EPA-600/3-76-067
September 1976 cJ Ecological Research Series
CHEMICAL AND PHOTOCHEMICAL
TRANSFORMATION OF
SELECTED PESTICIDES IN AQUATIC SYSTEMS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Athens, Georgia 30601
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-76-067
September 1976
CHEMICAL AND PHOTOCHEMICAL TRANSFORMATION OF
SELECTED PESTICIDES IN AQUATIC SYSTEMS
by
N. Lee Wolfe, Richard G. Zepp, George L. Baughman,
Robert C. Fincher, and John A. Gordon
Environmental Processes Branch
Environmental Research Laboratory
Athens, Georgia 30601
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
ATHENS, GEORGIA 30601
P.- .a-"- -•
-------
DISCLAIMER
This report has been reviewed by the Environmental
Research Laboratory, U. S. Environmental Protection Agency,
and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or
recommendation for use.
11
-------
CONTENTS
Page
List of Tables iv
List of Figures V1
Acknowledgments x
I Introduction 1
II Summary 3
III Conclusions 4
IV Recommendations 6
V Background 7
VI Materials and Methods 34
VII Photochemical Screening Studies 53
VIII Results and Discussion: Malathion 55
IX Results and Discussion: 2,4-D Esters 77
X Results and Discussion: Methoxychlor 87
XI Results and Discussion: Captan 102
XII Results and Discussion: Carbaryl 119
XIII Results and Discussion: Atrazine 125
XIV Results and Discussion: Diazinon 129
XV Results and Discussion: Parathion 133
XVI Results and Discussion: Toxaphene 140
111
-------
LIST OF TABLES
Number Page
1 Z Values for Latitude 40°N. 16
A
2 Z Values for the Summer Season 17
A
3 Triplet State Energies of Several Pesticides 24
4 Source and Purification Techniques for
Pesticides 35
5 Relative Direct Photolysis Rates of Selected
Pesticides in Distilled Water-Screening Study
Results 54
6 Carbon-13 Chemical Shifts for Malathion and
Related Compounds 60
7 Malathion Acid-Catalyzed Degradation Kinetic
Data 62
8 Temperature Effect on the Malathion Alkaline
Degradation Rate Constant 64
9 Rate Constants for Malathion Elimination and
Carboxyl Ester Hydrolysis Reactions 68
10 Alkaline Degradation Rate Constants for
Malathion Monoacids and Malathion Diacid in
water at 27° 70
11 Kinetic Data for the Acid and Base Hydrolysis
of Methyl and n-Butoxyethyl Esters of 2,U-D 78
12 Kinetic Data for Hydrolysis of 2,U-D Esters in
Water at 28° 80
13 Disappearance Quantum Yields for Direct Photo-
lysis of 2,4-D Esters at 313 nm 83
14 Comparison of Photolysis Data for 2,4-D
Butoxyethyl Ester 85
xv
-------
15 Kinetic Data for Dehydrochlorination of
Methoxychlor and DDT 88
16 Rate Constants for Methoxychlor Degradation in
Water 88
17 Half-lives for Methoxychlor Degradation with
Varying Amounts of Hydrogen Peroxide Added
(65°C) 90
18 Quantum Yields for Direct Photolysis of
Methoxychlor 93
19 Kinetic Parameters for the Direct Photolysis
of DDT and Methoxychlor in the Central United
States 95
20 Half-lives for Photodecomposition of Methoxy-
chlor (40 ppb) under Sunlight in Various River
Waters 99
21 Pseudo-first-order Rate Constants for Captan
Hydrolysis at Several pH's and Temperatures 106
22 Singlet Oxygen Reactivities of Captan, 4-
Cyclohexene-1,2-Dicarboximide, and Cyclo-
hexene. 114
23 Kinetic Parameters for hydrolysis of Carbaryl
and Several Other Carbamate Pesticides 119
24 Hydrolysis Half-lives for Carbaryl at pH
Values Usually Found in the Aquatic Environ-
ment 120
25 Quantum Yields for Photolysis (313 nm) of
Carbaryl in Water at 25°C 123
26 Calculated Direct Photolysis Half-lives of
Carbaryl at Different Seasons and Latitudes in
the Northern Hemisphere 123
27 Half-lives and Rate Constants for Hydrolysis
of Atrazine 126
28 Hydrolysis Half-lives and Rate Constants for
Diazinon and Diazoxon 130
29 Specific Sunlight Absorption Rates of Para-
thion and Other Selected Pesticides during
Midsummer and Midday at Latitude 40°N 138
v
-------
LIST OF FIGURES
Number Page
1 Ten representative materials selected for
study by the U.S. EPA 2
2 Typical pH-rate profiles for the hydrolysis of
pesticides containing ester moieties 8
3 Midday solar irradiance for summer, latitude
40°N 18
4 Dependence of short-wavelength solar uv irrad-
iance upon season and latitude 19
5 Dependence of long-wavelength solar uv irradi-
ance upon season and latitude 20
6 Penetration of ultraviolet light into two
natural waters and pure waters 21
7 Mechanisms for the riboflavin-sensitized oxi-
dation of 2,4-dichlorophenol 25
8 Mechanism for generation of singlet oxygen in
the aquatic environment 27
9 Photosensitized oxidation of cis-resmethrin 27
10 Kinetics equation for photosensitized oxida-
tion involving singlet oxygen 29
11 Mass spectra of methyl 2-chloro-4-hydroxy-
phenoxyacetate (A) and lactone derived from
thermal decomposition of esters of 4-chloro-2- 36-
hydroxyphenoxyacetic acid (B) . 37
12 Determination of cell pathlength for quantum
yield studies 50
13 Potential chemical pathways for malathion
degradation 56
14 Scheme for synthesis of malathion monoacids 57
VI
-------
15 Liquid chromatograms showing the relative
amounts of malathion monoacids formed in syn-
thesis 58
16 Scheme for synthesis of malathion diacid and
monoacids 59
17 Synthesis of malaoxon monoacid 59
18 Carbon-13 spectra of malathion and malathion
3-monoacid 61
19 Carbon-13 spectra of malaoxon and malaoxon 8-
monoacid 62
20 Alkaline degradation of malathion 65
21 Malathion disappearance and product formation
at 27°C 66
22 Malathion disappearance and product formation
at 0°C 67
23 Temperature effect on malathion degradation at
several pH values 68
24 Alkaline degradation of malathion monoacid 71
25 Time dependence of malathion disappearance and
product formation at 27°C 72
26 Pathways of alkaline degradation for malathion
and malathion acid derivatives at 27° 74
27 pH-rate profile for 2,4-D butoxyethyl ester at
67° in water 79
28 Hydrolysis of 2,U-D butoxyethyl ester in water
from the Withlacoochee River 81
29 Photoreactions of 2,4-D esters 82
30 Computed dependence of 2,4-D butoxyethyl ester
photolysis rate upon time of day in the South-
ern United States 84
31 Products from direct photolysis of methoxy-
chlor in hydrocarbon solvents 90
32 Photoproducts of methoxychlor in pure water 91
vxi
-------
33 Electronic absorption spectra of methoxychlor 94
and DDT in hexane
34 Computed midday half-lives for direct photo-
lysis of methoxychlor in water during summer 96
35 Calculated effects of ozone reduction upon
photolysis rates of DDT and methoxychlor 98
36 Chemical structures of captan, folpet, and
captafol 102
37 Standard curve for captan response to electron
capture detector 103
38 Captan concentration and pH vs. time in non-
buffered water at 28°C 104
39 Pseudo-first-order plots for the hydrolysis of
captan, folpet, and captafol 105
40 Plot of log k vs. pH for captan hydrolysis at
28°C in buffered aqueous solution 107
41 pH-half-life profile for captan hydrolysis in
water at 28°C 109
42 Degradation of captan in water from the
Tombigbee River 110
43 Major products for the hydrolysis of captan 111
44 Mechanism for hydrolysis of captan that
involves nucleophilic displacement of chloride 113
45 Mechanism for hydrolysis of captan involving
nucleophilic substitution at the sulfur atom 113
46 Postulated products for light-initiated auto-
oxidation of captan and 4-cyclohexene-1,2-di-
carboximide 114
47 Concentration dependence of the quantum yield
for reaction of captan with singlet oxygen in
acetonitrile 115
48 Effect of 1,4-diazabicyclooctane upon the
photosensitized oxygenation of captan 116
vxn
-------
49 Products from the reaction of captan and 4-
cyclohexene-1,2-dicarboximide with singlet
oxygen 117
50 Chemical structures of carbaryl and other car-
bamate pesticides 120
51 Photoreactions of substituted phenyl N-methyl
carbamates 122
52 Hydrolysis of atrazine 127
53 Photoreaction of atrazine 127
54 Hydrolysis products of diazinon 129
55 Postulated mechanism for photooxidation of
diazinon 131
56 Structures of parathion and some products de-
rived from its chemical transformations 133
57 Hydrolysis of parathion 135
58 Reported photoalteration products of parathion 137
ix
-------
ACKNOWLEDGMENTS
Our thanks go to Dr. Walter M. Sanders III of the Environ-
mental Research Laboratory, U.S. EPA, Athens, Georgia, for his
support and assistance. We also express appreciation to
William Loy and Donald Brown, Surveillance and Analysis
Division, Region IV, U.S. EPA, Athens, Georgia, and to the
Staff of the Analytical Chemistry Branch, Environmental
Research Laboratory, Athens, Georgia, for their help and use
of their instrumentation. We thank Drs. Richard Cox and
Richard Hautala, Department of Chemistry, University of
Georgia, Athens, Georgia, for their assistance in obtaining
spectral data, and Mr. Alfred Thurston, Environmental Research
Laboratory, Athens, Georgia, for his help with liquid
chromatographic analysis. Finally, special thanks go to David
M. Cline, Environmental Research Laboratory, U.S. EPA, Athens,
Georgia, for his assistance in writing a computer program for
computation of photolysis rates of pollutants and to Carlyn 3.
Haley for an outstanding job in typing this report.
x
-------
SECTION I
INTRODUCTION
Lack of data on the fate and impact of pesticides in
freshwater systems prompted Environmental Protection Agency
researchers in 1971 to select 10 representative materials
(Figure 1) for detailed study. One of the needs highlighted
at that time was for more reliable data on the rates, routes,
and products of transformation. Subsequently, Paris and
Lewis1 comprehensively reviewed the literature on microbial,
chemical, and photochemical transformations of these materials
and have since completed microbial studies with them.
This report describes laboratory research on the chemical
and photochemical behavior of the selected chemicals under
conditions expected in aquatic ecosystems. Polychlorinated
biphenyls (PCB's) were not included in these studies. The
results are offered not as a statement of what is important in
the environment but rather as a first approximation of what is
LIKELY TO BE important and should be considered. In this
sense, the results are complementary to other EPA studies
concerning toxicology, microbiology, or other aspects of the
environmental transport and transformation of these materials.
A major objective has been to obtain rate data that can be
used as a yardstick for assessing the relative importance of
various chemical and photochemical transformations of the
selected pesticides. Products and mechanisms of a reaction
were studied in detail when it appeared that the reaction
might be significant under environmental conditions. These
studies were pursued for several reasons. First, products
must be known because of their potential environmental impact,
in order to define the reaction mechanism, and frequently in
order to measure the rate. Second, information about reaction
mechanisms is required if reasonable predictions are to be
made concerning the effect of water quality parameters on
reaction rates or products, likelihood and nature of
catalysis, effect of temperature, etc. Knowledge of the
reaction mechanisms is also required to predict the probable
rate and products of transformation of related compounds.
A second objective of the project was to synthesize the
transformation products and make them available to others for
analytical, toxicological, and other studies.
-------
REFERENCES
1 Paris, D. F., and D. L. Lewis. Residue Reviews. 45:95,
1973.
(CH30)2P-S-CHCOOCH2CH3
CH2COOCH2CH3
MALATHION
CH30
CHCCU
METHOXYCHLOR
OCONHCH3
CARBARYL
(C2H50)2P-0
^
n
CH3
DIAZINON
TOXAPHENE
A mixture of polychlorobicyclic terpenes
with chlorinated camphene predominating.
(Structural formula is representative.)
Cl
Cl
BUTOXYETHYL ESTER OF 2,4-D
NHCH(CH3)2
ATRAZINE
)-P(OC2H5)2
PAR ATM ION
/r\jr\
POLYCHLORINATED BIPHENYLS
A mixture of chlorinated biphenyls;
x= 1, 2 10
Figure 1. Ten representative materials selected
for study by the U.S. EPA
-------
SECTION II
SUMMARY
This report presents the results of laboratory studies to
quantitatively predict chemical and photochemical
transformation rates and products of pesticides in water. It
includes a general discussion of relevant transformation
processes and associated kinetic expressions. The processes
treated in most detail are hydrolysis, direct photolysis, and
reaction with singlet oxygen. Implications of other processes
such as oxidation and sensitized photolysis are also
discussed.
Results of detailed studies are included for the
pesticides malathion, carbaryl, methoxychlor, captan, and 2,4-
D esters. The measured rate constants and half-lives indicate
that chemical and/or photochemical processes of these
compounds are likely to be important in the aquatic
environment.
Less extensive data are presented for the pesticides
atrazine, diazinon, parathion, and toxaphene, along with a
discussion of available literature data.
-------
SECTION III
CONCLUSIONS
Chemical degradation of malathion is likely to be the
major pathway for its transformation in basic natural
waters (pH greater than 7) . The products are a mixture of
malathion acids, fumaric acid and its ethyl esters, and
0,0-diethylphosphorodithioic acid. Photolysis of
malathion by sunlight is too slow to compete with chemical
degradation in pure water.
Hydrolysis of 2,4-D esters to 2,4-D is very rapid in basic
water but is slow in acidic water. Hydrolysis rates of
2,4-D esters depend greatly upon ester structure; esters
possessing ether linkages near the carboxyl group
generally hydrolyze more rapidly than hydrocarbon chain
esters. Photolysis of 2,4-D esters in pure water is a
slow process (half-life for the butoxyethyl ester is about
14 days), but it can be important in acidic natural
waters. The major photoproducts are chlorohydroxyphenoxy-
acetic acid esters and 2,4-dichlorophenol.
Hydrolysis of methoxychlor is very slow and is pH-
independent under reaction conditions that are usually
found in aquatic environments; the half-life is greater
than 200 days at 25°C. Photolysis of methoxychlor in pure
water is also very slow; the half-life is greater than 300
hours of summer sunlight. However, sensitized photolysis
may be rapid in natural waters.
Captan hydrolyzes very rapidly in water with a maximum
half-life of one-half day. The hydrolysis products are 4-
cyclohexene-1,2-dicarboximide, sulfur, carbon dioxide, arid
hydrochloric acid. Photolysis in pure water is too slow
to compete with hydrolysis, but in a river-water sample,
sensitized photooxidation of captan was very rapid. One
of the products of photooxidation by singlet oxygen was
found to be N-[ (trichloromethyl)thio]-3-cyclohexene-4-
hydroperoxy-1,2-dicarboximide. The imide derived from
captan hydrolysis is also rapidly photooxidized by singlet
oxygen. Abiotic transformations of captan are likely to
be the predominant processes in aquatic environments.
Hydrolysis of carbaryl is fast in basic waters but slow in
acidic waters. Hydrolysis half-lives range from 1.3 days
at pH 8 to 4.4 months at pH 6. Products of hydrolysis are
1-naphthol, methylamine, and carbon dioxide. Direct
-------
photolysis of carbaryl is pH-independent in the pH 5 to 7
range and may be important in acidic water; the minimum
photolysis half-life is about four days in the central
United States (latitude
6 These studies indicate that atrazine is very stable to
hydrolysis and direct photolysis in aquatic systems.
Literature data indicate that half-lives at pH 3 and pH 11
(25°C) are 66 days and 81 days, respectively. Photo-
chemical screening studies using uv light of wavelengths >
280 nm indicated that the photolysis rate of atrazine is
at least ten times slower than that of carbaryl.
7 Hydrolysis of diazinon is very slow at pH values normally
found in lakes and rivers (minimum half-life of one month
at 20°C). Hydrolysis is more rapid in acidic water (pH 5-
7) than in basic water (pH 7-9). Photolysis in pure water
is also a slow process; under high-intensity uv light (>
280 nm) the photolysis rate was about eight times slower
than that of carbaryl.
8 Data in the literature indicate that parathion hydrolyzes
slowly in pure water at pH values and temperature found in
the aquatic environment. The pH-independent hydrolysis
half-life in the pH 5 to 9 range is 243 days (20°C).
Photolysis in pure water was about ten times slower than
carbaryl when Pyrex-filtered (> 280 nm) light from a
mercury lamp was employed as light source.
9 The hydrolysis and direct photolysis of toxaphene is
extremely slow in water, even at temperatures and uv light
intensities that are much greater than those in the
environment.
10 Malathion, 2,4-D esters, methoxychlor, carbaryl, atrazine,
diazinon, parathion, and toxaphene are not readily
oxidized by singlet oxygen.
11 Oxidation of malathion, 2,4-D esters, methoxychlor, and
toxaphene by molecular oxygen in water is too slow to have
environmental significance.
-------
SECTION IV
RECOMMENDATIONS
The indirect (or "sensitized") photolysis of pollutants in
natural waters should be studied in more detail. These
studies should include (1) elucidation of the products
formed by sensitized photolysis in natural waters; (2)
characterization of the materials in natural waters that
are responsible for sensitized photolysis; (3)
determination of the rates and mechanisms of sensitized
photolysis.
Studies should be carried out to delineate sediment-
associated hydrolysis processes. These should include a
quantitative evaluation of its contribution and
characterization of the products.
The role of oxidative processes in natural waters should
be determined. Oxidizing agents should be identified, and
rates of oxidation with known pollutants should be
measured and products characterized. These studies should
include an investigation of light-initiated autooxidatiori
of pollutants on plant and soil surfaces.
-------
SECTION V
BACKGROUND
CHEMICAL
Hydrolysis
No pesticide review article is considered complete without
a section on the chemical degradation of pesticides. These
sections generally contain fragmentary information, often
contradictory, describing the persistence of a pesticide under
given reaction conditions. For example, a known amount of
pesticide might be added to water at an elevated temperature
and its persistence monitored. While such studies have some
limited value, this type of information will not allow
extrapolation of the observed chemical behavior to other
reaction conditions. Also in many cases the products of
degradation and the persistence of these products, even if
identified, are not known.
There are several common pathways of chemical degradation
under environmental conditions which are apparent from an
examination of the pesticide literature. These are
nucleophilic substitution, elimination, oxidation, reduction,
and autooxidation reactions. The term "hydrolysis" is loosely
used to describe reactions in which a bond of the molecule is
cleaved and a new bond is formed with the oxygen atom of a
water molecule.
Hydrolytic reactions are generally mediated by acid and/or
base. The extent to which these catalytic effects come into
play is dependent on the type of reaction and the chemical
structure of the compound.
Temperature effects vary for different reaction pathways
and the magnitude of these effects vary with chemical
structure. In fact, temperature may affect not only the rate
of the reaction, but the products as well (see Section VIII).
pH-rate profiles (Figure 2) are useful in understanding
and evaluating contributions of acid, alkaline, and neutral
hydrolysis. For purposes of example, pH-rate profiles of car-
boxyl ester hydrolysis are discussed. However, similar pH-
rate profiles are applicable to phosphate esters, amides,
carbamates, and other compounds labile to hydrolysis.
-------
0
o>
t/)
_Q
O
O
O
-4
3
7
11
PH
Figure 2. Typical pH-rate profile for the hydrolysis
of pesticides containing ester moieties:
A, less reactive; B, ester of intermediate
reactivity; C, highly reactive ester.
The hydrolytic behavior of carboxylic esters in the pH
range U to 8 varies with the hydrolytic reactivity of the
ester (for a more complete discussion, see reference 1). The
overall reaction for alkaline hydrolysis is
O
II
R-C-OR'
H O
-> R-C-O- + HOR1
(1)
HO-
8
-------
How readily the ester is converted to the acid is dependent on
the structure of the substituents R and R1 and the pH of the
water.
Acid hydrolysis generally requires more stringent reaction
conditions. The overall reaction is
O O
|| H20 ||
R-C-OR1 » R-C-OH + HOR« (2)
H+
This reaction occurs by a pathway in which acid is not con-
sumed and is termed an acid catalyzed reaction.
Esters can also undergo hydrolysis by reaction with water
in the absence of acid or base:
O 0
II II
R-C-OR + H O » R-C-OH + HOR (3)
This reaction is most significant in the cases of more re-
active esters.
The overall rate expression for the disappearance of the
ester is given by
-££- = k [RCOOR'][H+] + ku FRCOOR'] + k [RCOOR"][OH~] (4)
dt H+ H2° OH~
where kH+, kH2o, and k0H~ are tne rate constants for acid,
neutral, and alkaline degradation, respectively1.
Because most natural waters have considerable buffering
capacity and pollutant concentrations are low, the effective
concentration of acid or base does not change during the
reaction. Pseudo-first-order kinetics are observed and the
observed first-order rate constant is given by
k =k.[H] + k +k [-OH] (5)
obsd H+L J H20 OH-L J v '
and the half-life expression2 is
0-693
2 OH
t = _ -
35 k . [H+] + k + k JOH~]
+ H°
-------
Figure 2 is a plot of pH vs. log kobsdfor the reaction of
three different esters based on literature data. For ester A
the neutral hydrolysis contribution is negligible and the
profile consists of two straight lines with a slope of minus
one (-1) for acid catalysis and plus one (+1) for alkaline
reaction which intersect at the rate minimum. At this minimum
the rate of acid and base hydrolysis are equal.
More reactive esters give a pH-rate profile illustrated by
curve B. In this case, the minimum is indicated by a line
with a slope of 0 which is attributed to hydrolysis by water.,
For very reactive esters, no minimum is observed because
the reaction with water makes the reaction pH independent at
acid pH's (curve C) .
Oxidation-Reduction
The definition of oxidation-reduction is not adequately
explained in terms of gain or loss of oxygen. It is customary
to define these reactions in terms of electron loss
(oxidation), electron gain (reduction)3, and half-cell
reactions. However, there may be little relation between what
is shown in the half-cell reactions and what actually occurs
when two reagents are mixed.
An understanding of the oxidation-reduction reaction
requires knowledge of whether electrons or atoms are
transferred, how many electrons are involved, any inter-
mediates formed, and rate constants for the reactions in-
volved.
Oxidation-reduction reactions have received environmental
interest, but it has been from a thermodynamic point of view.
Work is needed to evaluate oxidation-reduction reactions of
pollutants under conditions common to aquatic ecosystems and
to determine their contribution as a degradative pathway.
Another oxidative pathway which is potentially important
for pollutant degradation is autooxidation. Autooxidation is
defined by Pryor* as the slow oxidation of an organic compound
by oxygen. Examples of autooxidation include air drying of
paints and varnishes, and deterioration of rubbers and
plastics. The general reaction scheme for autooxidation is
given below. There are three parts to the reaction:
initiation (steps 7-9), propagation (steps 10 and 11), and
termination (steps 12, 13, and 14).
10
-------
light or heat
A-B » A* + B* •>. (7)
A* + RH » R* + AH > Initiation (8)
\ Reaction
RH + 0_ > R" + HOO J (9)
k1
R* Q —* ROO' -v (10)
( Propagation
( Reaction
J (11)
kt
2ROO* » Non-reactive products N (12)
ROO* + R* —* Non-reactive products \ Termination (13)
Reaction
2R" ) Non-reactive products -/ (14)
Autooxidation involves free radicals and the reaction may
be initiated (steps 7-8) by peroxides, metal salts, or any
other free radical source A-B to give a pollutant-derived free
radical, R* . In the absence of a free radical source, the
reactions are initiated by slow reaction of oxygen with
pollutants (equation 9). The slow build-up of free radical
initiators, A-B, can occur by such pathways as photolysis or
thermolysis.
The propagation steps (10 and 11) involve reaction of the
pollutant-derived free radical with molecular oxygen (step 10)
which gives a peroxy radical. The peroxy radical reacts with
the pollutant (step 11) to give a peroxide and regenerate R".
There are three possible termination reactions (steps 12-
14) . At moderately high oxygen pressures which exist in
aerobic waters and low pollutant concentrations, step 12 is
the most likely termination step. Under these conditions and
making a steady-state assumption, the rate of radical
initiation (Ri) may be expressed as
Ri = 2 kt [ROO* ]2
where kt is the termination rate constant. The rate of
disappearance of pollutant (RH) is given by:
11
-------
k [RH]
= k [RHHROO-] = -2 (16)
[2 kt]*
This equation shows that the overall rate of disappearance is
proportional to the pollutant concentration and the square
root of the free radical initiation rate.
Extrapolation of autooxidation kinetic data to the
environment is complicated by lack of data on concentrations
of free radical initiators that are present under environ-
mental conditions. Rate constants for steps 10, 11, and 12
are readily available in the literature.* For example, for the
organic compound tetralin, k-| (6.7 x 107 M~* sec-*) and kt (2
x 107 M-i sec-i) approach diffusion control. However, k2 (13
M-1 sec-*) is slow and the magnitude varies with organic
structure.
Another complicating factor is the presence of naturally
occurring anti-autooxidants, compounds like phenols and
aromatic amines5 with readily extractable hydrogen atoms.
These compounds react with R* to form non-reactive free
radicals and halt the chain propagation steps (10 and 11).
The impact of autooxidation on the degradation of
pollutants in the environment has not been evaluated. Reports
in the literature indicate that autooxidation of pesticides
does occur. These investigations, however, deal with
pesticide degradation in thin films on soils, plants, and
glass.
PHOTOCHEMICAL
Direct Photolysis
Numerous studies concerning pesticide phot©decomposition
via direct absorption of light have appeared in the
literature. A great majority of these studies were concerned
with products derived from such direct photolysis. A review
of this work is well beyond the scope of this report.
Moreover, several excellent review articles have already been
published6"1*.
Our review of the pesticide literature indicated that
little'has been published about the rates of direct photolysis
under sunlight. Generally, information of this type can be
obtained only by careful inspection of the Experimental
sections of papers. Since rate data are essential to the
modeling of pollutant dynamics in the environment, one of the
12
-------
principal goals of this project was to obtain data that can be
used to predict direct photolysis rates under a variety of
environmental conditions.
The simplest way to determine direct photolysis rates is
irradiation of the pollutant by sunlight in an uncovered re-
action vessel. However, kinetic data obtained in this way
should not be extrapolated from one location to another,
because the intensity of solar ultraviolet radiation at the
earth's surface, particularly in the 300-320 nm region, is a
function of latitude, season, and elevationis. All of the
pesticides included in this study absorb sunlight most
strongly in the ultraviolet region. In the following section,
we will discuss a simple technique that employs laboratory
data to calculate direct photolysis rates. Other discussions
of direct photolysis kinetics in solution16 or the atmo-
sphere*7 provide a good background for this section.
Scientists realized long ago that only light which is
absorbed can effect chemical change in a system; this basic
law of photochemistry was first enunciated by Grotthus, then
Draper, back in the early Nineteenth Century. The average
photoreaction rate, (-d[P]/dt)\f at wavelength X in a com-
pletely mixed water body is directly proportional to the
amount of light absorbed in a unit volume per unit time. The
latter is defined by Lamberts law (equation 17) where a^ is
the absorbance of a one-centimeter thick layer of the water
body and I is the pathlength of the light.
-a> H
Fraction Absorbed =1-10 (17)
Underwater sunlight intensity on a horizontal plane is
derived from two sources, light intensity directly obtained
from the sun (I^x) an<^ the diffuse light intensity obtained
from the sky (IS^)1S»17. The average rate of light absorption
(IaX) for a water layer of depth D is defined by equation 18
-a, H, -a^c,
I.,(1-10 A d) + I.XU-10 A S)
I = —& ** (18)
aX D
where SL& and Zs are the average pathlengths for direct and sky
radiation respectively.
Under most natural conditions, a pollutant absorbs only a
fraction of the light absorbed by a water body. This fraction
is expressed by the ratio £xtp]/aX wnere £X ^s the molar
extinction coefficient of the pesticide at wavelength X in
units of 1000 cmz/mole and [P] is the pollutant concentration.
13
-------
The average rate of light absorption by a pollutant (1^) is
defined by equations 19 and 20
I- = I £X[P]
XaX aX — - (19)
jax
(20)
where ka^ = laXex'^X an<^ j is a constant that converts [P]
into units that are compatible with the intensity units (j
equals 6.02 x 102<> when [P] is expressed as moles/liter).
The equation for ka^ simplifies under two circumstances:
• If a^Jl and a.\Hs are both > 2, then essentially all
the sunlight responsible for photolysis is absorbed and k
can be expressed by equation 21
.
d. A
• If a^a and ou£s a*e both < 0.02, e.g., near the
surface of a water body, then the fraction of light absorbed
(equation 17) becomes approximately equal to 2.303 a,£ and k
is defined by equation 22 X aA
_ 2'303 £X[IdX*d + ^s1 (22)
aX JD
Note that the value of k , in this case is independent of the
nature of the water body.
Equation 22 can be re-expressed as
2.303 e,Z,
k = - LA (23)
aX j
where Z, is defined as
^ + i ,£ ]
d - ^_£- (24)
14
-------
The pathlengths , &d and £s, can be expressed in terms of D by
equations 25 and 26
a. = Dn - (25)
*
/ n - sin Z
Si = 1.2D (26)
s
where n is the refractive index of water and z is the angle
that the sun makes with the sky zenith, i« e. , the "solar
zenith angle." Values of Z \ for the mid-part of the four
seasons in the central United States (latitude 40°N) are
included in Table 1. All values pertain to midday solar
radiation at sea level. For most wavelengths (330 to 500 nm)
the Z\ values are expressed in photons cm- 2 sec-n averaged
over 10 nm intervals. The values of 297.5 to 320 nm are
averaged over 2.5 nm intervals and the 323.1 nm value
represents 323.125 ± 1.875 nm. Midsummer values of Z \ for
three latitudes (0°, 40°, 70°N) are listed in Table 2. Data
for the 297.5 to 380 nm regions were taken from Bener's
reportis. we calculated the other data by computer using data
and procedures described by Leighton**. The values take into
account reflection of sunlight at the water's surface.*8
Usually, only a fraction of the light absorbed by a
compound results in photore action. This fraction, or quantum
yield for reaction ( <(>) , can be determined by laboratory
experiments employing monochromatic light. The direct photo-
lysis rate of a pesticide is also proportional to its quantum
yield for reaction.
The complete photolysis rate expression is shown in
equation 27
_ d[P] =
-------
Table 1. Z, VALUES FOR LATITUDE 40°N
A
297.
300
382
305.
397.
310.
312.
315 .
31?
320.
3 2 3 .
338.
340.
358.
368.
370.
380 .
390.
480.
410.
420.
430.
440
458,
460.
478.
488
490
580.
525.
550.
575
688
625
650
675
700
750
800
. 5
. 8
5
8
5
8
5
0
5
8
1
0
0
8
0
8
0
8
8
8
0
8
0
, 0
, 0
. 0
8
8
8
0
0
8
8
0
. 8
. 8
. 8
. 8
. 8
8
8.
8,
a.
Q.
8.
8.
0.
8.
0.
8.
8.
a.
8.
8
8.
8.
0.
8.
8.
8.
0.
e.
8.
0.
0.
0.
0
8.
8.
8.
0.
e.
0
8.
8.
8.
8.
8 .
SP
274E+ 12
128E+13
419E+13
121E+14
2 2 3 E + 1 4
372E+ 14
584E+14
788E+ 1 4
992E+ 1 4
1 i. 7 E + 1 5
221E+15
761E+15
8 8 8 E •>• 1 5
942E+ 15
181E+16
1 12E+16
124E+ 16
143E+ 16
213E+ 16
280E+16
288E+16
277E+ 16
3 2 7 E + i 6
36SE* 1 6
371E+16
384E+ 16
392E+16
371E+ 16
378E+ 16
398E+ 16
413E+16
4 17E+16
421 E+ 16
4 2 2 E •* 16
424E+16
423E+16
419E+16
481E+16
385E+ 16
P H 0
8
0
0
8
0
8
8
0
8
0
PHO
8
PHO
M
8
8
8
8
8
0
8
8
8
8
e
8
8
8
0
8
0
0
8
8
0
e
8
8
8
8
8
T 0 H S ;
su
. 7 1 6 E
. 248E
723E
. 182 E
. 385E
. 495E
. 717E
. 9 3 3 E
. 1 15E
. 135E
TOMS<
. 252E
TONS'!
846E
963E
. 183E
. i 18E
. 122E
. 135E
. i 6 1 E
. 231 E
. 332E
. 318E
. 298E
351 E
. 3 9 4 £
398E
. 411E
. 420E
3?6E
. 404E
. 426E
. 442E
. 446E
450E
4 sen
. 451E
. 448E
. 443E
423E
. 4 0 5 E
C M - 2
+ 12
+ 13
+ 13
M 4
+ 1 4
+ 14
+ 14
+ 14
+ 15
+ i 5
C M - 2
+ 15
CH-2
M5
+ 1 5
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
•* i 6
+ 1 6
+ 16
+ 16
+ 1 6
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
+ 16
SEC
8.
8.
8.
0 .
8.
8 .
8.
8.
8.
0
SEC
0.
SEC
0 .
8 .
8.
8.
8.
0.
e.
9 .
8.
0.
0
y
0.
8.
0.
8.
e.
0.
0.
0.
0.
e
0
8.
8.
8.
8
8.
- 1 2 . 5
Fft
9 4 9 E + 1 1
524E+1 2
223E+13
670EM 3
i 3 5 E + 1 4
288E+14
371E+14
494E+14
e 4 1 E + 1 4
em + 14
-1375
144E+15
-1 18 N
588E+15
694E-M 5
645E+15
687E+1 5
754E+15
P22EM 5
I 0 £• E •* : 6
! 56E + 16
286E+16
212E+1 6
285E+16
•;; 4 4 £ 4 ; 6
275E+16
279E+16
289E+16
2 9 6 E + 1 6
;; 8 1 E + i 6
287E+16
305E+16
318E+1 6
322E+16
326E+lo
•> c* 3 L "* 1 6
332E+16
333E+16
338E+16
3 18E+16
3 tf 6 E + l 6
NH-i )
8
8 •
8.
8
8 .
8 .
8.
8.
8
0 .
H H - 1
8.
M-l)
8.
8
8
8 .
8.
0
9 .
8 .
8.
0.
0.
8.
0
8 .
0.
8 .
e.
8.
0.
8.
8.
8 .
8.
8.
8 .
8.
8
0 .
WIN
890E +
733E +
368E*
178E +
450E +
854E +
177E +
271E +
362E +
498E +
;,
986E +
342E +
428E +
449E +
479E +
529E +
562E +
805E +
116E-
154E-
159E +
154E +
184E +
288E +
2 1 1 E +
2 i 9 E +
225E +
213E +
218E +
232E +
241E +
243E +
247E +
252E +
256E +
259E +
258E +
250E +
242E +
00
11
12
13
13
13
14
14
14
14
14
15
15
15
15
15
15
15
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
16
-------
Table 2. Z, VALUES FOR THE SUMMER SEASON
A
48 70
297.
300.
302.
385.
307.
310.
312.
315.
317.
320.
323.
336.
346.
350.
360.
370.
380.
330.
480.
410.
420.
430.
440.
450.
460.
470.
480.
490.
500.
525.
550.
575.
600.
625.
650.
675.
700.
750.
800.
5
0
5
0
5
0
5
0
.5
8
1
0
0
0
0
0
8
0
0
0
0
0
0
0
0
, 0
0
0
. 0
0
0
0
0
0
0
0
0
0
0
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0,
8.
9.
0.
8.
8.
0.
0.
0.
8.
0.
8.
0.
0.
0.
0.
0.
8.
0.
0.
8.
0.
0.
0.
0.
0.
0.
0.
161E +
422E +
189E +
244E +
384E +
587E +
88GE +
181E +
124E +
143E +
263E +
858E +
9S3E +
183E+
1 18E +
122E +
135E +
161E +
231E +
382E +
318E +
298E +
351E +
394E +
398E +
411E +
420E +
397E +
485E +
427E +
445E +
458E-*-
454E +
453E +
453E +
458E +
444E +
424E +
485E +
13
13
14
14
14
14
14
15
15
15
15
15
15
16
16
16
16
16
16
16
16
16
IS
16
16
16
16
16
IS
16
16
16
16
16
16
16
16
16
16
8.
8.
0.
0.
0.
0.
0
8.
e.
0.
0.
8.
0.
0.
0.
0.
8.
8.
0.
716E +
248E +
723E +
, 181E +
385E +
. 495E +
717E +
S33E +
1 15E +
135E +
252E +
846E +
963E +
103E +
1 18E +
122E +
135E +
161 E +
2 3 1 E +
12
13
13
14
14
14
14
14
15
15
15
15
15
16
16
16
16
16
16
8. 382E+16
8.
8.
8.
8.
8.
8
8.
8,
8.
8
8
8 .
8.
0
0.
0.
8.
0.
8.
318E +
298E +
351E +
394E +
. 398E +
411E +
. 428E +
396E +
404E +
426E +
16
16
16
16
16
16
16
16
16
16
0.
0.
0
8.
0.
8.
0.
8.
0.
0.
0.
0.
0.
0.
8.
8.
8.
8.
8.
8.
0.
0.
8.
8.
8.
8.
0.
8.
8.
8.
442E+16 8.
. 446E +
450E +
458E +
451E +
448E +
. 443E +
423E +
485E +
16
16
16
16
16
16
16
16
e.
8.
8.
Q.
8.
8.
8.
8.
307E
233E
12 IE
417E
923E
157E
307E
436E
571E
731E
134E
497E
604E
645E
687E
754E
822E
188E
156E
286E
212E
285E
244E
275E
279E
288E
296E
288E
286E
383E
315E
318E
322E
326E
338E
331E
338E
318E
386E
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
•H
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
+ 1
•n
+ 1
+ 1
+ 1
+ 1
•u
+ i
+ 1
+ 1
+ i
+ 1
+ 1
+ 1
1
2
3
3
3
4
4
4
4
4
5
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
17
-------
As previously noted, all of the pesticides included in our
study absorb sunlight most strongly in the uv regions. The
short-wavelength intensity of solar uv light (297.5 to 320 nm)
is strongly influenced by both the solar zenith angle, z, and
the atmospheric ozone content (03). Ozone absorption causes a
sharp decrease in intensity at wavelengths < 320 nm (Figure
3). Both Z\ and (O3) depend upon season and latitude*5*, i?, t«.
- 3.0
oo
CM
O
00
C
O
"5
.cz
ex
2 5
f- • J
2.0 -
7 1.5
O
X
£ i.o
0.5 -
GO
O
O
0.0
290
310 330 350 370
WAVELENGTH, nm
390
410
Figure 3. Midday solar irradiance for summer,
latitude 40°N
The long wavelength portion of solar uv light (330-390 nm) is
unaffected by (03), but depends only on variations in z.
Seasonal variations of Z\ values for several latitutdes are
shown in Figures 4 and 5. Note that seasonal variations of
the short-wavelength Z\ values (Figure 4), because they are
influenced by (03), are much more pronounced than variations
18
-------
in the 330-390 run region (Figure 5). Generally, the amplitude
of the seasonal variation of Z\ becomes larger and the
magnitude of Z^ smaller with increasing Northern latitude,
especially with Z\ values for wavelengths < 320 nm. The
following conclusions can be derived from Figures 3-5.
7 18
o
CD
CO
I
o
f-H
X
M
16
14
E
o
CO
§ 12
•5 c
10
8
6
4
2
0
0°N
J FMAMJJASOND
MONTH
Figure 4. Dependence of short-wavelength solar uv
irradiance ppon season and latitude
• Direct photolysis rates of pollutants that absorb
sunlight most strongly at wavelengths < 320 nm vary
considerably from one season to another. Maximum
photolysis rates, however, generally occur during the
19
-------
seasons (spring and summer) of greatest pesticide
use.
Direct photolysis rates generally decrease with
increasing Northern latitude. During the summer,
however, there is less than a two-fold difference in
rates between the equator and latitude 50°N.
o
o>
CO
CM
E
o
CO
C.
O
•6
•&1
X
M
0
^— 0°N
0°N
I I
I I I I 1 I I 1 I
J FMAMJJASOND
MONTH
Figure 5. Dependence of long-wavelength solar uv
irradiance upon season and latitude
Other factors also influence direct photolysis rates of
pesticides. Ultraviolet intensity is decreased somewhat by
clouds, but even on an overcast day the midday ultraviolet
20
-------
intensity is attenuated only by a factor of two20. Reflection
of sunlight from the surface of a water body also decreases
its intensity, but the fraction of sunlight reflected is less
than 0.1 during most of the day*5.
One of the most important determinants of direct photo-
lysis rates is the penetration of ultraviolet light into the
water. Inland surface waters generally contain varying
amounts of dissolved organic or humic materials that are
derived primarily from decayed vegetation. As shown in Figure
6, these dissolved organics can have an appreciable effect
upon the penetration of ultraviolet light. The term 2/a in
Figure 6 is the depth at which 99X of direct sunlight at z =
0° is absorbed. In rivers such as the Suwannee River in north
Florida, direct photolysis, on the average, is very slow,
since it occurs at depths not greater than a few centimeters.
This is certainly not the case in all inland surface waters.
For example, ultraviolet light penetrates to much greater
depths in the Savannah River (Figure 6). Light scattering by
suspended particles in turbid natural waters can also markedly
attenuate ultraviolet penetration.21
a
CXJ
0
320 340 360 380
WAVELENGTH, nm
400
Figure 6.
Penetration of ultraviolet light into
two natural waters and pure water
21
-------
Some scientists have found that highly insoluble chlori-
nated pesticides such as DDT and dieldrin concentrate in
slicks that occur on the surface of natural waters. 22-2* This
phenomenon obviously accelerates the overall photolysis rates
of the pesticides. Unfortunately, there are insufficient data
to assess its general importance at the present time. It is
likely that the more water-soluble pesticides such as carbaryl
and malathion will not concentrate in surface slicks to a
great extent .
In conclusion, the above discussion demonstrates that the
direct photolysis of pesticides should obey a first-order rate
law and be directly proportional to the quantum yields for
reaction and the sunlight absorption rates. Near the surface
of water bodies, the latter are independent of the nature of
the water body and can be calculated from the ultraviolet
absorption spectra of the pollutants and tables of solar
radiation data. Direct photolysis rates can be calculated for
different locations and seasons. Photolysis rates in natural
waters are strongly dependent upon absorption and scattering
of ultraviolet light by materials in the water.
Sensitized Photolysis
Several studies have shown that the photodecomposition of
pesticides can be accelerated by the presence of other organic
compounds that absorb light more strongly than the pesticides
themselves25. Such acceleration, generally referred to as
photosensitization, can occur on solid surfaces such as plant
leaves,26 or in natural waters.27/28
Perusal of the literature suggests that several mechanisms
may account for pesticide photosensitization. One mechanism
involves light absorption by a sensitizer S (eg 29) , followed
by energy transfer from the sensitizer to the pesticide P (eq
S — ^» IS* (29)
13* - > 33* (30)
3S* + P - > S + 3P* (31)
- > Products (32)
Although light absorption initially puts the sensitizer into
its first excited singlet state, *S*, energy transfer most
often occurs from its first excited triplet state, 3S* (eq
30) . It should be noted that physical chemists usually
restrict the term "photosensitization11 to energy transfer
phenomena.
22
-------
Several studies have demonstrated that triplet energy
transfer (eq 31) occurs efficiently only when the triplet
state energy of the sensitizer is greater than or equal to
that of the energy acceptor.29 Our studies of the photosensi-
tized decomposition of phenylmercury compounds30 have demon-
strated that large decreases in quantum yield occur when
sensitizers with insufficient triplet state energies are em-
ployed. Since the triplet state energies of pesticides are
important determinants of the efficiency of triplet energy
transfer, we have listed approximate triplet state energies of
some pesticides and their hydrolysis products in Table 3.
Classical studies by Lewis and Kasha3* established that most
benzene derivatives have triplet state energies in the 75-85
kcal mole-1 range. The data in Table 3 indicate, as expected,
that most of the pesticides included in this study have
triplet energies > 75 kcal. Notable exceptions are the
nitrobenzene derivatives, parathion and p-nitrophenol, the
naphthalene derivatives, carbaryl and 1-naphthol, and 1,1-
diphenylethylene derivatives, DDE and the DDE analog of
methoxychlor. The relatively low triplet energies of these
compounds indicate that a wide range of sensitizers can
transfer energy to them. On the other hand, the high triplet
state energies of pesticides such as DDT-type compounds and
2,4-D derivatives indicate that their photodecomposition by an
energy transfer mechanism can only occur when high energy
sensitizers, such as acetone30, are employed. Various
naturally occurring compounds, such as some amino acids and
some common aromatic formulation components, such as xylenes,
have sufficiently high triplet energies to transfer energy to
all of the pesticides listed in Table 33*/35. The sunlight
absorption rates of these high-energy sensitizers are low,
however, and the quantum yields for energy transfer
photosensitization are sharply reduced by the presence of
competing energy acceptors such as oxygen.
Several studies in the literature indicate that mechanisms
other than energy transfer are also responsible for
photosensitized decomposition. Ivie and Casida26 found that
there was no good correlation between the effectiveness of
several sensitizers and their triplet state energies. More-
over, Plimmer and Kearney36 and Rosen and co-workers37 have
reported that chloroaniline decomposition is photosensitized
by benzophenone and riboflavin. Since the triplet energies of
benzophenone (68 kcal mole-1) and riboflavin (47 kcal mole-1)
are lower than those of anilines (> 75 kcal mole-1)3*,
mechanism(s) other than energy transfer must account for these
observations.
Another likely mechanism for photosensitization involves
chemical reaction between the electronically excited sensi-
tizer and the pesticide (eq 33)
23
-------
Table 3. TRIPLET STATE ENERGIES OF SEVERAL PESTICIDES
Pesticide or Hydrolysis Triplet Energy a'
Product „ , , . -1
KcaI/mole
DDT 79c'e
Methoxychlor 80c'f
DDE 54d
DDE Analog-Methoxychlor 53
Parathion 58
Diazinon 76
Carbaryl 60
2,4-D Butoxyethyl ester 77°>f
p-Nitrophenol 58
1-Naphthol 60
aExcept where noted, calculated from X of highest energy
band in phosphorescence spectrum.
Except where noted, taken from phosphorescence data reported
by Winefordner31.
cCalculated from onset of phosphorescence.
Reference 32.
eReference 33.
Measured in collaboration with Dr. R. Hautala, University of
Georgia.
24
-------
S*
Products
(33)
Of the variety of reactions that can occur, hydrogen atom
transfer from pesticide to sensitizer is one of the most
general. In the presence of oxygen, this reaction often leads
to oxidation of the pesticide by a free radical mechanism.
The effectiveness of rotenone as a sensitizer26 is probably
related to the fact that it is a substituted phenyl ketone;
photoexcited phenyl ketones are known to be efficient
hydrogen-atom abstractors38. Riboflavin, a widely used
pesticide photosensitizer, is also a good hydrogen abstractor
in the excited state39. The riboflavin-sensitized oxidation
of 2,4-dichlorophenol apparently involves hydrogen
abstraction40 (Figure 7)as products due to coupling of free
radicals are obtained.
Formation of ground-state or excited-state complexes*1
between naturally occurring substances and pesticides can also
lead to accelerated pesticide photodecomposition. The
enhanced sunlight photolysis rate of DDT observed in the
presence of high concentrations of amines has been attributed
to formation of excited-state charge-transfer complexes42.
Light-induced decomposition of dissolved pesticides in the
aquatic environment is not likely to occur via ground-state
amine-pesticide complexes; the weak complexes that amines form
with chlorinated hydrocarbons43 are completely dissociated at
low concentrations. Numerous studies in the chemical
literature have provided examples of photoreactions mediated
by excited-state complexes, or "exciplexes. II4» /44 Compounds
+ RIB
hv, 02
RIB-H-
RIB-H-
02
-»- RIB + HOO-
RIB= riboflavin
J. Plimmer and U. Klingebiel, Science, 174, 407 (1971).
Figure 7. Mechanisms for the riboflavin-sensitized
oxidation of 2,4-dichlorophenol
25
-------
containing alkylamino groups are particularly prone to
photoreact by this pathway*5 (eq 34-36); oxidation of the
amine and reduction of the sensitizer occurs via electron
transfer within the exciplex.
6- 6+
S* + R NCH,R > (S«««R NCH_R) * (34)
£. £ £, £,
Exciplex
Exciplex » S- + R NCH R » SH« + R NCHR (35)
R2NCHR —» R2N=CHR » R2NH + RCOH (36)
Ketones, such as benzophenone, and aromatic compounds, such as
naphthalene derivatives, are known to be photoreduced by
amines.
Because all aquatic environments contain dissolved mole-
cular oxygen, autooxidation, initiated by photochemical de-
composition of free radical initiators, may also be involved
in the light-induced decomposition of pesticides in natural
waters. Autooxidation was discussed in detail in an earlier
section of this report. Autooxidation could be initiated by
photoinduced hydrogen-abstraction reactions (eq 37), similar
to those previously discussed.
S* + P-H > SH« + P« (37)
The above discussion demonstrates that a variety of path-
ways are available for the light-induced decomposition of
pesticides. Thus, the direct photolysis half-life can be re-
garded only as a minimum estimate of the photoreactivity of a
pesticide under environmental conditions. Studies discussed
later in this report demonstrate that photolysis half-lives
for pesticides in natural waters are sometimes much shorter
than direct photolysis half-lives.
Photooxygenation Involving Singlet Oxygen
One other general mechanism for photosensitized oxidation
of pesticides deserves separate consideration. Singlet
molecular oxygen*6 can be generated by energy transfer from
sunlight-absorbing substances in water bodies (Figure 8). In
Figure 8, SENS and SENS* represent some singlet oxygen
photosensitizer in its ground and electronically excited
state, 3O2 and »O represent oxygen in its triplet ground
26
-------
state and first excited singlet state, and P represents a re-
active pesticide.
Sens
Sens* + 302
!02 + P
'0,
'Oj, + Q
hv
K
Q .
Sens"
Sens + !02
P + 302
Products
- 3o2
'02+Q
Figure 8. Mechanism for generation of singlet oxygen
in the aquatic environment
Merkel and Kearns*7 have shown that the rate constants for
reaction (Kr) are relatively insensitive to solvent changes,
but the rate constant for radiationless decay of singlet
oxygen (Kd) varies greatly from one medium to another. The
value of Kd is much larger in water than in organic solvents.
The lifetime of singlet oxygen may be shortened in natural
waters by the presence of "quenchersr" Q, such as amines, that
return it to its ground state (Figure 9).
COO
hv,
C00-
Rose Bengal
in methanol
K. Ueda, L. Gaughan, and J. Casida, J. Agr. Food Chem., 22, 212 (1974).
Figure 9. Photosensitized oxidation
of cis-resmethrin
Numerous publications have appeared concerning the role of
singlet oxygen in atmospheric*8 and other environmental
systems*9. These studies have shown that singlet oxygen
participates in three important reaction types; oxidation of
olefins (eq 38), addition to conjugated dienes or similar
27
-------
compounds (eg 39) , and oxidation of compounds containing
heteroatoms, such as alkyl sul fides (eq 40)
- -* RCH=CHCH2OOH (38)
RCH=CH-CH=CH2 - ^ RCH-CH=CH-CH 2 (39)
0—0
RSR » RSR (40)
Peroxides formed by reactions of singlet oxygen with
naturally-occurring substances*9 can also serve as free
radical initiators for autooxidation of pollutants.
Pollutants containing the above structural features are most
likely to react with singlet oxygen. For example, cis-
resmethrin, which contains a furan moiety, reacts with singlet
oxygen50 to form a product which results from methanol
addition to the expected endoperoxide intermediate (Figure
9)**. This is a good example of addition of singlet oxygen to
a dienoid moiety (eq 39). Other examples will be discussed
later in the report. Recent studies by Heitz and his co-
workers have demonstrated that singlet oxygen, generated
photochemically, readily oxidizes the acetylcholinesterase
from imported fire ants5*. Irradiation of fire ants that have
been fed singlet oxygen photosensitizers results in symptoms
similar to classical nerve poisoning52.
Kinetic equations describing photosensitized oxygenation
in a natural water are shown in Figure 10. The average rate
(-d£P]/dt) at depth 1 is proportional to the quantum yield for
reaction (s) , the fraction of light absorbed by the singlet
oxygen photosensitizer (£sXcs/2eXc) » an^ t^ie intensity of sun-
light (Gx). At low pesticide concentrations, the quantum
yield becomes directly proportional to the pesticide
concentration and the kinetics obey a first-order rate law.
Foote has expressed singlet oxygen reactivities in terms of 3
values where $(Kd/Kr) is the concentration of "acceptor"
required to react with half of the singlet oxygen formed in a
system*6. The rate is inversely proportional to the 3 value.
28
-------
d[p] =
dt / ^ £exc
K,[P]
* w I *• J
If Kd»Kf[P],
*=
C. S. Foote, Accounts Chem. Res., J., 104 (1968).
Figure 10. Kinetics equations for photosensitized
oxidation involving singlet oxygen
Although Plimmer11 and Crosby*2 have both suggested that
singlet oxygen may play a role in the photooxidation of
pesticides, no data concerning the kinetics of 'C^-pesticide
reactions have appeared in the literature. This lack of data
prompted the singlet oxygen studies that are discussed later
in this report.
REFERENCES
1 Kirby, A. J. In: Comprehensive Chemical Kinetics, Vol.
10, Bamford, C. H., and C. F. H. Tipper (eds.). New York,
Elsevier Publishing Co., 1972. Chapter 2.
2 Frost, A. A., and R. G. Pearson. Kinetics and Mechanisms,
2nd Edition. New York, John Wiley and Sons, Inc., 1961.
p. 27-46.
3 Basolo, R., and R. G. Pearson. Mechanisms of Inorganic
Reactions, 2nd Edition. New York, John Wiley and Sons,
Inc., 1967. p. 454-525.
4 Pryor, W. A. Free Radicals. New York, McGraw-Hill Book
Co., 1966. p. 288-295.
5 Nonhebel, D. C. , and J. C. Walton. Free-Radical Chemis-
try. Cambridge, Cambridge University Press, 1974. p.
305-416,
29
-------
6 Crosby, D. G. In: Fate of Organic Pesticides in the
Aquatic Environment. Advances in Chemistry Series, No.
111. Washington, American Chemical Society, p. 173-188.
7 Crosby, D. G., and M. Li. In: Degradation of Herbicides,
Kearney, P. C., and D. D. Kaufman (eds.). New York,
Marcel Dekker, Inc., 1969. Chapter 12.
8 Crosby, D. G. Residue Reviews. 25:1 (1969).
9 Crosby, D. G., K. W. Moilanen, M. Nakagawa, and A. S.
Wong. In: Environmental Toxicology of Pesticides,
Matsumura, F., G. M. Boush, and T. Misato (eds.). New
York, Academic Press, 1972. p. 423.
10 Plimmer, J. R. Residue Reviews. 3_3:47 (1971).
11 Plimmer, J. R. In: Degradation of Synthetic Organic
Molecules in the Biosphere. Washington. ISBN 0-309-02046-
8. National Academy of Sciences. 1972. p. 229-288.
12 Crosby, D. G. In: Degradation of Synthetic Organic Mole-
cules in the Biosphere. Washington. ISBN 0-309-02046-8.
National Academy of Sciences. 1972. p. 260-278.
13 Rosen, J. D. In: Organic Compounds in Aquatic Environ-
ments, Faust, S. D., and J. V. Hunter (eds.). New York,
Marcel Dekker, Inc., 1971. Chapter 17.
14 Paris, D. F., and D. L. Lewis. Residue Reviews. 45;95
(1973).
15 Bener, P. Approximate Values of Intensity of Natural
Ultraviolet Radiation for Different Amounts of Atmospheric
Ozone. Davos Platz, Switzerland. Report # DAJA37-68-C-
1017. U.S. Department of the Army. June 1972.
16 Balzani, V., and V. Carassiti. Photochemistry of
Coordination Compounds. New York, Academic Press, 1970.
p. 6-15.
17 Leighton, P. A. Photochemistry of Air Pollution. New
York, Academic Press, Inc., 1961. Chapters 2 and 3.
18 Hutchinson, G. E. A Treatise on Limnology, Vol. I. New
York, John Wiley and Sons, Inc., 1957. p. 372-376.
19 London, J. Beitrage zur Physik der Freien Atmosphare.
3_6:254 (1963).
30
-------
20 Schultze, R., and K. Grafe. In: The Biologic Effects of
Ultraviolet Radiation, Urbach, F (ed.). New York,
Pergamon Press, 1969. p. 359,
21 Esaias, W., W. H. Biggley, and H. H. Seliger. (Presented
in part at the 37th Annual Meeting of the American Society
of Limnology and Oceanography. Seattle. June 1971.)
22 Seba, D. B., E. F. Corcoran. Pestic. Monit. J. .3:190
(1969) .
23 (a) Duce, R. A., J. G. Quinn, C. E. Olney, S. R.
Piotrowicz, B. J, Ray, and T. L. Wade. Science. 176;161
(1972). (b) Bidleman, T. F. , and C. E. Olney. Science.
183:516 (1974).
24 Maclntyre, W. G., C. L. Smith, J. C. Munday, V. M.
Gibson, J. L. Lake, J. G. Windsor, J. L. Duprey, W.
Harrison, and J. D. Oberhottzer. Investigation of Surface
Films-Chesapeake Bay Entrance. Washington. EPA 760/2-73-
099. U.S. Environmental Protection Agency. February 1974.
p. 99-108.
25 LyJcken, L. In: Environmental Toxicology of Pesticides,
Matsumura, F., G. M. Boush, and T. Misato (eds.). New
York, Academic Press, 1972. p. 449-469.
26 Ivie, G. W., and J. E. Casida. J. Agr. Food Chem. 19;
405 (1971) .
27 (a) Ross, R. D., and D. G. Crosby. J. Agr. Food Chem.
21:335 (1973). (b) Ross, R. D., and D. G. Crosby. (Pre-
sented in part at the 167th National Meeting of the
American Chemical Society. Los Angeles. April 1974.)
28 Zepp, R. G., N. L. Wolfe, and G. L. Baughman.
Environmental Research Laboratory, U.S. EPA. (Presented
in part at the 168th National Meeting of the American
Chemical society. Atlantic City. September 1974.)
29 Turro, N. J. Molecular Photochemistry. New York, W. A.
Benjamin, Inc., 1965. Chapter 5.
30 Zepp, R. G., N. L. Wolfe, and J. A. Gordon. Chemosphere.
2:93 (1973) .
31 Moye, H. A., and J. D. Winefordner. J. Agr. Food Chem.
13:516 (1965).
32 Ullman, E. F., and W. A. Henderson. J. Amer. Chem. Soc.
.89:4390 (1967).
31
-------
33 Hornig, A. W. Identification of Polychlorinated Biphenyls
in the Presence of DDT-Type Compounds. Washington.
EPA#R2-72-004 W73-03515. U.S. Environmental Protection
Agency. October 1972. 66p.
34 Lewis, G. N., and M. Kasha. J. Amer. Chem. Soc. 66:2100
(1944) .
35 Lamola, A. A. Photochem. Photobiol. _8:601 (1968) .
36 Plimmer, J. R., and P. C. Kearney. (Presented in part at
the 158th National Meeting of the American Chemical
Society. New York. September 1969.)
37 Rosen, J. D., M. Siewierski, and G. Winnett. J. Agr.
Food Chem. 18:494 (1970).
38 Wagner, P. J. Accounts Chem. Res. .4:168 (1971).
39 (a) Merkel, J. A., and W. J. Nickerson. Biochem.
Biophys. Acta. 14:303 (1954). (b) Moore, W. M., J. T.
Spence, F. A. Ramond, and S. D. Colson. J. Amer. Chem.
Soc. 85:3367 (1963). (c) Kearns, D. R., R. A. Hollins,
A. U. Khan, and P. Radlick. J. Amer. Chem. Soc. 89:545e>
(1967) .
40 Plimmer, J. R., and U. Klingebiel. Science. 174:407
(1971).
41 Bowman, R. M., T. R. Chamberlain, C. Huang, and J. J.
McCullough. J. Amer. Chem. Soc. _96:692 (1974) and
references therein.
42 Miller, L. L. , and R. S. Narang. Science. 169;368
(1970) .
43 Biaselle, C. J., and J. G. Miller. J. Amer. Chem. Soc.
96:3813 (1974).
44 Wagner, P. J., and A. Kemppainen. J. Amer. Chem. Soc.
9J.:3985 (1969).
45 Cohen, S. G. Chem. Rev. T3:141 (1973).
46 Foote, C. S. Accounts Chem. Res. 1:104 (1968).
47 Merkel, P. B., and D. R. Kearns. J. Amer. Chem. Soc.
14:7244 (1972) .
32
-------
48 Pitts, J. N. In: Chemical Reactions in Urban Atmospheres,
Tuesday, C. S. (ed.)• New York, American Elsevier
Publishing Co., 1971. p. 3-31.
49 International Conference on Singlet Molecular Oxygen and
Its Role in Environmental Sciences, Trozzolo, A. M.
(ed.). Ann. NY Acad. Sci. .171:1-302 (1970).
50 Ueda, K., L. Gaughan, and J. Casida. J. Agr. Food Chem.
22:212 (1974).
51 Callahan, M. F., L. A. Lewis, J. R. Broome, M. E.
Holloman, and J, R. Heitz. (Presented in part at the
168th National Meeting of the American Chemical Society.
Atlantic City. September 1974.)
52 Broome, J. R., M. F. Callaham, L. A. Lewis, C. M. Ladner,
and J. F. Heitz. (Presented in part at the 168th National
Meeting of the American Chemical Society. Atlantic City.
September 1974.)
33
-------
SECTION VI
MATERIALS AND METHODS
MATERIALS
Sources and purification methods for the pesticides used
in this study are listed in Table 4. Synthetic procedures are
described below. Products of chemical and photochemical
reactions were obtained commercially whenever possible, but in
most cases they were synthesized as described.
Chlorohydroxvphenoxyacetic acid esters
The acids, 2-chloro-4-hydroxyphenoxyacetic acid and 4-
chloro-2-hydroxyphenoxyacetic acid, were synthesized as
described by Brown and McCall*. Esters were prepared by acid-
catalyzed reaction of the acids with appropriate alcohols.
Mass spectra of methyl 2-chloro-4-hydroxyphenoxyacetate and
that of the lactone obtained from thermal decomposition of
esters of 4-chloro-2-hydroxyphenoxyacetic acid are shown in
Figure 11. The lactone formed quantitatively during gas
chromatographic analysis of the esters.
O,O-Dimethyl phosphorodithioic acid
Phosphorus pentasulfide (Eastman) (35.2 g, 0.158 mol) was
added to 40 ml of benzene. The slurry was stirred and
maintained at 40°; 28.8 ml (0.713 mol) of absolute methanol
was added dropwise over a period of three hours. The reaction
mixture was filtered and the solvent removed by flash
evaporation. The OrO-dimethyl phosphorodithioic acid was
purified by distillation (12" Vigreux column), and the
fraction distilling at 34-35° (0.15 mm Hg) ; [lit. 42-44° (0.5
mm Hg)] was collected. The ir spectrum compared with the
literature spectrum2.
0,0-Dimethyl-S-n-carbethoxy-2-carboxy)ethyl phosphoro-
dithioate (g-monoacid)
Following a procedure similar to that of Chen et al.3,
O,O-dimethyl phosphorodithioic acid and ethyl hydrogen
maleate* were reacted in the presence of pyridine. The
reaction mixture was cooled and 10 ml of water was added. The
mixture was slowly titrated with sodium hydroxide (1 M) until
the residue dissolved (pH below 6). The solution was washed
with 10 ml of chloroform and separated; the resulting aqueous
layer was acidified to pH 2.5 (10St HCl) and extracted with
34
-------
CO
W
Q
H
u
H
EH
CO
W
CM
rt
O
Cu
CO
w
•3
a
H
K
o
H
EH
O
H
U
H
H
tt
PH
Q
J2;
rfj
CO
H
O
^
0
CO
*»
CU
X)
id
EH
(U
3
G1
•rl
f*
JJ-*
O
CU
EH
a
o
•H
-P
id
o
•H
m
•H
j-i
rj
P4
CU
O
J-l
J3
O
co
cu
'O
•H
O
•H
4J
CO
CU
CU
ro
rH
U
W
U
g
0
4-1
•a
_|
O
CU
>_j
M
0 rH
0
n3 d
cu id
rH A
H 4J
•H CU
•P g
CO
•H
Q
'O
CU
N
•H
CO
CU
-P
£*
^1
CO
CO
14
-------
1V101 JO 30ViN30H3d
-8
t
o
1
X!
I
^
O
M
o
O
I
g
M-i
O
-P
o
a)
a
CO
CO
CO
a)
S
(1)
M
>IV3d 3SV8 30 30ViN3Dd3d
36
-------
IVJOi JO 30ViN3DH3d
R
-S
o
t
-S
•§
•S
O
•a
o
.8
-S
g
-S?
1 1
>IV3d 3SV8 JO 30ViN3D«3d
PQ
a)
Cn
•H
37
-------
three 50-ml portions of chloroform. Workup yielded 0.78g
(85%) of a colorless oil that slowly crystallized on standing.
Analysis by Ic showed the mixture to consist of the B-isomer
(97%) and the a-isomer (3X) . Recrystallization from
chloroform-hexane gave white crystals [mp 42-45°; lit. 51-
52°].3 The ir5 and proton nmr3 compared with those reported in
the literature.
O,O-Dimethyl-S-M-carboxy-2-carbethoxy) ethyl phosphoro-
dithioate (g-monoacid)
To 0.158 g (1.0 mmol) of O,O-dimethyl phosphorodithioic
acid in 10 ml of 50% benzene-hexane v/v was added 0.144 g (1..0
mmol) of ethyl hydrogen maleate*. 11,1*,1'-Triphenyl-
benzeneazomethane (Eastman) (1 mg) was added; the solution was
cooled to 0° and irradiated through Pyrex with a Hanovia
medium pressure 450 watt lamp for one hour. The malathion
monoacid product mixture contained the a-isomer (95%) and the
g-isomer (5%) as shown by Ic analysis. The reaction mixture
was worked up as described above, yielding 0.120 g (40%) of a
white solid. Crystallization from chloroform-hexane gave
white crystals, mp 55-57° [lit. an oil].3 The ir and nmr
spectra compared with the literature spectra3'5.
O,O-Dimethyl-S-(1,2-dicarboxy)ethyl phosphorodithioate
fdiacid)
Following the above procedure, 23.7 g (0.150 mol) of O,O-
dimethyl phosphorodithioic acid was added to 14.7 g (0.150
mol) of maleic anhydride (Aldrich, recrystallized) and one
drop of pyridine. The reaction was maintained at 60° for
three hours. The residue was dissolved in 150 ml of chloro-
form, washed with three 100-ml portions of water, and dried
(Na2SO4) . To 11.8 g (0.046 mol) of the anhydride was added 40
ml of water; the resulting solution was heated at 70° for one
hour. The aqueous solution was washed with 10 ml of chloro-
form, acidified to pH 2 (cone. HC1), and extracted with four
50-ml portions of ether. Evaporation of solvent yielded a
white solid, which crystallized from chloroform to give 9.8 g
(71%) of the diacid: mp 115-118°. Recrystallization from
chloroform yielded crystals with a melting point of 127-129°.
The ir spectrum matched that reported in the literature6.
O,O-Dimethyl-S-(1-carbethoxv-2-carboxy)ethyl phosphorothiolate
(malaoxon g-monoacid)
A 10% solution of bromine in water was added dropwise with
stirring to malathion 8-monoacid (1 g, 3.3 mmol) dissolved in
100 ml of 50% aqueous ethanol (v/v) (27°) until a faint yellow
color persisted. The aqueous solution was extracted with
three 150 ml portions of chloroform, and the organic layer
38
-------
dried (Na2SO4). Concentration gave 0.91 g (96X) of a
colorless oil. The glc retention time, after methylation
(diazomethane)t was less than the retention time of methylated
starting material, and the ir spectrum was consistent with the
title compound. Attempts to crystallize or distill the oil
under reduced pressure failed.
2,4-D Esters
Commercial 2,4-D (Aldrich) was reacted with appropriate
alcohols in the presence of an acid catalyst (usually dilute
sulfuric acid), The methyl ester was subseguently purified by
recrystallization (nip 38-39°) and the liquid esters were
purified by vacuum distillation. Their mass spectra were
consistent with the assigned structures.
Methoxychlor
Chloral was condensed with anisole using aluminum tri-
chloride as catalyst in alcohol-free chloroform. After
stirring at room temperature for twelve hours, the purple
reaction mixture was hydrolyzed with a large excess of water
and the chloroform layer was dried over anhydrous sodium
sulfate. Upon evaporation a yellow oil was obtained which
crystallized upon standing. Pure methoxychlor was obtained by
chromatographing the product on Woelm neutral alumina, then
recrystallizing it five times from 95% ethanol [mp 88-89°;
lit.7 mp 87-88°].
2- and 4-Chlorophenoxyacetic Acid Esters
Methyl esters were synthesized from the commercially ob-
tained acids by Clinton and Laskowski's procedure8. Butoxy-
ethyl esters were prepared by the acid-catalyzed reaction of
the acids with 2-butoxyethanol. The mass spectra of the
esters were consistent with their assigned structures.
N-Methyl-1-hydroxy-2-naphtharnide
The acid chloride of 1-hydroxy-2-naphthoic acid was pre-
pared by refluxing the acid in thionyl chloride containing a
trace of dimethylformamide. An ethereal solution of the acid
chloride was slowly added to a stirred 1:1 mixture of ether
and HQ% agueous methylamine with occasional cooling by an ice
bath. The reaction mixture was neutralized by addition of
hydrochloric acid, then the ether layer was washed by water,
5% agueous sodium bicarbonate, and water. After drying over
magnesium sulfate, the ether was removed under vacuum to yield
the acid amide. Recrystallization from benzene-hexane yielded
the pure amide, [mp 122°; lit.® mp 123-121°].
39
-------
N-Methyl-4-hydroxy-1-naphthamide
The methyl ester of 4-hydroxy-1-naphthoic acid was pre-
pared by a procedure described elsewhere10. The ester was
dissolved in UOX aqueous methylamine and the mixture was
heated at 100°C for three days. Upon cooling, the amide
crystallized from the reaction mixture.
1I ,1-Bis(p-methoxyphenyl)-2,2-dichloroethylene (DMDE)
DMDE was synthesized by reaction of methoxychlor with
potassium hydroxide in boiling 95% ethyl alcohols [mp 111-
111.5°; lit.11 mp 109°]. The mass spectrum was the same as
that reported by MacNeil et al.12
1,l-Bis(p-methoxyphenyl)-2,2-dichloroethane fDMDD)
DMDD was synthesized as previously described13 [mp 116-
117°; lit.13 mp 114.5-115°]. The mass spectrum agreed with
the spectrum previously reported12.
1,1-Bis(p-methoxyphenyl)-2-chloroethvlene
1,1-Bis(£-methoxyphenyl)-2-chloroethylene was prepared
from DMDD by the same procedure used to synthesize DMDE from
methoxychlor [mp 76-77°].
Other Materials
Water used in most of the experiments was distilled,
passed through ion exchange columns, then redistilled twice.
The final distillation was carried out from permanganate in an
all-glass apparatus with retention of a middle cut. Natural
water samples were collected from inland surface waters in the
southeastern United States. The natural waters used in
photochemical experiments were sterilized by passage through
0.22-micron Millipore filters.
Benzene was acid-washed, dried, and distilled. Acetonit-
rile was purified following Mann's procedure,1* then distilled
on a spinning-band column. Other reagent-grade solvents were
used as received. Chemical Samples Co. 2,4-hexadien-1-ol and
cis-1,3-pentadiene were distilled and stored at -20° to
prevent oxidation. Singlet oxygen quenchers, 3-carotene and
1,4-diazabicyclo-(2,2,2)-octane were obtained from Sigma
Chemical Co. and Chemicals Procurement Laboratories,
respectively. Valerophenone was purified by vacuum
distillation.
40
-------
ANALYTICAL PROCEDURES
Gas-liquid chromatography (glc) using flame or electron-
capture detection was employed to analyze reaction mixtures in
most cases. Liquid chromatography (Ic) was used to analyze
mixtures containing carfcaryl15 and atrazine. Preparative Ic
on a Micropak SL-10 column with 595 methanol in dichloromethane
as mobile phase was employed to separate malathion a- and 8-
monoacids. Valerophenone actinometers** were analyzed by glc
on glass columns containing 10X SILAR 5CP on Gas Chrom Q;
methyl benzoate was employed as internal standard. Mixtures
containing cis and trans 1,3-pentadiene were analyzed by glc
on a 15-ft 20% 1,2,3-tris(2-cyanoethoxy)propane column.
APPARATUS
Glc analyses were performed on Tracer MT-220 and 550 gas
chromatographs. Glc peaks were integrated by Autolab 6300
Digital Integrators. The liquid chromatograph was a DuPont
Model 820 equipped with an ultraviolet (254 nm) photometric
detector.
Kinetic studies of thermal reactions were carried out in a
thermostated oil bath that regulated temperature within ±
0.05°C. Photolysis studies were conducted in a photochemical
apparatus described in detail elsewhere*7. A Hanovia 450 watt
mercury lamp was the light source. Mass spectra were obtained
on a Finnigan 1015 SL quadrupole mass spectrometer coupled
with a Varian Aerograph Model 1532-E gas-liquid chromatograph
and a Systems Industries 150 digital computer. Nuclear
magnetic resonance (nmr) and infrared (ir) spectra were
obtained, respectively, on a Varian HA-100 NMR Spectrometer
and a Perkin Elmer 621 Grating Infrared Spectrophotometer.
Carbon-13 magnetic resonance spectra were recorded on a JEOL-
PFT-100 Spectrometer with the JEOL EC-100 20K data system.
Ultraviolet-visible absorption spectra were obtained on a
Perkin-Elmer model 602 Digital Spectrophotometer.
HYDROLYSIS PROCEDURES
Malathion
Kinetics—The following kinetic procedure based on extraction
and glc analysis for the determination of the malathion
disappearance rate constant is representative of the pro-
cedures used throughout these studies. A buffer solution (99
ml) of 0.008 M boric acid and 0.002 M sodium hydroxide was
prepared. The flask was placed in a constant temperature bath
at 27.00° ± 0.02° and allowed to equilibrate. Then one ml of
41
-------
a stock solution of 1.41 x 10~2 M malathion in acetonitrile
was added to the buffer solution to give a 1,41 x 10-* M
malathion solution containing 1X acetonitrile. The pH of the
solution was determined to be 8.56, employing a pH meter
calibrated with standard buffer solutions.
A 5-ml aliquot was removed and the reaction quenched by
extraction of the malathion with a 5-ml aliquot of chloroform
containing 3,4,6,2*,5» pentachlorobiphenyl (PCB) (2.79 x 10~*
M) as an internal standard. The time was recorded and taken
as 0% reaction. Aliquots were withdrawn at various time
intervals through approximately two malathion disappearance
half-lives. At the end of ca 10 half-lives, an aliquot was
withdrawn and used as an infinity point. Analysis of this
infinity point showed that greater than 98% of the malathion
had reacted. The concentration of malathion was determined
for each aliquot based on PCB-malathion peak area ratios.
Pseudo-first-order rate constants were obtained using the
integrated first-order rate equation where
C — C
k = _L- in -_2 22_ (41)
C. - C
t oo
and
C = malathion concentration at zero time
o
C = malathion concentration at time t
C = malathion concentration after ten half-lives.
oo
The rate constant was taken as the slope of the line ob-
tained by a linear least squares analysis of the data. The
second-order rate constant was obtained by dividing the
pseudo-first-order rate constant by the hydroxide ion
concentration.
Following the same procedure, kinetic runs were carried
out at elevated temperatures in the constant temperature
baths. The temperatures were 47.00° ± 0.02° and 67.00° ±
0.05°. Kinetic data were obtained at 0.0 ± 0.5° using an ice
bath.
Activation parameters were calculated from reaction rate
constants obtained at two temperatures different by at least
20°.
Several kinetic studies dealt with the formation and
disappearance of malathion acid products. In these studies,
diphenylacetic acid was added to the reaction solution as an
42
-------
internal standard. Aliquots were removed, acidified to pH 2
with 10% hydrochloric acid, and extracted with ether. The
ether extracts were methylated with diazomethane. Malathion
acid concentrations were then calculated based on the internal
standard-malathion peak area ratios.
The rate constants for disappearance of malathion, (km(j) ,
formation of malathion monoacids (kmh), and O,O-dimethyl
phosphorodithioic acid (kme) were obtained by computer curve
fitting techniques. Pseudo-first-order rate constants were
assumed for formation of malathion acid and O,O-dimethyl
phosphorodithioic acid. The following constraints were
employed.
md mh me (42)
where k . and k are > 0.
mh me
The concentration of O,O-dimethyl phosphorodithioic acid
[E] was obtained by
[E] = [MQ] - [A] - [M]
where
[M ] = initial malathion concentration
o
[A] = monoacid concentration
[M] = malathion concentration
The rate of disappearance of malathion and the rate of forma-
tion of products were obtained from the following three
differential equations:
dt mn
d[E] = k
dt me
<«4)
(45)
[M] (46)
Product Studies—Product studies were carried out at the end
of one half-life. In general the solution (100 ml) was
acidified to pH 2 with 10% hydrochloric acid and extracted
with three 50-ml portions of ether. The organic fractions
were combined, dried (Na SO ), and concentrated. Thin-layer
43
-------
chromatography was used for qualitative identification of
malathion, malathion monoacids, malathion dicarboxylic acid,
and O,O-dimethyl phosphorodithioic acid. Quantitative deter-
minations were carried out by methylation (diazomethane) of
the ethereal extract followed by glc analysis. Confirmation
of products was obtained by glc-ms and nmr. The relative
amounts of malathion or and 3-monoacid isomers were determined
by Ic. Nmr and glc analysis of the reaction mixture was used
to verify the presence of diethyl fumarate, ethyl hydrogen
fumarate, and thiomalic acid.
2,4-D
Kinetics—Hydrolysis of the methyl and n-butoxyethyl esters of
2,U-D were carried out in buffered aqueous solutions con-
taining 0.1 to 195 acetonitrile. Starting ester concentrations
were 10~5 M and the concentration of ester was followed
through approximately two disappearance half-lives. At
recorded time intervals, aliquots of the hydrolysis solution
were extracted with an aliquot of benzene containing 10~5 M
PCB as an internal standard. The concentration of ester at
known times was determined by glc analysis from the peak area
ratio of ester to the PCB standard.
Psuedo-first-order rate constants were obtained using the
integrated form of the first-order rate equation. Second-
order rate constants were obtained by dividing the pseudo-
first-order rate constant by the hydroxide or hydrogen ion
concentration.
Product Studies—Product studies were carried out after ten
half-lives. The reaction solution was acidified to pH 2 and
extracted with ether. The ethereal extraction was methylated
(diazomethane) and analyzed by glc. In all kinetic runs 2,4-D
was found as the product in greater than 98% of theoretical
yield.
Methoxychlor
Acid degradation of methoxychlor was carried out at 1 x
10~7 M in distilled water or 1 x 10~6 M in 5% acetonitrile
water. A 25-ml portion was titrated to the desired pH with
hydrochloric acid. The acidified solution was placed in 5-ml
ampoules. At given time intervals the ampoules were opened
and the aqueous solution was extracted with benzene. The
benzene extract was then analyzed by glc(ec) and the methoxy-
chlor reaction determined by comparison with an ampoule which
had been stored at -10°.
Basic degradation was carried out in a manner similar to
acid degradation. A 25-ml portion of 1 x 10~7 M methoxychlor
44
-------
in distilled water or 1 x 10~6 M methoxychlor in 5%
acetonitrile-water was titrated to the desired pH with sodium
hydroxide. The ampoules were opened, extracted with benzene,
and analyzed by glc. Methoxychlor concentration was
determined by comparison with an ampoule which was stored at
-10°.
Elevated temperature studies were carried out by placing
the ampoules in an oil bath at 45.00° ± 0.02°, 65.00° ± 0.02°,
and 85.00° ± 0.05°.
Captan
Kinetics—The rate of disappearance of captan (1 x 10—5 M) was
followed by a glc method. Reactions were carried out under
buffered reaction conditions in water containing 0.3 to 1%
acetonitrile. pH's were determined with a pH meter which was
calibrated with standard buffer solutions. Solutions were
thermostated in a constant temperature bath at 27.00° ± 0.02°,
37.00° ± 0.02°, and 8.0° ± 0.5°. Five-mi aliquots were
withdrawn at recorded time intervals and the reaction quenched
by extraction with 5 ml of benzene.
Psuedo-first-order rate constants were determined em-
ploying the integrated form of the first-order rate ex-
pression.
Product Studies—Product studies were carried out at 27° at
the end of one and ten half-lives. Organic compounds were
tentatively identified by glc and confirmed by gc-ms and ir.
Chloride ion was confirmed by first extracting the ten
half-lives reaction solution with benzene and treating the
aqueous layer with silver nitrate. The change in hydrogen ion
concentration was determined by use of a pH meter. Sulfur was
qualitatively confirmed as a product by solid probe ms.
Carbaryl
Kinetics—The rate of alkaline hydrolysis of carbaryl under
buffered conditions was determined by an Ic technique.
Carbaryl (100 ml), 2 x 10—* M in water, was buffered at pH
9.42 (boric acid-sodium hydroxide). Aliquots were withdrawn
at recorded times and the reaction quenched by making the
solution slightly acidic with 10X HCl. Aliquots of the
aqueous reaction solution were then analyzed directly by Ic.
The concentration of carbaryl was determined by comparison
with a standard based on peak height. Dilution of a standard
solution showed the Ic response to be linear over the concen-
tration ranges used in the studies. Rate constants were
obtained as described above.
45
-------
Product Studies — Based on Ic and glc retention times, a-
naphthol was identified as a product of hydrolysis. Lc
analysis at one half- life and ten half -lives showed a-naphthol
to be formed in 48 and 98X yield, respectively.
Atrazine
Kinetics — Atrazine hydrolysis was carried out using the gen-
eral technique described for malathion. A 2 x 10~* M solution
of atrazine in water was made by stirring the calculated
amount of atrazine in water for 24 hours. The atrazine
solution was then adjusted to the desired pH with 10X HCl or 1
M NaOH as determined by a pH meter. The solution was hydro-
lyzed at 25.00° ± 0.02° and three aliquots removed at recorded
time intervals, extracted with chloroform and the atrazine
concentration determined by glc analysis.
The rate constants were determined by calculating the
half -life from a plot of log concentrations vs time. The
pseudo-first-order rate constants were obtained from the rela-
tionship
v 0.693 ,._.
kobsd = -t— (47)
Product Studies — Hydroxyatrazine was identified as a product
of both acid and alkaline degradation based on comparison of
Ic retention times with an authentic sample.
Diazinon
The hydrolysis rate of diazinon in 5% acetonitrile-water
was determined by the glc method. Alkaline and acid studies
were carried out at 0.0900 M NaOH and 0.002 M HCl,
respectively, with 3.8 x 10-* M diazinon at 25.00° ± 0.02°.
Aliquots were removed at recorded times and extracted with
chloroform. Diazinon concentrations were determined by
comparison with standards.
Pseudo-first-order rate constants were determined from
half-life values which were obtained from plots of log con-
centration vs time.
Parathion
Parathion hydrolysis was not investigated. Faust18 has
reported kinetic data for the hydrolysis of this compound over
a wide pH range at 20°.
46
-------
Toxaphene
A 100-ml stock solution of 1.3 ppm toxaphene in 5% ace-
tonitrile-water was prepared. A portion of this solution was
titrated to pH 3.7 with 10% HCl. Aliquots were sealed in
ampoules and placed in an oil bath at 65°. At recorded times,
ampoules were removed, opened, and the aqueous solution was
extracted with chloroform. The amount of degradation was
determined by comparison of the gas chromatographic
fingerprints with a standard.
The alkaline degradation studies were carried out in a
similar manner at pH 10.0.
OXIDATION PROCEDURES
Malathjon
Aliquots of a 3.06 x 10-* M solution of malathion in water
buffered at pH 5.25 (phosphate buffer system) were placed in
ampoules. The solutions were saturated with oxygen, the
ampoules sealed, and placed in the constant temperature bath
for recorded time intervals.
Sample analysis was similar to that described for hydro-
lysis products.
2,4-D
Oxidation studies were carried out in air-saturated water
solutions and in one study, hydrogen peroxide (1 x 10~3 M) was
added as an oxidant, A standard solution of 2,4-D (2 x 10~* M
in water) was used throughout the study. Aliquots were
allowed to react in sealed ampoules as described above.
The solutions were analyzed directly by glc.
Methoxychlor
A solution of methoxychlor (1 x 10~7 M) in water was
employed as a stock solution. Aliquots were saturated with
oxygen, sealed in ampoules, and placed in a thermostated bath
at U7° ± 2°, 67° ± 2°, and 87° ± 2°. One ampoule beside the
bath served as a standard. Ampoules were removed at given
time intervals and stored at -20°. At the end of the run, the
ampoules were opened, extracted with benzene and the
concentration of methoxychlor determined by glc analysis.
Studies involving hydrogen peroxide as a reactant were
carried out as described above.
47
-------
Product studies were carried out at 1 x 10-* M methoxy-
chlor in 30% 1-propanol-water solution. The products were
analyzed by glc.
PHOTOLYSIS PFOCEEURES
General
Screening studies were conducted by irradiating in par-
allel 13.0 mm Pyrex tubes containing 2.80 ml of saturated
aqueous solutions of the pesticides. The tubes transmitted
only wavelengths > 280 nm. First-order rate constants were
obtained from logarithmic plots of pesticide concentration vs
time.
Quantum yield studies were conducted by parallel irradia-
tion of pesticide and actinometer solutions using filtered
light from the mercury lamp. Filter solutions (1.0-cm thick}
were as follows: 313 nm - 0.001 potassium chromate in 3%
aqueous potassium carbonate; 436 nm - 40 g cupric sulfate, 68
ml concentrated ammonium hydroxide, and 500 g sodium nitrite
in one liter of water; 578 nm - 50 g cupric chloride
(hydrated), 150 g anhydrous calcium chloride and 150 g
potassium dichromate in 500 ml water acidified by hydrochloric
acid. Valerophenone*9, 1-phenyl-4-methyl-1,2-pentanedione20,
and ferrioxalate2* actinometers were used respectively for
313, 436, 578 nm light. Quantum yields for dilute solutions
of pesticides were calculated from slopes of first-order plots
of pesticide concentration vs time in seconds. For weakly
absorbing systems (absorbance < 0.02), the slope is equal to
ka^ where ka is the specific absorption rate and is the
quantum yield. The value of ka was calculated according to
eq. 48
k = 2.303 e,I,£ (48)
a A A
where £A is the molar extinction coefficient of the pesticide
at wavelength A, IA is a constant expressed as einsteins
liter™1 sec—1, and £ is the effective cell path length in cm.
IA is equal to I0S/V where I0 is the incident light intensity
(einsteins cm2 sec-*) at wavelength A, S is the surface area
exposed to light, and V is the cell volume. The value of IA
was measured directly by actinometers that absorbed all the
light. Generally, we assumed that eA in water was the same as
that measured in acetonitrile-water. The value of £ was
determined as described below.
-------
Determination of Cell Path Length for Quantum Yield Studies
A series of solutions containing 0.122 M cis-1,3-penta-
diene and 0.0012 to 0.05 M benzophenone in benzene were
prepared. Equal volumes (2.80 ml) of the solutions in 13.0 mm
tubes were degassed under high vacuum by several freeze-thaw
cycles. The solutions were then exposed to equal amounts of
313 nm light by parallel irradiation in the photochemical
apparatus. The percentage of trans-1,3-pentadiene formed by
the photosensitized isomerization22 was determined by glc
analysis of the solution. These percentages were then
corrected for back reactions of the trans to cis-diene as
described by Lamola and Hammond22. The cell path length was
determined from the slope of a plot of - log (1 - X) versus £C
for the various solutions (Figure 12) , where X is the ratio of
trans isomer formed at benzophenone concentration C to trans
isomer formed in the 0.050M benzophenone solution (percent
absorbed > 99), and e is the extinction coefficient of benzo-
phenone at 313 nm. Least squares analysis of the data
indicated that the cell path length was 10.0 ±0.2 mm.
Kinetic Studies under Sunlight
Air-saturated, aqueous solutions of the pesticides were
exposed to sunlight in tightly stoppered quartz tubes that
were suspended six inches above a black cloth parallel to the
earth's surface. Generally experiments were carried out
during the summer months of June through September. Dark
controls (tubes wrapped in aluminum foil) were also used in
all experiment. In most cases, no change in concentration of
pesticide occurred in the dark controls during the
experiments.
Photochemical Studies in Natural Waters
Saturated solutions of the pesticides in filter-sterilized
natural waters were photolyzed in the laboratory or under
sunlight as described above. Generally, solutions of
pesticide in distilled water were simultaneously irradiated
for comparison. Dark controls were also used in these
experiments.
Singlet Oxygen Studies
Singlet oxygen was generated photochemically using
methylene blue as photosensitizer. Air- or oxygen-saturated
solutions containing pesticide and methylene blue(MB) (10~s to
10-* M) were irradiated by 436 nm or 578 nm light. Under
these conditions singlet (*Ag) oxygen is generated with a
quantum yield near unity23. Acetonitrile and water were used
as solvents for these experiment. Prolonged irradiation of
49
-------
o
o
0.25
0.20
0.15
0.10
0.05
0
SLOPE = 10.0 ±0.2 MM
I
I
I
0 0.05 0.10 0.15 0.20 0.25
eC
Figure 12. Determination of cell pathlength
for quantum yield studies
most of the pesticides resulted in no detectable reaction.
Captan and its hydrolysis product, 4-cyclohexene-1,2-
dicarboximide, were notable exceptions and more detailed
studies of their reaction with singlet oxygen were carried
out. The photoreactions of captan and the imide were almost
completely inhibited by 1.00 x 10~3 Mr 1,4-diazabicyclooctane
(DABCO), a singlet oxygen quencher. The concentration
dependence of the reaction was determined by irradiating in
parallel a series of solutions containing 1.0 x 10~* MB and
varying amounts of captan or imide.
50
-------
Photolysis Product Studies
The same general procedure was followed for all the pesti-
cides. For direct photolysis product studies, the pesticide
solutions (4 £) were irradiated by Pyrex filtered light from a
mercury lamp until glc analysis indicated about 30% reaction
had occurred. The irradiated solutions were extracted by
three portions of 10% ether-chloroform (portion = 100 ml per
liter of solution). The combined organic extracts were dried
by anhydrous sodium sulfate, then concentrated to one ml. The
product mixtures were then characterized by glc-ms or by
infrared spectroscopy after isolation of the products by
preparative glc or Ic. Singlet oxygen products from captan or
imide were characterized by glc-ms after reduction of the
hydroperoxides to alcohols.
REFERENCES
1 Brown, J. P., and E. B. McCall. J. Chem. Soc. 3681
(1955) .
2 Nyquist, R. A. Spectrochim. Acta. 2J5:47 (1969).
3 Chen, P. R., W. P. Tucker, and W. C. Dauterman. J. Agr.
Food Chem. JT7:86 (1969) .
4 Dahlgren, G.r Jr., and F. A. Long. J. Amer. Chem. Soc.
£2:1303 (1960).
5 Welling, W., P. T. Blaakmeer, and H. Copier. J.
Chromatog. .47:281 (1970).
6 Walker, W. W. Ph.D. Thesis, Mississippi State University,
State College, MS, 1972.
7 Cristol, S. J. J. Amer. Chem. Soc. .67:1494 (1945).
8 Clinton, R. O. and S. C, Laskowski. J. Amer. Chem. Soc.
20:3135 (1948) .
9 Bass, C. R., G. H. Borwn, J. R. Thirtle, and A.
Weissberger. Photographic Science and Engineering. j>:195
(1961) .
10 de Montmillan, G. and J. Spieler. U. S. Patent 1474928
(1923) .
11 Schneller, G. H. and G. B. C. Smith. Ind. Eng. Chem.
4_1: 1027 (1949).
51
-------
12 MacNeil, J. P., R. W. Frei, S. Safe, and O. Hutzinger. J.
Ass. Offic. Anal. Chem. .55:1270 (1972).
13 Cristol, S. J., N. L. Hause, A. J. Quant, H. W. Miller, K.
R. Eilar, and J. J. Meek. J. Amer. Chem. Soc. 74:3333
(1952) .
14 O'Donnell, J. F., J. T. Ayres, and C. K. Mann. Anal.
Chem. .37:1161 (1965).
15 Thruston, A. D. Liquid Chromatography of Carbamate Pesti-
cides. EPA-R2-72-079. Nov. 1972. 15p.
16 Wagner, P. J. and A. E. Kemppainen. J. Amer. Chem. Soc.
14:7495 (1972).
17 Moses, R. G., R. S. H. Liu, and B. M. Monroe. Mol. Photo-
chem. J:245 (1969) .
18 Faust, S. D. Environ. Lett. 1:171 (1972).
19 Wagner, P. J., and A. E. Kemppainen. J. Amer. Chem. Soc.
.94:7495 (1972).
20 Zepp, R. G. and P. J. Wagner. J. Amer. Chem. Soc.
22:7466 (1970).
21 Hatchard, C. G. and C. A. Parker. Proc. Roy. Soc.
(London) A235;518 (1956).
22 Lamola, A. A. and G. S. Hammond. J. Chem. Phys. 41:2129
(1965) .
23 Merkel, P. B., and D. R. Kearns. J. Amer. Chem. Soc.
94:7244 (1972).
52
-------
SECTION VII
PHOTOCHEMICAL SCREENING STUDIES
To obtain a preliminary estimate of the relative direct
photolysis rates of the pesticides, aqueous solutions were
photolyzed in Pyrex reaction cells using a medium-pressure
mercury lamp. The short wavelength cutoff for the cells (1 mm
wall thickness) was about 280 nm. The effective cutoff for
sunlight is close to 297 nm. For most of the pesticides
studied, the most effective portion of the light was in the
280-305 nm range; irradiation through a 4-mm thick Pyrex
filter (cutoff 300 nm) generally caused only extremely slow
photoreaction (notable exceptions were parathion and
carbaryl).
Rate constants for the pesticide photolyses were
determined from the slopes of first-order plots. In each
experiment, the ultraviolet light intensity was determined
using a chemical actinometer that strongly absorbed only light
of wavelengths < 300 nm; i.e., the butoxyethyl ester of 2,4-D
in hexane (10~3 M) . Relative photolysis rates, normalized to
a uniform light intensity, are summarized in Table 5. The
light intensity from the mercury lamp, after Pyrex filtration,
was found to be about 80 times more intense than midday
sunlight^o in the 280-300 nm range (lat. 30°N, summer).
Accordingly, the experimental rate constants were divided by
this factor to give the normalized half-lives in Table 5.
This procedure gives only a crude estimate of the sunlight
photolysis half-lives because the spectral distribution of the
Pyrex-filtered light from the mercury lamp differs from that
of sunlight. The half-lives of pesticides such as parathion
and carbaryl, that absorb strongly at wavelengths > 320 nm,
are shorter than indicated in the table. The lamp intensity
was found to be only eight times greater than sunlight in the
280-370 nm range. On the other hand, the direct photolysis
half-life of methoxychlor is much longer than that indicated
by the screening procedure. The intensity of Pyrex (1 mm)
filtered UV light from the mercury lamp is much higher at
wavelengths < 295 nm than that of sunlight. Since
methoxychlor absorbs strongly from 280 to 290 nm, but absorbs
very weakly at wavelengths > 295 nm, the screening procedure
gave a false indication of its photoreactivity under sunlight.
We conclude that the screening studies can be a useful
tool for determining those pesticides that are not likely to
react efficiently under sunlight. However, if Pyrex-filtered
light efficiently degrades a pesticide, more detailed studies
53
-------
should be carried out to ascertain its direct photolysis rate
under sunlight.
The data in Table 5 indicated that methoxychlor, carbaryl,
and the butoxyethyl ester of 2,4-D were the most photoreactive
compounds. Accordingly, more detailed direct photolysis
studies of these three pesticides were carried out.
Table 5. RELATIVE DIRECT PHOTOLYSIS RATES OF SELECTED
PESTICIDES IN SCREENING STUDY
Pesticide
Atrazine
Captan
Carbaryl
Diazinon
2,4-D BEEb
Malathion
Methoxychlor
Parathion
Toxaphene
Relative
Rate
8
< 10
100
16
300
1
800
10
< 1C
Normalized Photolysis
Half-life (Hrs) a
3
> 2
2
1
1
2
3
2
> 2
X 103
X 103
X 102
X 103
X 102
x 10*
x 101
X 103
x 10*
aCalculated for a light intensity of 3 x 10*3 photons cm-2
sec-i from 280 to 300 nm.
Butoxyethyl ester of 2,4-D.
°No change in flc profile during period of exposure.
54
-------
SECTION VIII
RESULTS 6 DISCUSSION: MALATHION
HYDROLYSIS
Data pertaining to the breakdown of malathion in the
aquatic environment are not generally available. Paris et
al.* isolated a heterogeneous bacterial population capable of
utilizing malathion as a carbon source. At low concentrations
of malathion and bacteria, the rate of bacterial degradation
was described mathematically by a second-order rate
expression. The major metabolite was malathion 3-monoacid.
There are several potential pathways by which malathion
may be chemically degraded under reaction conditions common to
the aguatic environment (Figure 13). Carbon-sulfur cleavage
proceeding through an elimination reaction would give O,O-
dimethyl phosphorodithioic acid and diethyl fumarate as
products. Phosphorus-sulfur bond cleavage would give diethyl
thiomalate and O,O-dimethyl phosphorothionic acid. However,
OrO-dimethyl phosphorothionic acid would be in equilibrium
with its tautomer OrO-dimethyl phosphorothiolic acid. Carbon-
oxygen bond cleavage by carboxy1 ester hydrolysis would result
in two possible products, malathion ot- and 3-monoacids.
Continued carboxyl ester hydrolysis would result in malathion
dicarboxylic acid formation. Another possible reaction would
be oxidation of the phosphorus-sulfur double bond to a
phosphorus-oxygen double bond to give malaoxon.
Muhlmann and Schrader2 reported that malathion underwent
acid breakdown to give diethyl thiomalate and O,O-dimethyl
phosphorothionic acid. Cowart et al.3 studied the rates of
hydrolysis of seven organophosphorus pesticides (pH about 6)
and reported a half-life for malathion of about one week.
Ruzicka et al.* found malathion had a half-life of 7-8 hours
(20% ethanol-water, pH = 6 at 70°C); on the other hand,
Goldberg et al.5 reported that malathion did not undergo
reaction under either alkaline or acidic conditions.
Synthesis and characterization of potential degradation
products and detailed kinetic studies are reported in this
section.
A synthetic scheme was needed to selectively synthesize
each malathion monoacid. Chen et al.6 reported that O,O-
dimethyl phosphorodithiodic acid reacted with ethyl hydrogen
maleate to produce an oil, which, based on proton nmr
55
-------
m
8
O
V
CO
CO=O-
Cv
o
UJ
o
0
O
CM
— O
1
OACID
O
a
g
fc
^^^
UJ
O
O
O
+ v-
+ CO
CO=Q_
•M
o
-T-
0
O
O
CM
O
Q
O
<
O
o
O
°
9
co=-
I:
v-<
CO
I
CM
O
tn
:r
o
O
O
(-0 Z
CO
CO=Q_
CM
g
x
<
9. s£
= 92
o o o
GJ o <
o o 1
O O o
o *- =
Q
O
<
O
g
x
CO
>
X
o
§ 1
8 = 1
o—o _,
3: o >!
S i
O I-
ffi H
Q
UJ
i
Q
6
o'
c
o
•H
-p
a
T3
(0
M
Cn
0)
C
O
•rl
a
or
CO
>1
(0
I
m
a.
(0
o
•H
e
0)
45
O
•H
-P
G
V
-P
O
cu
ro
Q)
56
-------
analysis, was believed to be a single monoacid. Based on the
chemical shifts of the methylene groups, they designated this
product as the 3-monoacid. Following a modified procedure
(Figure 14), the reaction gave an oil which based on Ic
analysis consisted of the (3-monoacid (973S) and the ot-monoacid
<3X).
The predominance of g-monoacid formation is consistent
with an ionic mechanism in which 0,O-dimethyl
phosphorodithioic acid undergoes acid-catalyzed nucleophilic
addition to the carbon-carbon double bond. This addition
pathway is favored in preference to electrophilic addition
because of deactivation of the carbon-carbon double bond by
the carboxy and carbethoxy groups.
The a-monoacid was obtained by carrying out the addition
reaction in the presence of a free radical source. Bacon and
LeSuer7 reported that O,O-dimethyl phosphorodithioic acid in
the presence of peroxides added to alkenes in an anti-
Markovnikov addition. In the absence of peroxides, normal
Markovnikov addition was observed. Huang8 found that under
homolytic addition reaction conditions, the trichloromethyl
radical added to the carbethoxy group of ethyl hydrogen
maleate. 1•,1',1•-triphenylbenzeneazomethane (PAT) was
employed as a free radical initiator. The reaction was
carried out in a non-polar organic solvent at 0° and the free
radicals were generated by irradiation of PAT with Pyrex-
filtered light (Figure 14) . Liquid chromatographic analysis
showed the product mixture to consist of the a-monoacid (95%)
S
(CH30)?P-S-CHCOOH
CHjCOOEt
a-Monoacid
(95%)
S
(CH30)2P-SH
S
n
(CH30)2P-S-CHCOOEt
CH2COOH
p-Monoacid
hv, C6H5N=NC(C6H5)3
hexane-benzene, 0°
CHCOOH Pyidine catalyst
(3%)
p-Monoacid
(97%)
Figure 14. Scheme for synthesis of malathion monoacids
57
-------
and the 3-monoacid (5%) (Figure 15) .
is sensitive to reaction conditions.
selectivity for the -isomer was 95%.
The isomer distribution
The maximum relative
Malathion diacid [O,O-dimethyl S-(1,2-dicarboxy)ethyl
phosphorodithioate] was expected to be of value in making the
»3C nmr assignments in the monoacids9; addition of water to
the anhydride give the diacid (Figure 16). Addition of one
equivalent of ethanol instead of water to the intermediate
Ionic
Addition
Free Radical
Addition
Figure 15.
Liquid chromatograms showing the relative
amounts of malathion monoacids formed in
synthesis
58
-------
anhydride yielded a mixture of a- and 3-monoacids. No attempt
was made to separate the mixture.
Q f
0 f. 0 •>
^ CH-C^ (CH30)2P-S-CH-C^ H20 (CH30)2P-S-CHCOOH
(CH30)2P-SH + II ,D »» > »- |
CH2-CN CHjCOOH
. _.. ... . Diacid
1. Ethanol (1 eq)
2. H20
a-Monoacid + p-Monoacid
(53%) (47%)
Figure 16. Scheme for synthesis of malathion diacid
and monoacids
Malaoxon and its derivatives were obtained by oxidation of
the corresponding malathion compound. Both malaoxon and the
fj-malaoxon monoacid were readily obtained by oxidation with
bromine in aqueous ethanol10 (Figure 17).
S 0
(CH30)2P-S-CHCOOEt Br? (CH30)2P-S-CHCOOEt
I ^ I
CH2COOH H?0, CH3CH?OH CH2COOH
Figure 17. Synthesis of malaoxon monoacid
The carbon-13 chemical shifts and carbon-phosphorus
coupling constants obtained for malathion and several related
compounds are given in Table 6. Representative spectra are
reproduced in Figures 18 and 19. The proton noise-decoupled
spectrum of malathion exhibits nine carbon resonances as
expected. Assignment of these peaks to the carbons of
malathion was made from comparison with the spectra of the
related compounds in Table 6, off-resonance proton coupling,
and from expected trends in the chemical shifts and
phosphorus-carbon coupling constants.
59
-------
«J
en
Q
O
s
o
o
Q
W
EH
a
K
Q
<
2 m
O a
H ~.rj CM
a O
a 1
cn O
CM
.-3 a
f< r-U
CJ <">
rH a
a rou
w
o
co
rH
|
O
§
rj
.
vo
CD
rH
X)
id
EH
>
CO
^
f»
"2>
in
t/>
j.
•>
CM
;>
Xi
^
p>
rQ
C
0
g
O
U
vo *-~
CM VO
^r co
in -^
in
o
•
rH
H
H
t
rH
vo
o
00
Cn
VO
•-I
in —«
Cn rH
r- co
co ^
vo -^^
CN r-~
m co
•* ~
m ^
Cn 00
Cn *3*
VO -— '
rH
CO
O
•
CN
vo
in
o
•
«*
rH
U
C
O
•rH
r]
•P
(d
rH
(d
a
0 ^
VO CTi
^ ^
m -^
i
i
i
i
00
in
H
c —
rH
CM — ^
rH VO
OO CO
co ^
^* ^"x
in r^
in co
•*-'
OO ^-N
in cr>
t~~i *&
r- -^
rH
1
1
1
1
13
n3
•H
o
cd
•H
Q
C
0
•rH
r]
-P
n5
rH
td
a
o ^
tr in
^ m
in ^
l
l
i
l
vo
r*^
in
r^*
H
a\ ^-~
vo co
r^ ^
co -^
G\ *-*•
CTv rH
^3* CO
^~-
co *~.
OO CM
o> vo
VO "-'
rH
r^
CM
•
CM
VO
r-
CTl
•
co
rH
•d
•o
•H
U
rd
0
C
O
a
fl rH
O Q)
•H g
X) 0
•P CO
nj -H
rH 1
(d OQ
a *~~
•^
CM
^
in
rH
rH
•
, |
O
^l1
•
rH
VO
00
G\
o^
vo
H
o
r-
r-!
co
en
o
in
•*
H
in
in
r-
rH
l
l
l
l
0
nd
•rH
U
(d
O
C
0
a
^
o
•rl
C1,
-P
(d
rH
(d
a
^~.
en
^*
• —
f-^
VH
cu
g
0
CO
•H
1
<3
*-*
CO —X
^1* rH
** vo
in — •
t-^
rH
•
r-|
rH
rH
*
rH
VO
00
r-
en
vo
rH
rf — ,
CM en
00 ^^*
co ^
00 *~*
vo r~-
CM CO
-* —
*-*
en i — l
en vo
vo • —
rH
00
rH
•
CN
VO
r^
rH
•
«*
rH
•0
C
o
X
0
id
rH
(d
a
o
in
^
m
l
l
l
l
en
o
co
r-*
rH
m
en
[--
co
en
"^
CM
<*
f*«x
O^
Q}
VD
rH
r~-
CM
•
CN
vo
CO
en
•
co
rH
•a
r^
•H
O
rd
O
C
0
2J
C
O
X
0
rd
rH
rd
a
^^
00
<3<
^-^
s~i
en
^!
"— '
^-^
r-
<*
*"*
^-,
en
-^r
•— *
,-»
i_i
cu
g
0
CO
•H
1
CD.
^^
rH ^
CM rH
^ vo
in ^
l
l
l
l
l
1
l
i
i
l
l
l
l
i
l
0
S-l
o
A
a,
CO
o
r] rQ
P-i -rl
U
rH rt]
^1
x: o
-P -rH
Q) O
g-H
Q -P
1 -rl
o -a
•h
o
i
ig
b
^3
g
S
c
H
"^
•
M
-P
C
rd
-p
U]
G
0
L>
C
-H
3
0
U
M
O •
£ N
ftffi
CO
O CM
x; vo
f>[ •
1 O
fi +1
oa
Q
!H
rd •
O N
K
CU
,X3 CM
•P CM
>
rH rH
O +1
m o
0
i. 1 •
OS 03
a
CO EH
CO rH
cu rd
XJ C
1 i i_i
C CU
CU -P
rH C
(C "H
ft
g
fi O
•H M
H-l
CO
CO T3
3 rH
H CU
rd -H
>ll i
*t~1
60
-------
0 0
CHjCHzOCCHzCHCOCHzCHs
SP(OCH3)2
3(C9 'I
TMS
0 0
HOCCH2CHCOCH2CH3
C2
TMS
Figure 18. Carbon-13 spectra of malathion and
malathion 3-monoacid
Considerations similar to those above were used to assign
the carbons in the carbon-13 spectrum of malaoxon and
establish that the monoacid obtained from the chemical
hydrolysis of malaoxon is the 3-isomer (Figure 19) .
The stability of malathion in water at acidic pH's was
investigated. It was found that malathion was stable in water
at pH 2.5 and 25°C for up to ten days. Therefore the kinetic
studies were carried out at elevated temperatures (67° and
87°) . The extrapolated second-order rate constant (k ) is
(H.8 ± 0.2) x 10-s M-» sec-i at 27° (Table 7).
61
-------
0 0
CH3CHtOCCH2CHCOCH2CH3
Ci.C,
TMS
Cs.C,
2 C2|
DCCIj
uLi
L*jJ
JMuM
C7
C, C*
^
,j4L
C5
•LtMWUll**
jUM
CS
0 0
HOCCH2CHCOCH2CHj
SP(OCHj)2 OCCI3
0
*«*•
t«y*wS»ii
C9 64
TMS
Figure 19. Carbon-13 spectra of malaoxon and
raalaoxon B-monoacid
Table 7. MALATHION ACID-CATALYZED DEGRADATION KINETIC DATA
Temp (°C)
87
67
27 a
(2
<<*
(4
k
M-* sec— i
.88±0.01) x 10-2
.36±0.04) x 10-2
.8 ±0.20) x 10-s
AHf Asf
kca I/mole eu
22.3±0.1 -4.U0.4
Extrapolated rate constant.
Based on the extrapolated rate constant, the reaction of
malathion with hydrogen ion would have a half-life greater
than one year at pH 4. At temperatures and pH's common to the
62
-------
aquatic environment, this degradative pathway would be too
slow to be significant.
Product studies, carried out at the end of one malathion
degradation half-life at 87°, indicated that the malathion
monoacids formed in USX yield (eq. 49). No malathion diacid
was found although it was an anticipated product at longer
reaction times.
(CH30) 2-P-S-CHC02-Et
S PH = 2.59 CH2C02H
II
(CH3O) 2-P-S-CHC02Et - » + (49)
CH_CO.Et 87
2 2
(CH30) 2-P-S-CHC02H
The diacid was subjected to similar reaction conditions.
Although the analysis was not carried out quantitatively, glc-
ms analysis showed the presence of methyl O,O-
dimethylphosphorothionate and dimethyl thiomalate. Thus,
malathion diacid undergoes phosphorus-sulfur cleavage to give
O,o-dimethyl phosphorothionic acid and thiomalic acid (eq.
50).
(CH30)2-P-OH
Acid
(CH_0) -P-S-CH-CO,H > + (50)
I 87°
CH2C02H
H-S-CHC00H
I 2
HO CCH
63
-------
This reaction sequence is consistent With the results reported
by Muhlmann and Schrader2 who found malathion breakdown
products in aqueous ethanol solvent at elevated temperatures
to be O,O-dimethyl phosphorothionic acid and diethyl
thiomalate. Under their reaction conditions, however, they
would not have found malathion monoacids.
The kinetics of alkaline malathion degradation and the
resulting products were also investigated. Malathion is more
susceptible to basic degradation and significant chemical
breakdown would be anticipated under certain environmental
conditions. Temperature and in some cases chemical species
common to aquatic ecosystems might be expected to influence
malathion persistence.
The malathion second-order disappearance rate constant for
alkaline degradation (kmd) is 5.5 +; 0.3 M~» sec-i at 27°
(Table 8). Based on this rate constant, malathion would have
a 36-hour half-life in water at pH 8 and 27°. As anticipated,
an increase in temperature results in a decrease in malathion
half-life. At pH 8 and 40° the half-life is about one hour,
while at 0° (pH 8) the half-life is about twelve days. At
lower temperatures found in the environment, although somewheit
slower, malathion alkaline degradation could still be
significant.
Table 8. TEMPERATURE EFFECT ON THE MALATHION ALKALINE
DEGRADATION RATE CONSTANT
Temp. (°C)
0
27
47
67
k M-
md
0.067
5.5
HI
730
i
±
±
±
±
sec-1 a
0.003
0.3
1
20
Malathion disappearance rate constants.
Product studies were carried out (glc) after one malathion
half-life. Based on initial malathion concentration, the
products were malathion (50%), malathion monoacid (15X),
diethyl fumarate and ethyl hydrogen fumarate (35%), and O,O-
dimethyl phosphorodithioic acid, These products show that at
27° malathion is undergoing two competing reactions (Figure
20).
64
-------
Cu O
o o
O O
>—o
o
o o
o *•
5-o
CO
I
CO
10=0-
CM
o
o
o
V
CO
Cu
o
O
O
CO
i
co=Q-
8
O
o
•H
x:
O
•H
-P
(0
•d
0)
0)
c
•H
H
m
o
-------
The build-up of the products at 27°C as a function of time
is shown in Figure 21.
A MALATHION
D DPTA
O MONOACID
0
133
TIME, min
268
Figure 21. Malathion disappearance and product
formation at 27°C
A temperature effect on product formation became evident
when products were investigated at other temperatures. At 47°
after one half-life, analysis showed approximately 5%
malathion monoacid, at 67° there was a trace of monoacid
detected, and at 0° (Figure 22) the monoacid product is
important. It is apparent that carboxyl ester hydrolysis is
favored at lower temperatures and elimination is favored at
elevated temperatures (compare Figures 21 and 22).
66
-------
-5.42
A MALATHION
Q DPTA
O MONO AC ID
0
0
Figure 22.
245
TIME, min
Malathion disappearance and product
formation at 0°C
Second-order rate constants for elimination and carboxyl
ester hydrolysis at 27° and 0° were obtained by following
malathion disappearance and monoacid formation. Table 9
contains the rate data for the individual competing reactions
along with the activation parameters for the reactions. The
rate constant for elimination (kme) is 3.9 _+ 0.2 M-* sec-i and
for carboxyl ester hydrolysis (kmh) 1.4 _+ 0,1 M~i sec-i at
27°. Based on these rate constants malathion is undergoing
74% elimination and 26% ester hydrolysis at this temperature.
The effects of pH and temperature on degradation rate are
shown in Figure 23 where malathion half-life is plotted as a
function of pH at various temperatures. This figure shows
that malathion has a maximum half-life at pH 4.
67
-------
106
12 104-
10
2 -\
20° n J
30° * ox
40° °
I
2
I
4
I
6
10
Figure 23. Temperature effect on malathion degradation
at several pH values
Table 9. RATE CONSTANTS FOR MALATHION ELIMINATION AND
CARBOXYL ESTER HYDROLYSIS REACTIONS
Reaction
Elim
Ester
(kmh)
Temp.
°C
0
27
47
67
0
27
47
67
*-*
4.0
3.9
4.1
6.6
2.5
1.4
6.8
k AHt
sec—1 kcal/mole
X 10-2
x 0.2 27.8
x 101
x 102
x 10-2
x 0.1 20.7
6.7
x 101
ASt
eu
-2.8
-2.9
68
-------
There are two possible malathion monoacids which can
result from malathion carboxyl ester hydrolysis. Qualitative
determination of both isomers was made using thin-layer
chromatography (tic).*1 An estimate of the relative amounts
of each isomer present at the end of one malathion reaction
half-life was obtained by liquid chromatographic (Ic)
analysis. The relative percentages were a-monoacid, 85%, and
B-monoacid, 15%. Paris et al.* reported that microbial
degradation of malathion gave the 8-monoacid as the major
product (9956) .
Analysis of the monoacid required methylation prior to gas
chromatography. However, it should be noted that it was
necessary to analyze (glc) for diethyl fumarate prior to
methylation with diazomethane. This is because diazomethane
reacts with the carbon-carbon double bond to form a pyrolazine
(eq. 51) which was not detected under the glc condition
employed.
CH-C02Et
Et02CCH
To determine if this laboratory data could be extrapolated
to natural waters, the malathion half -life was determined in a
natural water sample obtained from a South Georgia river, the
Withlacoochee River (pH 8.21) . The half-life predicted from
laboratory rate data was 20 hours. The experimentally
determined half- life in Withlacoochee River water was 22
hours. This is in good agreement with laboratory rate data
obtained in distilled water experiments.
The effect of inorganic species common to the aquatic
environment which in some way might catalyze malathion
degradation was also investigated. Payne and Feisal nutrient
medium,12 which have a high concentration of inorganic salts,
did not effect malathion degradation at pH 6.8 in two weeks.1
Sodium chloride (0.05 M) had no apparent catalytic effect on
degradation of malathion at pH 6.5 as shown by no detectable
degradation at the end of three weeks.
General base catalysis13 might be expected to be important
in the elimination pathway. However, there was no detectable
difference in the rate of disappearance of malathion over
nearly a tenfold change in buffer concentration (boric acid-
sodium hydroxide) at constant ionic strength (sodium
chloride) . Thus, any general base catalysis contribution
would likely be insignificant in natural waters.
69
-------
Bender,** in fish toxicity studies, has investigated nine
potential malathion degradation products for toxicity to the
carp. His data showed that in some cases the breakdown
products were as toxic as the parent pesticide. However,
Bender did not investigate the toxicity of the malathion
monoacids. Because they are anticipated products, we examined
their persistence under alkaline reaction conditions.
The disappearance second-order rate constant (kad)
obtained on starting with a mixture of 5335 a-monoacid and 47%
3-monoacid was (3.1 .+ 0.2) x 10-* M-I (Table 10). Assuming no
large difference in reactivity for the two isomers (Figure
21), at pH 8 the monoacids would have a half-life of about 24
days. Thus malathion monoacids are about 18 times more stable
than malathion under the same alkaline conditions.
Half-life product studies by glc, after acidification and
extraction with ether showed the presence of ethyl hydrogen
fumarate. Methylation of the reaction mixture (diazomethane)
and glc analysis showed the presence of malathion monoacid
(50%), malathion dicarboxylic acid (15X), and O,O-dimethyl-
phosphorodithioic acid.
Table 10. ALKALINE DEGRADATION RATE CONSTANTS FOR
MALATHION MONOACIDS AND MALATHION DIACID IN WATER AT 21°.
Rate Constant
sec-*
k 1 = (3.1 ± 0.02) x 10-i
ad
k 2 (1.3 ± 0.1) x 10-i
cl il
k 3 (2.0 ± 0.1) x 10-i
ae
k* (1.8 ± 0.2) x 10-2
iMalathion monoacid disappearance rate constant.
2Malathion monoacid carboxyl ester hydrolysis rate constant,
3Malathion monoacid elimination rate constant.
*Malathion diacid elimination rate constant.
Figure 25 is a plot of malathion disappearance and
monoacid and dicarboxylic acid formation as a function of time
at 27°. The monoacid elimination rate constant (k ) is (2.0
3.6
70
-------
o 3:
0=0
o o
O «M
0-0
CO
oo=cj_
CM
O
«rt
O
CO
i
O-
•Sl
o
•d
•H
o
i
c
o
•H
,C
-P
rd
JL
CM
c
o
(0
n
Cn
0)
tJ
0)
(3
•H
H
rt
^
r-l
<
Cu §
0 o
o 8
0 «S1
v-S
CO
1 .
a. +•
§
0
o
c
8
M
•w*
0-0
CO
1
co=o_
CVJ
0)
O
o
71
-------
± 0.1) x 10~l M-» sec-* and the ester hydrolysis rate constant
is (1.3 ± 0.1) x 10-i M-» sec-1 at 27°. Based on rate
constants the monoacids are undergoing about 40% ester
hydrolysis and about 6Q% elimination.
— MALATHION
— • DPT A
— — MONOACID
— DIACID
Figure 25.
11.6
TIME, days
Time dependence of malathion disappearance
and product formation at 27°C
Temperature effects on the two competing reaction pathways
for monoacid degradation were not investigated. However, it
is anticipated that malathion dicarboxylic acid formation will
be favored at low temperatures and the elimination reaction to
give O,O-dimethyl phosphorodithioic acid and ethyl hydrogen
fumarate will dominate at elevated temperatures.
Because malathion diacid is anticipated to be a
significant breakdown product, we determined its stability
under alkaline reaction conditions. The diacid disappearance
second-order rate constant (kdd) is (1.8 ^ 0.2) x 10~2 M-*
sec-i at 27° (Table 10). These data indicate that at pH 9 the
degradative half-life would be about a year. Thus, under
72
-------
alkaline conditions malathion diacid is approximately 200
times less reactive than malathion.
Product studies at the end of one half-life were carried
out. Extraction and methylation (diazomethane) followed by
glc-ms analysis showed that O,O-dimethyl phosphorothionic acid
and thiomalic acid were products of hydrolysis (eq. 52).
(CH,0) -P-OH
k 32
S dd
|| pH = 11.30
(CH 0) -P-S-CHCO.H - >
32 2 27° + (52)
*'
HS-CHC02H
CH CO.H
4*
-------
^o
'c.
o
o
CO
•s
o
JC
a-
o
"§
O
JC
Q.
»/)
o
.c
Q.
O
co
o
'o
O
•H
A
-P
m
iH
(0
e
-o
O
o"
co
E
o
I
O
o>
to
o
i_
o
00
o
o
•H 01
-P (1)
m >
•o -H
0)
(U
t
2
-o
i
o
o"
+
o>
To
i_
CO
E
13
•H -0
rH -H
(tf O
x to
o
-H
CO
m
I g
-p -d
n) C
& ti
OJ
(D
co
O
74
-------
OXIDATION
Malathion is readily oxidized to malaoxon by a variety of
oxidizing reagents in the laboratory. Therefore, one might
anticipate oxidation by molecular oxygen in the environment.
However, experiments showed that malathion was stable in
oxygen-saturated water at acidic pH's for up to two weeks.
Therefore, oxidation by molecular oxygen does not appear
environmentally significant.
PHOTOCHEMISTRY
Malathion was distinguished fcy the fact that its direct
photolysis was the slowest of all the pesticides examined.
Since direct photolysis cannot compete with the hydrolysis of
malathion in the pH 5 to 9 range, no attempt was made to
characterize the direct photolysis products.
The decomposition of malathion was photosensitized by
acetone and by natural materials dissolved in water from the
Suwannee River (pH 4.7). Wavelength studies established that
ultraviolet light of wavelengths less than 340 nm caused the
decomposition in the river water. The photolysis half-life of
malathion in the river water was 15 hours under September
sunlight (lat. 34° N). Photolysis was not sensitized in water
from two other rivers, the Withlacoochee and the Santa Fe in
north Florida, suggesting that photosensitized decomposition
of malathion may not be a general phenomenon in natural
waters. Products from the sensitized transformations were not
determined, although it was established that malaoxon was not
a product. Malathion was found to be unreactive towards
sinqlet oxygen in both acetonitrile and distilled water (3
value > 37.5 in water).
REFERENCES
1 Paris, D. F., D. L. Lewis, and N. L. Wolfe. Environ. Sci.
Technol. ^:135 (1975).
2 Muhlmann, R., and G. Schrader. Z. Naturforsch. 12b;196
(1957).
3 Cowart, R. P., F. L. Bonner, and E. A. Epps, Jr. Bull.
Environ. Contain. Toxicol. j6:231 (1971).
4 Ruzicka, J. H., J. Thompson, and B. B. Wheals. J.
Chromatog. .31:37 (1967).
75
-------
5 Goldberg, M. C., H. Babad, Groothius, and H. R.
Christiansen. Geological Survey Prof. Paper 600D20
(1968) .
6 Chen, P. R., W. P. Tucker, and W. C. Dauterman. J. Agr.
Food Chem. V7: 86-90 (1969).
7 Bacon, W. E., and w. M. LeSuer. J. Amer. Chem. Soc.
76:670-676 (1954).
8 Huang, R. L. J. Chem. Soc. 1749-1755 (1956).
9 March, R. B., T. R. Fukuto, R. L. Metcalf, and M. G.
Maxon. J. Econ. Entomol. .49:185^195 (1956).
10 Fallscheer, H. R., and J. W. Cook. Ass. Off. Agr. Chem.
_39:691-697 (1956).
11 Welling, W., P. T. Blaakmeer, and H. Copier. J.
Chromatog. 45:281-283 (1970).
12 Payne, W. J., and V. E. Feisal. Appl. Microbiol. 11:339-
344 (1963) .
13 Frost, A. A., and R. G. Pearson. Kinetics and Mechanisms.
2nd Ed. New York, John Wiley and Sons, Inc. 1961.
14 Bender, M. E. Water Res. J:571 (1969) .
76
-------
SECTION IX
RESULTS S DISCUSSION: 2,4-D ESTERS
The widespread application of phenoxy herbicides has
prompted numerous studies of their biological,* chemical, * and
photochemical2 degradation. However, few publications
concerned with the chemical behavior of the esters of 2,4-D
have appeared, although 2,U-D usually is introduced into the
environment as an ester. The esters are often sprayed onto
surface waters to control aquatic weeds; they also enter lakes
and rivers in the runoff from field application.3
HYDROLYSIS
In accord with previous extensive studies of ester
hydrolysis,* three reactions considered in kinetic studies of
the 2,4-D esters were acidic, neutral, and alkaline
hydrolysis. As expected, product studies showed that 2,4-D
was formed quantitatively in these reactions.
Detailed studies of the reactions of the 2-butoxyethyl and
methyl esters indicated that the hydrolysis rate (-d[E]/dt)
could be expressed by equations 4 and 5.
A plot of log kobs vs pH for the butoxyethyl ester was
nearly V-shaped (Figure 27), indicating that neutral
hydrolysis is unimportant relative to acidic and basic
hydrolysis.* Table 11 presents rate constants for acidic and
alkaline hydrolysis in water, measured by hydrolyzing the
esters at low and high pH. Activation parameters were
determined by measuring hydrolysis rates at different
temperatures (Table 11).
The kinetic data in Table 11 can be used to calculate
hydrolysis rates as a function of pH and temperature.
Reaction kinetics become pseudo-first-order when the water is
buffered, as often occurs in natural water. Since kQH- is
much larger than kH+, the half-life for the pH 5 to 9 range
can be adequately expressed by
= JK693 (53)
* k [OH~]
OH"
Using equation (53) and structure-reactivity relationships
for the alkaline hydrolysis of various acetic acid esters,5 we
77
-------
CO
H
CO
>H
|_3
O
OH Q
Q 1
^H 1H
X
OH O
O EH
En !D
ffl
rfj 1
EH C
frf
Q Q
S
U <
H
EH rH1
is >H
53 a
H EH
§
• En
rH O
rH
(U
rH
Q
(Tj
EH
•H-
CO 3
< OJ
0)
rH
O
•H- g
a "^
< rH
(Tj
O
o
,—
1
o
(1)
CQ
t—
1
gj
«-^
1
EC
O
O
T—
1
O
(1)
U
T-
1
S3
-r
K
y
O*
g U
0) o
EH
CO
Q M
1 (U
^P -P
- CO
CN W
CM
• CM
•^r +|
+1 C\
00 (N
•<* rH
rH CM
I
i i
^* ^
• •
rH 0
+1 +1
rH VD
• •
o r-
CN H
»- tN t-
o o o
rH rH rH
M V* luil
i"i ^i tN
<--» *^** *— *.
ro ro ro
CN i— 1 CN
* * •
o o o
+1 +1 +1
CNJ in ro
o co t~-
• • •
CO CM rH
'~-* -~^ *-^
•O
•<* ro in
I I I
O O O
rH rH rH
XXX
-— . ^-~
CO 00 --»
rH rH in
• • .
O 0 O
+1 +1 +1
o^ vo
in or> o
• • •
VD (N (N
•**• *—r N_x
oo r^ t*^ r*^ oo oo
{N ^* VD 00 CM CM
(0
ft] ,Q
PQ W
03 §
CO
.p
C
(0
-P
CQ
O
U
(1)
-P
(0
^
t i
0)
•d
^|
o
i
•P
CO
•H
MH
O
^3
0)
CQ
g
O
J_l
UH
*d
0)
c
•H
g
^1
0)
(U
Q
O
•
rH
>-,
fH
4J
o;
g
XI
•
rH
t*1
r\
•P
(1)
X
O
JJ
3
PQ
nt
•
0)
^
H
5
'd
0)
-p
n)
rH
O
n
M^
<0
^l
4J
X
H
•d
78
-------
have estimated hydrolysis half-lives for commonly used 2,4-D
esters (Table 12). These data demonstrate that the chemical
hydrolysis rate for a given ester varies greatly within the pH
range normally found in natural waters (pH 5 to 9) and that
hydrolysis rates are much higher in basic than in acidic
waters. Ester structure also strongly influences hydrolysis
rates. The esters possessing ether linkages near the ester
carboxyl group generally hydrolyzed more rapidly than the
hydrocarbon chain esters. For example, the hydrolysis rate of
the 2-butoxyethyl ester was about an order of magnitude
greater than that of the 1-octyl ester.
-2
-3
o
o>
co
to
-4
O
O
-5
-7
0
Figure 27.
8
10
pH-rate profile for 2,4-D butoxyethyl
ester at 67° in water
The rapid hydrolysis of 2,4-D esters in basic distilled
water suggests that chemical hydrolysis is often the major
pathway for degradation of the esters in basic natural waters,
79
-------
Table 12. KIKETIC DATA FOR HYDROLYSIS OF 2,4-D ESTERS
IN WATER AT 28°C.
kOH
Ester
(days)
Methyl a
2-Propyl b
1 -Butyl b
1-0ctylb
2-Octylb
2-Butoxyethyla
2-Butoxy-
M-i sec-1
17.3
1.1
3.7
3.7
0.52
30.2
4.3
pH 9 (hrs)
1.1
17
5.2
5.2
37
0.6
4.4
pH 6
44
710
220
220
1500
26
180
methylethyl
Calculated from data in Table 11. Calculated assuming
structure-reactivity relationship for 2,4-D esters is the
same as that for acetic acid esters.5
To check this point, the disappearance O-f the butoxyethyl
ester was studied in water collected from the withlacoochee
River in South Georgia. As shown in Figure 28, the observed
disappearance rate matched the hydrolysis rate calculated from
data in Table 11, using the pH of the river water (pH 8.1).
Since 2,4-D esters hydrolyze much more slowly in acidic
waters, other processes such as those discussed below are
probably more important in such waters.
PHOTOCHEMISTRY
Conflicting reports concerning the photoproducts of 2,4-D
esters in water have appeared in the literature. Aly and
Faust6 reported that several 2,4-D esters yielded 2,4-
dichlorophenol as the photoproduct. Identification of this
photoproduct, however, was based upon a colorimetric technique
that would not have distinguished 2,4-dichlorophenol from
other phenolic products. Binkley and Gates7 have recently
reported that only monochlorophenoxyacetic acid (2- and 4-CPA)
esters result from photolysis of 2,4-D esters in water.
80
-------
However, other studies have shown that photolysis of
chlorinated aromatics in water results in replacement of
chlorine by a hydroxyl group.8
1.0
0.9
0.8
0.7
0.6
0 5
\
^ • Calculated
O Observed
r \
\0
\
\
1 1 1 1
0 100 200 300 400
TIME, min
\
50
Figure 28. Hydrolysis of 2,4-D butoxyethyl ester
in water from the Withlacoochee River
Our product studies were carried out in both water and
organic solvents. The latter studies were prompted by the
suggestion that organic films are a likely site for a
photoalteration of pesticides.'
In the hydrocarbon media, photoreaction involved
quantitative replacement of one of the chlorines by hydrogen,
likely via free radical intermediates.»o In both hexane and
hexadecane, replacment of the ortho chlorine accounted for >
90% of the photoreaction.
Studies in water were conducted at several concentrations,
At concentrations well exceeding their solubility limits (>
300 ppm)r the esters formed emulsions and the major
photoproducts, 2- and 4-CPA esters, were the same as those
formed in the hydrocarbon solvents. However, irradiation of
81
-------
very dilute (< 1 ppm), air-saturated solutions yielded
products very similar to those resulting from photolysis of
the 2,4-D acid.** Major photoproducts were 2,4-dichlorophenol
and compounds resulting from replacement of one chlorine by
hydroxyl. The lactone formed by elimination of alcohol from
the ester of 4-chloro-2-hydroxyphenoxyacetic acid was elucted
from the gas chromatograph. The alcohol elimination was
found, however, to be the result of thermal degradation in the
injection port (180°C). Results of these product studies are
summarized in Figure 29.
OCHjCOOR
Cl
OCHjjCOOR
OCHjfCOOR
Cl
Solution in organic solvent or
emulsion in H20
, air-saturated
OCHjCOOR
Cl
Heat
Figure 29. Photoreactions of 2,4-D esters
Kinetics of phenoxy herbicide photoalteration have
received very little study. Photolysis of 2,4-D and its
esters was reported to be rapid when short wavelength, high
intensity light is employed;6 however, Crosby and Wong12 have
reported that photoalteration of 2,4-D is relatively slow when
less intense, longer wavelength light is used.
Ultraviolet spectra of 2,4-D esters in water were very
similar to that of 2,4-D itself. Spectra of the methyl and
82
-------
butoxyethyl esters were identical. Although two absorption
maxima occur in the 280-295 nm wavelength range, the point of
maximum spectral overlap with midday summer sunlight occurs at
300 nm.
Quantum yields for photoalteration of 2,4-D esters (Table
13) were far lower than unity and were considerably lower in
water than in hydrocarbon solvents. Variations of pH (between
5 and 8) caused no change in quantum yield although the
phenolic photoproducts were further photodegraded at higher
rates in basic water. The quantum yields in hexane and
hexadecane, solvents with very different viscosities, were
identical, although photoreactions involving formation of free
radicals are often viscosity-dependent.13 The lack of
efficient photosensitization by acetone, a triplet sensitizer,
indicated that the direct photolysis proceeds from the excited
singlet state of 2,4-D esters. The extremely short lifetimes
of the excited states that lead to 2,1-D ester photolysis
preclude any significant quenching by species dissolved in
natural waters. Ester structure also affected the photolysis
rate. The methyl ester had a lower quantum yield and
therefore a lower photolysis rate than did the butoxyethyl
ester.
Table 13. DISAPPEARANCE QUANTUM YIELDS FOR PHOTOLYSIS OF
2.4-D ESTERS AT 313 nm (28°C).
Ester
BEE3
MEb
ME
BEE
Solvent
* Water, pH
Water, pH
Quantum
Yield, $
5.3
6.6
Water, pH 7.8
| n-Hexane
L n-Hexadecane
Water, pH
6.6
n-Hexadecane
n-Hexane,
Acetone-s ensitized
0
0
0
0
0
0
0
0
.056
.052
.056
.17
.17
.031
.13
.0080
aButoxyethyl. bMethyl.
83
-------
Photolysis rates of the butoxyethyl ester were calculated
as a function of time for a clear September day in the
Southern United States (Figure 30). Because ka varies
throughout the day, we calculated the photolysis rates of the
butoxyethyl ester integrated over the 12-hour period of
daylight. As shown in Table 14, measured photolysis rates for
September agreed closely with the calculated values.
o
f—I
X
CD
10
9
8
7
6
4
3
2
1
AM
678
9 10 11 12 1
TIME OF DAY
2345
6
PM
Figure 30. Computed dependence of 2,4-D butoxyethyl ester
photolysis rate upon time of day in the Southern
United States
We also investigated the photoaIteration of an ester in a
natural water sample. The butoxyethyl ester (1.0 ppm) was
irradiated (> 290 nm) in filter-sterilized water from the
Suwannee River (pH 4.7). The major photoproducts found in
distilled water (Figure 29) were also found in the river water
although the yield of 2,4-dichlorophenol was much higher and
the yields of the hydroxylated phenoxy esters were lower. As
with distilled water, no monochlorinated phenoxy esters were
formed in the river water. The overall photolysis was
actually more rapid in the river water than in distilled
84
-------
water. This acceleration is attributed to photosensitization
by humic materials dissolved in the river water. Since
photolysis of 2,4-D esters by direct absorption of sunlight is
slow, sensitized photolysis may be more important in many
natural waters.
The sensitized photolysis of 2,4-D esters in river water
cannot involve singlet oxygen. The 3 value for the
butoxyethyl ester was > 37.5 in water, indicating that singlet
oxygen reacts very slowly with 2,4-D esters.
Table 14. COMPARISON OF PHOTOLYSIS DATA FOR
2,4-D BUTOXYETHYL ESTER
Data
intg' **y-*
*
intg*' ^y1
Calculated half-life,
days
Empirical half-life,
days
Water
1.04
0.056
0.058
12
14
Medium
Hydrocarbon
1.07
0.17
0.18
3.8
4.0
REFERENCES
1 Loos, M. A. In: Degradation of Herbicides, Kearney, P.
C. and D. D. Kaufman (eds). New York, Marcel Dekker.
1961. p. 1.
2 Crosby, D. G., and A. S. Wong. J. Agr. Food Chem.
11:1052 (1973).
3 Hindin, E., D. S. May, and G. H. Dunstan. Residue Rev.
2:130 (1964).
4 Kirby, A. J. In: Comprehensive Chemical Kinetics, Vol.
10, Bamford, C. H., and C. F. H. Tipper (eds). New York,
Elsevier Publishing Co. 1972. Chapter 2.
85
-------
5 Jones, R. W. A., and J. D. R. Thomas. J. Chem. Soc.
B:661 (1966).
6 Aly, 0. M., and S. D. Faust. J. Agr. Food Chem. 12;541
(1964).
7 Binkley, R. W., and T. R. Gates. J. Org. Chem. 39:83
(1974) .
8 Pliminer, J. JR. Residue Rev. 33^47 (1971).
9 Rabson, R., and J. R. Plimmer. Science. 180;1204 (1973)
10 Lemal, D. M. , M. Fox, and W. C. Nichols. J. Amer. Chem.
Soc. .95:8164 (1973).
11 Crosty, D. G., and H. O. Tutass. J. Agr. Food Chem.
J_4:596 (1966).
12 Crosby, D. G., and A. A. Wong. J. Agr. Food Chem.
21:1052 (1973).
13 Pryor, W., and K. Smith. J. Amer. Chem. Soc. 92:5403
(1970).
86
-------
SECTION X
RESULTS 6 DISCUSSION: METHOXYCHLOR
HYDROLYSIS
Crosbyi has reported that at alkaline pH's, methoxychlor
eliminates hydrogen chloride to give the corresponding
diphenylethylene derivative. Merna et al.2 reported the half-
life of methcxychlor in distilled water buffered at pH 7 and 8
to be 270 days. In distilled water which had previously
contained fish, the half-life was reduced to 8 days. In
natural water samples, the half-life was 7-18 days depending
on the water sample employed.
Cristol3 reported the results of a kinetic study of the
dehydrochlorination (Eq. 54) of substituted diphenylchloro-
ethanes related to DDT in 95% ethanol-water. (Table 15) Based
on these data, DDT is about 200 times more reactive than
methoxychlor in the hydrogen chloride elimination reaction at
30°. The half-life predicted using these data is 12 days for
DDT and 7 years for methoxychlor at pH 9. Although these
studies were carried out in aqueous ethanol, they should
provide a good indication of relative reactivity in water.
HO
(CH.OC..H .) 0CHCC1 > (CHOCH ) C=CC1
J o * *• -3 OH~ 3642 2
methoxychlor 1, 1-bis (p_-methoxyphenyl)
2,2-dichloroethylene
Data from our studies with methoxychlor in water and 5%
acetonitrile-water are given in Table 16. Based on the rate
constants methoxychlor degradation is pH independent over the
pH range 3 to 10 due to reaction with water. Above pH 10 the
reaction rate increases with pH indicating dehydrochlorination
as the dominant pathway.
In agreement with this, at pH 13, 1,1-bis(£-methoxyphenyl)
-2,2-dichloroethylene (DMDE) is the major product. Below pH
10, only 2-436 theoretical DMDE was found. The major product
was not identified.
The pseudo-first-order rate constant for DDT, under
similar reaction conditions (pH 7) is 7.6 x 10~6 sec-1 at 85°.
This corresponds to a half-life of 25 hours. Thus DDT is
about 3 times less reactive than methoxychlor at this
temperature.
87
-------
Table 15. KINETIC DATA FOR DEHYDROCHLQRINATION OF
METHOXYCHLOR ANE DDT.a
Compound
DDT
Methoxy chl or
Temp . °
20.11
30.37
20.11
30.37
k M-»sec-»
2.48 x
6.96 x
9.18 x
3.06 x
10-2
10-2
10-5
10-*
E - kcal
cl
18.2
19.1
Taken from Reference 3.
While methoxychlor hydrolysis may be slow in distilled
water, at environmental temperatures its chemical degradation
may be fast in some natural waters. Methoxychlor degradation
was studied in natural river water samples at 65°. In water
samples from the Tomtigbee River (pH 8.0) and Alabama River
(pH 7.7), there was an apparent rate acceleration. However
the amount of degradation as a function of time was very
erratic and factors other than hydrolysis appeared operative.
Table 16. RATE CONSTANTS FOR METHOXYCHLOR DEGRADATION
IN WATER.a
pH
k -(M-»sec-M
25°
65°
kH 0(sec-»)
85°
3
5
7
9
11
13
2. 1 x
3.3 x 10-* 2.6 x
4.0 x 10-6 2.4 x
2.6 x 10-« 2.6 x
4.5 x
3.6 x 10-*
10-s
10-s
10-s
10-5
10-s
Rate constants were obtained either in water or 5% acetonitrile-
water. bSecond-order rate constant. cPseudo-first~order rate
constant.
-------
OXIDATION
Preliminary experiments with methoxychlor in natural water
samples at 65° indicated methoxychlor was being degraded at a
rate too fast to be accounted for by hydrolysis.
We therefore investigated the degradation of methoxychlor
in the presence of various concentrations of hydrogen peroxide
to determine if free radical degradation might account for the
increased reactivity (Table 17).
The reaction appears to be a free radical reaction based
on the catalysis by hydrogen peroxide and the effect of
species in the waters employed. The significance of free
radical oxidative degradation under environmental reaction
conditions is hard to assess at this time. However, the
importance of this pathway should not be overlooked.
PHOTOCHEMISTRY
Direct Photolysis
Products. Several studies have established that methoxy-
chlor, like DDT, is photodecomposed to a wide variety of
products. Other investigators have reported that the major
photoproducts in aqueous alcohol were p,£1-dimethoxybenzo-
phenone (MDCO) , jo-methoxybenzoic acid (MBA) and JD-
methoxyphenol (MP)*'S. Photolysis of methoxychlor in
butteroil6 reportedly yielded the above products along with
several dimeric products similar to those derived from
photolysis of concentrated solutions of DDT7. MacNeil and co-
workers8 found that 1,1-bis(j3-methoxyphenyl)-2,2-
dichloroethane (DMDD) and, to a lesser extent, 1,1-bis(p-
methoxyphenyl)-2,2-dichloroethylene(DMDE) were the major glc
amenable photoproducts in nitrogen-saturated heptane.
Given the multiplicity of products found in these studies
we were not surprised to find that the photolysis of
methoxychlor is highly dependent upon its reaction medium. In
agreement with the above studies, we found that the major
photoreaction in oxygen-free hexane involves stepwise
replacement of chlorine by hydrogen to initially give DMDD;
small amounts of DMDE were also formed (Figure 31). The
presence of air altered the nature of the products; MDCO
became the major product and DMDD, DMDE, and MP were minor
products (Figure 31).
89
-------
Table 17. HALF-LIVES FOR METHOXYCHLOR DEGRADATION WITH VARYING
AMOUNTS OF HYDROGEN PEROXIDE ADDED AT 65°.
HO Concentration (M)
Half-life (hrs)
0.0
0.0
1.0 x 10-1
8.0 x 10-2
8.0 x 10-3
8.0 x 10-*
8.0 x 10-5
58
< 1.7
2.01
17b
25b
Carried out in water at 1 x 10—7 M methoxychlor.
bCarried out at 1 x 10~6 M methoxychlor in water containing 5%
acetonitrile.
hv
CH,0
Degassed
CHCCI,
hi'
Air saturated
CH,0
CHCCI2H + CH30
DMDD
-+• CH30
C=CCI2
DMDE
+ DMDD + DMDE
OCH:
Figure 31. Products from direct photolysis of methoxychlor
in hydrocarbon solvents
Direct photolysis of methoxychlor (25 ppb) in air-
saturated distilled water by > 280 nm light yielded DMDE as
the major product (Figure 32). After partial reaction (^
conversion of methoxychlor), DMDE accounted for 60% of the
90
-------
products. The yield decreased with increasing conversion
because of subsequent photolysis of DMDE. After further
reaction (30 to 50% conversion) we also found £-
methoxybenzaldehyde but no MDCO in organic extracts of
photolyzed solutions (Figure 30). Control experiments
established that MDCO should have teen detected if it had
formed. Experiments were also conducted at higher
methoxychlor concentrations (0.3 g/liter) in an air-saturated
solution of 1.1 M water in acetonitrile; under these
conditions DMDE (60% yield) and a yellow non-volatile product
were formed. In degassed water-acetonitrile, DMDE was the
major glc amenable product but its yield was lower (^ 20%).
Addition of the free radical scavanger, 2-mercaptoethanol
(0.036 M), to the degassed water-acetonitrile mixture resulted
in a sharp decrease in the yield of DMDE and formation of a
new product, DMDD, the major product that formed in degassed
hexane.
CHO
CH30
CHCCh
hv
«-cH30-/ \-C=CCI2 +
DMDE
OCR.
Figure 32. Photoproducts of methoxychlor in pure water
These results suggest that the primary photochemical
process for methoxychlor is the same as that for DDT —
homolysis of one of its carbon-chlorine bonds to form free
radical intermediates [Eq. 55]9/»°. Subsequent chemical
reactions of the free radicals with themselves, oxygen, or the
solvent determine the nature of the products. Some possible
reactions that account for the major products in water are
shown in Equations 56-59 where Ar represents a jg-methoxyphenyl
group and RSH is a thiol.
Ar 2CHCC1 3
Ar2CHCCl2. + Cl-
Ar CHCC1 ' + 0
222
Ar CHCC1 • + 'OH
222
Ar CHCC1 ' + RSH
2 2
hv
Ar2CHCCl2' + Cl"
HC1
Ar2C=CCl2
Ar C=CC12
RS-
(55)
(56)
(57)
(58)
(59)
91
-------
The thiol scavenqed the free radical intermediate as shown in
Equation 59, forming DMDD and inhibiting DMDE formation.
Plimmer and his co-workers* ° suggested that the free radical
intermediate obtained from DDT photolysis reacts with oxygen
according to Eguations 60 and 61.
Ar2CHCCl2' + 02 - » Ar2CHCCl202- (60)
Ar2CHCCl202- - » - » Ar2CHCOCl (61)
In the case of methoxychlor in aqueous media, such a reaction
sequence would ultimately lead to formation of bis (jD-methoxy-
phenyl) acetic acid. This product was not found presumably
because reactions such as those in equations 56-58 occur much
more rapidly in the case of methoxychlor. Other studies have
demonstrated that abstraction of fcenzylic hydrogen atoms by
chlorine atoms and oxygen free radicals is markedly
accelerated by para-methoxy relative to para-chloro ring
substituents * * .
No detailed studies were made of the photooxidation of
methoxychlor in hydrocarbons. The observed formation of
dimethoxybenzophenone can be accounted for by a mechanism
similar to one suggested by Plimmer et al. >o for
photooxidation of DDT to dichlorobenzophenone (eq. 62-65) .
The alkylperoxy radicals, RO2., result from reaction of
solvent-derived free radicals with oxygen.
Ar CHCC1 + RO . - > RO,H + Ar_CCCl, (62)
2. 3 2 ^ Z o
i'-
Ar2CCCl3 + 02 - *• Ar2CCCl3 (63)
O.' O-
I2 I
2Ar CCC1 - > 2Ar CCC1 + O (64)
23 ^ -5 2.
O'
I
Ar2CCCl3 - > Ar2CO + 'CC13 (65)
Rates. The dearth of information concerning the rate of
direct photolysis of methoxychlor under sunlight prompted us
to examine the kinetics in some detail. Preliminary
experiments under natural sunlight indicated that the
photolysis of methoxychlor (30 ppb) in either hexane or
distilled water was quite slow. Exposure to U8 hours of
92
-------
midday (1000 to 1400 E.D.T.) sunlight (June, lat. 34°N)
resulted in less than 10% decomposition in both solvents.
Assuming first-order kinetics for the photolysis, these data
indicated that at least 300 hours of midday sunlight would be
required to decompose half of the pesticide.
Quantum yields for direct photolysis of methoxychlor were
measured in hexane, 1.1 M water in acetonitrile, and distilled
water (Table 18). The quantum yield in hexane was similar to
the value of 0.16 previously reported for DDT in hexane9. The
quantum yield in a degassed 1.1 M water in acetonitrile
mixture was not decreased by the addition of 0.036 M 2-
mercaptoethanol, a free radical scavenger, or 0.004 M cis-1,3-
pentadiene, a triplet-state quencher. The latter experiment
indicates that dissolved substances in natural waters are too
dilute*2 to quench the photolysis by energy transfer
processes*3. The low quantum yields for DDT and methoxychlor
may in part be due to recombination of the free radicals
initially formed in the photolysis (eq 55, reverse reaction).
Table 18. QUANTUM YIELDS FOR DIRECT PHOTOLYSIS OF METHOXYCHLOR
Disappearance
Solvent Quantum Yield
n-Hexaneb 0.12
1.1M Water in acetonitrileb 0.32
Water0 0.3
^Maximum transmittance of chemical filter was 277 nm.
Methoxychlor concentration was 0.00300 M.
Methoxychlor concentration was 30 ppb.
The magnitude of the specific sunlight absorption rates of
pesticides depends upon the degree of spectral overlap between
their electronic absorption spectra and the spectrum of
sunlight at the earth's surface. The cutoff for solar
radiation is about 295 nm»*. The spectra of carefully
purified samples of DDT and methoxychlor in hexane are
compared in Figure 33. In water-acetonitrile mixtures the
extinction coefficients of methoxychlor at wavelengths > 295
nm are slightly higher than in hexane; the reverse is true for
DDT. However, even in the polar solvents, the extinction
coefficients of methoxychlor were very low. Specific sunlight
93
-------
ti
(d
X
0)
C
•H
CQ
EH
Q
D
id
O
iH
X
U
&
O
45
•P
M-l
O
rd
M
4J
O
QJ
£4
cn
O
•H
-P
O
Ul
XI
•H
C
O
O
0)
•H
W
n
0)
94
-------
absorption rates for both compounds were calculated from the
above spectral data and Bener* s table of intensities of
natural ultraviolet radiation1*.
Kinetic parameters for the direct photolysis of DDT and
methoxychlor during mid-summer in the central United States
(lat. 40°N) are compared in Table 19. The half-lives, t^ ,
in Table 19 represent the period of midday sunlight required
to decompose the pesticides to half of their original
concentration near the surface of a water body [t^ = 0.693
(k (j))"1]. The following conclusions were derived from
examination of these parameters:
• Direct photolysis of methoxychlor is at least 300
times more rapid than that of DDT in water.
• Direct photolysis of methoxychlor is nearly six times
more rapid in aqueous than in hydrocarbon media.
However, the hydrocarbon data are probably relevant
only to oil slicks, not surface films formed by
naturallyoccurring surface-active substances.
Hautala*5 has found that the electronic absorption
spectra and quantum yields for reaction of pesticides
are different in micelles formed by surfactants in
water than in pure water or hydrocarbons. Similar
differences probably occur when pesticides are
intimately associated with surface-active films.
• Both DDT and methoxychlor are photolyzed very slowly
by sunlight in pure water or hydrocarbons. The data
in Table 2 pertain to mid-summer, midday sunlight;
the half-lives are longer during other seasons and in
the morning or afternoon. Midday half-lives for
methoxychlor and DDT, averaged over all seasons, are
1,100 hours and > 460,000 hours, respectively, in
water at lat. 40°N. Assuming conservatively that the
average daily half-lives are SOX longer than the
midday half-lives, the direct photolysis half-lives
(in 12-hours days) of methoxychlor and DDT in water
are 4.5 months and greater than 150 years
respectively (lat. 40°N).
The calculated photolysis half-life of methoxychlor for
midsummer is an increasing function of northern latitude
(Figure 34). For example, the half-life in the central U.S.
is approximately 30% longer than in the tropics.
95
-------
Table 19. KINETIC PARAMETERS FOR THE DIRECT PHOTOLYSIS OF DDT
AND METHCXYCHLOR IN THE CENTRAL UNITED STATES.3
Pesticide
Methoxychlor
DDT
Medium
Hydrocarbon
{
Water
Hydrocarbon
£
Water
10«k
(sec-*)
390
870
4.3
0.85
(sec-*)
0.12 470
0.3 2800
0.16b Z6.8
c
(hrs)
4,100
690
280,000
>227,000d
Calculated for midday sunlight during midsummer at latitude
40°N.
Taken from Mosier et al.9
°Data not available.
Minimum half-life calculated assuming quantum yield of unity.
3200 -
./. 2400 -
1600 -
800 -
0
20 40 60
LATITUDE. °N
80
Figure 34.
Computed midday half-lives for direct photolysis
of methoxychlor in water during summer
96
-------
Direct photolysis of the 1,1-diphenylethylene derivative,
DMDE, was found to be much more rapid than that of
methoxychlor. Quantum yields (313 nm) for photolysis of DMDE
were found to be 0.20 in hexane and 0.30 in distilled water;
Mosier et al.» reported that the quantum yield for DDE in
hexane is siirdlar, 0.26 at 254 nm. The calculated midday
half-lives for DMDE (summer, central United States) were 45
and 60 minutes in water and hydrocarbons, respectively.
Photolysis rates measured under natural sunlight agreed
closely with the calculated values. DDE also photolyzes much
more rapidly than DDT in hydrocarbon media. The calculated
midday half-life of DDE is about four hours under the above
conditions whereas that of DDT is 280,000 hours. These
results provide an interesting contrast to the findings of
Moilanen and Crosby*' that DDT photolyzes more rapidly than
DDE in the vapor phase.
The solar intensities employed in the above calculations
are based upon average atmospheric ozone amounts (O,) for a
given latitude and season. In fact, natural variations of ±
5% in (O,) occur over a period of years and (O3) varies
longitudinally within a given latitude*7. Recently, several
scientists have expressed concern that certain human
activities may be leading to depletion of stratospheric
ozone18. Our calculations indicate that variation in ozone
amount affects the direct photolysis rate of DDT more than
that of methoxychlor (Figure 35). Generally, each 5X
reduction in ozone amount should result in about a 10%
increase in rate for EOT and a 6% increase for methoxychlor.
Sensitized Photolysis
Although the photolysis of methoxychlor in pure water is
slow, we found that photolysis under sunlight was rapid in
certain filter-sterilized natural waters (Table 20). The
natural waters were collected from rivers in South Carolina,
Alabama, Georgia, and Florida; pH values ranged from 4.7
(Suwannee River) to 8.2 (Withlacoochee River). Dark controls
showed no decomposition in any of the water samples over the
period of sunlight exposure.
Detailed product studies were not carried out. However,
the glc traces of organic extracts of the photolyzed solution
all had peaks with the same retention time as DMDE. In the
Suwannee River water, DMDE formation almost completely
accounted for methoxychlor disappearance at 30% conversion of
the methoxychlor. The DMDE photolyzed more slowly than
methoxychlor in the Suwannee River water; its low photolysis
rate was attributed to light screening by the ultraviolet-
absorbing materials in the river water.
97
-------
<
CtL
00
o
Ol
O
I—
O
10 20 30 40
OZONE REDUCTION, percent
50
Figure 35.
Calculated effects of ozone reduction upon
photolysis rates of DDT and methoxychlor
Methoxychlor photodecomposed rapidly in distilled water
containing a commercially available derivative of decayed
plant materials, Aldrich "humic acid" (Table 20). This
material was used because its electronic absorption spectrum
was very similar to that of materials dissolved in our natural
water samples and because it is likely to contain many of the
same chemical moieties. The observed sunlight photolysis rate
using 20 ppm of this model substance was roughly comparable to
that found in several natural waters.
At present, we know very little about the mechanism(s) for
sensitized photolysis in natural waters. However, energy
transfer photosensitization cannot account for our results.
The triplet state energy of methoxychlor, determined from its
phosphorescence spectrum, was found to be 80 kcal mole-1.
Since triplet energy transfer occurs efficiently only when the
triplet energy of the sensitizer equals or exceeds that of the
energy acceptor13 only high energy sensitizers can transfer
energy to methoxychlor. Such sensitizers may be present in
98
-------
natural waters, but they absorb only the short wavelength
ultraviolet component of sunlight, so their rates of sunlight
absorption are not rapid. The quantum efficiencies of energy
transfer under natural conditions are also decreased by the
presence of competing energy acceptors such as dissolved
oxygen. Moreover, studies using filtered light from a mercury
lamp indicated that visible light (> 400 nm) caused
decomposition of methoxychlor in the Suwannee River water; the
energy of visible light is < 71 kcal mole-* i8, far lower than
the minimum energy required to electronically excite
methoxychlor (80 kcal mole-*).
Table 20. HALF-LIVES FOR PHOTCDECCMPOSITION OF METHOXYCHLOR
(40 ppb) UNDER SUNLIGHT IN VARIOUS RIVER WATERS3
Photolysis
Water pH Half-life, hrb
Distilled
Suwannee River
Tombigbee River
Alabama River
Withlacoochee River
South Georgia Stream
20 ppm "humic acid" in
6.3
4.7
7.6
7.7
8.2
7.2
5.2
> 300
2.2
5.4
2.9
— c
_ c
7.3
distilled water6
aSolution in sealed quartz cells exposed to midday May
sunlight, latitude 34°N.
Calculated assuming first-order kinetics, expressed as hours
of midday sunlight.
^No detectable photolysis after two hours exposure.
Colleted near Thomasville, GA.
eHumic acid obtained from Aldrich Chemical Company.
Also, the possibility was investigated that singlet
oxygen, an excited form of oxygen19 (Trozollo, 1971), may play
a role in the light-induced decomposition of methoxychlor in
natural waters. Our results indicated that both methoxychlor
and DMDE are unreactive towards singlet oxygen.
99
-------
Other possible mechanisms were not investigated, although
the exciplex mechanism suggested by Miller and Narang2<> for
the amine-induced photolysis of DD1 is one attractive
possibility. Alternatively, decomposition of methoxychlor may
result from attack by free radicals generated by photolysis of
materials in the natural water. The latter is a common
mechanism for indirect photolysis in the atmosphere21.
REFERENCES
1 Crosby, D. G. Ann. NY Acad. Sci. 160:82 (1969).
2 Merna, J. w., M. E. Bender, and J. R. Novy. Trans. Amer.
Fish. Soc. 101;298 (1972).
3 Cristol. S. J. J. Amer. Chem. Soc. 67:1494 (1945).
4 Crosby, D. G. Presented at the 158th National Meeting of
the American Chemical Society, New York, NY, September
1969.
5 Fernadenz, M. M. S. Thesis. University of California,
Davis, CA. 1966.
6 Bradley, R. L., and C. F. Li. J. Dairy Sci. 52:27
(1969) .
7 Fleck, E, E. J. Amer, Chem. Soc. T\:1034 (1949).
8 MacNeil, J. D., R. W. Frei, S. Safe, and O. Hutzinger. J.
Assoc. Offic. Anal. Chem. _55:1270 (1972).
9 Mosier, A. F., W. D. Guenzi, and L. L. Miller. Science.
164; 1083 (1969).
10 Plimmer, J. R., U. I. Klingebiel, and B. E. Hummer.
Science. 167:67 (1970).
11 Pryor, W. A. Free Radicals. New York, McGraw-Hill Book
Co. 1968. p. 170-177.
12 Hutchinson, G. E. A Treatise on Limnology. Vol. I. New
York, John Wiley and Sons, Inc. 1957. p. 878-902.
13 Turro, N. J. Molecular Photochemistry. New York, W. A.
Benjamin, Inc. 1965. Chapter 5.
14 Bener, P. Approximate Values of Intensity of Natural
Ultraviolet Radiation for Different Amounts of Atmospheric
100
-------
Ozones. U.S. Army Report DAJA 37-68-C-1017, Davor Platz,
Switzerland. June 1972.
15 Hautala, R. Personal Communication. University of
Georgia, Athens, GA. 1975.
16 Maugh, T. H. Science. ^60z51Q (1973).
17 London, J. Beitrage zur Physik der freien Atmosphare.
36:254 (1963) .
18 Hammond, A. L. , and T. H. Maugh. Science. 186;335
(1974).
19 Trozollo, A. M. (ed.) International Conference on Singlet
Molecular Oxygen and its Role in Environmental Sciences.
Ann. NY Acad. Sci. 171; 1 (1970).
20 Miller, L. L., and R. S. Narang. Science. 169;368
(1970).
21 Altschuller, A. P., and J. J. Bufalini. Environ. Sci.
Tech. 5:39 (1971).
101
-------
SECTION XI
RESULTS fi DISCUSSION: CAPTAN
HYDROLYSIS
Several reports dealing with the stability of captan in
water have appeared in the literature. Melnikov* reported the
products of captan hydrolysis are 4-cyclohexene-1,2-
dicarboximide (THP), carbon dioxide, hydrochloric acid, and
elemental sulfur. Daines et al2. reported similar products,
but instead of sulfur, they reported hydrogen sulfide as a
product. Daines et al2. also reported that hydrolysis of
captan in a 2% slurry occurred rather slowly at temperatures
below 110°F, but above 110°F hydrclysis was rapid. Von Rumker
and Horay3 reported that captan half-life decreased with
increasing pH. The half-life at pH 4 (20°) was 4 hours and at
pH 10 (20°) less than two minutes. They found decreased half-
lives at 40°.
Interest in the chemical reactivity of captan is
exemplified by the considerable attention placed on the
reaction of captan with thiols. Several of these studies will
be reinterpreted in view of our hydrolysis studies. We
investigated the rate of hydrolysis of captan in detail and
compared it with two other trichlorothiomethyl containing
fungicides, folpet and captafol (Figure 36).
I\I-S-CCI3
N-S-CCI3
0
Figure 36.
N-S-CCI2CHCI2
0
CAPTAFOL
Chemical structures of captan,
folpet, and captafol
102
-------
Kinetic studies employed a glc method to determine the
concentration of captan at varous time intervals. With column
temperatures above 200°, a second peak with a shorter
retention time appeared in the chromatogram; since both peak
sizes varied with each sample injection, captan was assumed to
be decomposing. By using a short column (21) and a column
temperature of 160°, the anomalous peak was eliminated. The
linearity of response under these conditions over the
concentration range of 1.2 x 10—5 M to 1.6 x 10~6 M is shown
in Figure 37 as a plot of peak area vs. concentration.
1450
PEAK AREA
2350
Figure 37.
Standard curve for captan response to
electron capture detector
It was necessary to add captan to the aqueous solutions
with a carrier solvent (acetonitrile) because the rate of
solution for water was slow compared to the rate of
103
-------
hydrolysis. In a separate kinetic experiment the
concentration of acetonitrile was reduced with no detectable
effect on the rate constant. Thus the *\% organic solvent was
not anticipated to have a pronounced effect on the rate
constant when compared to water.
Kinetic Studies
Captan (1 x 10-* M) was first hydrolyzed in an open vessel
in non-buffered water (595 acetonitrile) and captan
concentration along with pH was followed as a function of
time. The results are shown graphically in Figure 38. As
captan hydrolyzes the pH of the solution decreases indicating
acidic product formation. Also the decrease in pH apparently
retards the rate of captan degradation.
- 8
- 7
I £
- 5
0
100 200 300
TIME, min.
Figure 38. Captan concentration and pH vs. time
in non-buffered water at 28°C
Because acidic products altered the pH of the reaction
solution, captan kinetic studies were done under pH buffered
reaction conditions. These studies were carried out over the
104
-------
range of pH 2 to pH 8 to evaluate the contributions of acid,
base, and water.
Plots of log concentration vs. time are shown in Figure 39
for captan, folpet, and captafol in the buffered pH range of
7.0-7.2. These plots are linear through at least one
pesticide degradation half-life and indicate pseudo-first-
order reaction conditions. Similarity of the chemical
structure of these three pesticides suggests the rates of
hydrolysis would be similar. The pseudo-first-order rate
constants at 28° are: captan, kobsd = 6.5 x 10~5 sec-* (pH
7.07); folpet, kobsd = 1.4 10-* sec~» (pH 7.1U); captafol,
k . , = 7.7 x 10-s sec-* (pH 7.17).
obsd
0
30
60
90 120
TIME,min.
150
170
Figure 39. Pseudo-first-order plots for the hydrolysis
of captan, folpet, and captafol
The hydrolysis studies for captan were extended to other
pH's to determine the contribution of acid and alkaline
degradation and to evaluate the contribution of water. Table
21 lists the pseudo-first-order rate constants at the various
pH values studied and the buffers employed. The data are
plotted in Figure HO as log k vs. pH to give a pH-rate
105
-------
Table 21. PSEUDO-FIRST-ORDER RATE CONSTANTS FOR CAPTAN HYDROLYSIS
AT SEVERAL pH's AND TEMPERATURES
Temp pH
8° 8.
28° 1.
2.
3.
4.
5.
6.
7.
7.
7.
7.
7.
8.
8.
48° 3.
39
97
48
39
83
16
10
07
45
66
82
93
08
25
3
(1.
(1.
(1.
(1.
(1.
(1.
(4.
(6.
(1.
(2.
(3.
(4.
(6.
(1.
(3.
49
79
67
87
97
87
6
5
5
81
9
8
5
1
57
k(sec )
± 0
± 0
± 0
± 0
± 0
± 0
± 0
± 0
+ 0
± 0
± 0
± 0
± 0
± 0
± 0
.03)
.06)
.08)
.08)
.08)
.09)
.3)
.5)
.1)
.2)
.1)
.2)
.1)
.1)
.09')
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
Buffer or Acid
~4 H BO /NaOH
~5 HC1
-5
-5
" 5 NaOAc/HOAc
-5
~5 Na^HPOVNaHPO,
24 4
-5
-4
-4
~4 H2 BO /NaOH
-4
-4
-3
_4 a
4 HC1
Acid
28.8 ± 0.7 kcal/mole
ASf 13.0 ± 2.5 eu
Base
AHf 22.8 ± 1.6 kcal/mole
ASt 2.2 + 5.4 eu
Average of two values.
106
-------
profile. Between pH«s 2 and 5 there is no change in the
pseudo-first-order rate constant with pH as the slope of the
line is essentially zero. Above pH 7 the pseudo-first-order
rate constant increases with increasing pH and the slope of
the line is +1. In the pH range 6-7 the plot is not linear
due to competing reactions of captan with water and hydroxide.
-2
o
OJ
1/1
o
o
-3
-4
o-o-o-
I
11
PH
Figure 40. Plot of log k vs. pH for captan hydrolysis
at 28°C in buffered aqueous solution
Based on this interpretation of hydrolysis the rate of
disappearance of captan under buffered reaction conditions is
given by
d[captan]
dt
= k Fcaptan] + k [captan][OH ]
H2° OH"
(66)
where [captan] is the captan concentration, [OH] is the
hydroxide ion concentration, kH_o is the pseudo-first-order
rate constant for reaction with water and k is the second-
on'
107
-------
It was not possible to determine the solubility of captan
in water even at low pH's because the rate of hydrolysis is
fast enough over the pH range investigated to compete with the
rate of solution. When 10 mg of finely divided captan was
stirred in 100 ml of water for 16 hours at room temperature,
no captan could be detected in the water down to 10~6 M
concentration. Thus, it appears that for captan the
hydrolysis half-life in aquatic systems will be influenced by
its rate of solution.
Hydrolysis was also studied in the presence of thioethanol
at 1.2 x 10-s M captan, 1 x 10~3 M thiol, and pH 3.64. The
reaction followed first-order kinetics through at least one
half-life, but then deviated from first-order kinetics as was
shown by an increase in the rate constants during the second
half-life. The effect of the thioalcohol was not further
investigated and has not been intepreted.
Natural Water Study
To determine if any chemical species in the aquatic
environment would alter the kinetics of reaction, we carried
out the hydrolysis in a natural water sample obtained from the
Tombigbee River, South Carolina (pH 7.0). A plot of the datei
(Figure 42) shows that degradation follows pseudo-first-order
kinetics. While this is only one natural water sample, the
half-life is in good agreement with the laboratory data (170
min. measured vs. 155 min. calculated).
Product Studies
To facilitate analysis for products the concentration of
captan was increased to 2 x 10-* M by employing 10%
acetonitrile-water as the solvent. At the end of 10 half-
lives at pH 9,0 an aliquot of the reaction solution was
qualitatively analyzed for chloride. The addition of silver
nitrate resulted in formation of a white precipitate
interpreted to be silver chloride, thus indicating chloride as
a reaction product. The addition of mercuric chloride to a
separate aliquot did not produce a precipitate. Thus sulfide
or thiols are not products. Also hydrogen sulfide and carbon
disulfide could not be detected in the reaction solution by
glc-ms analysis of the toluene extracts of a separate aliquot.
Extraction at pH 9.0 with chloroform gave 4-cyclohexene-
1,2-dicarboximide (THP). This was identified by glc retention
time, melting point, ir spectra, and mass spectral data.
108
-------
order rate constant for reaction with hydroxide ion. The
average pseudo-first-order constant (kn2o) £°r t^6 reaction of
captan with water in the pH range 2-6 is (1.8±0.1) x 10~s
sec~». The average second-order-rate constant (koH~) ^or
alkaline hydrolysis in the pH range 7-9 is (5.7±0.4) x 102 M-I
secr-i.
Computed from the above rate constants and equation 6r a
pH-half-life profile is shown in Figure 41. It is apparent
from this diagram that the maximum hydrolysis half-life for
captan is about 1/2 day. While folpet and captafol were not
included in these detailed pH studies, it is anticipated that
they would give similar pH-rate profiles.
500
E
£ 100
S 50
10
O—-I
5
PH
Figure 41. pH-half-life profile for captan hydrolysis
in water at 28 °C
The temperature effect was determined for both the neutral
and alkaline contributions. The activation parameters are
included in Table 21. As anticipated the rates of both
reactions increase with increasing temperature.
109
-------
o
o
<
Q_
O
O
O
10
-6
CALCULATED ty2 = 155 min.
I
0
40
80 120
TIME, min.
160
200
Figure 42. Degradation of captan in water
from the Tombigbee River
Subsequent acidification (pH 2.5) and extraction of the
reaction mixture with chloroform gave a residue which resisted
crystallization. The ir spectrum of this residue differed
from that of the THP, but the solid probe mass spectra were
compatable with a mixture of elemental sulfur and imide. The
gas-liquid chromatogram of the residue had two peaks, whose
mass spectra were identical to that of THP and 4-cyclohexene-
1,2-dicarboxylic anhydride. The latter may have formed as a
result of thermal decomposition of proposed intermediates.
Major products that were identified are summarized in Figure
43.
Pathway of Degradation.
Captan has three different functional groups, the
trichloromethyl, sulfenimide, and carboxyl groups, which are
labile to hydrolysis; the reactivity of these groups should be
considered when invoking a degradative pathway. Based on
hydrolysis data for phthalimide*, the carboxyl functional
110
-------
0
N-S-CCI3 + 2H20
0
Figure 43. Major products for the hydrolysis of captan
groups of captan would not react under our experimental
conditions. On the other hand other compounds with
sulfenimide and trichloromethyl groups undergo hydrolysis
readily.
We have considered two reaction pathways for captan
hydrolysis similar to the heterolytic mechanism proposed by
Owens and Blaak5 for the reaction of captan with thiols
(Figures 44 and 45). The first pathway (Figure 44) involves
hydrolysis of the trichloromethyl moiety by nucleophilic
substitution of chloride by water or hydroxide*.
Consideration of this pathway is based on literature
hydrolytic data for similar trichloromethyl compounds which
are readily hydrolyzed. For example, methyl trichloromethyl
ether7, trichloromethyl mercaptan8, and trichloromethyl
benzene9 react by this mechanism. In the case of captan rapid
nucleophilic substitution and subseguent decarboxylation would
give the THP and sulfur as products. The general reaction
pathway is outlined in Figure 44 where we have considered only
the overall pathway.
The second pathway to be considered (Figure 45) involves
mucleophilic substitution at sulfur10. Sulfenimides are
cleaved by acid and alkali to give the free imide and sulfenic
acid11 ,12/13. For captan the primary products would be THP
and the unstable trichloromethyl sulfenic acid intermediate11,
which would rapidly decompose to hydrogen sulfide and carbon
dioxide via the reactive intermediates, thiophosgene and
carbonyl sulfide. Trichloromethyl sulfenic acid was the
postulated intermediate in the hydrolysis of trichloromethyl
sulfenyl chloride to hydrogen sulfide and carbon dioxide11.
Based on our kinetic data one cannot distinguish between
the two mechanisms outlined above. Both pathways are
compatable with the rate determining step as the reaction of
water or hydroxide with either the trichloromethyl or
sulfenimide sulfur.
Ill
-------
The products that we found are consistent with the
mechanism in Figure 44 and are the same as those reported by
Melnikov*. However, we were not able to verify this pathway
by isolating or spectrally identifying any of the proposed
intermediates. Any intermediates formed by the reaction with
water or thiols are very reactive and would have short life
times in the aquatic environment. The end products of
degradation with the exception of the THP are naturally
occurring compounds and would likely have little impact on the
environment.
Conclusions
Captan undergoes degradation in water with a maximum
hydrolysis half-life of 1/2 day. The reaction is pH
independent over the pH range 2-6 and pH dependent from pH 6
to 9. The products of the reaction were identified as sulfur,
chloride, and THP. Hydrolysis of captan, as well as of folpet
and captafol, is very fast and is likely to be the predominate
pathway of degradation in the aquatic environment.
PHOTOCHEMISTRY
Our studies indicated that direct photolysis of captan
cannot compete with its rapid hydrolysis, even when we
employed ultraviolet light that is 80 times more intense than
sunlight. This teing the case, no attempt was made to
characterize direct photolysis products. The hydrolysis
product of captan, THP, is also unreactive due to lack of
sunlight absorption.
Although their direct photolysis is unimportant, light may
play a role in the decomposition of captan and THP. Sunlight
may accelerate the autooxidation of these compounds (Figure
46); cyclohexene itself is autooxidized to an allylic
hydroperoxidei*. Captafol, another cyclohexene derivative
(Figure 36), may be oxidized in a similar fashion.
Although cyclohexene itself reacts very slowly with
singlet oxygen15, our studies revealed that captan and THP are
quite reactive. 3 values for both compounds were determined
by measuring quantum yields (578 nm) at various concentrations
(Figure 10)t5; methylene blue was employed as photosensitizer.
A representative plot of the data for captan is shown in
Figure 47; 8 values are determined from the slopes of such
plots (Figure 10). The photosensitized oxygenations of captan
and THP were efficiently quenched by 1,4-diazabicyclooctane
(DABCO),*6 and by $-carotene, * 7 confirming that the reactions
were mediated by singlet oxygen (Figure 48). Comparison of
our data with those of Foote*5 (Table 22) indicates that
112
-------
0
I\I-S-CCI3
H,O
0
0
I\I-S-CCI2OH
0
+ HCI
I
H2O
0
s +
+
C02
N-H -*•
0
0
N-S-COOH
0
2HCI
Figure 44. Mechanism for hydrolysis of captan that involves
nucleophilic displacement of chloride
0
0
IM-S-CCI3
•»
OH'
0
2HCI + COS -«-
N-H + HOSCCI3
H2O
S
II
C
ci
X
x
+ HCI
HoO
HSCCI
HoO
H2S + C02
Figure 45. Mechanism for hydrolysis of captan involving
nucleophilic substitution at the sulfur atom
113
-------
hv
02
X=-SCCI3( H
Figure 46. Postulated products for light-initiated
autooxidation of captan and 4-cyclohexene-
1,2-dicarboximide
captan and THP are 42 times more reactive than cyclohexene.
The relatively high reactivities of captan and THP are
attributed to conformational effects.
Table 22. SINGLET OXYGEN REACTIVATES OF CAPTAN,
4-CYCLOHEXENE-1,2-DICARBOXIMIDE, AND CYCLOHEXENE
3. M
Acceptor Acetonitrile
Captan 0.30
4-Cyclohexene- 0.30
1 , 2-dicarboximide
Cyclohexene a
Methanol
d-3)b
d.3)b
55
Water
4.5
4.5
-
n from Reference 15. Calculated assuming $ values is
4.3 times greater than in acetonitrile than in methanol (Ref.
19).
Photooxygenation products of captan and its hydrolysis
products included allylic hydroperoxides derived from the well
dol documented "ene" reaction of olefins15 (Figure 49). The
hydroperoxides were reduced to corresponding alcohols by
triethylphosphite. Note that these products differ from the
114
-------
0)
J J I I
I I
o
o
LTV
O
CD
<
Q_
,0,
CVJ
vO
CVJ
CO
-P
H-H
o c
IH o
-P
t3 0)
•H u
0) (0
•H
-P (U
C DI
(0 >i
3 X
tJ1 O
0) 4J
JC 0)
-P H
Cn
m c
O -H
03
0)
O A
G -P
OJ -H
ft (0
0) -P
'd a.
rtJ
c o
o
•H m
-P O
10
H C!
•P O
fi-rH
(U -P
o o
a nJ
O (U
U M
0)
M
D
tJI
•H
115
-------
autooxidation products (Figure 46). Folpet, shown in the
lower part of Figure 46, was not readily oxidized by singlet
oxygen (3 > 11 in acetonitrile), indicating that the
trichloromethylthio moiety in captan is unreactive towards
singlet oxygen. Photoxygenation of captan in acetonitrile
also led to formation of THP and other minor unidentified
products. We postulate that these products result from
reaction of unreacted captan with the hydroperoxide products,
50
40
s30
o
UJ
Q_
20
10
NO DABCO ADDED
5.0xlO"4MDABCO
ADDED
0 **•
0
Figure 48.
30 60 90
TIME, min
120
150
Effect of 1,4-diazabicyclooctane upon the
photosensitized oxygenation of captan
Because singlet oxygen reactions can be sensitized by
substances in the environment such as chlorophyll*8 that
adsorb the intense visible and/or near infrared portion of
sunlight, it is possible that these reactions are important
under natural conditions.
We found that the half-life of captan (10~s M) in water
(pH 5) containing 10~* M methylene blue was only 45 min, under
midday, summer sunlight. The hydrolysis half-life measured in
a dark control was about 9 hours (Figure 41), indicating that
116
-------
photooxielation accounted for most of the degradation under
sunlight. In a highly colored natural water sample (pH 4.7)
from the Suwanee River, the degradation rate of captan under
midday sunlight was about 30% more rapid than in the dark.
The lower photoxidation rate in the river water was probably
mainly due to the presence of phenolic humic materials;
phenols efficiently inhibit photooxidations involving singlet
oxygen*7. The latter results indicate that hydrolysis of
captan is probably more rapid than photooxidation in natural
waters. Photooxidation of captan on leaf, fruit, or soil
surfaces may be important, however.
[H]
X = H, SCCI3
NSCCh
'02
-X-
Figure 49. Products from the reaction of captan and
4-t:yclohexene-l, 2-dicarboximide with
singlet oxygen
REFERENCES
1 Melnikov, N. N. Chemistry of Pesticides, New York,
Springer-Verlog. 1971. p. 247.
2 Daines, R. H., R. J. Lukens, E. Brennan, and I. A. Leone.
Phytopathology. JT7:567 (1957).
3 von Rumker, R., and F. Huray. Pesticide Manual. Vol. 1.
U.S. Department of State, Agency for International
Development. 1972.
4 Zerner, B., and M. L. Bender. J. Amer. Chem. Soc.
8^:2267 (1961).
5 Owens, R. G., and G. Blaak. Contribs. Boyce Thompson
Inst. 2j6:475 (1960).
117
-------
6 March. J. Advanced Organic Chemistry Reactions,
Mechanisms, and Structure. New York, McGraw-Hill Book Co.
1968.
7 Hine, J., and R. Rosscup. J. Amer. Chem. Soc. 82;6115
(1965) .
8 Bohme, H., and H. J. Gran. Ann. 581;133 (1953).
9 Bensley, B., and G. Kohnstam. J. Chem. Soc. 4747 (1957),
10 Pryor, W. A. Mchanisms of Sulfur Reactions. New York,
McGraw-Hill Book Co. 1960.
11 Kharasch, K.r S. J. Potempa, and H. L. Wehrmeister. Chem,
Rev. 39:269 (1946).
12 Riesg, E. Bull. Soc. Chem. (Prance). 1449 (1966).
13 Brown, D., and B. T. Grayson. In: The Mechanisms of
Reactions of Sulfur Compounds. 1970. Chapter 5.
14 Pryor, W. A. Free Radicals. New York, McGraw-Hill Book
Co. 1966. p. 288.
15 Foote, C. S. Accounts Chem. Res. J:104 (1968).
16 Cannes, C., and T. Wilson. J. Amer. Chem. Soc. 90:6528
(1968) .
17 Foote, C. S., R. W. Denny, L. Weaver, Y. Chang, and J.
Peters. Ann. NY Acad. Sci. J7Jki:39 (1970).
18 Rawls, H. R., and P. J. van Santem. Ann. NY Acad. Sci.
171:135 (1970) .
19 Merkel, P. B., and D. R. Kearns. J. Amer. Chem. Soc.
94:7244 (1972).
118
-------
SECTION XII
RESULTS AND DISCUSSION: CARBARYL
HYDROLYSIS
Aly and El-Dib1 reported the hydrolysis of carbaryl along
with several other carbamate pesticides, whose structures are
shown in Figure 50. They carried out hydrolysis studies under
non-buffered reaction conditions over the pH range 4.0 to 10.0
and at various temperatures. The second-order rate constants
for these four compounds at 20° are given in Table 23. They
reported that carbaryl was stable over the pH range of 3.0-
6.0.
Wauchope and Hague2 also determined the second-order rate
constant for carbaryl in unbuffered water. They report a
second-order rate constant of 3.4 x 102 M-1 min-1 at 25° which
is in good agreement with the value reported by Aly and El-
Dib, corrected for the temperature difference. Karinen et
al.3 reported a second-order rate constant of 1.4 x 102 M~J
min-1 in sea water at 20°. We carried out the hydrolysis of
carbaryl under buffered pseudo-first-order reaction conditions
at 27° and pH 9.52 (phosphate buffer). The second-order rate
constant was found to be (3.61±0.02) x 102 M"1 min-1, in good
agreement with the value reported by Aly and El-Dib.1 Using
these data we calculated hydrolysis half-lives for carbaryl at
pH values normally found on lakes and rivers (Table 24).
Hydrolysis is anticipated to be an important degradation
pathway in basic natural waters, but in acidic natural waters
hydrolysis is extremely slow.
Table 23. KINETIC PARAMETERS FOR HYDROLYSIS OF CARBARYL
AND SEVERAL OTHER CARBAMATE PESTICIDES AT 20°.1
Pesticide k (M~* min-1) E (kcal/mole)
OH' a'
Carbaryl 2.04 x 102 16.9
Baygon 3.04 x 10 15.8
Pyrolan 7.0 x 10~» 13.7
Dimetilan 3.4 x 10-3 14.0
119
-------
OCONHCH:
CARBARYL
OCONHCH(CH3)2
BAYGON
DIMETILAN
Figure 50. Chemical structures of carbaryl and
other carbamate pesticides
Table 2U. HYDROLYSIS HALF-LIVES FOR CARBARYL AT pH-VALUES
USUALLY FOUND IN THE AQUATIC ENVIRONMENT (27°C).
PH
Half-life
9
8
7
6
5
3.2 hrs
1.3 days
13 days
U.4 months
3.6 years
OXIDATION
The oxidation of carbaryl under environmental reaction
conditions was not anticipated to be significant in view of
120
-------
its chemical and photochemical reactivity. The product of
hydrolysis 1-napththol, however, would be very readily
oxidized.
Wauchope and Hague2 reported that the 1-naphthoxide ion (3
x 10~* M) in room light undergoes photooxidation to give 2-
hydroxy-1,4-naphthoquinone.6
PHOTOCHEMISTRY
Considering the widespread usage of carbaryl and its
reported alteration by sunlight,f we were surprised to find
that relatively little is known about its direct photolysis.
Crosby and co-workers reported that carbaryl photoproducts
include 1-naphthol, methyl isocyanate, and other unidentified
cholinesterase inhibitors.7'8 Aly and El-Dib reported that
some of the photoproducts from carbaryl in water have the same
chromatographic retention time as photoproducts of 1-
naphthol.9 One of the photoproducts from 1-naphthol in air-
saturated basic water was found to be 2-hydroxy-1,4-
naphthoquinone*.
The lack of data concerning carbaryl photoproducts
prompted us to look elsewhere for data concerning photolysis
of aryl carbamates. Trecker, Foote, and Osborn1" reported
that p_-tolyl-N-methylcarbamate undergoes a photoreaction
similar to the photoFries reaction of aryl esters and
anilides1* (Figure 51). Recently, in a series of papers,
Silk, Unger, and their co-workers have published results which
indicate that the photo-Fries type rearrangement occurs with
several carbamate insecticides12"15 (Figure 51). These
photolyses also yielded brown polymeric substances which were
not identified.
The above studies prompted us to synthesize the potential
photo-Fries products of carbaryl, N-methyl-1-hydroxy-2-
naphthamide and N-methyl-4-hydroxy-1-naphthamide. Careful
examination of photolysis mixtures revealed that no more than
trace amounts of these products were formed in either methanol
or water. Our studies also indicated that 1-naphthol and
methyl isocyanate are not products of the direct photolysis (>
290 nm) of carbaryl in either degassed or air-saturated
methanol and water (pH 5.5). We analyzed for 1-naphthol by
liquid chromatography on a Permaphase ODS column.16
Control experiments indicated that methyl isocyanate reacts
rapidly with methanol to form methyl N-methylcarbamate (eq.
67). However, no methyl N-methyl carbamate was detected (by
gas-liquid chromatography), in photolyzed solutions of
carbaryl (0.01 M in methanol).
121
-------
OCONHCH3
hv
OH
CONHCH3
OCONHCH3
hv
OH
OH
CONHCH3
CH3 CH3
CH3
CONHCH3
Figure 51. Photoreactions of substituted phenyl
N-methyl carbamates
CH3N=C=O
CH3OH
CH3NCOCH
(67)
At present, we cannot explain the differences between our
results and those reported by others?-*. We did not identify
the photoproducts but we have found the following:
• The products from photolysis in methanol and water
are non-volatile. Methylation by ethereal
diazomethane produces no products detectable by gas
chromatography.
• Highly-colored products form in air-saturated water.
The colored products do not form in air-free
solutions.
Kinetic studies of carbaryl indicated that its photolysis
rate under sunlight is rapid compared to other pesticides that
we examined. The quantum yields under various reaction
conditions are summarized in Table 25. The data indicated
that the photolysis is slowed by the presence of oxygen and is
pH-independent in the pH 5 to 1 range. Calculated half-lives
for the direct photolysis of carbaryl near the surface of a
water body are summarized in Table 26. Experiments under
midday summer sunlight (June, lat. 34° N) yielded data that
were in reasonable agreement with the calculated half-lives
for lat 30° and 40° N; the half-life was found to be about 45
hours in distilled water buffered at pH 5.5. Dark controls
showed no decomposition. Although the quantum yield for
carbaryl photolysis is low, its rapid absorption of sunlight
results in facile photoalteration.
122
-------
Other studies indicated that carbaryl is not readily
photooxidized by singlet oxygen in water (3 > 37.5).
Table 25. QUANTUM YIELDS FOR PHOTOLYSIS OF CARBARYL
AT 25°C (313nm) IN WATER
Reaction conditions Disappearance Quantum Yields
Degassed, pH 5.5 0.016
Air-saturated, pH 5.5 0.0056
Air-saturated, pH 5.2 0.0052
Air-saturated, pH 6.1 0.0058
Air-saturated, pH 6.9 0.0055
Table 26. CALCULATED DIRECT PHOTOLYSIS HALF-LIVES OF CARBARYL
AT DIFFERENT SEASONS AND LATITUDES IN THE NORTHERN HEMISPHERE.
Latitude, °N
0
30
40
50
70
Spring
43
51
64
90
292
Half-life
Summer
45
46
52
61
121
t_ hrsa
Fall
41
68
102
186
1760
Winter
43
103
200
571
43800
a
Calculated for mid-season and for depths of < 10 cm in pure
water.
123
-------
REFERENCES
1 Aly, O. M.r and M. A. El-Dib. Water Research. j>:1191
(1971).
2 Wauchope, R. D. , and R. Hague. Bull. Environ. Contain.
Toxicol. 2:257 (1973).
3 Karinen, J. F., J. G. Lamberton, N. E. Stewart, and L. C.
Terriere. J. Agr. Food Chem. 15:147 (1967).
4 Bender, M. L. , and R. B. Homer. J. Org. Chem. 3_0_:3975
(1965).
5 Bender, M. L. Chem. Rev. j>0_: 53 (1960).
6 Tomkiewicz, M. A., A. Groen, and M. Cocivera. J. Amer.
Chem. Soc. 93:7102 (1971).
7 Crosby, D. G., E. Leitis, and W. L. Winterlin. J. Agr.
Food Chem. J_3:204 (1965).
8 Crosby, D. G. In: Symposium on Pesticides in the Soil,
Ecology Degradation, and Movement. East Lansing, Michigan
State University. 1970. p. 86.
9 Aly, O. M., and M. A. El-Dib. In: Organic Compounds in
Aguatic Environments, Faust, S. J. and J. W. Hunter (ed.).
New York, Marcel Dekker, Inc. 1971. Chapter 20.
10 Trecker, D. J., R. S. Foote, and C. L. Osborn. Chem.
Commun. 1034 (1968).
11 Bellus, D.r and P. Hrdlovic. Chem. Rev. 6_7:599 (1967).
12 Silk, P. J., and I. Unger. Inter. J. Environ. Anal. Chem.
2:213 (1973) .
13 Kumar, Y. , G. P. Semeluk, P. J. Silk, and I. Unger.
Chemosphere. 23 (197U).
14 Addison, J. B., P. J. Silk, and I. Unger. Bull. Environ.
Contain. Toxicol. VI: 250 (1974).
15 Addison, J. B., P. J. Silk, and I. Unger. Intern. J.
Environ. Annal. Chem. _3:73 (1973) .
16 Thurston, A. D. Liquid Chromatography of Carbamate Pesti-
cides. U.S. EPA fEPA-R2-72-079. November 1972. p. 1-15.
124
-------
SECTION XIII
RESULTS AND DISCUSSION: ATRAZINE
HYDROLYSIS
Reports of atrazine hydrolysis in the literature have been
primarily concerned with soil degradation studies. Several
studies indicate that chemical degradation appears to be a
significant breakdown route. Obien and Green1 found that the
rate of degradation of atrazine was dependent on soil pH and
temperature and invoked chemical degradation as being more
significant than biological degradation. Zimdahl et al.2
investigated atrazine degradation in soils and postulated that
atrazine was first converted non-enzymatically to the 2-
hydroxy derivative as the first step in soil degradation.
Skipper et al.3 evaluated chemical vs. microbial degradation
in soils and concluded that chemical degradation of atrazine
was largely dependent on chemical hydrolysis to
hydroxyatrazine.
Armstrong and co-workers*-5 investigated atrazine
hydrolysis in soils and determined the rates of degradation of
atrazine under acidic and alkaline conditions. Table 27
contains the pseudo-first-order rate constants calculated from
half-life data obtained from the graphs in reference 4. The
average value of the second-order rate constant for acid
hydrolysis is 3.4 x 10-s M~* sec-*. For alkaline degradation,
the average value for the second-order rate constant is 6.5 x
10~s M-1 sec-1. It was not possible to evaluate any
contribution by water from these data. Based on work by
Harrobin6, who studied some chloro-1r3,5-triazines, the
hydrolysis contributions by water is probably negligible.
Our preliminary studies on the hydrolysis of atrazine were
in agreement with the data reported by Armstrong et al. For
acid and alkaline hydrolysis, the second-order rate constants
were 3.9 x 10~s M-1 sec-* and 7.6 x 10~5 M~l sec-1,
respectively. Thus, it appears that alkaline hydrolysis is
about two times more rapid than acidic hydrolysis. The
overall reaction is given in Figure 52.
PHOTOCHEMISTRY
Most of the work on direct photolysis of j3-triazine
pesticides has been reported by Zabik and his co-workers7-9,
125
-------
rt
W
53
H
N
£
H
O
Q
W
0^
o
fa
CO
EH
rfj
EH
cn
53
O
U
W
njj
K
Q
rfl
CO
w
H
1
1
.
r-
CN
0)
rH
.Q
EH
-d
^^
T—
1
O
0)
CO
r-
'5
i
K
0
TJ
T—
|
O
0)
CO
1
2J
*^^
+
ffi
^
u
•s
XI
«—
1
o
0)
CO
*~"
-o
CO
XI
o
,y
0)
MH ^^«
•H CO
rH, ^^1
1 tO
m T3
rH %~'
nj
X
W
in in in
i i I
0 0 O
rH rH rH
XXX
vo r~ oo
r^ ^D in
m in 'i1
i i i
o o o
H rH rH
XXX
•H in
• • •
CO VO rH
VD r- r- oo r- VD
i i i i i i
o o o o o o
rH rH rH rH rH rH
X X X X X X
vo ro CN cr* n CN
• •••••
i — 1 "^* rH CT\ in ^*
rH ^* ^* i — 1 (N O\
..<•••
in oo vo H in CN
rH VO CO rH rH
CO CM H rH O\ O\
• • • • • t
rH
-------
H20, H+ H
or OH"
CH3CH2NH N
ATRAZINE HYDROXYATRA2INE
Figure 52. Hydrolysis of atrazine
Photolysis of atrazine in hydroxylic solvents, e. g. , water and
methanol, results in nucleophilic displacement of the chlorine
(Figure 53) . Other j3-triazines undergo similar
photo re actions. The photolysis rates for various s_-triazines
were found to vary according to the nature of the halogen and
alkylamino substituents on the ring7. Ruzo and co-workers
reported surprising differences in the position of the long-
wavelength absorption bands of simazine, atrazine, and
propazine7. Other data published by Sadtler Research Corp.
indicate that the ultraviolet absorption spectra of these
three pesticides are superimposable* °, Our spectral data are
in agreement with the Sadtler spectra. The spectral data
indicate that the extremely low direct photolysis rates of
atrazine and other si-triazines are due mainly to lack of
sunlight absorption. Direct photolysis of atrazine is so slow
that it is unlikely to have environmental importance.
Cl OR
CH3CH2NH
R = Hf CH3
Figure 53. Photoreaction of atrazine
The sensitized photolysis of atrazine was not
investigated. Alkylamino substituents of atrazine may be
oxidized via the exciplex mechanism discussed earlier in this
report (See Section V) .
Atrazine was found to be very unreactive towards singlet
oxygen. The 8 value in water was found to be > 37. 5. The
lack of reactivity of atrazine may in part be due to its
127
-------
alkylamino substituents; amines are known to rapidly quench
singlet oxygen without reacting11.
REFERENCES
1 Obien, S. R., and R. E. Green. Weed. Sci. JT7:509 (1969).
2 Zimdahl, R. L., V. H. Freed, M. L. Montgomery, and W. R.
Furtick. Weed. Res. J_0:18 (1970).
3 Skipper, H. D., C. M. Gilmour, and W. R. Furtick. Soil
Sci. Soc. Amer. Proc. 3^:653 (1967).
4 Armstrong, D. E., G. Chesters, and R. F. Harris. Soil
Sci. Soc. Amer. Proc. .3J.:61 (1967).
5 Armstrong, D. E. Environ. Sci. Tech. 2!:686 (1968).
6 Horrobin, S. J. Chem. Soc. 4130 (1963).
7 Ruzo, C., M. J. Zabik, and R. D. Schultz. J. Agr. Food
Chem. 2J: 1047 (1973) .
8 Pape, B. E., and M. J. Zabik. J. Agr. Food Chem. 18:202
(1970).
9 Pape, B. E., and M. J. Zabik. J. Agr. Food Chem. JO;316
(1972).
10 Ultraviolet Spectra of Agricultural Chemicals. Sadtler
Research Laboratories, Philadelphia, PA.
11 Cannes, C., and T. Wilson. J. Amer. Chem. Soc. 90:6528
(1968) .
128
-------
SECTION XIV
RESULTS AND DISCUSSION: DIAZINON
HYDROLYSIS
Melnikov1 has reported that diazinon is not as resistant
to hydrolysis as is parathion. Under acidic conditions, it is
reportedly hydrolyzed 12 times faster than parathion. Under
alkaline reaction conditions, diazinon hydrolyzes at the same
rate as parathion. The hydrolysis reaction is given in Figure
54.
Cowart et al.2 determined the rates of hydrolysis of
several pesticides (low concentration) in water at pH 6.0.
Plotting their data as log % cone, remaining of diazinon vs
time gives a half-life of 264 hours (k = 7.3 x 10~7 sec-1).
The reaction appears to follow first-order kinetics through
one half-life.
Ruzicka et al.3 determined the half-life of diazinon in
20% ethanol-water at 70° under buffered reaction conditions
(pH 6.0). Based on these data, diazinon has a half-life of 37
hours or a pseudo-first-order rate constant of 1.3 x 10~s
sec-1 at 70°.
Sethunathan and MacRae* reported a diazinon degradative
half-life of 43.8 days in sterilized soil at pH 4.7. Konrad
et al.5 observed that diazinon did not hydrolyze in water at
pH 6, but was readily hydrolyzed at pH 2.
Mortland and Raman6 have reported that copper (II)
catalyzes the hydrolysis of diazinon in 50% aqueous ethanol in
s
0-P(OCH2CH3)2
H20, H+
or OH' CH3
S
HO-P(OCH2CH3)2
Figure 54. Hydrolysis products of diazinon
129
-------
the pH range 5-6. Unfortunately, they did not give a rate
constant for the uncatalyzed hydrolysis. Lichtenstein et al. 7
determined the effect of sodium azide on the degradation of
diazinon in water. Calculation from his data gives a half-
life of about 12 days at pH 4.9.
Faust and Gomaa8 carried out a study on the effect of pH
on the rate of hydrolysis of diazinon and diazoxon. Their
data are given in Table 28. The average second-order rate
constant (20°) is 5.6 x 10~2 M~* sec-1 for alkaline hydrolysis
and 2.3 x 10~2 M~l sec-* for acid degradation. The
contribution to hydrolysis by water is apparently too slow to
be important. Diazoxon reacts somewhat faster than diazinon.
Based on the data of Comma and Faust,8 the average alkaline
and acid second-order rate constants are 6.0 x 10-2 M~2 sec-1
and 6.U x IO-1 M-1 sec-1, respectively. At alkaline pH's,
diazinon and diazoxon react at the same rate. Under acidic
conditions diazoxon is about 27 times more reactive than
diazinon.
In view of these studies, diazinon hydrolysis was not
examined in detail. We did carry out a limited investigation
with diazinon and based on half-life data calculated the
alkaline and acid second-order rate constants (27° c. ) as 2.4
x 10~3 M~4 sec-1 and 7. 3 x 10~2 M~» sec-1, respectively.
These approximate values are in agreement with Faust and
Gomaa.8
Table 28. HYDROLYSIS HALF-LIVES AND RATE CONSTANTS
FOR DIAZINON AND DIAZOXON AT 20°C.a
Diazinon
PH
3
5
7
9
10
k . ^(sec 1)
obsdv '
.1
.0
.5
.0
.4
1
2
4
5
1
.6 x
.6 x
.3 x
.9 x
.3 x
io-5
io-7
io-8
io-8
io-6
todays)
0.49
31
185
136
6.0
Diazoxon
k , , (sec"1 )
obsd
5
6
2
4
1
.1
.3
.8
.4
.9
x 10
x 10
x 10
x 10
x 10
-4
-6
-7
-7
-5
t, (days)
'5
0.
1.
29
18
0.
017
27
42
Reference 8.
130
-------
OXIDATION
Rails et al.9 reported that diazinon when sprayed on
tomatoes give within five days 2-isopropyl-4-methylpyrimidin-
6-ol. The authors suggested that diazinon was oxidized to
diazoxon which in turn was hydrolyzed to give the above
product.
Studies in our laboratory showed that diazinon was stable
to molecular oxygen in air-saturated water for several days at
room temperature. Thus, oxidation by molecular oxygen of
dilute solutions of diazinon is not likely to be an
environmental degradative pathway.
PHOTOCHEMISTRY
No products due to the direct photolysis of diazinon have
been reported in the literature. Although this pesticide
absorbs sunlight much less strongly than its organophosphorus
relative, parathion, our screening studies suggest that it
undergoes direct photolysis in water at a similar rate. The
slow photolysis rate of diazinon is due principally to weak
absorption of sunlight.
Studies by Pardue and co-workers*o indicate that the
isopropyl group of diazinon is susceptible to light-initiated
autooxidation. The oxidation product "hydroxydiazinon" that
formed upon sunlight irradiation of diazinon-treated kale
leaves or ultraviolet irradiation of a diazinon film probably
resulted from reduction of a hydroperoxide intermediate that
is analogous to the hydroperoxide formed upon autooxidation of
isopropyl-benzene11 (Figure 55).
(CH3)2C^N^OP(OCH2CH3)2
Figure 55. Postulated mechanism for photooxidation of diazinon
131
-------
Diazinon, like parathion,12 was unreactive towards singlet
oxygen (3 > 37.5 in water). These and Grunwell's data12
indicate that singlet oxygen is not directly involved in the
photochemical conversion of phosphorothionates to the more
toxic oxons.
REFERENCES
1 Melnikov, N. N. Res. Rev. 36:1 (1971).
2 Cowart, R. P., F. L. Bonner, and E. A. Epps, Jr. Bull.
Environ. Contam. Toxicol. jS:231 (1971).
3 Ruzicka, J. H., J. Thompson, and B, B. Wheals. J.
Chromatog. 31:37 (1967).
4 Sethunathan, N., and I. C. MacRae. J. Agr. Food Chem.
17:221 (1969).
5 Konrad, J. G., D. E. Armstrong, and G. Chesters. Agron.
J. 50:531 (1967).
6 Mortland, M. M., and R. V. Raman. J. Agr. Food Chem.
15:163 (1967).
7 Lichtenstein, E. P., T. Fuhremann, and K. R. Schulz. J.
Agr. Food Chem. V6:870 (1968).
8 Faust, S. D., and H. M. Gomaa. Environ. Letters. 3_:171
(1972) .
9 Rails, J. W., D. R. Gilmore, and A. Cartes. J. Agr. Food
Chem. 1_4:387 (1966).
10 Pardue, J. R., E. A. Hanson, R. P. Barron, and J. T. T,
Chen. J. Agr. Food Chem. 10:405 (1970).
11 Pryor, W. A. Free Radicals. New York, McGraw Hill, 1966.
Chapter 18.
12 Grunwell, J. R., and R. H. Erickson. J. Agr. Food Chem.
21:929 (1973).
132
-------
SECTION XV
RESULTS & DISCUSSION: PARATHION
HYDROLYSIS
Although comprehensive studies concerning the fate of
parathion in the environment are not available in the
literature, hydrolysis is believed to be the most important
process for degradation even though the rate of hydrolysis is
slow in water. In addition, two other compounds, the oxygen
analog, paraoxon and the isomer 0-p-nitrophenyl
phosphothiolate (Figure 56) are associated with the biological
effects and therefore their hydrolytic data, when available,
is worth scrutiny.
Thermal stability of parathion was investigated by Metcalf
and March*. When heated to 150° for 24 hours parathion was
85% degraded and gave 0,S-diethyl-O-p_-nitrophenyl
phosphorothiolate as the major product. Although these
compounds might be found as impurities in parathion as a
result of manufacturing or storage it is not likely that they
will be formed at most environmental temperatures.
Ruzika et al^2 in laboratory studies investigated the
half-life of parathion along with 36 other pesticides.
Parathion was determined to have a half-life of 43.0 hours at
70° in 20% ethanol-water buffered at pH 6.0. The oxygen
analog paraoxon had a half-life of 28.0 hours. ^Parathion
s o
(CH3CH20)2P-0-^VN02 (CH3CH20)2P-0
PARATHION PARAOXON
CH3CH2S
O,S-DI ETHYL O-(p-NITROPHENYL)
PHOSPHOROTHIOLATE
Figure 56. Structures of parathion and some products
derived from its chemical transformations
133
-------
half-life was also determined in two river water samples
containing 2095 ethanol. Parathion had a half-life of 65 hours
in River Thames water (pH 8.0) and 68 hours in River Irthing
water (pH 7.5).
Cowart et al.3 investigated the hydrolysis of seven
organophosphate pesticides. Using these data, a semilog plot
of parathion concentration vs. time gives a straight line
through only one half-life and then deviates.
Muhlman and Schrader* determined the rate of hydrolysis of
parathion over the temperature range of 0-70° in the pH range
1-5 and reported a pseudo-first-order rate constant (k ) of
1.2 x 10-8 sec-1 at 20°. H2°
Aldridge and Davison3 reported a first-order hydrolysis
rate constant (kobsd) of 1.6 x 10~8 sec*1 at pH 7.6 at 37° and
1.8 x 10~5 sec*1 for the S-ethyl isomer of parathion under the
same conditions.
In more detailed laboratory studies Hassan et al.6
investigated the hydrolysis of parathion and paraoxon over the
pH range 3.1 to 10.4 at 20°. Their data shows no pH
dependence below pH 7 which corresponds to a hydrolysis
contribution by water with a pseudo-first-order rate constant
(kH2o) of about 5 x 10-a sec-1 for both parathion and
paraoxon. Based on rate data at pH 10.4 parathion has an
alkaline second-order rate constant (kQa-) of 2.3 x 10~2 M~»
sec-* compared to 1.3 x 10-* M-1 sec-* for paraoxon.
Pet* reported rate data for parathion hydrolysis at 25°
and 35°. Extrapolation of his data to 20° gives a pseudo-
first order rate constant for water (kH2o) of 3.3 x 10~« sec—1
and a secondorder rate constant for alkaline hydrolysis (k )
of 4.6 x 10-* M-1 sec-1.
Ketelaar8'* also studied alkaline hydrolysis of paratiori
and methyl parathion in water. They found the second-order
alkaline hydrolysis rate constant (kOH-)r extrapolated to 20°,
was 5.9 x 10~* M-1 sec-1 while for methyl parathion it was 2.4
x 10-3 M-1 sec-1.
Scrutiny of the above data reveals that the second-order
rate constant for water hydrolysis (kn2o) is about (3-4) x
10~8 sec-1 and the second-order alkaline hydrolysis rate
constant (kOH-) is (4.6-5.9) x 10~* M-1 sec-1. The rate data
of Hassan et al.6 for alkaline hydrolysis is out of line with
the other data. It is also noted that at alkaline pH's their
data does not appear to follow second-order kinetics, contrary
to other reports. Products of the hydrolysis are shown in
Figure 57.
134
-------
s
(CH3CH20)2P-OH
Figure 57. Hydrolysis of parathion
OXIDATION
Hassan et al. have reported the oxidation of parathion and
paraoxon in water by Cl2r C1O2, and KMnO4.6 The oxidation of
parathion by KMnO4 at neutral pH1s gave paraoxon as the
product. Also at pH 7.U, C12 and ClO2 oxidized parathion to
paraoxon. Gunther has shown that parathion is oxidized by
ozone to paraoxon.10 However, it is not anticipated that these
processes will be important in most aquatic ecosystems.
Koivistoninen and Merilainen1» showed that parathion as a
thin film on glass gave paraoxon as a product in the dark.
The effect of uv radiation was to increase the amount of
paraoxon. An explanation for these results might well be
autooxidation where the effect of uv radiation is to provide
free radical initiation.
PHOTOCHEMISTRY
Interest in the photolysis of parathion has stemmed
primarily from observations that its photoproducts are
cholinesterase inhibitors. These observations were first
recorded by Payton12 and were later corroborated by Frawley
and cook.*3 Early work by Cook and Pugh indicated that one of
the cholinesterase inhibitors was paraoxon.14 Koivistoninen
and Merilainen11 exposed parathion to uv light and sunlight
and reported paraoxon, S-ethyl and S-phenyl isomers of
parathion and other unidentified products. Methyl parathion
on the other hand gave only methyl paraoxon. El-Refai and
Hopkins15 observed paraoxon as one of the products when
parathion was deposited on bean plant leaves and glass plates
and irradiated by light. Recently, Joiner and co-workers have
reported that parathion forms numerous products upon exposure
to high intensity ultraviolet light16"18 (Figure 58). Several
of these products were found in field studies conducted by
Joiner and Baetke;18 paraoxon was the major product found in
these studies. Grunwell and Erickson19 have reported that
O,O,S-triethylthiophosphate (shown below) is the major
photoproduct from parathion at high concentrations in dioxane-
water.
135
-------
(C2H50)2-P-S-C2H5
Products reported by Joiner and Grunwell are summarized in
Figure 58.
Most of the above product data were obtained from
photolysis of parathion films or high concentrations of
parathion in aqueous emulsion or organic solvents. According
to Grunwell,19 the presence of water is essential to the
formation of paraoxon in solution; no photolysis occurred in
dry organic solvents. Grunwell has suggested that the
mechanism shown below accounts for the photoalteration
products at high concentrations20 (eq. 68-70). This mechanism
suggests that formation of paraoxon may be less important at
lower concentrations and that the photolysis rate of parathion
may increase with increasing concentration.
s v», S
II hv I
ArOP(OC2H5)2 > ArOP-(OC2H5)2 (68)
PAR PAR*
SC_H_ S
I 2 5 II n"
PAR* + PAR > ArOP(OC.H_). + ArOP" (69)
2 5 2
sc „ r*
CH H I
PARAOXON + C H SH
]25 HO | 2 5
ArOP(OC^H.)0 -=—* ArOP(OC,H_), (70)
b 2. t t s t Q
ArOH
Our studies of the direct photolysis of parathion were not
detailed. They did establish that photoalteration of
parathion at wavelengths > 280 nm is very slow in water (pH
5.5) at dilute concentrations (2.8 ppm) . The slow rate was
not due to lack of light absorption at wavelengths > 300 nm.
Parathion absorbs sunlight much more rapidly than the other
136
-------
0
OP(CH2CH3)2
N02
PARAOXON
N02
SP(CH2CH3)2
N02
02N
N02 02N
XOCH2CH3
N02
0
II
OP(CH2CH3)2 OP(OCH2CH3)2 OH
N02
OH
0
0
(CH3CH2)2P-OH CH3CH2OP(OH)2
NH2
(CH3CH20)3P=0
(CH3CH20)3P=S
0
I-
(CH3CH20)2PSCH2CH3
Figure 58. Reported photoalteration products of parathion
137
-------
pesticides that were included in this report (Table 29). The
quantum yield for parathion photolysis must be very low,
certainly less than 0.001. Other workers have also reported
that the quantum yield for photoinduced hydrolysis of p_-
nitrophenylphosphate is very low,21"22 although m-
nitrophenylphosphate photohydrolyzes very efficiently.
Grunwell has concluded that parathion and other thiophos-
phates are unreactive toward singlet oxygen.19"20 Our data are
consistent with his observations.
Table 29. SPECIFIC SUNLIGHT ABSORPTION RATES OF PARATHION
AND OTHER SELECTED PESTICIDES DURING MIDSUMMER AND MIDDAY
AT LATITUDE 40°N
Pesticide
Parathion
Carbaryl
2,U-D Ester
Atrazine
Diazinon
Malathion
Methoxychlor
ka x 106 (sec-1)
cl
6,930
623
194
6.6
6.2
1.2
0.87
Relative k_
cl
8,000
720
22
7.6
7.1
1.3
1
REFERENCES
1 Metcalf, R. L. and R. B. March, J. Econ. Entomol.
^16:288 (1953).
2 Ruzicka, J. H., J. Thompson, and B. B. Wheals. J.
Chromatog. H:37 (1967).
3 Cowart, R. P., F. L. Bonner, and E. A. Epps, Jr. Bull.
Environ, contam. Toxicol. .6:231 (1970).
U Muhlmann, R., and G. Schrader. Z. Naturierstz. 12:196
(1957).
138
-------
5 Aldridge, W. N., and A. N. Davison. Biochem. J. 52:663
(1952) .
6 Hassan, M., H. M. Gomaa, and S. D. Faust. In: Advances
in Chemistry Series, Fate of Organic Pesticides in the
Aquatic Environment, Gould, R. F. (ed.). 1972. p. 189.
7 Pet, D. R. Chem. and Ind. 526 (1948) .
8 Ketelaar, J. A. A. Pec. Trav. Chem. .6_9:649 (1950) .
9 Ketelaar, J. A. A., H. R. Gersmann, and K. Koopmons. Rec.
Trav. Chem. .71:1253 (1952).
10 Gunther, F. A., D. E. Ott, and M. Ittig. Bull. Environ.
Contain. Toxicol. 1:87 (1970).
11 Koivistoninen, P., and M. Merilainen. Acta Ag.
Scandanavia. JH3:267 (1963).
12 Payton, J. Nature. V7_l:355 (1953).
13 Frawley, J. P., and J. W. Cook. J. Agr. Food Chem. 6^:28
(1958).
1U Cook, J. W., and N. D. Pugh. J. Assoc. Offie. Anal. Chem.
40:277 (1957) .
15 Fl-Refai, A., and T. L. Hopkins. J. Agr. Food Chem.
14:588 (1966).
16 Joiner, R. L. , H. W. Chambers, and K. P. Baetke. Bull.
Environ. Contam. Toxicol. £:220 (1971).
17 Joiner, R. L., and K. P. Baetke. J. Assoc. Offic. Anal.
Chem. .56:338 (1973).
18 Joiner, R. L., and K. P. Baetke. J. Agr. Food. Chem.
H:391 (1973).
19 Grunwell, J. R., and R. H. Erickson. J. Agr. Food Chem.
2.1:929 (1973) .
20 Grunwell, J. R., and R. H. Erickson. Presented in part at
the 165th National Meeting of the American Chemical
Society, Dallas, TX, April 1973.
21 Havinga, E., R. O. deJongh, and W. Dorset. Rec. Trav.
Chim. 7J5:378 (1956).
22 Letsinger, R. L., and O. B. Ramsay. J. Amer. Chem. Soc.
86:1447 (1964).
139
-------
SECTION XVI
RESULTS & DISCUSSION: TOXAPHENE
HYDROLYSIS AND OXIDATION
The persistence of toxaphene, when applied to lakes and
rivers, has been related to the time required for detoxification
of the water body. This persistence has been suggested to be
dependent on such factors as pHr alkalinity, temperature,
sunlight and dissolved oxygen. The subject has been recently
reviewed by Hughes1 and there does not appear to be an
understanding of the relationship of environmental factors to
toxaphene persistence.
Toxaphene is dehydrohalogenated in strong alkaline medium.
Archer and Crosby2 reported partial dehydrochlorination in
approximately 2 M alcoholic KOH in 15 min. at 75-80°.
It is reported that toxaphene is stable to sulfuric acid and
a 1:1 mixture of sulfuric-nitric acid.» However, it is surprising
that any double bonds in the toxaphene are not protonated by the
strongly acidic medium.
In our studies toxaphene was found to be stable in air-
saturated dilute acidic and alkaline medium at 65°. After two
days under these reaction conditions, there was no detectable
change in the gas chromatographic fingerprint. These results
indicate that under environmental reaction conditions the
degradation of toxaphene by oxygen, acid or alkali would be too
slow to be environmentally significant.
PHOTOCHEMISTRY
Our screening studies indicated that photolysis of toxaphene
under sunlight is a very slow process in pure water. Also, it is
not readily oxidized by photochemically-generated singlet oxygen
(3 > 37.5 in water). Thus, no detailed studies were carried out.
REFERENCES
1 Hughes, R. A. Ph.D. Dissertation. University of Wisconsin,
Madison, Wisconsin (1970).
2 Archer, T. E., and D. G. Crosby. Bull. Environ. Contam.
Toxicol. 1:70 (1966).
140
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-76-067
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
CHEMICAL AND PHOTOCHEMICAL TRANSFORMATION OF
SELECTED PESTICIDES IN AQUATIC SYSTEMS
5. REPORT DATE
September 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
N. Lee Wolfe, Richard G. Zepp, George L.
Baughman, Robert C. Fincher, & John A. Gordon
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Research Laboratory
Office Research and Development
U.S. Environmental Protection Agency
Athens, Georgia 30601
10. PROGRAM ELEMENT NO.
1BA609
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
Same as above
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report presents the results of laboratory studies to quantita-
tively predict chemical and photochemical transformation rates and pro-
ducts of pesticides in water. It includes a general discussion of
relevant transformation processes and associated kinetic expressions.
The processes treated in most detail are hydrolysis, direct photolysis,
and reaction with singlet oxygen. Implications of other processes such
as oxidation and sensitized photolysis are also discussed.
Results of detailed studies are included for the pesticides, mala-
thion, carbaryl, methoxychlor, captan, and 2,4-D esters. The measured
rate constants and half-lives indicate that chemical and/or photo-
chemical processes of these compounds are likely to be important in
the aquatic environment.
Less extensive data is presented for the pesticides, atrazine,
diazinon, parathion, and toxaphene, along with a discussion of avail-
able literature data.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Pesticides
Hydrolysis
Photolysis
6F
7C
7E
3. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
151
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
: 1976 — 657-695/6105 Region 5-1
141
------- |