-2-(methy1thio)propiona1dehyde O-(methy1carbamoy1)oxime in the rat. J. Agric. Food Chem. 14:573-578. Lemley, A.T., and W.Z. Zhong. 1983. Kinetics'of aqueous base and acid hydrolysis of aldicarb. sulfoxide, and aldicarb sulfone. J, Environ, ;Sci, Health B18:l89.-206. Light, T.S. 1972. Standard solution for redox potentialllleasurements. Anal. Chem. 44:1038-1039. Maitlen, J .C. J L.N. McDonough, and 11. Beroza, 19.68, Determination of residues of 2-Methyl-2-Clnethylthio) propionaldehyde O-Clnethylcarbrunoyl) oxi.me (UC-2ll49, Temik), its sulfoxide, and its sulfone by gas chromatography. J. Agr. :Food Chem. Patrick, W.H., Jr., and I.C. Mahapatra. 1968. Transformations and availability to rice of nitrogen and phosphorus in water-logg-ed' soils. Advances in Agronomy 20:323-359. Perdue, E.N., and N.L. Wolfe. 1983. Prediction of buffer catalysis in field and laboratory studies of pollutant hydrolysis reactions. Environ. Sci. Technol.17;635-642. Porter, K.S., A.T. Lemley, H.B. Hughes, and R.L. Jones, 1984. Developing information on aldicarb levels in Long Island groundwater .. In; Proceedings on the Second International Conference on Groundwater Quality Research, .March 26-29., 1984, Tulsa, Oklahoma. 11.S. Environmental Protection Agency. (in press). Rothschild, E. R., R.J, Manser, and M,P. Anderson. 1'9.82-, Investigation of aldicarh in ground water in selected areas of the Central Sand Plain of Wisconsin. Ground Water 20;437-445. 63

PAGE 70

Sharom, 11.S., J.R.W. MilesJ C.R. Harr"is, and .F.L. McEwin. 1980. Persistence of 12 insecticides in water. Water Res, l4d089-l093. Smelt, J .H" M, Leistra, W,H. Raux, and A, Dekker. 1978a Conversion .rates of aldicarb and its oxidation products in soils. I. Aldicarb su1phone. Pestic. Sci. 9;279-285. 1978b. Conversion rates of -----------------------------------------------------aldicarb and its oxidation products in soils. II. Aldicarb su1phoxide. Pestic. Sci, 9!286-292. 1978c. Conversion rates of -----------------------------------------------------aldicarb and its oxidation products in soils, III. Aldicarb. Pestic, Sci. 9:293-300, Trehy, 11.L., R.A. Yost, and J.J. McCreary. 1984. Determination of aldicaro, aldicarb oxime and aldicarb nitrile in water by GC/MS. Anal Chern. 55 (in press) Uhlinger, D. J., and D.C. White. 1983. Relationship between the physiological status and the formation of extracellular polysaccharide glycoca1yx in Pseudomonas atlantica, Appl. Environ. 45:64-70. Union Carbide Corporation, Agricultural Products Company, Inc. 1980, A method for the determination of aldicarb .. residues in water. Unpublished. 1983. Temik -----------------------------------------------------------------aldicarb pesticide: A scientific assessment. Research Triangle Park, N. C. ]1 pp. Walter-Echols, G., and E.P. Lichtenstein. 1977. Microbial reduction of phorate sulfoxide to phorate in a soil-lake mud-water microcosm, J, Econ, Entoml 70:505-509. 64

PAGE 71

White, D,C., J.S. Nickels, J.H, Parker, R.H. Findlay, N.J, Gehron, G.A, Smith, and R. F. Martz'1984. Biochemical measures of the biomass connnunity and_metabolic the ground water microbiota. Chapt. XX. In: C,H. Ward (ed) Proc. First Intern. Conf. Ground Water Quality Research. Wiley Interscience, N.Y Zaki, N.H.., D. 'Moran, and D. Harris. Pesticides:in groundwater: The a1dicaro story in County, N,Y. kD.. J, Public Health 72:1391-1395 65

PAGE 72

APPENDIX I SUMMARY OF COMPARISONS BETWEEN LABORATORIES IN THE SPLIT SAMPLING PROGRAM 66

PAGE 73

IFAS FDER a Pesticide Research Labb Sample FIT Tallahassee Gainesville Ryal well-B Aldicarb amended Non-sterile T = 0 days 101 ppb 120 ppb Aldicarb hydrolysis pH = 10 Time = 5 days 49 ppb 60 ppb Aldicarb hydrolysis-A pH = 10 Time = 15 days 18 ppb 16 ppb Luther well-B Aldicarb amended Non-sterile Time = 90 days 5l ppb 36 ppb a TTR determination using a Hewlitt-Packard model 5700 GC with a nitrogen-phosphorus detector and a 1% SPIOOO on Carbopack B 60/80 mesh column; injector, column and detector temperatures were 250, 200, and 300oC, respectively. b TTR determination using a Perkin-Elmer Series 4 high pressure liquid chroma to-graph with a Gilson l21 fluorometer (excitation: 305-395 nm; emission: 430-470 nm) and a Zorbax C-8 (15 em x .. 4.6 mm) stationary phase. The mobile phase consisted of a 10 min. linear gradient with an initial composition of 4% CH3 CN, 16% CH30H, 80% H 2 0 gnd ending with a final composition of 14% CH3 CN, 56% CH30H, 30% H 20. A post-column derivitization step included 0.5 mL/min of 0.05 M NaOH at 950C followed by 0.5 mL/min of OPA at ambient temperature. 67

PAGE 74

APPENDIX II EFFECT OF BUFFER CATALYSIS ON THE HYDROLYSIS ,. OF ALDICARB IN STERILE pH-BUFFERED SOLUTIONS 68

PAGE 75

En'ECT OF BUFFER CATALYSIS ON THE HYD;aOLYS.rS OF ALDICARB IN -STERILE, pH.,BUFFERED SOLlITIONS In laboratory studies of hydrolysis of the aldicarb, conditions of con-stant pH are desired to simplify kinetic interpretations. Pseudo-first-order kinetics are usually observed only at a constant pH for acid-base-catalyzed hydrolysis of a pollutant, P: dJpJ = -k Ip] dt obsed (A-l) where K b d = k O[H20]+ oi=IH30+] + +:? (k. CHB.] + k [B.]) (A-2) o se -112 3 1. -1fB1. 1. -'Bi 1. and RBi and Bi are the ith Bronsted acid-base pair in solution. Eq. A-2 states that hydrolysis reaction rates are a function not only of pH (the first three terms) but also of the nature and concentration of the acid-base system used to buffer the pH, called buffer catalysis. The first three terms of Eq. A-2 can contribute to k b d in all aqueous solytions, their contribution o se being predictable if the second-order rate constants (kR _O 0+' and kOH-) 3 and pH are known. At constant pH, the combined contributions of H20, H30+, and OH to.k b d are constant and can be represented by a pseudo-first-order o se rate constant, k : w (A-21 Thus, t!le observed pseudo-first-order rate constant (.It b d) equals the o se pseudo-first-order rate constant for catalysis by solvent species (k ) plus w the buffer catalysis contribution (the last two terms of Eq-. A-2). Perdue and Wolfe (1983) developed a theoretic basis for assessing a maximum contribution of buffer cat.alysis for hydrolysis reactions in aqueous systems. They derived a buffer catalysis factor (BCF) for the buffers commonly employed to Duffer solutions to a constant pH. The potential significance of 69

PAGE 76

buffer ctalysis in aqueous solutions can be expressed; k b d/k :::; 1 + CBCBCFl o se w where C B is tne concentration of the buffer catalyst. CA-4) When k b d/k then o se w the contribution from buffer catalysis to the rate of hydrolysis is negligible. Perdue and Wolfe suggested that when k b d/t{ >1.10 (i.e." buffer catalysis o se w contribution that equals or exceeds 10% of the combined kinetic contribution of H 20, H 3ct, and OH-) then a 10% or greater increase in kobsed results from buffer catalysis, and should be viewed as having potential significance. Substituting the molar concentrations and published BCFs for the buffers used in the aldicarb hydrolysis investigation yQelded k b d/k ratios ranging o se w from 1,23 to 2.67 CTable A,;,.l). Even though these represent a potential for significant buffer catalysis of 23 to 167% in kb d' the pH buffers which o se exhibited the highest ratios (i,e., pH = 6.02, 7.06, and 7.96) corresponded to the slowest kobsed, while those.buffers which. had faster k.obsed (i,e!, pH = 3.95, 8.85, 9,85) were associated with lower k b d/k ratios, indicating o se w that the potential for buffer catalYSiS was not realized. It is important to recognize that the relative contribution of buffer catalysis to k b d is o se a function df pH, and substrate, but that the preceding calculations predict the maximum contribution of buffer catalysis for a particular catalyst and pH only (ignoring the type of substrate), Buffer catalysis would be somewhat less important for any real substrate, such as aldicarb, and our data indicate a negligible contribution from the buffers at the strength used in our hydrolysis experiment::. Fukuto et a1. (1967) f-ound varying concentrations of phosphate buffer anions did not participate in the hydrolytic reaction of p-nitrophenyl N-methycarbamate, which supports our conclusion that.buffer catalysis is probably unimportant :in hydrolyzing 70

PAGE 77

Table A-l. Potential for Maximum Contribution of Buffer Catalysis to the Observed Pseudo-first-order Rate (kobsed) in the Hydrolysis of Aldicarb Using Various pH Buffers in Distilled water. Factora kobsed/kw Buffer Concentration Buffer Catalysis pH (C B ) (BCF) (=1 + C B (BCF) ) 3.95 1.67 x 10-2 M phthalate 30 1.50 6.02 1.67 x 10-2 M phthalate 50 1. 84 7.06 1.67 x 10-2 M phosphate 100 2.67 7.96 1.67 x 10-2 M phosphate 80 2.34 8.85 4.10 x 10-3 M borate 55 1. 23 9.85 1.67 x 10-2 M carbonate 25 1.42 aperdue and Wolfe (1983) 71

PAGE 78

APPENDIX III RAW DATA 72

PAGE 79

Table A-2 Hydrolysis for A1dicarb in Sterile pH-buffered Distilled Water at 200 + 20 C. A1dicarb Remaining, ppb .E!! Day 0 Day 5 Day 15 Day 37 Day 89 3.95-A 111 121 112 120 72 3.95-B 117 118 128 106 76 6.02-A 133 133 122 127 120 6.02-B 139 150 130 124 7.06-A 108 121 128 130 132 7.06-B 117 130 137 127 124 7.96-A l30 l33 127 100 7.96-B 127 150 134 120 8. 85-A 88 84 72 54 32 8.85-B 120 95 78 57 32 9.85-A 108 59 18 <3 9. 85-B 108 62 18 <3 73

PAGE 80

Table A-3 Well Ryall -A RyalL -B Ryall -A Ryall -B Sexton -A Sexton -B Sexton -A Sexton -B Luther -A Luther -B Luther -A Luther -B Raw Data For Aldicarb Degradation in Duplicate Sterile .and Non-sterile Groundwaters at 24 + 2 oe. Sterile Yes No No Yes Yes No No Yes Yes No No Day 0 94 99 99 101 83 92 80 99 97 97 97 95 Day 5 92 90 95 107 85 118 89 104 98 118 118 124 74 TTR Remaining, Day 11 84 81 91 98 78 78 88 84 95 91 95 98 ppb Day 34 73 70 57 80 80 70 73 73 77 53 97 87 Day 90 54 56 58 54 51 47 51 49 56 49 66 51

PAGE 81

Table A-4. Raw Data For Aldicarb Sulfoxide and Aldicarb Sulfone Degradation in Duplicate Non-Sterile Groundwaters at 24 .i 2 oe. TTR Remaining, ppb Aldicarb-Well Day 0 Day 5 Day 20 Day 42 Ryall -A Sulfoxide 97 85 54* Ryall -B Sulfoxide ll05 92 4* Ryall -A Sulfone 81 68 43* <3* Ryall -B Sulfone 97 79 28* <3* Sexton -A Sulfoxide 97 83 34* 4* Sexton -B Sulfoxide 98 84 35* 4* Sexton -A Sulfone 80 63 12* <3* Sexton -B Sulfone 92 81 37* <3* Luther -A Sulfoxide 95 92 43* 9* Luther -B Sulfoxide 92 82 40* 13* Luther -A Sulfone 90 65 20* <3* Luther -B Sulfone 94 60 30* <3* Nitrile removed from sample. 75

PAGE 82

Table A-5. Raw Data For Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone Degradation in Duplicate'Non-sterile Groundwaters with and Wi thout Subsoil From the BBC and Lindsey Wabasso Groves. TTR Remaining, ppb Well Substrate Subsoil Day a Day 23 BBC-A Aldicarb Absent ,89 77 BBC-B Aldicarb Absent 100 87 BBC-A Aldicarb Present 73 20 BBC-B Aldicarb Present 93 24 BBC-A Aldicarb Sulfoxide Absent 82 54 BBC-B Aldicarb Sulfoxide Absent 104 59 BBC-A Aldicarb Sulfoxide Present 73 37 BBC-B Aldicarb Sulfoxide Present 72 42 BBC-A Aldicarb Sulfone Absent 67 38 BBC-B Aldicarb Sulfone Absent 88 46 BBC-A Aldicarb Sulfone Present 80 37 BBC-B Aldicarb Sulfone Present 90 34 Day 0 Day 30 Lind. Wab.-A Aldicarb Absent 100 89 Lind. ,Wab.-B Aldicarb Absent 100 89 Lind. Wab.-A Aldicarb Present 62 <3 Lind. Wab.-B Aldicarb Present 65 <3 Day a Day 25 Lind. Wab. -A Aldicarb Sulfoxide Absent 79 54* Lind. Wab.-B Aldicarb Sulfoxide Absent 80 53* Lind. Wab. -A Aldicarb Sulfoxide Present 51 12* Lind. Wab. -B Aldicarb Sulfoxide Present 62 10* Lind. Wab.-A Aldicarb Sulfone Absent 90 30* Lind. Wab. -B Aldicarb Sulfone Absent 90 29* Lind. Wab.-A Aldicarb Sulfone Present 64 10* Lind. Wab.-B Aldicarb Sulfone Present 78 16* *Nitrile removed from sample. 76

PAGE 83

Table A-6. Bacterial Counts on Groundwater Samples. Pour Plate ( cells/mL for water and Sampling Date Well cells/g (dry wt) for sediments) 9 December 1983 Ryall start: 6; 5; 4 x=5 finish: 10; 12 x=ll Sexton start: 2 I 3 ; 2 x=2 finish: 7 I 8 x=7 Luther start: 58; 51; 55 x=55 finish: 42; 33 x=37 26 March 1984 Ryall start: 3; 5 x=4 finish: 6; 8 x=7 sexton start: 1; 4 x=3 finish: 6; 7 x=6 Luther start: 75; 79 x=77 finish: 63; 64 x=63 30 January 1984 Wab. (sed. ) start: 1.2 x 105 ; 1.0 x 105 x=l.l x .10 5 finish: 1.4 x 104 ; 1.6 x 105 x=1.5 x 104 Lind. (water) start: 800; 700 x=750 finish: 228; 219 x=223 23 April 1984 Lind. Wab. (sed.) start: 6.3 x 104 ; 5.7 x 104 x=6.0 x 104 finish: 8.0 x 103 ; 8.0 x 103 x=8.0 x 103 Lind. Wab. (water) start: 1500i 1200 x=1350 finish: 800; 1100 x=950 104 ; 104 104 22 February 1984 BBC (sed. ) start: 3.9 x 5.4 x x=4.6 x finish: 9.0 x 103 ; 9.0 x 103 x=9.0 x 103 BBC (sater) start: 31,000; 42,000 x=36,000 finish: 1800 ; 2300 x=2,050 77