SEPA
EPA/600/R-08/089 I September 2008 | vwvw.epa.gov/athens
United States
Environmental Protection
Agency
Fate of High Priority Pesticides
During Drinking Water
Treatment
Ecosystems Research Division, Athens, GA 30605
National Exposure Research Laboratory
Office of Research and Development
-------
EPA/600/R-08/089
September 2008
Fate of High Priority Pesticides During
Drinking Water Treatment
by
Stephen E. Duirk, ^Lisa M. Desetto, and ^Gary M. Davis
National Exposure Research Laboratory
Ecosystems Research Division
Athens, GA
^Student Services Authority
U.S. Environmental Protection Agency
Office of Research and Development
Washington D.C. 20460
-------
NOTICE
This work has been subject to external peer and administrative review, and has been approved
for publication as an EPA document. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
-------
ABSTRACT
The fate of organophosphorus (OP) pesticides in the presence of chlorinated oxidants was
investigated under drinking water treatment conditions. In the presence of aqueous chlorine,
intrinsic rate coefficients were found for the reaction of hypochlorous acid (kHoci,op) and
hypochlorite ion (k0ci,op) with eight OP pesticides. The reaction of hypochlorous acid (HOC1)
with each OP pesticide was relatively rapid at near neutral pH, kHoci,op = 0.86 - 3.56 x 106 M^h"
l. The reaction of HOC1 with OP pesticides occurs at the thiophosphate (P=S) moiety resulting
in the formation of the corresponding oxon (P=O), which is more toxic than the parent OP
pesticide. Hypochlorite ion (OC1") was found not to oxidize the pesticide but act like a
nucleophile accelerating hydrolysis, k0ci,op = 37.3 - 15,908.9 M^h"1. Both the kHoci,op and the
koci,op were found to correlate well with molecular descriptors within each subgroup of OP
pesticide class. The most commonly detected OP pesticides in drinking water sources were then
investigated in the presence of monochloramine (NH2C1). Monochloramine was found not to be
very reactive with OP pesticides, kNH2ci,op = 10.6 - 21.4 M^h"1. Dichloramine (NHC^) was
found to be two orders of magnitude more reactive with the OP pesticides investigated than
monochloramine, kNHci2,op = 1995.0 - 2931.9 M^h"1. The reactivity of the three chlorinated
oxidants was then found to correlate with half-wave potentials (£1/2) for each OP pesticide
respectively. A model was developed to predict the transformation of OP pesticides in the
presence of chlorinated oxidants. With hydrolysis rate coefficients, the transformation of OP
pesticides under drinking water treatment condition was found to be adequately predicted over
the pH range of 6.5-9. The structure-activity relationships and mechanistic models developed
here could be used by regulators to determine if drinking potable water contaminated with OP
pesticides represents significant risk to a receiving population serviced by a community water
system.
ii
-------
ACKNOWLEDGEMENT
The authors would like to thank Mr. Jimmy Avants and Dr. Lidia Samarkina their technical
assistance, Dr. Paul Winget for calculating highest occupied molecular orbital (EHOMO) energies
for the organophosphorus pesticides, Dr. Wayne Garrison and Dr. Jackson Ellington for their
consultation and expertise, as well as Dr. Craig Adams at the Missouri University of Science and
Technology and Dr. David Sedlak at the University of California, Berkeley for agreeing to
review this report. Finally, the authors would like to thank Dr. Mac Long and Dr. Eric Weber for
their continued support of this work at the Ecosystems Research Division in Athens, GA.
in
-------
TABLE OF CONTENTS
LIST OF TABLES v
LIST OF FIGURES vi
EXECUTIVE SUMMARY ix
1 INTRODUCTION 1
2 EXPERIMENTAL PROCEDURES 5
2.1 Materials 5
2.2 Methods 5
2.2.1 Chlorination and Hydrolysis of OP Pesticides 5
2.2.2 Chloramination of OP Pesticides 7
2.3 OP Pesticide Analysis 8
2.4 Numerical Calculations and Parameter Estimation 8
3 RESULTS AND DISCUSSION 9
3.1 Hydrolysis of select OP pesticides 9
3.2 OP Pesticide Degradation in the Presence of Aqueous Chlorine 11
3.3 OP Oxon Chlorine-Assisted Hydrolysis 16
3.4 OP Pesticide-Chlorine Model Validation 17
3.5 OP Pesticide Degradation in the Presence of Monochloramine 17
4 CONCLUSION 24
5 REFERENCES 27
TABLES 31
FIGURES 39
IV
-------
LIST OF TABLES
Table 1 Stoichiometric equations and coefficients in the monochloramine autodecomposition
model 32
Table 2 Structures and some chemical properties of select OP Pesticides (48) 33
Table 3 Neutral and alkaline hydrolysis rate coefficients for 8 OP pesticides and 3 oxon
transformation products. 95% confidence intervals shown in parentheses 34
Table 4 Stoichiometric equations used in the chlorine-OP pesticide transformation pathway
model 35
Table 5 Oxidation and chlorine-assisted hydrolysis rate coefficients for 8 OP pesticides and 3
oxon transformation products. 95% confidence intervals shown in parentheses 36
Table 6 Differential equations for the monochloramine autodecomposition model 37
Table 7 Intrinsic rate coefficients for monochloramine and dichloramine reacting with CP,
DZ, and MA. 95% confidence intervals shown in parentheses 38
-------
LIST OF FIGURES
Figure 1 Hydrolytic behavior of chlorethoxyfos (CE) over the pH range of 3-11. Insert: slope
represents the alkaline hydrolysis rate coefficient. [CE]0 = 0.5 |jM, [Bufferjx = 10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.... 40
Figure 2 Hydrolytic behavior of tebupirimfos (TE) over the pH range of 3-10. Insert: slope
represents the alkaline hydrolysis rate coefficient. [TE]0 = 0.5 |jM, [Bufferjx =10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.... 41
Figure 3 Hydrolytic behavior of methidathion (ME) over the pH range of 3-11. Insert: slope
represents the alkaline hydrolysis rate coefficient. [ME]0 = 0.5 |jM, [Buffer]T =10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.... 42
Figure 4 Hydrolytic behavior of phosmet (PM) over the pH range of 2-10. Insert: slope
represents the alkaline hydrolysis rate coefficient. [PM]o = 0.5 |jM, [Buffer]T = 10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.... 43
Figure 5 Observed first-order rate of CE loss in the presence of aqueous chlorine at pH 6.5.
[CE]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 44
Figure 6 Observed first-order rate of DZ loss in the presence of aqueous chlorine at pH 6.5.
[DZ]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 45
Figure 7 Observed first-order rate of MA loss in the presence of aqueous chlorine at pH 6.5.
[MA]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 46
Figure 8 Observed first-order rate of ME loss in the presence of aqueous chlorine at pH 6.5.
[ME]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 47
Figure 9 Observed first-order rate of PA loss in the presence of aqueous chlorine at pH 6.5.
[PA]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 48
Figure 10 Observed first-order rate of PM loss in the presence of aqueous chlorine at pH 6.5.
[PM]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 49
Figure 11 Observed first-order rate of TE loss in the presence of aqueous chlorine at pH 6.5.
[TE]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate 50
VI
-------
Figure 12 The reaction order of chlorine with 8 OP pesticides at pH 6.5. [OP]0 = 0.5 |jM,
[PO4]T= 10mM, Temperature = 25±1°C, and [HOC1]T = 10-100 |iM ...................... 51
Figure 13 Second-order apparent rate coefficient for the OP pesticides at pH 6.5. [OP]0 = 0.5
HM, [PO4]T = 10 mM, Temperature = 25±FC, and [HOC1]T = 0-100 [iM. Error bars
represent 95% confidence intervals [[[ 52
Figure 14 The pH dependency of the first-order observed rate coefficients for the OP pesticides.
[OP]0 = 0.5 nM, [HOC1]T = 25 |iM, [Buffer]T = 10 mM, and Temperature = 25 °C.
Error bars represent 95% confidence intervals [[[ 53
Figure 15 Second-order apparent rate coefficient for the OP pesticides at pH 9. [OP]0 = 0.5
[CO3]T = 10 mM, Temperature = 25±FC, and [HOC1]T = 0-100 |iM. Error bars
represent 95% confidence intervals [[[ 54
Figure 16 Relationship between the kHoci,op with EHOMO as a function of OP pesticide subgroup.
Error bars about the regression line represent 95% confidence intervals ................... 55
Figure 17 Relationship between the k0ci,op with alkaline hydrolysis rate coefficient as a function
of OP pesticide subgroup. Error bars about the regression line represent 95%
confidence intervals [[[ 56
Figure 18 Second-order apparent rate coefficient of 3 OP pesticide oxons at pH 9. [OP]0 = 0.5
HM, [CO3]T = 10 mM, Temperature = 25±1°C, and [HOC1]T = 0-200 [iM. Error bars
represent 95% confidence intervals for both data points and regression lines ........... 57
Figure 19 Experimental and model results for DZ loss in the presence of chlorine at pH 7.0.
[DZ]0 = 0.6 nM, [HOC1]T = 20 |iM, [PO4]T = 10 mM, and Temperature = 25±1°C.
Error bars represent 95% confidence intervals. Lines represent model results ......... 58
Figure 20 Experimental and model results for PA loss in the presence of chlorine at pH 8.0.
[PA]0 = 0.57 nM, [HOC1]T = 25 |iM, [PO4]T = 10 mM, and Temperature = 25±1°C.
Lines represent model results [[[ 59
Figure 21 First-order observed loss of CP and DZ in the presence of monochloramine as a
function of pH. [NH2C1] = 50 |iM, [OP]0 = 0.5 |iM, Cl/N = 0.7 mol/mol, [Buffer]T =
10 mM, Temperature = 25±1°C, and pH 6.5-9.0. Error bars represent 95%
-------
Figure 23 Observed first-order observed loss of three OP pesticides in the presence of
monochloramine at pH 8.5. [NH2C1] = 50 |iM, [OP]0 = 0.5 |iM, Cl/N = 0-0.7
mol/mol, [HaBOs]! =10 mM, and Temperature = 25±1°C. Error bars represent 95%
confidence intervals 62
Figure 24 Experimental and model results for CP loss in the presence of monochloramine.
[CP]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH 6.5-
9, and Temperature = 25±1°C. Error bars represent 95% confidence intervals. Lines
represent model results 63
Figure 25 Experimental and model results for DZ loss in the presence of monochloramine.
[DZ]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH 6.5-
9, and Temperature = 25±1°C. Error bars represent 95% confidence intervals. Lines
represent model results 64
Figure 26 Experimental and model results for MA loss in the presence of monochloramine.
[MA]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH
6.5-8.5, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
Lines represent model results 65
Figure 27 Structure-activity relationship relating the reaction of monochloramine, dichloramine,
and hypochlorous acid with CP, DZ, and MA. Piela and Wrona (49) found half-wave
potentials for monochloramine (-0.2 V), dichloramine (0.4), and hypochlorous acid
(0.8 V). These values were then converted using the Nernst equation to the intensive
property of the log ratio of the activities of the reduced products to the oxidized forms
of the oxidant 66
Vlll
-------
EXECUTIVE SUMMARY
Environmental regulations require that all relevant routes of human exposure to
anthropogenic chemicals be considered in risk assessments. Community water systems (CWSs)
serve approximately 95% of the US population and potable water is considered a relevant route
of exposure to anthropogenic chemicals. There is available monitoring data for high priority
pesticides and toxic chemicals in drinking water sources (both surface and ground water).
However, there is very little monitoring data for these chemicals or their transformation products
in finished drinking water. Limited experimental studies show that some chemicals are partially
removed by physical water treatment processes (e.g., filtration, flocculation, etc.), and some are
transformed by reactions that occur during chemical treatment (e.g., disinfection and softening).
Transformation products of some contaminants have been shown to be more toxic than the
parent compound.
This report is in partial fulfillment of the National Exposure Research Laboratory Task #
ERD08103-1, "Fate of Pesticides and Toxic Chemicals During Drinking Water Treatment",
under Safe Pesticides/Safe Products (SP2) Long-Term Goal 1.4.1. The goals of this research
task are to: 1) provide chemical-specific information on the effects of water treatment for high-
priority pollutants, 2) provide physicochemical parameters for transformation products, and 3)
develop predictive models for forecasting treatment effects that cross chemical class and
treatment conditions.
The work reported here demonstrates a "proof-of-concept" towards the development of a
comprehensive modeling tool for OP pesticide fate in drinking water treatment plants and
distribution systems. Our objective was to develop predictive models that associate intrinsic
rates of reactivity to structural variability. This will allow decision makers to rank and prioritize
chemicals found in drinking water sources according to potential risk. For this purpose, eight OP
ix
-------
pesticides and three oxon transformation products were investigated. The 8 OP pesticides
chosen for this study contained the thiophosphate moiety (P=S) from the phosphorothioate
(chlorethoxyfos (CE), chlorpyrifos (CP), diazinon (DZ), parathion (PA), and tebupirimfos (TE))
and the phosphorodithioate (malathion (MA), methidathion (ME), and phosmet (PM))
subgroups. The three oxon (i.e., P=O) products were from the phosphate subgroup (diazoxon
(DZO), malaoxon (MAO), and paraoxon (PAO)). It was originally thought that by investigating
a class of pesticides that structure-activity relationships could easily be developed. However,
reactivity with chlorinated oxidants could only be correlated with subgroups and not across the
class of OP pesticides.
The phosphorodithioate subgroup was found to hydrolyze much more rapidly than the
phosphate or the phosphorothioate subgroups, and oxons are more susceptible to alkaline
hydrolysis when compared to the corresponding thionates. Also, ethyl esters appear to hinder
nucleophilic attack by hydroxide ions (i.e., alkaline hydrolysis) compared to a structurally
similar pesticide with methyl ester linkages at the phosphorus atom; however, there is a need for
more hydrolysis studies to be conducted at environmentally relevant conditions to build
structure-activity relationships in order to predict intrinsic hydrolysis rate coefficients.
In the presence of chlorine, intrinsic rate coefficients for both hypochlorous acid (HOC1)
and hypochlorite (OC1") were calculated, knoci,op and koci,op respectively. Since multiple
reaction pathways occur simultaneously over the pH range of 6.5-9, a chlorine-OP pesticide
reaction model previously developed for CP was used. The reaction with HOC1 and the OP
pesticides resulted in the rapid formation of oxons, which are more toxic than the parent OP
pesticide. Hypochlorite accelerated the hydrolysis of each OP pesticide but did not result in
oxon formation. The energy of the highest occupied molecular orbital (£HOMO) showed that
-------
knoci,op correlated within the two subgroups. The alkaline hydrolysis rate (kB,op) was found to
correlate with koci,op within each OP subgroup, which also implies that the reaction mechanism
of OC1" and OH" are similar. Therefore, the cross correlation demonstrates that OC1" does act as
a nucleophile accelerating the hydrolysis of OP pesticides. Oxons did not undergo further
oxidation by aqueous chlorine, but they were susceptible to chlorine assisted hydrolysis.
Three of the most commonly detected OP pesticides (CP, DZ, and MA) in drinking water
supplies were investigated in the presence of chloramines. The loss of all three pesticides under
chloramination conditions was found to be highly pH dependent. At pH 6.5, the primary
degradation pathway for OP pesticides was due to reaction with dichloramine and hypochlorous
acid, which are formed during monochloramine autodecomposition. The direct reaction of
monochloramine with each OP pesticide was found to be a minimal loss pathway. The order of
reactivity for the three chlorinated oxidants was HOC1:» NHC^ » NH2C1. The reactivity of the
three chlorinated oxidants was found to correlate with half-wave potentials (£1/2) for each OP
pesticide.
The stability of OP pesticides was investigated under conditions similar to drinking water
treatment. The critical reaction parameters could not be correlated across the OP pesticide class,
but correlations were made within subgroups indicating that minute structure differences
significantly impact reactivity. The structure-activity relationships developed here could allow
for the prediction of critical reaction parameters of OP pesticides in the presence of chlorinated
oxidants. Along with the presented model, regulators could access the potential risk associated
when consuming potable water contaminated with OP pesticides.
XI
-------
1 INTRODUCTION
The United States Geological Survey (USGS) conducts a national reconnaissance survey
known as NAWQA (National Water-Quality Assessment) Program to help define the effect
contaminants have on drinking water supplies and aquatic ecosystems (1). 90 pesticides and
some selected metabolites were chosen as target chemicals to monitor in US drinking water
sources. However, there is a relative dearth of information on occurrence of pesticide residuals
and pesticide metabolites in finished drinking water. Two surveys have been conducted for a
few community water systems examining pesticide concentrations in the source and finished
drinking water (2, 3). Neither of these studies thoroughly examined the effect of each treatment
process on a single slug of water, hence only the influent and effluent of each treatment facility
could be qualitatively discussed with respect to overall removal efficacy. Also, these studies did
not account for the treatment plant hydraulic retention time, thus it was not possible to ensure
that influent and effluent samples were properly paired.
The Food Quality Protection Act of 1996 (FQPA) requires that all pesticide chemical
residuals in or on food be considered for anticipated human exposure. Drinking water is
considered a potential pathway for dietary exposure, but there is reliable monitoring data for only
the source water. When assessing potential pesticide exposure due to drinking potable water, all
potential transformation pathways need to be addressed. Under drinking water treatment
conditions, hydrolysis and chemical oxidation are the most relevant transformation pathways for
the class of organophosphorus (OP) pesticides (4). The OP pesticides were chosen for their
wide-spread use and measured concentrations in drinking water supplies (1,3,5). However,
there is little quantitative rate coefficient information for either pathway.
Only 6 OP pesticides have been studied in-depth to reveal their hydrolytic behavior at
acidic, neutral, or alkaline conditions. Malathion has been extensively studied over the pH range
1
-------
of 1-10. In the pH range of 1-4, hydrolysis occurs at an ether linkage resulting in the formation
of malathion monoacid (6). Above pH of 4, nucleophilic attack at the tetrahedral phosphorus
atom by hydroxide ion was found to be twice as fast as carboxyl ester hydrolysis (7). Malathion
appears to be the most stable at pH 4, with an estimated half-life greater than one year.
Diazinon, diazoxon, parathion, and paraoxon have been studied over the pH range of 3.1-10.4
(8). All four of these OP pesticides were found to have an acidic, neutral, and alkaline
component to their hydrolytic behavior. Also, chlorpyrifos and its corresponding oxon were
studied over the pH range of 3-11, which neutral and alkaline hydrolysis rate coefficients were
determined for both pesticides (4, 9). However, OP pesticide hydrolysis is very dependent on
structure allowing for rate coefficient comparisons only to be made within each subgroup (i.e.,
phosphorothioate, phosphorodithioate, phosphorothiolate, and phosphate) (8).
Chlorination is the most commonly used chemical disinfection process for community
water systems (10), and it is known to react with numerous pesticides. For example, four s-
triazines were found to degrade in the presence of aqueous chlorine ([HOCl]x = HOC1 + OC1")
(11, 12). Atrazine was also found to be significantly degraded by ozone (13); however,
subsequent chlorination of the ozonated effluent had very little effect on the concentration of
residual atrazine or its ozone degradation products (2). Also, some carbamate pesticides have
been shown to react with chlorine while other members of this pesticide class were found to be
stable in chlorinated water. For example, carbaryl and propoxur do not react with chlorine; but
aldicarb, methomyl, and thiobencarb do exhibit significant reactivity (14-16). These findings
demonstrate that chlorine reactivity with different members in a specific class of pesticides can
vary significantly due to chemical structure variations. Therefore, it is prudent to study the fate
and transformation pathways of entire chemical classes, using class members that exhibit
-------
systematic structural variations and employing carefully selected experimentation and numerical
modeling.
When chlorine reacts with the phosphorothioate subgroup of organophosphorus (OP)
pesticides, the thiophosphate functionality (P=S) can be oxidized to its corresponding oxon
(P=O) (17-19). The resulting oxons are typically more potent than the parent as an inhibitor of
acetlycholinesterase, an enzyme necessary for regulating nerve impulse transmission between
nerve fibers (19). Duirk and Collette, (4) elucidated the fate of chlorpyrifos (CP) and its
transformation products over the pH range of 6-11. They were able to model the loss of CP and
chlorpyrifos oxon (CPO) to the stable end-product of 3,5,6-trichloro-2-pyridinol (TCP) over this
pH range in buffered deionized water systems, as well as in the presence of naturally occurring
aqueous constituents such as bromide and natural organic matter (NOM) (20).
Chloramines are a common secondary disinfectant alternative for many drinking water
utilities. In some cases, monochloramine is used for both primary and secondary disinfection
when excessively high background ammonia concentration in the source water do not allow for
breakpoint chlorination (10). However, very little is known about the reactions of
monochloramine with anthropogenic chemicals. In the presence of aromatic-ether and amine-
containing pharmaceuticals, aqueous chlorine was found to react more rapidly than
monochloramine (21). Also, monochloramine was found to react with diuron, a phenyl urea
herbicide, resulting in the formation of N-nitrosodimethylamine (NDMA) (22), which is a known
disinfection byproduct of chloramination (22, 23).
However, it is difficult to directly observe monochloramine reactions with anthropogenic
compounds due to its own autodecomposition, i.e., a series of reactions ultimately resulting in
monochloramine loss (Table 1) (24). This mechanism is highly pH dependent and allows for
-------
multiple oxidants to coexist (i.e., monochloramine (NH2C1), dichloramine (NHCb), and HOC1),
which confounds the interpretation of observational results. Using this well understood
autodecomposition reaction mechanism, the reactions of monochloramine with natural organic
matter resulting in chlorinated and brominated disinfection byproduct formation (DBF) were
resolved (25-27). This same technique has been used to investigate triclosan reactivity in
chloraminated water (28). It was found that dichloramine was approximately three orders of
magnitude more reactive with triclosan than monochloramine, but aqueous chlorine was found to
be three orders of magnitude faster than dichloramine. Similar results have been see with sulfite
and cyanide (29). Therefore, reaction pathway models can be used as a tool to propose and
elucidate reaction pathways in complex systems with parallel reactions.
The purpose of this study was to further elucidate the kinetics and transformation
pathways of OP pesticides in the presence of chlorinated oxidants. The 8 OP pesticides chosen
in this study contained the thiophosphate moiety from the phosphorothioate (chlorethoxyfos
(CE), chlorpyrifos (CP), diazinon (DZ), parathion (PA), and tebupirimfos (TE)) and the
phosphorothioate (malathion (MA), methidathion (ME), and phosmet (PM)) subgroups (Table
2). Hydrolysis neutral and alkaline rate coefficients were determined for CE, ME, PM, and TE.
Since all these OP pesticides can form oxons in the presence of chlorine, reaction rate
coefficients HOC1 and OC1" were correlated to molecular descriptors. For both correlations,
subgroup differences were observed. CP, DZ, and MA were investigated in the presence of
monochloramine over the pH range of 6.5-9. These three OP pesticides were chosen due to their
frequency of detection in drinking water supplies (3).
-------
2 EXPERIMENTAL PROCEDURES
2.1 Materials
Chlorpyrifos (CP), chlorpyrifos oxon (CPO), 3,5,6-trichloro-2-pyridinol (TCP), parathion
(PA), paraoxon (PAO), 4-nitrophenol (n-Ph), diazinon (DZ), diazoxon (DZO), 2-isopropyl-6-
methyl-4-pyrimidinol (IMP), malathion (MA), malaoxon (MAO), methidathion (ME),
tebupirimfos (TE), phosmet (PM) and chlorethoxyfos (CE) were purchased from ChemService
(West Chester, PA). Commercial 10-13% sodium hypochlorite (NaOCl), purchased from
Aldrich (Milwaukee, WI), contained equal-molar amounts of OC1" and Cl". Aqueous stock
solutions and experiments utilized laboratory prepared deionized water (18 MQ cm"1) from a
Barnstead ROPure Infmity™/NANOPure ™ system (Barnstead-Thermolyne Corp., Dubuque,
IA). Filters with pore size of 0.45 |j,m were purchased from Millipore (Billerica, MA).
Phosphate and carbonate salts used for buffer solutions were dissolved in deionized water and
filtered through a 0.45 |j,m filter, which was pre-rinsed with deionized water. The pH for the
experiments was adjusted with either 1 N H2SO4 or NaOH. All other organic and inorganic
chemicals were certified ACS reagent grade and used without further purification. The
glassware and polytetrafluoroethylene (PTFE) septa used in this study were soaked in a
concentrated free chlorine solution for 24 hours, rinsed with deionized water, and dried prior to
use. All chlorination, hydrolysis, and chloramination experiments were conducted at constant
temperature (25±1°C) and pH was maintained to ±0.1 of the initial pH for the duration of each
experiment.
2.2 Methods
2.2.1 Chlorination and Hydrolysis of OP Pesticides
For all OP oxidation and hydrolysis experiments, each OP was spiked by adding 1 mL of
4 mM stock in ethyl acetate into an empty 4 L borosilicate glass Erlenmeyer flask. A gentle flow
-------
of nitrogen gas was used to evaporate the ethyl acetate and 4 L of deionized water was then
added to the flask. The solution was slowly stirred and allowed to dissolve for 12 hours resulting
in an aqueous OP concentration of 1 |jM.
OP chlorination kinetic experiments were conducted under pseudo first-order conditions:
total chlorine, [HOC1]T, to OP molar ratios of 20:1, 50:1, 100:1, and 200:1. Chlorine was added
to solutions under rapid mix conditions achieved with a magnetic stir plate and a PTFE coated
stir bar. All chlorination experiments were conducted until at least 87% loss in the pesticide
initial concentration was achieved. Above pH 8, 10 mM carbonate [CO3]T buffer was used to
maintain pH. The commercial aqueous chlorine solution was first diluted to 250 mM and then
added to the aqueous system containing 0.5 |jM OP pesticide and carbonate buffer in a 2 L
Erlenmeyer flask. Thirteen aliquots from the large 2 L reactor were then placed into 120 mL
amber reaction vessels with PTFE septa and stored in the dark. In the pH range of 6.5-8, the rate
of OP loss in the presence of chlorine was very fast. Therefore, thirteen 100 mL aliquots of the 2
L aqueous system containing 10 mM phosphate buffer, [PO^x, and 0.5 |jM OP were placed in
250 mL amber Erlenmeyer flasks. Each flask was individually dosed with chlorine.
At each discrete sampling interval, two reaction vessels were sacrificed in their entirety.
One vessel was used to determine total chlorine concentration ([HOC1]T = [HOC1] +[OC1"]) via
Standard Method 4500-C1 F DPD-FAS titrimetric method (30). Free chlorine residuals were
quenched in the second reaction vessel with sodium sulfite in 20% excess of the initial free
chlorine concentration and the pH of the 100 mL sample was then adjusted to 7. Sulfite had no
effect on the recovery of any of the analytes as previously shown (4).
Hydrolysis experiments were conducted in a similar manner to the chlorination
experiments. Phosphate buffer was used over the pH range of 2-8; while carbonate buffer was
-------
used in the experiments at alkaline pH. Since hydrolysis or most of the OP pesticides was
relatively slow, these experiments were conducted in 2-L Erlenmeyer flaks with 10 mM of
buffer. However, the hydrolysis rate of PM was extremely fast at pH 9-10. The experimental
setup used to study the chlorination of OP pesticides at neutral pH was modified in order to study
PM hydrolysis at alkaline pH. Instead of spiking a 100 mL solution under rapid mix conditions
with a minuscule amount of chlorine, 98 mL of a [PM]0 = 0.5 |jM solution was spiked with 2 mL
of concentrated carbonate buffer. The concentrated buffer solution was the proper proportion of
bicarbonate/carbonate, which immediate adjusted the 100 mL sample to the target pH of 9, 9.5,
or 10. The rapid rate of PM hydrolysis at alkaline pH was then stopped by immediately
adjusting the pH to 7 with the appropriate amount of sulfuric acid. Hydrolysis experiments were
conducted until at least 70% loss in the initial pesticide concentration was achieved.
2.2.2 Chloranimation of OP Pesticides
Monochloramine kinetic experiments utilized additions of preformed monochloramine to
avoid potential artifacts caused by reactions of excess free chlorine that may briefly exist if
monochloramine was formed in-situ. The stock solution was prepared by mixing 5.64 mM
ammonia with 3.7 mM hypochlorous acid to achieve the desired 0.7 Cl/N molar ratio. The
solution was aged for 30 minutes in 10 mM bicarbonate buffer, pH 8.5, prior to use in any of the
experiments. Kinetic experiments were performed in a 2 liter Erlenmeyer flask in which
monochloramine ([NH2Cl]o = 0.05 mM) was added to a buffer ([Bufferjx = 10.0 mM) solution
containing [OP]0 = 0.5 |jM under rapid mix conditions, which was then poured into sixteen 128
mL batch reactors. At each sample interval, two reactors were used in their entirety and
monochloramine, pH, as well as the pesticide and its known degradation products were measured
for at least 140 hours.
-------
2.3 OP Pesticide A nalysis
Parent OP, oxon products, and other degradation products were extracted using C-18
solid phase extraction cartridges purchased from Supleco (Bellefonte, PA). Prior to extraction,
the sample was spiked with 1 |jM of phenthorate (internal standard), mixed thoroughly by hand
for two minutes, passed through the SPE cartridge at an approximate flow-rate of 7 mL/min, and
eluted with 3 mL of ethyl acetate. Quantification for each analyte was compared to eight
extracted standards over the concentration range of 0.01 to 1 |jM. A Hewlett-Packard 6890 GC
equipped with a 5973 MSD was used to analyze the OPs and their degradation products. GC
conditions were as follows: 30-m Restek Rtx-200 column with a 0.25-mm ID and 0.5-|j,m film
thickness. The temperature profile was: 100°C for 5 minutes, 100 to 250°C at 10 °C/minute, and
then held at 250°C for 25 minutes.
2.4 Numerical Calculations and Parameter Estimation
Scientist™ by Micromath (Salt Lake City, UT) was used to solve the systems of stiff
ordinary differential equations. The model was used to estimate rate coefficients using non-
linear regression analysis techniques. Scientist uses a modified Powell algorithm to minimize
the unweighted sum of the squares of the residual error between the predicted and experimentally
observed values to estimate specific parameters in the model. Final parameter estimates were
based on global pooling of all experimental data for each OP pesticide. All molecular
descriptors were calculated using GAMESSPLUS software.
-------
3 RESULTS AND DISCUSSION
3.1 Hydrolysis of select OP pesticides
Understanding the hydrolytic behavior of water soluble pesticides, such as the OP
pesticides, is paramount in order to determine their fate in the environment or engineered
systems. The phosphorothioate subgroup examined contained five pesticides. Of these, very
little information was known about CE or TE. CE was first investigated at pH 3 and 7 in order to
determine the acidic and neutral hydrolysis rate coefficients. The hydrolytic behavior was
similar to CP over this pH range and CE exhibited no statistical difference in the first order
observed rate coefficient at pH 3 or 7 (Figure 1). At alkaline pH, hydroxide ion appears to
significantly accelerate CE hydrolysis (Figure 1, insert). TE was also investigated over the pH
range of 3-10 with only slightly different results (Figure 2). At pH 3 and 7, the observed first-
order rate coefficients of TE hydrolysis were statistically similar. However, the observed rate of
TE loss at pH 10 was found to be statistically larger than the observed rate at either pH 3 or 7.
The presence of hydroxide ion at pH 10 appears to slightly increase the observed rate of TE loss.
Since the observed hydrolysis rate of TE at pH 7 was only minimally different when compared to
the observed hydrolysis rate at pH 10, the second order alkaline hydrolysis rate coefficient was
estimated from these two pHs (Figure 2, insert).
The neutral hydrolysis rates for both TE and CE were found to be very similar with CP,
DZ, and PA. Neutral hydrolysis rates for the phosphorothioate subgroup ranged from 1.56 x 10"4
- 1.10 x 10~3 h"1. However, the range in second-order rate coefficients for alkaline hydrolysis
was significantly greater but with most of the OP pesticides relatively close in magnitude (Table
3). CP, DZ, PA, and TE second order hydrolysis rate coefficients ranged from 6.0-37.1 M"1 h"1.
However, CE was found to be 2 orders of magnitude faster than CP. Hydroxide ion attacks the
tetrahedral phosphorus atom via an SN2 nucleophilic substitution releasing the best leaving group
-------
and resulting in the formation of diethyl thiophosphate. CP, DZ, PA, and TE all have aryl
functional groups as leaving groups. CE has an alkyl leaving group with four chlorine atoms.
The highly electronegative chlorine atoms may draw the electrons associated with the phospho-
ether linkage creating a slightly more partial positive charge at the tetrahedral phosphorous atom,
which would allow CE to be more susceptible to SN2 attack by hydroxide ion. CP also has
chlorine atoms associated with the aryl structure and is more susceptible to nucleophilic attack
than the other phosphorothioates; however, the aryl structure appears to significantly negate the
electron withdrawing effects of the three chlorine atoms. The oxons of DZ, PA, and CP (i.e.,
phosphate subgroup) have all previously been examined (4, 8). The alkaline hydrolysis rate for
the oxons in the phosphate subgroup is an order of magnitude faster than their corresponding
phosphorothioates (Table 3), which has been previously reported (4, 8, 31). This demonstrates
that the alkaline hydrolysis rate of OP pesticides can be significantly different with minor
changes in molecular structure.
In the phosphorodithioate subgroup, MA was the only pesticide in this subgroup in
which its hydrolytic behavior has been well characterized (6, 7). The observed rate coefficients
for ME at pH 3 and 7 were statistically indistinguishable (Figure 3). At pH 10 and above, ME
alkaline hydrolysis was very rapid (Figure 3, insert). PM was found to be relatively stable at pH
2; however, its hydrolysis increased rapidly with increasing hydroxide ion concentration (Figure
4). Alkaline hydrolysis of PM was found to be very rapid above pH 8 (Figure 4, insert). For the
phosphorodithioate subgroup, ME was the only OP pesticide in this subgroup found to be stable
at neutral pH. Neutral hydrolysis for the phosphorodithioate subgroup was then operationally
defined as the pH where each pesticide was the most stable. By this definition, the neutral
hydrolysis rate coefficients are comparable to the phosphorothioate and phosphate subgroups.
10
-------
Alkaline hydrolysis was generally found to be more rapid than the other two subgroups. All the
phosphorodithioate pesticides in Table 2 have methyl esters linkages at the tetrahedral
phosphorus atom. Alkaline hydrolysis may be faster due to the fact that SN2 nucleophilic attack
by the hydroxide ion is hindered more by the ethyl esters than the methyl esters. Chlorpyrifos
and parathion alkaline hydrolysis rates have been found to be slower when compared to
chlorpyrifos-methyl and parathion-methyl (8). Also, the thiol ester could be more susceptible to
nucleophilic attack by hydroxide ion than phenyl esters. This makes it difficult to directly
compare the phosphorodithioates to the phosphorothioates at environmentally relevant
conditions.
3.2 OP Pesticide Degradation in the Presence of Aqueous Chlorine
In the presence of aqueous chlorine, the oxidation of CP was found to be first order
resulting in an overall second order reaction (4). Therefore, observed loss of each OP pesticide
in the presence of aqueous chlorine was assumed to be first order with respect to the OP
pesticide. If ln([OP]/[OP]o) versus time (t) plots are linear, then this assumption would be valid
when there is a molar excess of chlorine. The observed first-order rate coefficients (k0bs) for all
the OP pesticides were determined from these plots via the slope of the regression line as shown
in the following expression.
[OP]
In-—— = -kht (1)
[OP]0
Under pseudo first-order chlorination conditions, all the OP pesticides from CE-TE exhibited a
first order dependency with respect to the OP pesticide in the presence of excess aqueous
chlorine at pH 6.5 (Figures 5-11). The reaction order of the aqueous chlorine reacting with each
OP pesticide was determined by plotting the log of k0bs versus the log of the initial chlorine
11
-------
concentration at pH 6.5 (Figure 12). Since the slope of the regression line is approximately 1 for
all OPs, this indicates that the loss of OP pesticides can be described as a second-order reaction.
The apparent loss of each OP pesticide in the presence of aqueous chlorine at a specific
pH could then be described by the following rate expression where kapp is the apparent second-
order rate coefficient at a specific pH (equation 2). The observed first-order rate coefficients at
^P = -kapp[HOCl]T[OP] (2)
each pH were assumed to linearly increase with increasing free chlorine concentration (equation
3). The kapp for each OP pesticide was determined by plotting k0bs versus the initial
kobs=kapP[HOCl]T (3)
total aqueous chlorine concentration. Figure 13 shows that kobs increased linearly with
increasing chlorine concentration at pH of 6.5. Diazinon and phosmet degraded the fastest at this
pH, while the rest of the OP pesticides exhibited similar reactivity with aqueous chlorine. Since
approximately 91% of the active chlorine is in the HOC1 form, the apparent rate coefficient will
be very close to the intrinsic rate coefficient of hypochlorous acid reacting with each OP
pesticide. As pH increased from 6.5 to 9, k0bs decreased for all the OP pesticides (Figure 14).
When the pH of the aqueous system shifts from neutral to alkaline pH, the chlorine species also
shifts from HOC1, which has a pKa of 7.5 (32), to OC1". The decrease in the k0bs would be
expected if hypochlorous acid is the dominant reacting species resulting in oxon formation.
Oxons for all the parent OP pesticides are not commercially available; therefore, transformation
products were identified by their mass spectra. For all the OP pesticides, their corresponding
oxons were the only transformation products identified by NIST mass spectra database at pH 6.5.
HOC1 appears to be the active chlorine species over the pH range investigated responsible for OP
oxidation and oxon formation.
12
-------
The transformation of OP pesticides can become relatively complicated at alkaline pH.
Several transformation pathways occur simultaneously resulting in unique transformation
products for each pathway. At pH 9, approximately 97% of the active chlorine will be in the
hypochlorite form. However, all the OP pesticides degraded rapidly in the presence of aqueous
chlorine (Figure 15). Phosmet degraded faster than diazinon due to PM being more susceptible
to alkaline hydrolysis. Just like at pH 6.5, the observed rate coefficients increased linearly with
increasing chlorine concentration. However, oxon products were not necessarily the major
transformation product. CP and its corresponding oxon were both found to be susceptible to
chlorine-assisted hydrolysis (4). Since tetrahedral phosphorus atoms are known to be more
susceptible to hydrolysis by supernucleophiles (33), and hypochlorite is a supernucleophile,
intrinsic rate coefficients for both the chlorine-assisted hydrolysis and the HOCl-oxidation
pathways need to be determined for each OP pesticide.
Degradation pathway models have been shown to be an effective tool determining rate
coefficients in complex systems. Previously, researchers have used the monochloramine
autodecomposition model to determine rate coefficients for the reactions of monochloramine and
dichloramine with triclosan and NOM (27, 28). More applicable to this reaction system, Duirk
and Collette (4) developed a degradation pathway model for CP in order to determine the
intrinsic rate coefficients for both HOC1 and OC1". Knowing that the pH dependency on the rate
of OP loss is due to chlorine speciation, pooled data at pH 6.5 and 9 will be used to parameterize
the intrinsic rate coefficients for HOC1 (knoci,op) and OC1" (koci,op) reacting with each OP
pesticide using the following system of ordinary differential equations. In the following
equations, koci,opo is the intrinsic rate coefficient for hypochlorite assisting in the hydrolysis of
13
-------
= -5kHOCl!OP[HOCl][OP] - kocloP[OCr][OP] -
d[HOCl]T
1 . ~" n*^A_-l, ^Jf LJL ~ J W^l, ^Jf L JL ~ J / A\
k0ci,opo[OCr][OPO]
= -kHoci,op [HOC1] [OP] - kh CP [OP] - kocl;OP [OCr ] [OP] (5)
UL
^[P = kHoci,oP [HOC1][OP] - khjOPO [OPO] - koaopo [OCr ][OPO] (6)
^S = k [0P]+it [ocr ][OP]+kh;OPO [OPO] +
dt (7)
k0ci,opo[OCr][OPO]
the OPO (oxon transformation product), and kh,op and kh,opo are the rate coefficients for both OP
and OPO hydrolysis, and OPH represents the hydrolysis product. Rate equations were written
based on the stoichiometric equations in Table 4. For most of the OP pesticides, only equations
4 and 5 are applicable due to the inability to accurately quantify the oxon product.
The knoci,op were found to cluster in the range of 1.7-2.2 X 106 M^h"1, with CE and DZ
being notable exceptions (Table 5). The k0ci,op appear to show differences between the
subgroups. For the phosphorothioate subgroup, koci,op appears to be an order of magnitude
faster than kB,op (Tables 3 and 5). However, the OC1" does not appear to increase the rate of
hydrolysis as significantly for the phosphorodithioate subgroup. This could be due to the fact
that they are generally found to be extremely susceptible to nucleophilic attack from hydroxide
ion. DZ is the only OP pesticide in which an intrinsic rate coefficient has been previously
reported (17), kHOci,Dz = 4.68 x 105 M'V1 and kOCi,Dz =972 M'V1. Although the hypochlorite
intrinsic rate coefficients were similar, HOC1 rate coefficients were found to be an order of
magnitude different. The difference can be explain because kHoci,Dz was approximated from
data gathered only over the pH range of 9.5-11, and it was thought that both HOC1 and OC1"
14
-------
resulted in oxon formation. Since parallel reaction pathways can occur at any pH, the most
accurate method to parameterize reaction rate coefficients of OP pesticides in the presence of
chlorine would be to use a comprehensive transformation pathway model and data sets relevant
to drinking water treatment conditions.
Frontier molecular orbital theory has been used to correlate oxidation rate coefficients
with the easily calculated energy of highest occupied molecular orbital (EHOMO) (34). EHOMO was
initially thought to be able to correlate all 8 OP pesticides; however, subgroup differences were
quickly unveiled (Figure 16). The phosphorodithioate subgroup was found to have a slightly
higher potential than the phosphorothioate subgroup. This is most likely due to the sulfur
linkage and the methyl esters at the tetrahedral phosphorus atom allowing them to be more easily
oxidized by chlorine than the phosphorothioate subgroup, which primarily had ethyl and phenyl
esters. EHOMO was found to be a good molecular descriptor describing the oxidation of OP
pesticides within each subgroup.
Other molecular descriptors were needed in order to correlate the reactivity of
hypochlorite ion with each OP pesticide. Cross correlations use rate coefficients as molecular
descriptors with a well understood reaction mechanism (i.e., SN2 alkaline hydrolysis) and infer a
mechanistically similar reaction for a different reactant (i.e., chlorine-assisted hydrolysis) (35).
Therefore, correlating koci,op with alkaline hydrolysis rate coefficients for all 8 OP pesticides
would then confirm that OC1" acts as a nucleophile accelerating OP pesticide hydrolysis.
Subgroup differences were evident as the phosphorothioates generally hydrolyze slower at
alkaline pH than the phosphorodithioates (Figure 17); however, chlorine appeared to have a
greater effect accelerating OP pesticide hydrolysis as indicated by the slope of the regression line
being an order of magnitude greater then the slope for the phosphorodithioates. Even with the
15
-------
differences between the two OP pesticide subgroups, this cross correlation validates that OC1"
acts like a nucleophile attacking the tetrahedral phosphorus atom accelerating the hydrolysis of
OP pesticides. However, it is important to note that HOC1 still rapidly transforms OP pesticides
to their corresponding oxon at pH 9. Oxon stability in the presence of aqueous chlorine still
needs to be investigated.
3.3 OP Oxon Chlorine-Assisted Hydrolysis
Chlorine assisted hydrolysis was first observed by Edwards et al., (33) investigating the
factors that determine the reactivity of nucleophiles, which are basicity, polarizability, and the
presence of unshared pairs of electrons on the adjacent atom to the nucleophilic atom (i.e., the
alpha effect). The three lone-pairs of electrons on both the chlorine and oxygen atom enhance
hypochlorite's nucleophilicity towards specific moieties such as a tetrahedral phosphorus (i.e.,
phosphoesters) and carbonyls. Hypochlorite is considered to be a supernucleophile towards
phosphoesters because of these repulsions (36). At pH 9, DZO, MAO, and PAO loss was
observed in the presence of aqueous chlorine (Figure 18). With subgroup differences aside, the
loss of each oxon increased linearly proportional with increasing chlorine concentrations. DZO
and PAO were significantly more stable than MAO at pH 9 because the phosphorothiolates are
more susceptible to alkaline hydrolysis (8).
The rate coefficients for chlorine-assisted hydrolysis for all four oxons were found only
to be slightly larger than their corresponding parents (Table 5). This was first reported with CP
and CPO in the presence of chlorine and bromine (4, 20). This is due to the structure of the
nucleophile, OC1", and not minor structural difference between a phosphorothioate and phosphate
analogs (36). Since tetrahedral phosphate moieties are susceptible to SN2 attack by OC1" (33),
the difference in the partial positive charge at the tetrahedral phosphorus atom between (P=S)
and (P=O) does not significantly influence the rate of nucleophilic attack. Therefore, chlorine-
16
-------
assisted hydrolysis rates for oxons can then be assumed to be approximately the same as their
corresponding parents. This could aid risk assessors estimate exposure to the more toxic oxon
products as well as direct future exposure research activities related to oxon stability in drinking
water distribution systems.
3.4 OP Pesticide-Chlorine Model Validation
With the available rate coefficients determined here and in cited literature, OP pesticide
transformation pathways in the presence of aqueous chlorine can be predicted under drinking
water conditions. Diazinon (DZ) and parathion (PA) were chosen because their corresponding
oxons and hydrolysis products are commercially available. Diazinon was found to react the
fastest among the OP pesticides selected (Table 5). At a pH and chlorine concentration typical
of drinking water treatment, DZ was rapidly transformed to diazoxon (DZO) (Figure 19). Also,
DZO is stable in the presence of chlorine for over 48 hours at neutral pH (18). At pH 8, PA was
also rapidly transformed to paraoxon (PAO) (Figure 20). PAO was also found to be relatively
stable at alkaline pH, as shown previously (Figure 18). This model has already been shown to
adequately predict CP transformation and degradation of CPO in the presence of natural
occurring aqueous constituents (i.e., bromide and natural organic matter) (20). Regulators could
potentially use this model to access potential exposure to OP pesticides and their more toxic
oxon products using the structure-activity relationships in Figures 16 and 17 to estimate reaction
rate coefficients across the phosphorothioate and phosphorodithioate subgroups.
3.5 OP Pesticide Degradation in the Presence of Monochloramine
Observing the direct reaction of monochloramine with analytes in solution can be very
difficult at environmentally relevant concentrations. Previous work elucidating the reaction
mechanism of monochloramine with bromide, cyanide, or nitrite was performed at high
millimolar concentrations of the reactants (29, 37, 38). Due to the autodecomposition
17
-------
mechanism of monochloramine (24), it is nearly impossible to observe the direction reaction of
monochloramine with specific analytes when at environmentally relevant concentrations.
Therefore, the monochloramine autodecomposition model can be used as a tool to validate
proposed transformation pathways with comprehensive data sets over a range of relevant
conditions.
The observed loss of each OP pesticide was assumed to be first-order with respect to each
OP. Monochloramine, dichloramine, and hypochlorous acid will exist simultaneously and vary
in concentration as a function of pH. Therefore, the k0bs presented in Figures 21 and 22 should
be viewed as a function of total oxidant in the aqueous system. The observed first-order rate
coefficients for CP and DZ both decreased as pH increased from 6.5 to 9 (Figure 21). This was
found to be very similar to the reaction of chlorine with both CP and DZ (Figure 14). MA
followed the same trend initially until approximately pH 7.5 when alkaline hydrolysis became
the dominant degradation pathway (Figure 22). However, it appears that dichloramine and
hypochlorous acid are primarily responsible for the loss of the OP pesticides at pH 6.5 and 7. As
pH grows more alkaline, the direct reaction between monochloramine and the OP may be
inconsequential compared to alkaline hydrolysis.
OP pesticide degradation pathways in the presence of monochloramine will be intimately
intertwined with monochloramine autodecomposition. Since alkaline hydrolysis of MA was a
significant loss pathway, further discussion on OP pesticide degradation pathways will primarily
refer to CP and DZ. As pH decreased so did the observed rate of OP pesticide loss. The initial
monochloramine concentration was constant for all the experiments; therefore, the direct reaction
of monochloramine with OP pesticides appears to be minimal. However, near neutral pH three
possible oxidants are formed due to monochloramine autodecomposition that could be
18
-------
responsible for the loss of OP pesticides: hypochlorous acid (HOC1), dichloramine (NHC12), and
monochlorammonium ion (NH3C1+). Hypochlorous acid forms primarily due to the hydrolysis
of monochloramine (Table 1 reaction 2), and it does react with all three OP pesticides. However,
HOC1 can either react with ammonia to reform monochloramine or with monochloramine to
form dichloramine (Table 1 reaction 3). Dichloramine formation is also due to monochloramine
disproportionation, which is acid catalyzed Table 1 reaction 5. This acid-catalyzed reaction
involves the formation of monochlorammonium ion (39).
NH2C1+H3O+ - NH3C1+ H2O K = 28 M'1 (8)
Even though phosphate and carbonate salts participate as proton donors at high concentrations,
they are several orders of magnitude less effective in the formation of monochlorammonium ion
at environmentally relevant concentrations. Therefore, the primary dichloramine formation
pathway will be due to HOC1 reacting with NH2C1. This can be significant because dichloramine
has been found to be more reactive with cyanide, sulfite, and triclosan than monochloramine (28,
29, 38). The following are proposed additional reactions of chlorinated oxidants with OP
pesticides in the presence of monochloramine.
NH2C1 + OP >• products kNH2ci,op (9)
NHC12 + OP >• products kNHcL2,op (10)
Since hypochlorous acid is a product of monochloramine autodecomposition, all known OP
degradation pathways will also be included in the partial rate expressions, equations 11-17,
which are added to the full monochloramine autodecomposition model (Table 6).
d[NH2Cl]
dt
d[NHC!2]
dt
= -kNH2ci,op[NH2Cl][OP] + kNHCL2;OP[NHCl2][OP] (11)
= -kNHci2,op[NHCl2][OP] (12)
19
-------
d[H°C1]T = -kHOci OP [HOC1][OP] - kocl OP [OC1- ][OP] -
at (13)
kocLopo[OCl-][OPO]
d[NH3]T
dt
d[OP]
= kNHC2ci,op[NH2Cl][OP] (14)
dt
d[OPO]
dt
d[OPH]
= -kHoci,op [HOC1][OP] - kh CP [OP] - kocl OP [OCr ][OP] (15)
= kHOci,op[HOCl][OP] - khoPO[OPO] - kocl OPO[OCr ][OPO] (16)
= khoP[OP] + kocloP[OCr][OP] +
k0ci,opo[OCr][OPO]
Additional assumptions are that in the monochloramine-OP pesticide reaction model the
reaction of monochloramine and dichloramine with each OP pesticide do not lead to oxon
formation. Also, that the reactions of monochloramine and dichloramine are 1-to-l elemental
stoichiometric reactions. The reaction of dichloramine and OP pesticides results in the
regeneration of monochloramine, which is similar to dichloramine reacting with inorganics (29,
38). The direct reaction of monochloramine with OP pesticides results in ammonia formation.
Finally, rate coefficients determined using the monochloramine-OP pesticide reaction model
exhibit no pH dependency.
Two rate coefficients need to be determined for both monochloramine and dichloramine
with these OP pesticides. Since dichloramine rapidly decays above neutral pH (40), experiments
to determine kNH2ci,op were conducted in the presence of excess ammonia at pH 8.5 to promote
monochloramine stability and reformation (41). In the presence of excess ammonia ([NH3]x =
0.02-0.45 or Cl/N = 0.7-0.1 mol/mol), the rate of OP pesticide degradation decreased as
ammonia concentrations increases (Figure 23). Due to MA being susceptible to alkaline
20
-------
hydrolysis, MA loss at pH 8.5 was significantly faster than CP and DZ. For each pesticide, these
data sets were pooled for each OP pesticide and an intrinsic rate coefficient for NF^Cl with each
OP pesticide was determined using nonlinear regression analysis. The rate coefficients found for
the direct reaction of monochloramine with each OP pesticide were relatively small in magnitude
(Table 7) compared to kHoci,op (Table 5). This is not unusual because another study has reported
similar results with triclosan (28). Knowing the knoci,op of each OP pesticide, the remaining
data for each OP pesticide was pooled in order to determine kNHci2,op. Dichloramine was found
to be two orders of magnitude great than monochloramine with each OP pesticide (Table 7),
which is consistent with previously reported results (29, 38). Dichloramine appears to play a
significant role in the loss of OP pesticides near neutral pH.
Using the final parameter estimates in Table 7, the monochloramine-OP pesticide
reaction model was used to predict the loss of each OP pesticide in the presence of
monochloramine over the pH range of 6.5-9. Hydrolysis was not a major loss pathway for either
CP or DZ over this range; therefore, oxidation was the primary degradation pathway for both
pesticides in the presence of monochloramine (Figures 24 and 25). The loss of CP and DZ in the
presence of monochloramine was found to be relatively fast at pH 6.5 due to the formation of
dichloramine and hypochlorous, which then react with the OP pesticides. As pH increased to 9,
the rate of CP and DZ loss in the presence of monochloramine decreased due to: 1) the rapid
disproportionation of dichloramine above neutral pH (42), 2) speciation of ammonium to
ammonia which rapidly reacts with hypochlorous acid reforming monochloramine, 3)
hypochlorous acid speciation to hypochlorite which does not participate in monochloramine
reformation, and 4) monochloramine not being as reactive with either pesticide compared to
other chlorinated oxidants. Since hydrolysis of MA is a significant degradation pathway at pH
21
-------
7.5 and above, modeling MA hydrolysis was very important in order to correctly parameterize
kNH2ci,op and kNnci2,op. As shown in Figure 26, the model was found very capable of predicting
both monochloramine and MA concentrations. The excellent correlation between the
experimental results and the model predictions indicate that the fundamental reactions
responsible for the loss of both monochloramine and the OP pesticides were incorporated into
the monochloramine-OP pesticide reaction model.
Unlike the chlorination experiments, only a select few OP pesticides were used in the
chloramination experiments. Since CP and DZ are from the phosphorothioate subgroup and MA
is from the phosphorodithioate subgroup, it is not possible to correlated the reactivity of
dichloramine and monochloramine to EHOMO as with knoci,op. However, the reactivity of the
three chlorinated oxidants can be correlated to each OP pesticide using half-wave potentials
(Ei/2). Half-wave potentials have been shown to be a good descriptor for the electrophilicties of
HOC1, M^Cl, and NHCb with organic and inorganic compounds (28); however, using half-
wave potentials to correlate structure to reactivity is not as robust as thermodynamic parameters
(43). Figure 27 shows a good correlation for the three oxidants with the three pesticides. It is
interesting to note that the slope for each pesticide was found to be the same (slope = 0.30) with
just slight differences in the intercept. This appears to indicate that degree of electrophilicity of
each oxidant with each OP pesticide may be proportional across the OP pesticide class and not
dependent on subgroup. Therefore, estimates for the reactivity of monochloramine and
dichloramine with other OP pesticides could be made from knowing only knoci,op (Table 5), or
the kNH2ci,op and kNncL2,op for CP or MA could be used to quickly estimate the fate of other OP
pesticides in the presence on monochloramine..
22
-------
Although the reaction rate coefficients of the chlorinated oxidants were found to correlate
with each OP, they appear not to react via the same mechanism (i.e., reaction center). Product
studies have shown that only the HOC1 reaction pathway resulted in oxon formation, and that the
resulting oxons also appear not to be stable in the presence of monochloramine. The reaction of
monochloramine and dichloramine do not appear to result in oxon formation; however, they may
react with each OP pesticide resulting in a difference transformation product while not mitigating
toxicity. Additional work is needed in order to understand the fate of OP pesticides and their
oxon transformation products in the presence of monochloramine.
23
-------
4 CONCLUSION
Eight OP pesticides and three of the oxon transformation products were investigated in
the presence of chlorinated oxidants. It was originally thought that by investigating a class of
pesticides that structure-activity relationships could easily be developed. However, it quickly
became apparent that subgroups could be correlated within each OP pesticide class; however,
correlations could not be made across the chemical class. This became evident looking at
hydrolysis rate coefficients for the phosphorothioate, phosphate, and the phosphorodithioate
subgroups. Chlorination of all eight pesticides led to a similar conclusion. Chloramination of
three of the most commonly detected OP pesticides provided some insight into how chlorinated
oxidants react with OP pesticides. Overall, OP pesticide structure was found to dictate reactivity
in all three systems.
Hydrolysis rate coefficients were determined for 4 pesticides due to lack of reported
intrinsic rate coefficients. Pesticides from the phosphorodithioate group hydrolyzed rapidly
under alkaline conditions. Also, oxons were found to be more susceptible to alkaline hydrolysis
compared to the corresponding thionates due to the oxygen being more electronegative than
sulfur, which facilitates the nucleophilic attack by hydroxide ion at the slightly more positively
charged phosphorus atom. Also, ethyl esters appear to hinder nucleophilic attack by hydroxide
ion compared to a structurally similar pesticide with methyl esters at the phosphorus atom;
however, no direct comparison can be inferred for the difference in hydrolysis rates between the
phosphorothioate and phosphorodithioate subgroups investigated here. Therefore, more
hydrolysis studies need to be conducted at environmentally relevant conditions to determine the
hydrolytic behavior of OP pesticides, which intrinsic hydrolysis rate coefficients could then be
predicted via structure-activity relationships.
24
-------
In the presence of aqueous chlorine, eight OP pesticides were investigated and intrinsic
rate coefficients for both HOC1 and OC1" were calculated. Since multiple reaction pathways
occur simultaneously over the pH range of 6.5-9, a chlorine-OP pesticide reaction model
previously developed for CP was used. Intrinsic rate coefficients were first found for HOC1 at
pH 6.5 then OC1" at pH 9. The reaction of HOC1 with OP pesticides results in the rapid
formation of the corresponding oxon. Hypochlorite acts as a nucleophile accelerating the
hydrolysis of the OP pesticide and not resulting in oxon formation. When the intrinsic rate
coefficients were correlated to molecular descriptors, subgroup differences were readily
apparent. The energy of the highest occupied molecular orbital (£HOMO) showed that kHoci,op
correlated within the two subgroups. The phosphorothioates were found to be lower in energy,
and the phosphorodithioates were more reactive as a group as indicated by the greater slope of
the regression line. The alkaline hydrolysis rate (kB,op) was found to correlate to koci,op within
each OP subgroup, which also implies that the mechanism of OC1" and OH" are similar.
Therefore, OC1" does attach the tetrahedral phosphorus as a nucleophile accelerating the
hydrolysis of all eight OP pesticides and three oxons.
Three of the most commonly detected OP pesticides (CP, DZ, and MA) in drinking water
supplies were examined in presence of chloramines. The loss of all three pesticides under
chloramination conditions was found to be highly pH dependent. At pH 6.5, the primary
degradation pathway for OP pesticides was due to reaction with dichloramine and hypochlorous
acid, which are formed during monochloramine autodecomposition. The direct reaction of
monochloramine with each OP pesticide was found to be a minimal degradation pathway, which
was confirmed when parameterizing kNH2ci,op for each OP pesticide in the presence of excess
ammonia. The order of reactivity for the three chlorinated oxidants with each OP pesticide was
25
-------
HOC1:» NHCb » NH2C1. The reactivity of the three chlorinated oxidants correlated with half-
wave potentials (£1/2) for each OP pesticide.
The stability of OP pesticides was investigated under conditions similar to drinking water
treatment. The eight pesticides in this study showed that reaction parameters could not be
correlated across the OP pesticide class, but correlations could be made within subgroups
indicating that minute structure differences can significantly impact reactivity. Structure-activity
relationships developed here can allow for the prediction of critical reaction parameters of OP
pesticides in the presence of chlorinated oxidants. Along with the presented model, regulators
could access the potential risk associated when consuming potable water contaminated with OP
pesticides.
26
-------
5 REFERENCES
1. Kolpin, D. W.; Barbash, J. E.; Gilliom, R. J., Pesticides in ground water of the United
States, 1992-1996. Ground Water 2000, 38, (6), 858-863.
2. Verstraeten, I. M.; Thurman, E. M.; Lindsey, M. E.; Lee, E. C.; Smith, R. D., Changes in
concentrations of triazine and acetamide herbicides by bank filtration, ozonation, and
chlorination in a public water supply. Journal of Hydrology 2003, 276, (1-4), 287-288.
3. Coupe, R. H.; Blomquist, J. D., Water-soluble pesticides in finished water of community
water supplies. Journal American Water Works Association 2004, 96, (10), 56-68.
4. Duirk, S. E.; Collette, T. W., Degradation of Chlorpyrifos in Aqueous Chlorine
Solutions: Pathways, Kinetics, and Modeling. Environmental Science and Technology
2006,40,546-551.
5. Pehkonen, S. O.; Zhang, Q., The degradation of organophosphorus pesticides in natural
waters: A critical review. Critical Reviews in Environmental Science and Technology
2002, 32, (1), 17-72.
6. Wolfe, N. L.; Zepp, R. G.; Baughman, G. L.; Gordon, J. A., Kinetic Investifation of
Malathion Degradation in Water. Bulletin of Environmental Contamination and
Toxicology 1975,13, (6), 707-713.
7. Wolfe, N. L.; Zepp, R. G.; Gordon, J. A.; Baughman, G. L.; Cline, D. M., Kinetic of
Chemical Degradation of Malathion in Water. Environ. Sci. Technol. 1977,11, (1), 88-
93.
8. Faust, S. D.; Gomaa, H. M., Chemical Hydrolysis of Some Organic Phosphorus and
Carbamate Pesticides in Aquatic Environments. Environmental Letters 1972, 3, (3), 171-
201.
9. Macalady, D. L.; Wolfe, N. L., New Perspectives on the Hydrolytic Degradation of the
Organophosphorothioate Insecticide Chlorpyrifos. Journal of Agricultural and Food
Chemistry 1983, 31, (6), 1139-1147.
10. Community Water System Survey. United States Environmental Protection Agency: 1997;
Vol. l,EPA815-R-97-011a.
11. Lopez, A.; Mascolo, G.; Tiravanti, G.; Santori, M.; Passino, R., Oxidation of Sulfur-
Containing S-Triazines During Groundwater Hypochlorination. Water Science and
Technology 1994, 30, (7), 53-59.
12. Mascolo, G.; Lopez, A.; Passino, R.; Ricco, G.; Tiravanti, G., Degradation of Sulfur-
Containing S-Triazines During Water Chlorination. Water Res. 1994, 28, (12), 2499-
2506.
27
-------
13. Adams, C. D.; Randtke, S. J., Removal of Atrazine from Drinking-Water by Ozonation.
Journal American Water Works Association 1992, 84, (9), 91-102.
14. Miles, C. J., Degradation of Aldicarb, Aldicarb Sulfoxide, and Aldicarb Sulfone in
Chlorinated Water. Environ. Sci. Technol. 1991,25, (10), 1774-1779.
15. Mason, Y. Z.; Choshen, E.; Ravacha, C., Carbamate Insecticides - Removal from Water
by Chlorination and Ozonation. Water Res. 1990, 24, (1), 11-21.
16. Kodama, S.; Yamamoto, A.; Matsunaga, A., S-Oxygenation of thiobencarb in tap water
processed by chlorination. Journal of Agricultural and Food Chemistry 1997, 45, (3),
990-994.
17. Zhang, Q.; Pehkonen, S. O., Oxidation of diazinon by aqueous chlorine: Kinetics,
mechanisms, and product studies. Journal of Agricultural and Food Chemistry 1999, 47,
(4), 1760-1766.
18. Magara, Y.; Aizawa, T.; Matumoto, N.; Souna, F., Degradation of Pesticides by
Chlorination During Water-Purification. Water Science and Technology 1994, 30, (7),
119-128.
19. Wu, J. G.; Laird, D. A., Abiotic transformation of chlorpyrifos to chlorpyrifos oxon in
chlorinated water. Environmental Toxicology and Chemistry 2003, 22, (2), 261-264.
20. Duirk, S. E.; Tarr, J. C.; Collette, T. W., Chlorpyrifos transformation by aqueous chlorine
in the presence of bromide and natural organic matter. Journal of Agricultural and Food
Chemistry 2008, 56, (4), 1328-1335.
21. Pinkston, K. E.; Sedlak, D. L., Transformation of aromatic ether-and amine-containing
Pharmaceuticals during chlorine disinfection. Environ. Sci. Technol. 2004, 38, (14),
4019-4025.
22. Chen, W. H.; Young, T. M., NDMA formation during chlorination and chloramination of
aqueous diuron solutions. Environ. Sci. Technol. 2008, 42, (4), 1072-1077.
23. Chen, Z.; Valentine, R. L., Modeling the formation of N-nitrosodimethylamine (NDMA)
from the reaction of natural organic matter (NOM) with monochloramine. Environ. Sci.
Technol. 2006, 40, (23), 7290-7297.
24. Jafvert, C. T.; Valentine, R. L., Reaction Scheme for the Chlorination of Ammoniacal
Water. Environ. Sci. Technol. 1992, 26, (3), 577-586.
25. Duirk, S. E.; Gombert, B.; Croue, J. P.; Valentine, R. L., Modeling monochloramine loss
in the presence of natural organic matter. Water Res. 2005, 39, (14), 3418-3431.
28
-------
26. Duirk, S. E.; Valentine, R. L., Modeling dichloroacetic acid formation from the reaction
of monochloramine with natural organic matter. Water Res. 2006, 40, (14), 2667-2674.
27. Duirk, S. E.; Valentine, R. L., Bromide oxidation and formation of dihaloacetic acids in
chloraminated water. Environ. Sci. Technol. 2007, 41, (20), 7047-7053.
28. Greyshock, A. E.; Vikesland, P. J., Triclosan reactivity chloraminated waters. Environ.
Sci. Technol. 2006, 40, (8), 2615-2622.
29. Schurter, L. M.; Bachelor, P. P.; Margerum, D. W., Nonmetal Redox Kinetics: Mono-,
Di-, and Trichloramine Reactions with Cyanide Ion. Environ. Sci. Technol. 1995, 29, (4),
1127-1134.
30. Standard Methods for the Examination of Water and Wastewater. 20th ed.; APHA,
AWWA and WEF: Washington D. C., 1998.
31. Wolfe, N. L., Organo-Phosphate and Organophosphorothionate Esters - Application of
Linear Free-Energy Relationships to Estimate Hydrolysis Rate Constants for Use in
Environmental Fate Assessment. Chemosphere 1980, 9, (9), 571-579.
32. Snoeyink, V. L.; Jenkins, D. A., Water Chemistry. John Wiley & Sons: New York, NY,
1980.
33. Edwards, J. O.; Pearson, R. G., Factors Determining Nucleophilic Reactivities. Journal of
the American Chemical Society 1962, 84, (1), 16-24.
34. Hu, J. Y.; Morita, T.; Magara, Y.; Aizawa, T., Evaluation of reactivity of pesticides with
ozone in water using the energies of frontier molecular orbitals. Water Res. 2000, 34, (8),
2215-2222.
35. Canonica, S.; Tratnyek, P. G., Quantitative structure-activity relationships for oxidation
reactions of organic chemicals in water. Environmental Toxicology and Chemistry 2003,
22, (8), 1743-1754.
36. England, W. B.; Kovacic, P.; Hanrahan, S. M.; Jones, M. B., Molecular-Orbital Theory
of Supernucleophiles - Complementary Criteria and Supporting Evidence. Journal of
Organic Chemistry 1980, 45, (11), 2057-2063.
37. Margerum, D. W.; Schurter, L. M.; Hobson, J.; Moore, E. E., Water Chlorination
Chemistry - Nonmetal Redox Kinetics of Chloramine and Nitrite Ion. Environ. Sci.
Technol. 1994, 28, (2), 331-337.
38. Yiin, B. S.; Walker, D. M.; Margerum, D. W., Nonmetal Redox Kinetics: General Acid
Assisted Reactions of Chloramine with Sulfite and Hydrogen Sulfite. Inorganic
Chemistry 1987, 26, (21), 3435-3441.
29
-------
39. Gray, E. T.; Margerum, D. W.; Huffman, R. P., Chloramine Equilibria and the Kinetics
of Disproportionation in Aqueous Solution. In Organometals andOrganometalloids:
Occurance and Fate in the Environment, Brinkman, F. E.; Bellama, J. M., Eds. ACS
Books: Washington D.C., 1978; pp 264-277.
40. Vikesland, P. J.; Ozekin, K.; Valentine, R. L., Monochloramine decay in model and
distribution system waters. Water Res. 2001, 35, (7), 1766-1776.
41. Qiang, Z. M.; Adams, C. D., Determination of monochloramine formation rate constants
with stopped-flow spectrophotometry. Environ. Sci. Technol. 2004, 38, (5), 1435-1444.
42. Jafvert, C. T.; Valentine, R. L., Dichloramine Decomposition in the Presence of Excess
Ammonia. Water Res. 1987, 21, (8), 967-973.
43. Tratnyek, P. G.; Hoigne, J., Kinetics of Reaction if Chlorine Dioxide (OC1O) in Water: 2.
Quantitative Structure-Activity Relationships for Phenolic Compounds. Water Res. 1994,
28, (1), 57-66.
44. Morris, J. C.; Isaac, R. A., A Critical Review of Kinetic and Thermodynamic Constants
for the Aqueos Chlorine-Ammonia System. In Water Chlorination Chemistry:
Environmental Impact and Health Effects, Jolley, R. L.; Burngs, W. A.; Coutruvo, J. A.;
Gumming, R. B.; Mattice, J. S.; Jacobs, V. A., Eds. Ann Arbor Science: Ann Arbor, MI,
1981; Vol. 4, pp 49-62.
45. Margerum, D. W.; Gray, E. T.; Huffman, R. P., Chlorination and the Formation of N-
Chloro Compounds in Water Treatment. In Organometals and Organometalloids:
Occurance and Fate in the Environment, Brinkman, F. E.; Bellama, J. M., Eds. ACS
Books: Washington D.C., 1978; pp 278-290.
46. Hand, V. C.; Margerum, D. W., Kinetics and Mechanisms of the Decomposition of
Dichloramine in Aqueous-Solution. Inorganic Chemistry 1983, 22, (10), 1449-1456.
47. Leao, S. F. Kinetics of Combined Chlorine: Reactions of Substitution and Redox.
University of California Berkeley, Berkeley, CA, 1981.
48. CRC Handbook of Pesticides. CRC Press: Boca Raton, Fl, 1995.
49. Piela, B.; Wrona, P. K., Electrochemical behavior of chloramines on the rotating
platinum and gold electrodes. Journal of the Electrochemical Society 2003, 750, (5),
E255-E265.
30
-------
TABLES
-------
Table 1 Stoichiometric equations and coefficients in the monochloramine
autodecomposition model.
Reaction Stoichiometry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Note:
HOC1 + NH3 > NH2C1 + H2O
NH2C1 + H2O > HOC1 + NH3
HOC1 + NH2C1 > NHC12 + H2O
NHC12 + H2O > HOC1 + NH2C1
NH2C1 + NH2C1 > NHC12 NH3
NHC12 + NH3 > NH2C1 + NH2C1
NH2C1 + NHC12 > N2+ 3H++3Cr
NHC12 + H2O > bl + 2HC1
bl + NHC12 > HOC1+N2+HC1
bl + NH2C1 > N2+H2O+ HC1
HOCI - — H+ + ocr
NH4+ ^ H+ + NH3
H2CO3 ^ H+ + HCCV
HCO3- ^=± H+ + CO32-
ak5 = kH+ [H+] + kH2C03 [H2C03] + kHCQ.
103 M'V, and k = 800 M'V at 25 °C. bl
Rate/Equilibrium Reference
Coefficients (25 °C)
ki= 1.5 x 10 M" h"
k2=7.6 x 10"2 h"1
k3= l.Ox lO'TVrV
k4= 2.3 x 10"3 h"1
ak5= pH dependent
k6=2.16xl08M-V
k7=55.0M'1h'1
k8=4.0x lO5^!"1^1
kg= 1.0 x 10 M" h"
kio=3.0xl07M-V
pKa = 7.5
pKa = 9.3
pKa = 6.3
pKa=10.3
[HCO3']: kH+ =2.5xl07M'2h'1,
: reactive intermediate.
(44)
(44)
(45)
(45)
(40)
(46)
(47)
(42)
(47)
(47)
(32)
(32)
(32)
(32)
k = 4j
32
-------
Table 2 Structures and some chemical properties of select OP Pesticides (48).
Structure and Name
LogK0
Vapor Pressure Water
(mm Hg) Solubility
At 20°C
(mg/L)
Chlorethoxyfos (CE)
S 0—CHj—CH3
T X
Cl—C—CH—0 0—CH2—CH3
Chlorpyrifos (CP)
Cl S O—CH2—CHS
= N P
Cl—( *) O O—CH,—CH,
Diazinon (DZ)
CH,
CH,—CH
CH,
Malathion (MA)
CH3 - CH2 - 0^_ ^O
— CH2— CH3
0 0 — CH2 — CH3
O - CH3
x
CH2 Xp/
CH - S O - CH3
C
CH3 - CH2 - O O
Methidathion (ME)
0 S
— CH3
N — CH2 — S 0 — CH3
CH3-0'
Parathion (PA)
— CH2— CH3
0,N
Phosmet (PM)
P
•0 0—CH,—CH,
0—CH,
N CH2—S O—CH3
Tebupirimfos (TE)
CH,
CH,—
CH,
0—CH2—CH3
-V /
p
0 0—CH—CH,
CH,
4.59
5.11
i.81
2.36
2.5
3.83
2.95
aNA
8.25 xW4 2.0x10°
1.85 x 10"
1.2x10
-2
7.9x10''
1.88 x 10
,-6
9.7x10
-6
3.75X 10"
2.0xlOu
4.0 xlO1
1.4x10"
2.0x10'
6.5xlOu
6.4 xlO'1 2.5 xlO1
5.5xlOu
aNA means not available.
33
-------
Table 3 Neutral and alkaline hydrolysis rate coefficients for 8 OP pesticides and 3 oxon
transformation products. 95% confidence intervals shown in parentheses.
OP Pesticide
Chlorethoxyfos (CE)
Chlorpyrifos (CP)
Chlorpyrifos oxon (CPO)
Diazinon (DZ)
Diazoxon (DZO)
Malathion (MA)
Methidathion (ME)
Parathion (PA)
Paraoxon (PAO)
Phosmet (PM)
Tebupirimfos (TE)
kNOpat25°C
' (h-1)
4.68(±0.69)xlO'4
3.72X 10'4
2.13 x 10'3
1.56 x 10'4
9.99 x 10'4
a7.92x!0'5
a!.42(+1.01)x 10'3
2.66 x 10'4
2.00 x 10'4
b5.55(±0.11)x 10'4
1.10(±0.0.01)xlO'3
kB!op at 25 °C
(M'V)
1.25(+0.30)x 103
37.1
230.2
18.9
165.6
1.98 xlO3
2.22(±0.04)x 102
4.3
46.1
2.73(±0.08)x 105
6.0(±0.1)
References
this work
(9)
(4)
(8)
(6)
this work
(8)
this work
this work
aMA hydrolysis rate coefficient at pH 4
bPM hydrolysis rate coefficient at pH 2
34
-------
Table 4 Stoichiometric equations used in the chlorine-OP pesticide transformation pathway
model.
Reaction Stoichiometry
Rate/Equilibrium Reference
Coefficient (25 °C)
1 5HOC1 + OP kHOC1 op > OPO + 5H+ + SCI" +
OP
4
5
6
OPO khopo
op + ocr
opo + ocr
HOC1^=
koci
oci,op
od,OPO
H+ + ocr
knoci,op = Table 5 this work
kh,OP = kN,OP
+ kB;0p[OH-]
kN,op = Table 3
kB,op = Table 3
kh,opo = kf^opo
+ kB,0po[OH-]
kN,opo = Table 3
kB!opo = Table 3
koci,op = Table 5 this work
koci,opo = Table 5 this work
pKa = 7.5(HOCl/OCr) (32)
35
-------
Table 5 Oxidation and chlorine-assisted hydrolysis rate coefficients for 8 OP pesticides and
3 oxon transformation products. 95% confidence intervals shown in parentheses.
OP pesticide
kHoci,op at 25 °C koci,op or koci,opo References
V) at25°C (M'V)
Chlorethoxyfos (CE)
0.86(+0.18)xl06 15,900+2100
this work
Chlorpyrifos (CP)
Chlorpyrifos oxon (CPO)
1.72 x 10b
990
1340
(4)
Diazinon (DZ)
Diazoxon (DZO)
3.56(±0.65)xl06
627+30
914.1+54.2
this work
Malathion (MA)
Malaoxon (MAO)
1.72(+0.36)xlOb 382+26
565+99
this work
Methidathion (ME)
1.89(+0.12)xl06 252+47
this work
Parathion (PA)
Paraoxon (PAO)
2.20(+0.53)xl06 37+10
48+10
this work
Phosmet (PM)
Tebupirimfos (TE)
2.84(+0.80)xl06 1000+100
1.76(+0.43)xl06 71+13
this work
this work
36
-------
Table 6 Differential equations for the monochloramine autodecomposition model.
- k2[NH2Cl] - k3[HOCl][NH2Cl] + k4[NHC!2] -
1) dt
2k5[NH2Cl]2 + k6[NHC!2 ][NH3] - k7[NHC!2 ][NH2C1] - k10[I][NH2Cl]
= k3[HOCl][NH2Cl]-k4[NHCl2] + k5[NH2Cl]2 -k6[H+][NH3][NHC!2]
2) dt
k7[NH2Cl][NHCl2]-k8[NHCl2][OH-]-k9[I][NHCl2]
d[TOTOCl] = .ki[HOCl][NH3] + k2[NH2Cl] - k3[HOCl][NH2Cl] + k4[NHC!2] +
3) dt
k9[I][NHC!2]
dt
1 = 3k7[NH2Cl]2 +2k8[NHCl2] + k9[I][NHCl2] + k10[I][NH2Cl]
dt
dt
= k8[NHCl2][OH-]-k9[I][NHCl2]
dt
= -(3k7[NH2Cl][NHCl2 ] + 3k9[I][NHCl2 ] + 3k10[I][NH2Cl]
dt
8) k1[NH3][HOCl](«0NH4 - ^001") + k.^OCl - «0NH4)[NH2C1]
k3[NH2Cl][HOCl](-a1OCr ) + k4[NHCl2](«1OCr ) +
+ k6[H+][NH3][NHCl2](aoNH4))//?
37
-------
Table 7 Intrinsic rate coefficients for monochloramine and dichloramine reacting with CP,
DZ, and MA. 95% confidence intervals shown in parentheses.
OP Pesticide kNH2ci!op kNnci2,op
(M'V) (M'V)
Chlorpyrifos (CP) 11.2(±1.2) 2700(+100)
Diazinon(DZ) 21.4(±1.9) 2930(+120)
Malathion (MA) 10.6(+0.6) 2000(+200)
38
-------
FIGURES
39
-------
1.25
1.00 -
0.75 -
0.50 -
0.25 -
1.25
1.00 -
0.75 -
0.25 -
0.00
y(x) = 1252.9(x) -0.0252 r = 0.99
0
—r~
2
300x10-'
600x10-'
[OH-] (M)
900x10
1—
4
—I—
6
PH
-•—•—r
—i—
10
12
Figure 1 Hydrolytic behavior of chlorethoxyfos (CE) over the pH range of 3-11. Insert:
slope represents the alkaline hydrolysis rate coefficient. [CE]0 = 0.5 (jM, [Buffer]T
= 10 mM, and Temperature = 25±1°C. Error bars represent 95% confidence
intervals.
40
-------
u.uuu
0.005 -
0.004 -
0.003 -
0.002 -
0.001 -
,n -
0.002 -I
^ 0.001 •
vi
o
0.000 -
(
y(x) = 6.0 l(x)- 1.05 x 10"3 -r
r2 = 1.00 ^^^^
^^^^
) 25xlO-6 50xlO-6 75xlO-6 lOOxlQ-6 125xlQ-6
[OH-] (M)
•
S
6
PH
10
12
Figure 2 Hydrolytic behavior of tebupirimfos (TE) over the pH range of 3-10. Insert: slope
represents the alkaline hydrolysis rate coefficient. [TE]0 = 0.5 |jM, [Buffer]T =10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
41
-------
0.25
0.20 -
0.15 -
0.10 -
0.05 -
0.25
0.20 -
0.15 -
y(x) = 222.5l(x) -0.01 r = 1.00
SOOxlO-6 600x10-'
[OH-] (M)
900x10
6
PH
10
12
Figure 3 Hydrolytic behavior of methidathion (ME) over the pH range of 3-11. Insert: slope
represents the alkaline hydrolysis rate coefficient. [ME]0 = 0.5 |jM, [Buffer]T =10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
42
-------
30
25 -
20 -
15 -
10 -
5 -
y(x) = 2.73 x 105(x) - 6.36 x 102
0 25xlO-6 50xlO-6 75xlO-6 lOOxlO'6 125x10
[OH-] (M)
6
PH
12
Figure 4 Hydrolytic behavior of phosmet (PM) over the pH range of 2-10. Insert: slope
represents the alkaline hydrolysis rate coefficient. [PM]o = 0.5 |jM, [Buffer]! =10
mM, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
43
-------
[HOC1]T = 10
[HOC1]T = 25
[HOC1]T = 50
[HOCl]T=100nM
-3.0
0.00
0.25 0.50
Time (hours)
0.75
Figure 5 Observed first-order rate of CE loss in the presence of aqueous chlorine at pH 6.5.
[CE]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
44
-------
N
a
N1
Q
[HOCl]T=10nM
A [HOC1]T = 25
• [HOC1]T = 50
[HOCl]T=100nM
0.000 0.025 0.050 0.075 0.100
Time (hours)
0.125
0.150
Figure 6 Observed first-order rate of DZ loss in the presence of aqueous chlorine at pH 6.5.
[DZ]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
45
-------
[HOC1]T= 10nM
[HOC1]T = 25
[HOC1]T = 50
[HOC1] = lOO^M
-6
0.00 0.05 0.10 0.15 0.20
Time (hours)
0.25
0.30
Figure 7 Observed first-order rate of MA loss in the presence of aqueous chlorine at pH 6.5.
[MA]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
46
-------
-0.5 -
-1.0 -
>! -1.5 -
[HOCl]T=10nM
A [HOC1]T = 25
• [HOC1]T = 50
[HOCl]T=100nM
-2.0 -
-2.5 -
0.00
0.05
0.10 0.15
Time (hours)
0.20
0.25
Figure 8 Observed first-order rate of ME loss in the presence of aqueous chlorine at pH 6.5.
[ME]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
47
-------
[HOC1]T = 10
[HOC1]T = 25
[HOC1]T = 50
[HOC1]T = 100
-7
0.0
0.1
0.2 0.3
Time (hours)
0.4
0.5
Figure 9 Observed first-order rate of PA loss in the presence of aqueous chlorine at pH 6.5.
[PA]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
48
-------
[HOCl]T=10nM
A [HOC1]T = 25
• [HOC1]T = 50
[HOCl]T=100nM
-3.0
0.00
0.05
0.10 0.15
Time (hours)
0.20
0.25
Figure 10 Observed first-order rate of PM loss in the presence of aqueous chlorine at pH 6.5.
[PM]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
49
-------
[HOC1]T = 10
[HOC1]T = 25
[HOC1]T = 50
[HOCl]T=100nM
-3.0
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35
Time (hours)
Figure 11 Observed first-order rate of TE loss in the presence of aqueous chlorine at pH 6.5.
[TE]0 = 0.5 nM, [HOC1]0 = 10-100 |iM, [PO4]T = 10 mM, and Temperature =
25±1°C. Experiments performed in triplicate.
50
-------
0.0
-0.5 -
-1.0 -
-1.5 -
-2.0 -
-2.5 -
-3.0
•
D
CP
CE
A DZ
v MA
• ME
O PA
T PM
A TE
slopeav = 1.05+0.09
3.8 4.0 4.2 4.4 4.6
-log[HOCl]T
4.8
5.0
5.2
Figure 12 The reaction order of chlorine with 8 OP pesticides at pH 6.5. [OP]0 = 0.5 |jM,
[PO4]T = 10 mM, Temperature = 25±FC, and [HOC1]T =10-100
51
-------
-T3
M
Figure 13 Second-order apparent rate coefficient for the OP pesticides at pH 6.5. [OP]0 = 0.5
HM, [PO4]T = 10 mM, Temperature = 25±1°C, and [HOC1]T = 0-100 |iM. Error
bars represent 95% confidence intervals.
52
-------
J3
M
125
100 -
75 -
50 -
25 -
0
0
6.0 6.5 7.0 7.5 8.0
PH
8.5
D
O
T
A
CP
CE
DZ
MA
ME
PA
PM
TE
9.0
9.5
Figure 14 The pH dependency of the first-order observed rate coefficients for the OP
pesticides. [OP]0 = 0.5 |iM, [HOC1]T = 25 |iM, [Buffer]T = 10 mM, and
Temperature = 25 °C. Error bars represent 95% confidence intervals.
53
-------
15.0
12.5 -
10.0 -
7.5 -
5.0 -
2.5 -
0.0
0
25
50
75
100
125
[HOC1]T
Figure 15 Second-order apparent rate coefficient for the OP pesticides at pH 9. [OP]0 = 0.5
HM, [CO3]T = 10 mM, Temperature = 25±FC, and [HOC1]T = 0-100 \)M. Error
bars represent 95% confidence intervals.
54
-------
15.5
15.0 -
14.5 H
o
8
- 14.0
13.5 -
13.0
• Phosphorothioate
• Phosphorodithioate
y(x) =
y(x) = 95.22(x) +47.09
r2 = 0.99
-0.37
-0.36 -0.35 -0.34
^OMO-HFMIDI V6 V)
-0.33
Figure 16 Relationship between the knoci,op with EHOMO as a function of OP pesticide
subgroup. Error bars about the regression line represent 95% confidence intervals.
55
-------
4 -
o
o
00
.2 2 H
1 -
Phosphorothi oate
Phosphorodithi oate
y(x) = 1.00(x)+1.22
r = 0.99
log (kROP)
Figure 17 Relationship between the koci,op with alkaline hydrolysis rate coefficient as a
function of OP pesticide subgroup. Error bars about the regression line represent
95% confidence intervals.
56
-------
0.4
0.
0
50
y(x) = 565.29(x) + 0.21 r = 0.99
y(x) = 914.13(x) + 0.00 r = 0.99
y(x) = 48.35(x) + 0.00 r = 0.99
100
150
200
250
[HOCl]TxlO'6(M)
Figure 18 Second-order apparent rate coefficient of 3 OP pesticide oxons at pH 9. [OP]0 =
0.5 nM, [CO3]T = 10 mM, Temperature = 25±FC, and [HOC1]T = 0-200 |iM.
Error bars represent 95% confidence intervals for both data points and regression
lines.
57
-------
o
'I
-4—»
(L>
O
O
O
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Figure 19
Time (hours)
Experimental and model results for DZ loss in the presence of chlorine at pH 7.0.
[DZ]0 = 0.6 nM, [HOC1]T = 20 \)M, [PO4]T = 10 mM, and Temperature = 25±FC.
Error bars represent 95% confidence intervals. Lines represent model results.
58
-------
o
'I
-4—»
(L>
O
O
O
0.6
0.5 -
0.4 -
0.3 -
0.2 -
0.1 -
0.0
0.0
0.1
0.2
o.:
0.4
0.5
0.6
Figure 20
Time (hours)
Experimental and model results for PA loss in the presence of chlorine at pH 8.0.
[PA]0 = 0.57 nM, [HOC1]T = 25 |iM, [PO4]T = 10 mM, and Temperature = 25±FC.
Lines represent model results.
59
-------
0.025
0.020 -
0.015 -
0.010 -
0.005 -
O CP
• DZ
• CP Control
D DZ Control
8
10
Figure 21 First-order observed loss of CP and DZ in the presence of monochloramine as a
function of pH. [NH2C1] = 50 |iM, [OP]0 = 0.5 |iM, Cl/N = 0.7 mol/mol, [Buffer]T
= 10 mM, Temperature = 25±1°C, and pH 6.5-9.0. Error bars represent 95%
confidence intervals.
60
-------
0.04
0.03 -
0.02 -
0.01 -
0.00
6.0
• MA
O MA Control
I
n
6.5
7.0
7.5
8.0
8.5
9.0
Figure 22
First-order observed loss of MA in the presence of monochloramine as a function
of pH. [NH2C1] = 50 nM, [MA]0 = 0.5 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10
mM, Temperature = 25±1°C, and pH 6.5-9.0. Error bars represent 95% confidence
intervals.
61
-------
75xlO-3
SOxlO-3 -
25xlO-3
2xlO
IxlO'3 -
0.5x10
0.00
0.25
0.50
0.75
Cl/N (mol/mol)
Figure 23 Observed first-order observed loss of three OP pesticides in the presence of
monochloramine at pH 8.5. [NH2C1] = 50 |iM, [OP]0 = 0.5 |iM, Cl/N = 0-0.7
mol/mol, [H3BO3]x =10 mM, and Temperature = 25±1°C. Error bars represent
95% confidence intervals.
62
-------
60.0
o
O
O
O
O pH6.5,NH2Cl
A pH 7, NH2C1
n pH 8, NFLC1
pH6.5, CP
pH 7,CP
pH 8, CP
pH 9, CP
pH 9, NFLC1
0
50
100
150
200
250
300
350
Figure 24
Time (hours)
Experimental and model results for CP loss in the presence of monochloramine.
[CP]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH
6.5-9, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
Lines represent model results.
63
-------
o
O
O
O
O pH 6.5, NH2C1
A pH 7, NH2C1
n pH 8, NH2C1
v pH 9, NFLC1
0
• pH 6.5, DZ
pH 7, DZ
pH 8, DZ
pH 9, DZ
100
200
300
400
Figure 25
Time (hours)
Experimental and model results for DZ loss in the presence of monochloramine.
[DZ]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH
6.5-9, and Temperature = 25±1°C. Error bars represent 95% confidence intervals.
Lines represent model results.
64
-------
o
O
O
O
O pH6.5,NH2Cl
A pH 7, NH2C1
n pH 8, NFLC1
pH8.5,NFLCl
pH 6.5, MA
pH 7, MA
• pH 8, MA
T pH 8.5, MA
0
50
100
150
200
250
300
350
Figure 26
Time (hours)
Experimental and model results for MA loss in the presence of monochloramine.
[MA]0 = 0.5 nM, [NH2C1]0 = 50 |iM, Cl/N = 0.7 mol/mol, [Buffer]T = 10 mM, pH
6.5-8.5, and Temperature = 25±1°C. Error bars represent 95% confidence
intervals. Lines represent model results.
65
-------
5 -
4 -_
3 :_
2 -
1 -
0
CP: y(x) = 0.30(x)+1.88 r = 0.97
DZ: y(x) = 0.30(
-5.0 -2.5 0.0 2.5 5.0 7.5 10.0
E1/2 (V vs. SCE)/0.059 V
12.5
15.0
Figure 27 Structure-activity relationship relating the reaction of monochloramine,
dichloramine, and hypochlorous acid with CP, DZ, and MA. Piela and Wrona (49)
found half-wave potentials for monochloramine (-0.2 V), dichloramine (0.4), and
hypochlorous acid (0.8 V). These values were then converted using the Nernst
equation to the intensive property of the log ratio of the activities of the reduced
products to the oxidized forms of the oxidant.
66
-------
vvEPA
United States
Environmental Protection
Agency
Office of Research
and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
EPA 600/R-08/089
September 2008
Please make all necessary changes on the below label,
detach or copy, and return to the address in the upper
left-hand corner.
If you do not wish to receive these reports CHECK HERE
D; detach, or copy this cover, and return to the address in
the upper left-hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Recycled/Recyclable
Printed with vegetable-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chlorine free
------- |