Towson University Office of Graduate Studies

Home , Bisulfide

TOWSON UNIVERSITY OFFICE OF GRADUATE STUDIES

ANAEROBIC ABIOTIC REDUCTION OF DICHLOROACETAMIDE SAFENERS:

EFFECTS OF MANGANESE OXIDES AND

AGROCHEMICAL CO-FORMULANTS

Allison Ricko

A thesis

Presented to the faculty of

Towson University

in partial fulfillment

of the requirements for the degree

Master of Science

Environmental Science

Towson University Towson, MD 21252

August 2015

ACKNOWLEDGEMENTS

First and foremost, I would like to thank my advisor, Dr. John Sivey, for his support and guidance throughout my graduate career. His input has been instrumental in both the completion of this research and my education. I cannot thank him enough for his willingness to take on a graduate student who didn’t have a chemistry background and for his patience as I learned the fundamentals necessary to successfully complete this research. The input I’ve received throughout this process has been invaluable.

I would like to thank the thesis committee, Dr. Joel Moore, Dr. Tim Brunker, and

Dr. Ryan Casey, for their input on the experimental design and writing. The entire committee provided useful insight into each of their areas of expertise, contributing to the coherency of the final document. Dr. Tim Brunker performed most of the NMR analyses reported here, and I am grateful for his interpretation of the results.

Several undergraduate students provided support and assistance throughout the research process: Mark Burton, Mark Bickley, Dan Victor, Nick Sapienza, and Nick

Race. I would also like to thank Dave Szymanik and Sarah Krause, who performed additional NMR analyses.

The continued encouragement from my family and friends has been instrumental in the success of my academic career. I will always be grateful for their advice, and their willingness to help proofread and to act as a practice audience has been invaluable.

iii ABSTRACT

Anaerobic Abiotic Reduction of Dichloroacetamide Safeners: Effects of Manganese Oxides and Agrochemical Co-Formulants

Allison Ricko

Safeners are added to herbicide formulations to protect crops from toxic effects of active herbicides. Evidence suggests that as dichloroacetamide safeners transform, they can become more biologically active. Dichloroacetamide safeners can transform via reductive dechlorination in anaerobic abiotic systems containing iron oxy-hydroxide minerals. Manganese oxides are important redox-active species that frequently co-occur with iron oxides. This study examines the anaerobic reduction of three dichloroacetamide safeners in the presence of agrochemical co-formulants, in changing pH buffer and ionic strength conditions, and in Fe(II)-amended, abiotic mixed-mineral systems. The safener dichlormid did not transform appreciably over the sampling period (up to 6 hours).

Transformation of the safeners benoxacor and furilazole was not appreciably affected by the presence of co-formulants, pH buffer, or changes in ionic strength. The molar ratio of

Fe(II)-to-Mn(IV) oxide had an appreciable effect on the transformation rate of benoxacor and furilazole.

iv TABLE OF CONTENTS

LIST OF TABLES ...... viii

LIST OF FIGURES ...... x

CHAPTER ONE: Introduction ...... 1

1.1 Safeners in Agrochemical Formulations ...... 2

1.2 Dichloroacetamide Safeners ...... 6

1.3 Environmental Fate of Herbicides and Safeners ...... 8

1.4 Surfactants in Agrochemical Formulations ...... 10

1.5 Environmental Fate of Surfactants...... 12

1.6 Iron and Manganese Oxides in Soils ...... 12

1.7 Influence of Iron and Manganese Oxides on Agrochemicals ...... 16

CHAPTER TWO: Materials and Methods ...... 18

2.1 Reagents ...... 19

2.2 Experimental Methods ...... 19

2.2.1 Effects of Herbicide and Surfactant Co-Formulants ...... 19

2.2.2 Effects of pH Buffer and Ionic Strength ...... 20

2.2.3 Effects of Mixed-Mineral Systems ...... 20

2.3 Analytical Methods ...... 21

2.3.1 Instrumentation ...... 21

2.3.2 Rate Constant Determination ...... 22

2.3.3 Furilazole Product Quantification ...... 24

v CHAPTER THREE: Effects of Agrochemical Co-Formulants, pH Buffer, and Ionic

Strength on the Reductive Dechlorination of Dichloroacetamide Safeners ...... 26

3.1 Effects of Agrochemical Co-Formulants ...... 27

3.1.1 Effects of S-Metolachlor on the Reductive Dechlorination of

Benoxacor ...... 27

3.1.2 Effects of Surfactants on the Reductive Dechlorination of

Benoxacor ...... 29

3.2 Effects of pH Buffer and Ionic Strength ...... 31

CHAPTER FOUR: Effects of Iron Oxide and Manganese Oxide Mineral Systems on the

Reductive Dechlorination of Dichloroacetamide Safeners ...... 34

4.1 Oxidation of Fe(II) by Mn(IV) Oxide ...... 35

4.2 Effects of Mixed-Mineral Systems on Dichloroacetamide Reduction ...... 37

4.2.1 Effect of Mn(IV) Oxide ...... 37

4.2.2 Effect of Addition Order ...... 41

4.3 Broader Impacts ...... 43

APPENDICES ...... 45

APPENDIX A: Reagents ...... 46

APPENDIX B: Preparation and Standardization of Aqueous Fe(II) and Cr(II) ...... 48

B.1 Preparation and Standardization of Aqueous Fe(II) ...... 49

B.2 Reduction and Standardization of Aqueous Cr(II) ...... 49

vi APPENDIX C: Synthesis and Characterization of Monochlorinated Analogue

of Benoxacor ...... 50

C.1 Synthesis of Monochlorinated Analogue of Benoxacor ...... 51

C.2 Characterization of Monochlorinated Analogue of Benoxacor ...... 52

APPENDIX D: GC Method Details ...... 55

D.1 Instrument Parameters...... 56

D.2 Calibration ...... 58

APPENDIX E: Quality Assurance and Quality Control ...... 63

E.1 Extraction Efficiencies ...... 64

E.2 Recoveries Following Centrifugation...... 65

E.3 Adsorption to Fe and Mn Oxides ...... 66

APPENDIX F: Summary of Observed Reduction Rate Constants ...... 68

APPENDIX G: Transformation of Dichloroacetamide Safeners in the Presence

of Free Chlorine and Free Bromine ...... 73

G.1 An HPLC Method for Dichloroacetamide Safeners ...... 74

G.2 Chlorination and Bromination Rates of Dichloroacetamide

Safeners ...... 77

G.3 Phosphate Catalysis of Benoxacor Bromination ...... 78

REFERENCES ...... 79

CURRICULUM VITA ...... 87

vii LIST OF TABLES

TABLE 1.1. Compound Class, Structure, and Crops Associated with Some Herbicide

Safeners ...... 3

TABLE 1.2. Structure and Properties of Dichloroacetamide Safeners and Their

Associated Herbicides ...... 7

TABLE 1.3. Structure and Properties of Some Surfactants Used in Herbicide

Formulations ...... 11

TABLE 1.4. Properties of Common Oxide Minerals ...... 13

TABLE A.1. List of Reagents, Their Purity, and Vendor ...... 47

TABLE D.1. GC Oven Temperature Program for Both µECD and MS Detection ...... 56

TABLE D.2. GC Retention Times for All Analytes with Both µECD and MS

Detection, and Quantitation Ions Used with Selected Ion Monitoring (SIM)

for MS Detection...... 57

2 TABLE D.3. Limits of Detection (LODs) and R Values for All Compounds,

Corresponding to Calibration Curves in Figure D.3 ...... 62

TABLE E.1. Extraction Efficiency for Safeners in the Absence and Presence of

Surfactants...... 64

TABLE E.2. Centrifugation Recovery for Safeners in the Absence and Presence of

Surfactants...... 66

Table F.1. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor in the Presence of the Herbicide S-Metolachlor ...... 69

viii Table F.2. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor in the Presence of the Surfactant SDS ...... 69

Table F.3. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor in the Presence of the Surfactant MyTAB ...... 70

Table F.4. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor in the Presence of the Surfactant Triton X-100® ...... 70

Table F.5. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor and Furilazole with Changing MOPS and NaCl Concentrations ...... 71

Table F.6. Reaction Conditions and Observed Rate Constants for the Reduction of

Benoxacor and Furilazole in Mixed-Mineral Systems ...... 72

TABLE G.1. Wavelengths of Maximum Absorbance (λmax) and HPLC Retention

Times for All Analytes...... 75

TABLE G.2. Observed Pseudo-First-Order Rate Constants (± 95% Confidence

Intervals) for Chlorination and Bromination of Safeners at 20.0 °C ...... 77

ix LIST OF FIGURES

FIGURE 1.1. Observed growth of corn at varying concentrations of an herbicide

(imazameth) with and without a safener (1,8-naphthalic anhydride) ...... 2

FIGURE 1.2. Generalized chloroacetamide herbicide metabolism within a plant cell...... 4

FIGURE 1.3. Generalized structure of the cytochrome P450 monooxygenase heme

group ...... 5

FIGURE 1.4. Reductive dechlorination reactions for a. dichlormid and b. benoxacor

under anaerobic conditions, as determined by Sivey and Roberts ...... 9

FIGURE 1.5. Relative abundance of manganese solids (i.e., manganese oxides) and

dissolved Fe(II) in flooded soils as conditions change from aerobic to

anaerobic ...... 14

FIGURE 1.6. Manganese enrichment, and use of metolachlor and acetochlor in the

continental United States ...... 15

FIGURE 2.1. a. Example time course data for benoxacor reduction, b. natural log plot

of parent safener loss ...... 24

FIGURE 2.2. Hypothesized reductive dechlorination reaction for furilazole ...... 24

FIGURE 2.3. GC/µECD chromatogram from a reaction of furilazole (FZ) with

Fe(II)-amended hematite and birnessite after 5.5 hours ...... 25

FIGURE 3.1. Time course for the reduction of benoxacor (BN) into a monochlorinated

analogue ...... 28

x FIGURE 3.2. Observed reduction rate constant (kobs) for benoxacor (BN) as a function

of the concentration of the herbicide, S-metolachlor (SM) ...... 28

FIGURE 3.3. Observed transformation rate constant (kobs) for reductive dechlorination

of benoxacor (BN) with changing concentration of a. sodium dodecyl sulfate

(SDS), b. myristryltrimethylammonium bromide (MyTAB), and

c. Triton X-100® ...... 30

FIGURE 3.4. Observed reduction rate constant (kobs) for benoxacor (BN) and furilazole

(FZ) with changing concentration of MOPS ...... 32

FIGURE 3.5. Observed reduction rate constant (kobs) for benoxacor (BN) and furilazole

(FZ) with changing concentration of NaCl ...... 33

FIGURE 4.1. Oxidation of Fe(II) by Mn(IV) oxide over an approximately 3.5-hour

period ...... 35

FIGURE 4.2. Fraction of Fe(II) recovered (i.e., Fe(II) dissolved in the aqueous phase),

calculated as [Fe(II)]aq/[Fe(II)]i after 3 hours ...... 36

FIGURE 4.3. Observed reduction rate constant (kobs) for a. benoxacor (BN) and

b. furilazole (FZ) with changing molar ratio of Fe(II) to Mn(IV) oxide ...... 39

FIGURE 4.4. Comparison of the reduction of benoxacor (BN) in the absence of and

after the addition of birnessite...... 40

FIGURE 4.5. Comparison of benoxacor (BN) reduction with safener and Fe(II)

addition as time zero ...... 42

xi FIGURE 4.6. Comparison of benoxacor (BN) reduction with safener and Mn(IV) oxide

addition as time zero ...... 42

1 FIGURE C.1. H NMR spectrum for the monochlorinated analogue of benoxacor ...... 53

FIGURE C.2. GC/MS chromatogram for the monochlorinated benoxacor synthesis

product ...... 54

FIGURE D.1. GC/µECD chromatogram of a calibration standard ...... 59

FIGURE D.2. GC/MS chromatogram of a calibration standard ...... 60

FIGURE D.3. Calibration curves for all analytes with a. µECD and b. MS detection ...... 61

FIGURE E.1. Adsorption of safeners benoxacor (BN), dichlormid (DL), and

furilazole (FZ) to Fe(III) oxide and Mn(IV) oxide ...... 67

FIGURE G.1. HPLC chromatogram of a calibration standard ...... 75

FIGURE G.2. Calibration curves and limits of detection (LODs) for: a. benoxacor (BN),

b. dichlormid (DL), and c. furilazole (FZ) using HPLC ...... 76

FIGURE G.3. Observed pseudo-first-order rate constants for benoxacor bromination as

a function of the concentration of phosphate buffer at varying pH ...... 78

xii 1

CHAPTER ONE

Introduction

2 1.1 Safeners in Agrochemical Formulations

Safeners are added to herbicide formulations in order to improve selectivity and to protect the desired crop plant from the adverse effects of the active herbicide

(Figure 1.1).1-5 Safeners are referred to in the literature under a variety of additional names, including antidotes, antagonists, and protectants; however, safener is the most common.1 Commercial use of safeners began in 1971, with the development of

1,8-naphthalic anhydride, and these agrochemicals have since grown to include a wide variety of compounds (Table 1.1).1-3,5 Functional groups on each compound are used to define the ‘class’ of the safener (e.g., dichloroacetamides). Each safener class is typically paired with a specific herbicide or group of herbicides (often of similar structure to the safener) to optimize both weed control and crop protection.1-6

a. b. c.

Figure 1.1. Observed growth of corn at varying concentrations of an herbicide (imazameth) with and without a safener (1,8-naphthalic anhydride). Growth with no safener is on the left and with 0.5% safener on the right for a. 0 μM herbicide, b. 2 μM herbicide, and c. 8 μM herbicide. Adapted from Davies.3

Safeners are applied in combination with herbicides as either a seed treatment or spray mixture.1-5 Application as a spray mixture can occur before or after the crop plant has begun growing (pre- or post-emergence, respectively).2-5 Safeners have the potential

3 to work through a variety of mechanisms within the crop plant. Safeners function by enhancing the metabolic pathways that convert the active herbicide into a chemical form that is less biologically-active or completely inactive within the plant.1-5

Table 1.1. Compound Class, Structure, and Crops Associated with Some Herbicide Safeners.2,5 Compound Class Safener Crops

Dichloroacetamides Benoxacor Corn

Dichlormid Corn

Furilazole Cereal crops

AD-67 Corn

Oxime ethers Cyometrinil Sorghum

Oxabetrinil Sorghum

1,8-naphthalic Other Safener Classes Corn anhydride

Fenclorim Rice

Flurazole Sorghum

Fenchlorazole-ethyl Cereal crops

Mefenpyr-diethyl Cereal crops

4 Herbicides are typically metabolized by three pathways in crops (Figure 1.2).2,7

Initially, during Phase I metabolism, a structural group on the herbicide is oxidized.2-4,7

Phase II metabolism is a conjugation step and can begin with the parent herbicide or the products from Phase I metabolism.2-4,7 Phase II conjugation commonly involves reduced glutathione (GSH), catalyzed by glutathione S-transferase (GST) enzymes.2-4,7 Finally, the herbicide metabolites are transported into vacuoles within the plant cell for elimination or sequestration.2,7 Safeners can enhance Phase I or II metabolism and thereby inhibit herbicidal activity in crop plants.1-5

O Cl = – R–C–CH2

1 2

O O O GS = – = = R–C–C–OH R–C–CH2

3 3

O O O GS = – = = R–C–C–OH R–C–CH2

Vacuole

Plant Cell

Figure 1.2. Generalized chloroacetamide herbicide metabolism within a plant cell. 1. Oxidation with cytochrome P450 monooxygenase; 2. conjugation with reduced glutathione, catalyzed by glutathione-S-transferase; 3. transport of metabolites into cell vacuole for sequestration or elimination. Adapted from Cobb and Reade.2

5 Phase I oxidative metabolism in plants includes many possible pathways and is facilitated by a group of enzymes collectively referred to as cytochrome P450 monooxygenases.2-5,7 Cytochrome P450 enzymes consist of a heme group complexed by a cysteine thiolate ligand.2,7 The heme group active site consists of ferric or ferrous iron surrounded by a porphyrin ring (Figure 1.3).2 Safeners likely stimulate the production of cytochrome P450 enzymes in the plant, thereby increasing the extent of oxidation reactions involving herbicides.1-4 This effect has been observed with the safeners

1,8-naphthalic anhydride, dichlormid, and benoxacor, among others.4,5 In some cases, presence of a safener caused an increase in monooxygenase-mediated metabolism by more than a factor of ten.3,4 Safeners can also increase the rate and extent of herbicide conjugation, particularly by GSH/GST; specifically, dichloroacetamide safeners promote this pathway during metabolism of thiocarbamate and chloroacetamide herbicides.4,5

Figure 1.3. Generalized structure of the cytochrome P450 monooxygenase heme group. The protein portion of the biomolecule bonds at the central iron by a cysteine thiolate ligand.

6 1.2 Dichloroacetamide Safeners

Dichloroacetamides are one of the most commonly used classes of safeners.5 The three major compounds in this class, benoxacor, dichlormid, and furilazole, are commonly paired with chloroacetamide and thiocarbamate herbicides (Table 1.2).2-5,8

Benoxacor is paired with the chloroacetamide herbicide S-metolachlor in commercial formulations (e.g., Dual II Magnum®, Syngenta) and is applied to corn as a pre-emergent spray mixture.5,8 The safener dichlormid is paired with the chloroacetamide herbicide, acetochlor (e.g., Surpass®, Dow AgroSciences), but can also be combined with thiocarbamate herbicides.2,5,8 Dichlormid is applied to corn as a pre-plant or pre-emergent spray.2-5 Benoxacor induces GSH activity by promoting GST production and thereby increases herbicide metabolism in crop plants.3,4,8,9 Similar to benoxacor, dichlormid acts by inducing synthesis of GST in order to increase herbicide metabolism through GSH conjugation.3,8 Furilazole is paired with the chloroacetamide herbicide acetochlor

(e.g., Degree®, Monsanto) and formulations are applied to corn as a pre-emergent spray mixture.2-5,8 Furilazole increases herbicide metabolism by inducing P450 oxidation, but the exact mechanisms causing this effect are unclear.4,8

7 Table 1.2. Structure and Properties of Dichloroacetamide Safeners and Their Associated Herbicides.

a b c Dichloroacetamide Safeners Water Solubility (mM) Boiling Point (⁰C) LogKow

Benoxacor 0.395 408.9 2.70

Furilazole 0.914 376.8 2.12

Dichlormid 5.13 253.8 1.84

a b c Chloroacetamide Herbicides Water Solubility (mM) Boiling Point (⁰C) LogKow

Metolachlor 0.179 406.8 2.90

Acetochlor 0.176 391.5 3.03

a b c Thiocarbamate Herbicides Water Solubility (mM) Boiling Point (⁰C) LogKow

Vernolate 0.109 272.8 3.84 a Water solubility values are estimates at 25 ⁰C obtained from US EPA’s EPISuite software as reported at www.chemspider.com b Boiling point values are reported at 760 mm Hg and obtained from the ACD Labs Percepta Platform, PhysChem Module as reported at www.chemspider.com c Log KOW values were obtained from US EPA’s expkow database as reported at www.chemspider.com

8 1.3 Environmental Fate of Herbicides and Safeners

The environmental fate and transformation of many commercial herbicides have been previously studied.10-16 Along with metabolism in plants, herbicides can undergo transformations in soil and aqueous systems.11,16 Chloroacetamide herbicides

(e.g., metolachlor and acetochlor) have several known transformation products, including sulfonic acid-, hydroxyl-substituted-, and dechlorinated-analogues.11,13,14,16

Chloroacetamide herbicides and their metabolites have been detected in surface and groundwater,12-15 suggesting that the associated safeners may also be found, and transform, in these systems. A suite of agrochemicals (including chloroacetamides) have been detected in groundwaters across the United States in both agricultural and urban land use areas.15 Higher detection frequencies were associated with agricultural areas, but urban areas still contained agrochemical contaminants at quantifiable levels.15

A series of studies also examined the occurrence of chloroacetamide herbicides and their analogues in both pre- and post-treatment drinking waters.12,14 These studies detected chloroacetamides and their degradates in pre-treatment surface and groundwaters used as drinking water sources and found that standard water treatment practices typically were not effective at removing these compounds.12,14

Chloroacetamide herbicides and selected metabolites have been detected in the

Chesapeake Bay and were found to persist with water depth up to 25 m.13 Surface waters contained between 15 and 50 ng/L herbicides and deep water samples (20 – 25 m) contained between 5 and 20 ng/L herbicides.13 We anticipate, from the presence of the herbicides, that associated safeners are likely to be present in surface and groundwaters proximate to agricultural fields to which these chemicals are applied. Currently, no

9 studies have examined the environmental occurrence of safeners or their transformation products, and their environmental fate is not well studied.

Safeners are classified as “inert” constituents in agrochemical formulations.17

Studies of safeners suggest they can transform in the environment into products that may be more biologically active than the parent compounds.10,17 The photo-transformation of dichlormid under ultraviolet light has been studied in both aqueous and organic solvents.10 Dichlormid was shown to photolyze and transform via hydrolysis, dealkylation, and dechlorination.10 A study by Sivey and Roberts17 explored the reduction of dichloroacetamide safeners in Fe(II)-amended hematite and goethite suspensions and found that these so-called “inert” compounds are reduced to products closely resembling active herbicide compounds (Figure 1.4). In the case of dichlormid, one transformation product was identified as the formerly-used herbicide N,N-diallyl-2-chloroacetamide

(CDAA).17

a. I

Dichlormid CDAA III

Benoxacor IV V Figure 1.4. Reductive dechlorination reactions for a. dichlormid and b. benoxacor under anaerobic conditions, as determined by Sivey and Roberts.17

10 Examining chemical mixtures is important when assessing the fate of environmental contaminants, as mixtures may impart synergistic and/or antagonistic effects.18 Studies have examined the potential toxicity of benoxacor in combination with metolachlor as an application-ready formulation and as a technical grade mixture, but none have examined the transformation of the safener in these mixtures.19

1.4 Surfactants in Agrochemical Formulations

Surfactants are often added to agrochemical formulations to improve uptake of herbicides into weeds.20-25 Surfactants also help agrochemical formulations stick to plant surfaces during application.20-22 A variety of surfactants are in use (Table 1.3), falling into four general classes: anionic, cationic, nonionic, and amphoteric.20,21 Regardless of the surfactant class, the compounds contain both a non-polar (i.e., hydrophobic) region and a polar or ionic (i.e., hydrophilic) region.20,21

A combination of factors is important in controlling the effects of surfactants on herbicidal activity.20-23 Surfactants increase the surface area on the plant covered by the agrochemical mixture, facilitating increased uptake due to an increase in uptake locations.20,22 It is also possible that surfactants keep the formulation moist longer after contact with plants, providing more time for uptake to occur.20,22 Lastly, surfactants can facilitate transport through leaves and into plants cells, promoting transport of active agents to locations where they can exert a biological effect.20,22,26

The hydrophile/lipophile balance (HLB) is likely an important surfactant property that controls facilitated transport of herbicides.26 HLB is determined by separate methods for ionic and non-ionic surfactants.26 Non-ionic surfactant HLB is calculated by the

11 weight percent of the molecule that is hydrophilic, divided by 5; ionic surfactant HLB can be estimated using a visual inspection of the compound in solution, or it can be calculated using a method developed by Davies.26,27 The Davies method uses a numerical value, assigned to different groups on the molecule, and sums them to obtain the overall

HLB.27 Evidence suggests that a high HLB surfactant promotes transport of water-soluble herbicides, and low HLB surfactant promotes transport of herbicides with low water solubility.26 Optimal surfactant effect for post-emergence spray applications typically occurs at a surfactant concentration of approximately 0.5% (w/w) in the mixture.23

Table 1.3. Structure and Properties of Some Surfactants Used in Herbicide Formulations.a

Surfactant Class CMC (mM) b HLB c

Sodium dodecyl sulfate (SDS) Anionic 7-10 40

Myristryltrimethylammonium bromide (MyTAB) Cationic 4-5 n.a. d

Triton® X-100 Nonionic 0.2-0.9 13.5 a All data obtained from Sigma-Aldrich, as reported on www.sigmaaldrich.com b Critical micelle concentration (CMC) values are reported for 20-25 ⁰C c Hydrophile/lipophile balance d Data not available (unknown)

12 1.5 Environmental Fate of Surfactants

Surfactants can affect herbicides’ physical properties (e.g., solubility and volatility), transformation rates, and the rate and/or extent of adsorption onto soils.21,24,25

Surfactants can either enhance or attenuate rates of herbicide transformation, depending on the specific surfactant and herbicide combination.21,24,25 The effect of surfactants on adsorption also varies, depending on how the surfactant interacts with both the soil and the herbicide.21 The surfactant sodium dodecyl sulfate (SDS), for example, did not appreciably affect adsorption of the herbicide atrazine in a Spanish soil.21 A similar surfactant, sodium dodecylbenzenesulfonate, did not affect adsorption of the herbicide acetochlor in a Chinese soil.21 The effect of surfactants on the transformation and adsorption of safeners has not been previously explored. Surfactants have the potential to affect safener transformation in multiple ways, including: (1) their presence in solution could change transformation rates due to intermolecular interactions between safener and surfactant; and (2) surfactants could affect the rate and/or extent of adsorption of safeners, thereby changing the rate of surface-mediated transformations.

1.6 Iron and Manganese Oxides in Soils

Iron oxy-hydroxide and manganese oxide minerals are abundant secondary soil minerals that occur across a wide range of soil types and underlying bedrock compositions.28,29 Iron oxy-hydroxides form as weathering products of iron-rich minerals.29 Manganese oxides can form through abiotic weathering when manganese substitutes for iron in the atomic structure of iron-rich minerals; however, manganese oxides are most commonly formed in natural soils through biomineralization by bacteria

13 and other microbial processes.29,30 Recent studies have also shown that vegetation can produce manganese oxides from Mn(II) dissolved in soil pore water.31,32 This process, along with the low solubility of manganese oxides, has led to enrichment of iron and manganese oxides in soils, particularly in areas where iron ore has been smelted.29,33,34

Iron oxy-hydroxides and manganese oxides typically occur in soils as either clay-sized particles or as coatings on the surface of larger particles.28,29 Iron in soils mainly occurs as hematite (an Fe(III) oxide, Table 1.4). Manganese typically occurs as birnessite (an Mn(IV) oxide, Table 1.4).28,29

Table 1.4. Properties of Common Oxide Minerals.a Oxidation State of BET Surface Area c Mineral Chemical Formula ZPC b Metal (m2g-1)

Hematite Fe2O3 +III 8 40

Birnessite MnO2 +IV 2 32 a All data obtained from Maurice28 b Zero point of charge c Brunaver, Emmet, and Teller surface area

Due to multiple stable oxidation states, iron and manganese influence the reduction-oxidation (redox) characteristics of soils.28,29,35 Iron can shift between Fe(III) oxy-hydroxides and dissolved Fe(II), and manganese can shift between Mn(IV) oxides and dissolved Mn(II).29,35 In natural systems, changing redox conditions (e.g., soils becoming flooded) can result in the co-occurrence of Fe(II) with manganese oxides

(Figure 1.5).29,35

Agricultural soils are one of the primary environments where manganese oxides, iron oxy-hydroxides, and dissolved Fe(II) can co-occur due to changing redox conditions as the soils shift from tilled (and well aerated) to flooded (and anaerobic).29 Recent

14 evidence has shown that manganese enrichment, where the concentration of manganese is increased in soils relative to the bedrock material, is influenced by anthropogenic inputs

(e.g., vehicle and industrial emissions, and metal ore processing) as well as by vegetation.31,33,34 Trees can uptake dissolved Mn(II) from soil pore water. Mn(II) is subsequently stored in plant biomass and eventually redeposited to soils as manganese oxide upon decomposition of leaf litter.30 Manganese enrichment in the continental

United States is highest in Iowa and the surrounding region (Figure 1.6a). This manganese enrichment directly overlaps with the “corn belt,” including agricultural regions where the herbicides acetochlor and metolachlor are applied in the highest quantities (Figure 1.6b-c).31,33,36

Figure 1.5. Relative abundance of manganese solids (i.e., manganese oxides) and dissolved Fe(II) in flooded soils as conditions change from aerobic to anaerobic. Time on the x-axis is in days after flooding. Adapted from Langmuir.35

- 2 m Mn,w (mg cm )

No estimated use 0 – 0.019 0.020 – 0.082 0.083 – 0.222 0.223 – 0.963 c.

No estimated use 0 – 0.027 0.028 – 0.108 0.109 – 0.228 0.229 – 0.361 Figure 1.6. Manganese enrichment, and use of metolachlor and acetochlor in the continental United States. a. Manganese enrichment of soils, relative to bedrock, adapted from Herndon and Brantley;33 b. Average annual application of metolachlor in pounds per square mile for the years 1992 – 1995, adapted from Thelin and Gianessi;36 c. Average annual application of acetochlor in pounds per square mile for the years 1992 – 1995, adapted from Thelin and Gianessi.36

16 1.7 Influence of Iron and Manganese Oxides on Agrochemicals

The presence of iron oxy-hydroxides or manganese oxides alone is typically insufficient to appreciably affect redox reactions of agrochemicals in soils.37-40 The oxide minerals provide the surface chemistry, but a reducing agent is necessary for reactions to proceed.37-40 The changing redox state of agricultural soils can generate dissolved Fe(II) near iron oxy-hydroxides and manganese oxides.29,35 Dissolved Fe(II) can bind to hydroxyl groups on some oxide mineral surfaces.37,39,41 Fe(II), bound to surface sites on the oxide minerals, can serve as a reductant in reactions with environmental contaminants,37,42-45 including agrochemicals.17,38,46-48

Surface-bound Fe(II) on Fe(III) oxide is a stronger reducing agent than free Fe(II) in solution.37,40,41,45 A study exploring the role of hematite crystal structure in electron transfer showed that as redox processes transform iron between the +II and +III oxidation states, electron transfer occurs through the hematite crystal.49 As Fe(II) adsorbed onto the basal (top and bottom) surfaces is oxidized to form Fe(III), Fe(III) is reduced to release

Fe(II) on the sides of the crystal.49 This results in a flow of electrons from basal surfaces to side surfaces while no net reduction or oxidation occurs.49

Fe(II), complexed by dissolved ligands or iron oxides, is capable of reducing some agrochemicals.17,37,39,43-48,50-52 Studies have also examined the ability of these systems to dehalogenate organic compounds.17,44,50,52 Sivey and Roberts studied the effect of Fe(II)-amended iron oxy-hydroxide systems on the reductive dechlorination of dichloroacetamide safeners, but no work so far has examined these agrochemicals in systems containing manganese oxides.17

17 Despite the lack of direct literature on the influence of manganese oxides in combination with iron oxy-hydroxides on agrochemicals, some of the likely results can be postulated. In systems with Fe(II), birnessite can oxidize Fe(II) to Fe(III), forming

Fe(III) hydroxides and releasing dissolved Mn(II) into solution.53-55 Under anaerobic conditions, the formation of dissolved Fe(II) along with dissolved Mn(II) and precipitation of Fe(III) hydroxides has been observed.54 The reduction of Mn(III/IV) oxides by phenols to form dissolved Mn(II) has been observed.56 Nowack and Stone57 also showed that Mn(III) minerals can be reduced to Mn(II) during the oxidation of nitrilotrismethylenephosphonic acid (NTMP). Neither work included dissolved Fe(II) or iron oxy-hydroxide minerals, so the influence of these components on the overall chemistry is unknown.56,57

The different redox characteristics of Fe(III) oxides and Mn(IV) oxides demonstrate the necessity for examining the effect of both minerals on agrochemical reactions.28,29,35 The overlap of manganese-rich soils with the areas where dichloroacetamide safeners are routinely applied points to the importance of studying the transformation of these agrochemicals in the presence of both solid phases.29,31,33,35,36

This study examines the reduction of the dichloroacetamide safeners benoxacor, dichlormid, and furilazole in order to: (1) determine the influence of herbicide and surfactant co-formulants on transformation rates; (2) identify effects of laboratory solution conditions (e.g., concentration of pH buffer and ionic strength) on transformation rates; and (3) determine the influence of Mn(IV) oxide on the transformation rates. This study is, to our knowledge, the first to quantify the abiotic transformation of the safener furilazole.

CHAPTER TWO

Materials and Methods

19 2.1 Reagents

A comprehensive list of reagents, their purities, and vendor information is given in Appendix A. All aqueous solutions were prepared with deionized water purified with a Nanopure Analytical UV system (18 MΩ•cm, Thermo Scientific). Stock solutions of

CrCl2 and FeCl2 were standardized via UV-vis spectrophotometry before use (see

Appendix B).58,59 The monochlorinated analogue of the safener benoxacor was synthesized for use in calibration standards; the method and characterization results for this synthesis are given in Appendix C.

2.2 Experimental Methods

All reactions examined herein were performed in 40-mL amber glass vials with

Teflon-lined caps held at room temperature in an anaerobic chamber (3% H2, 97% N2).

The total volume of all reactors was 35 mL. Initial safener concentration in all reactions was 20 µM, added as a methanolic spike at t = 0. Reactors were mixed continuously on a vial rotator, except when aliquots were removed for analysis.

2.2.1 Effects of Herbicide and Surfactant Co-Formulants

Reactions with safeners and agrochemical co-formulants (surfactants and herbicides) were performed in aqueous solutions of sulfuric acid (5.0 mM). Chromium was used as a reductant in these reactions (500 µM, as CrCl2), so that no solid phase was needed for the reaction to proceed. For herbicide reactions, benoxacor was monitored in the presence of the herbicide S-metolachlor. Concentrations of S-metolachlor ranged from 0 to 40 µM, added as a methanolic spike. For surfactant reactions, surfactant

20 concentration ranged from 0 to 10 µM, added as an aqueous spike of sodium dodecyl sulfate (SDS), myristryltrimethylammonium bromide (MyTAB), or Triton® X-100.

Aliquots (0.5 mL) of reactors were immediately added to 0.5 mL of 10 mM NaOH, to quench the reaction, and extracted into 1.0 mL of toluene prior to analysis by gas chromatography (GC).

2.2.2 Effects of pH Buffer and Ionic Strength

Reactions to determine the effects of pH buffer and ionic strength were performed in an aqueous slurry of 10 g/L hematite (as Fe(III) oxide, Fe2O3) at pH 7.0. Fe(II) was added as a reductant (as a concentrated aqueous spike of FeCl2, 5.0 mM). For reactions used to determine the influence of pH buffer, the concentration of 3-morpholinopropane-

1-sulfonic acid (MOPS) ranged from 10.0 to 50.0 mM, with the concentration of NaCl fixed at 50.0 mM. For reactions performed as a function of ionic strength, the concentration of NaCl ranged from 10.0 to 200.0 mM, with the concentration of MOPS fixed at 30.0 mM. Samples were obtained as 1.0 mL aliquots, immediately centrifuged

(13,000 rpm for ten minutes), and extracted into 2.0 mL of toluene for analysis via GC.

2.2.3 Effects of Mixed-Mineral Systems

Reactions with safeners in mineral slurries were performed in an aqueous slurry containing MOPS buffer (30.0 mM, pH 7.0), with 50.0 mM NaCl to maintain constant ionic strength. Solids were fixed at 10 g/L total solids loading and added as a concentrated slurry to achieve a mixture of 9 g/L hematite and 1 g/L birnessite. Fe(II) served as a reductant (2.0 to 22 mM) and was added to reactors as a concentrated aqueous

21 spike of FeCl2. Samples were obtained as 1.0 mL aliquots and immediately centrifuged at

13,000 rpm for ten minutes and extracted into 2.0 mL of toluene prior to analysis by GC.

Experiments to examine the rate of Fe(II) oxidation by birnessite had a total solids loading of 1 g/L birnessite and an initial Fe(II) concentration of 6.0 mM, added as FeCl2.

Samples were extracted as 1.0 mL aliquots into 9.0 mL of 1% HNO3 in order to remove any Fe(II) adsorbed to the mineral phase. The sample solution was subsequently filtered with a 0.2 µm syringe filter (nylon membrane, Whatman) prior to analysis with UV-vis spectrophotometry.

Experiments examining the competitive adsorption of Fe(II) and Mn(II) onto hematite included a total solids loading of 10 g/L as hematite. Fe(II) was added to reactors as FeCl2 at either 1.5 or 3.0 mM. Mn(II) was added to reactors as MnCl2 at either

1.5 or 3.0 mM. Reactors were prepared in which the Fe(II)/Mn(II) molar ratio ranged from 2:1 to 1:2. Samples (approximately 2 mL) were filtered with a 0.2 µM nylon syringe filter prior to subsequent analyses of Fe(II) with UV-vis spectrophotometry.

2.3 Analytical Methods

2.3.1 Instrumentation

Reactions monitoring the reductive dechlorination of safeners were analyzed via

GC with either mass spectrometry (MS) or micro-electron capture (µECD) detection.

Both detectors were used with an Agilent DB-5MS+DG column (30 m, 250 µm i.d.,

0.25 µm) with He as the carrier gas. A full discussion of GC/MS and GC/µECD parameters, including temperature programs and retention times, is included in

Appendix D. Chlorobenzonitrile (CBN) was used as an internal standard (10.2 µM in

22 toluene) for all GC analyses. Analyte loss and extraction efficiencies were quantified for each step of the method (Appendix E). Analysis of Fe(II) oxidation and adsorption, in the absence of safeners, was performed using an Agilent Cary 60 UV-visible spectrophotometer after combining with ferrozine indicator, following the method reported by Stookey.59

2.3.2 Rate Constant Determination

Rate constants for all reactions were determined by the Langmuir-Hinshelwood-

Hougen-Watson (LHHW) model for surface-mediated reaction kinetics. The LHHW model assumes that surface-mediated reactions proceed in three steps (eqs 1-3, using

Fe(III) oxide as the solid):

III II III II >Fe −O−Fe + iox (aq) >Fe −O−Fe ‐‐‐iox [1]

III II III III >Fe −O−Fe ‐‐‐iox → >Fe −O−Fe ‐‐‐ired [2]

III III III III >Fe −O−Fe ‐‐‐ired >Fe −O−Fe + ired (aq) [3]

III II where >Fe –O–Fe represents Fe(II) adsorbed on the surface of Fe(III) oxide and iox and ired represent the oxidized and reduced form of a dichloroacetamide, respectively. Under the assumption that the surface reaction (eq 2) is rate-limiting, the reaction rate is calculated by eq 4:

d i k FeII i ox ox ads ox [4] dt 1 ox[iox] red ired

-1 where k (s ) is the rate constant for the surface reaction (eq 2); [iox] and [ired] (mol/L) are

II the concentrations of oxidized and reduced analyte, respectively; [Fe ]ads (mol/L) is the concentration of adsorbed Fe(II); and Kox and Kred (L/mol) are the adsorption equilibrium constants for iox and ired, respectively.

23 Following the derivation by Sivey and Roberts17 (assuming

II II Kox[iox]+Kred[ired] << 1, [Fe ]ads >> [iox], and [Fe ]ads remains constant throughout the reaction), eq 4 simplifies to a pseudo-first-order rate equation (eq 5):

d i ox k i [5] dt obs ox

-1 II where kobs (s ) is a pseudo-first-order rate constant equal to kKox[Fe ]ads.

Previous work demonstrated that rates of reductive dechlorination of benoxacor in

Fe(II)-amended goethite slurries were first-order in the initial concentration of benoxacor, consistent with our assumption that surface reaction is rate-determining.17

Integration and natural log transformation of eq 5 yields a linear equation:

ln iox t kobst ln iox [6] where [iox]t is the concentration of the parent safener at any given time and [iox]0 is the initial concentration of the parent safener. This linear model was used along with time course data to obtain kobs values. Over the course of a reaction, the loss of the parent safener and the formation of the monochlorinated product were monitored (Figure 2.1a).

A natural log transformation of parent compound concentrations over time resulted in a graph with slope equal to –kobs (Figure 2.1b). For most reactors, kobs values were determined by analysis of a single reactor. In order to confirm reactor-to-reactor reproducibility, a set of four reactors was analyzed and showed low variability (kobs relative standard deviation of 4%). All statistical analyses discussed herein consisted of an F-test, followed by a two-tailed t-test.

a. b. 1 0

0.8 -0.1

0.6 -0.2

0.4 -0.3

0.2 -0.4 Normalized Concentration Normalized

0 concentration) ln(normalized -0.5 0 5 10 15 20 0 5 10 15 20 Time (min) Time (min) Figure 2.1. a. Example time course data for benoxacor reduction, b. natural log plot of parent safener loss. Time course data shows parent safener loss (blue) and product (monochlorinated analogue) formation (white). The natural log transformation of the parent safener data yields a slope with absolute value equal to the observed rate constant. Concentrations were normalized to the mass balance, assuming that benoxacor and its monochlorinated analogue were the only species present.

2.3.3 Furilazole Product Quantification

In order to account for the temporal variability in µECD response values, analyte concentrations were normalized to the mass balance (total concentration of parent dichloroacetamide and monochlorinated product). For furilazole, reference material was not available for the monochlorinated analogue of this safener. We hypothesize that the transformation of furilazole in the studied systems follows the reductive dechlorination observed by Sivey and Roberts for benoxacor (Figure 2.2).

Furilazole VI VII Figure 2.2. Hypothesized reductive dechlorination reaction for furilazole.

25 During GC analysis of furilazole reaction samples, a peak was observed in the chromatogram that was consistent with a monochlorinated analogue of furilazole

(Figure 2.3). The peak consistently appeared at the same retention time (10.69 min) prior to furilazole (11.02 min). The peak area for the monochlorinated analogue consistently increased as that of furilazole decreased throughout a time course. In order to approximate a mass balance for reactions containing furilazole, the peak area of the assumed monochlorinated analogue was normalized to the internal standard using the following equation:

[ ] [7]

where AreaMF,corrected (µM) is the corrected peak area of monochlorinated analogue of furilazole; AreaMF and AreaCBN (counts) are the peak areas of the monochlorinated analogue of furilazole and the internal standard, respectively; and [CBN] (µM) is the concentration of the internal standard.

2.6 e+7 C

2.4 e+7

2.2 e+7

2.0 e+7

1.8 e+7

1.6 e+7 1.4 e+7

Abundance 1.2 e+7

1 e+7

8 e+6

6 e+6 A

4 e+6 B 2 e+6 5.00 10.00 15.00 Time (min) Figure 2.3. GC/µECD chromatogram from a reaction of furilazole (FZ) with Fe(II)-amended hematite and birnessite after 5.5 hours. Peak A is the internal standard (CBN, 10.2 µM), peak B is assumed to be the monochlorinated analogue of furilazole, and peak C is furilazole. Reactor conditions: [FZ]i = 20 µM, [Fe(II)]i = 18 µM, Fe(III) oxide loading = 9 g/L, Mn(IV) oxide loading = 1 g/L.

CHAPTER THREE

Effects of Agrochemical Co-Formulants, pH Buffer, and Ionic Strength

on the Reductive Dechlorination of Dichloroacetamide Safeners

27 3.1 Effects of Agrochemical Co-Formulants

The effects of agrochemical co-formulants on the abiotic reduction of dichloroacetamide safeners have not been previously studied. In order to determine the effects of co-formulants on the reduction of dichloroacetamide safeners, benoxacor was monitored over time in systems amended with either an herbicide or a surfactant.

Herbicide-amended systems contained varying concentrations of the chloroacetamide herbicide, S-metolachlor. Surfactant-amended systems contained varying concentrations of either the anionic surfactant SDS, the cationic surfactant MyTAB, or the non-ionic surfactant Triton X-100®.

3.1.1 Effects of S-Metolachlor on the Reductive Dechlorination of Benoxacor

The reductive dechlorination of benoxacor in systems containing Cr(II) follows the same pseudo-first-order reaction as expected for mixed-mineral systems (Figure 3.1).

In the presence of S-metolachlor, the transformation rate constant of benoxacor did not change, except at an herbicide concentration of 40.0 µM (Figure 3.2). Even at this concentration of herbicide, the observed rate constant decreased by only 12%. These findings suggest that at environmentally relevant concentrations in Fe(II)-amended mineral systems, the presence of an herbicide is unlikely to influence the reductive dechlorination of dichloroacetamide safeners. For a tabulated summary of the observed reduction rate constants and reaction conditions, see Appendix F.

0.8

0.6

0.4

Concentration Normalized 0.2

0 0 5 10 15 20 Time (min)

Figure 3.1. Time course for the reduction of benoxacor (BN) into a

monochlorinated analogue. Reactor conditions: [BN]i = 20 µM, [Cr(II)]i = 500 µM, [H2SO4] = 5.0 mM. Error bars represent 95% confidence intervals. Concentrations were normalized to the mass balance, assuming that benoxacor and its monochlorinated analogue were the only species present.

5.0E-04 * 4.0E-04

) 3.0E-04 1 - (s obs k 2.0E-04

1.0E-04

0.0E+00 0 10 20 30 40 Concentration of S-Metolachlor (µM)

Figure 3.2. Observed reduction rate constant (kobs) for benoxacor (BN) as a function of the concentration of the herbicide, S-metolachlor (SM). Reaction conditions: [BN]i = 20 µM, [Cr(II)]i = 500 µM, [H2SO4] = 5.0 mM, [SM]i = 0 – 40 µM. Error bars represent 95% confidence intervals. The rate constant at [SM] = 40.0 µM (indicated with asterisk) is statistically different (at the 95% confidence level) from those at lower concentrations of S-metolachlor.

29 3.1.2 Effects of Surfactants on the Reductive Dechlorination of Benoxacor

In the presence of SDS, MyTAB, and Triton X-100®, the observed reductive dechlorination rate constant of benoxacor did not change appreciably as the concentration of surfactants changed (2-10 µM, Figure 3.3). In the presence of SDS and MyTAB, although the observed rate constant did not change with changing surfactant concentration, the observed rate constant was significantly lower (at the 95% confidence interval) than in the absence of surfactant. In the presence of SDS, the observed rate constant changed from 0.026 ± 0.002 s-1 to an average of 0.0211 ± 0.0005 s-1 (an 18% reduction). In the presence of MyTAB, the average observed rate constant was

0.0234 ± 0.0008 s-1 (a 10% reduction, compared to the absence of surfactant). While these small differences are not likely to have an impact on the reduction of benoxacor in an environmental setting, it suggests a possible effect specific to ionic surfactants.

Appendix F contains a summary of observed reduction rate constants and solution conditions.

30 a. 5.0E-04

4.0E-04 * ) 1

- 3.0E-04 (s obs

k 2.0E-04

1.0E-04

0.0E+00 0 2 4 6 8 10 Concentration of SDS (µM)

b. 5.0E-04

4.0E-04 * ) 1

- 3.0E-04 (s obs

k 2.0E-04

1.0E-04

0.0E+00 0 2 4 6 8 10 Concentration of MyTAB (µM)

c. 5.0E-04

4.0E-04 ) 1

- 3.0E-04 (s obs

k 2.0E-04

1.0E-04

0.0E+00 0 2 4 6 8 10 Concentration of Triton X-100® (µM)

Figure 3.3. Observed transformation rate constant (kobs) for reductive dechlorination of benoxacor (BN) with changing concentration of a. sodium dodecyl sulfate (SDS), b. myristryltrimethylammonium bromide (MyTAB), and c. Triton X-100®. Reaction conditions: [BN]i = 20 µM, [Cr(II)]i = 500 µM, [H2SO4] = 5.0 mM. Surfactant concentration ranged from 0 – 10 µM. Error bars represent 95% confidence intervals. For SDS and MyTAB, the rate constant with no surfactant present (indicated with asterisks) is significantly different (at the 95% confidence level) than those in the presence of added surfactant.

31 3.2 Effects of pH Buffer and Ionic Strength

In laboratory settings, experimental systems frequently incorporate pH buffers to control pH. This is particularly important for systems with a mineral phase that has a circum-neutral zero point of charge, as changes in the pH around this value can affect the charge of surface sites on the mineral. A change in mineral surface charge can strongly influence sorption, particularly for charged species in solution, like Fe(II). MOPS is frequently used as a pH buffer in systems containing Fe(II) because of the low affinity of

MOPS for metals. The influence of MOPS on the reduction of dichloroacetamides in hematite systems has not been examined previously. In order to test the influence of ionic strength changes on the reductive dechlorination of dichloroacetamide safeners, NaCl was used as a model salt compound. Reactions were performed with varied concentration of MOPS and NaCl in order to determine their influence on the transformation of benoxacor and furilazole.

As the concentration of MOPS increased, the observed reaction rate constant did not appear to change (Figure 3.4). Buchholz et al.60 observed that increasing concentration of MOPS decreased the observed reaction rate constant for the reduction of

CCl4 in Fe(II)-amended goethite. The authors propose that the MOPS effect in these systems resulted from competitive adsorption of MOPS on the goethite surface with concurrent release of adsorbed Fe(II). In the Fe(II)-amended hematite systems containing dichloroacetamide safeners examined herein, no change in rate constant in the presence of MOPS was observed. The difference in results may be due to interactions observed by

60 Buchholz et al. that are specific to CCl4, goethite, or both.

32 Reactions examining the effect of ionic strength were prepared in the same manner as those for MOPS. In reactors containing varying concentrations of NaCl, the observed reaction rate constant for both benoxacor and furilazole did not change appreciably (Figure 3.5). A complete summary of observed reduction rate constants for benoxacor and furilazole, along with reaction conditions, is provided in Appendix F.

3.0E-05 BN FZ

2.0E-05 ) 1 - (s obs k 1.0E-05

0.0E+00 0.0 10.0 20.0 30.0 40.0 50.0 Concentration of MOPS (mM)

Figure 3.4. Observed reduction rate constant (kobs) for benoxacor (BN) and furilazole (FZ) with changing concentration of MOPS. Reaction conditions: [BN]i = 20 µM, [FZ]i = 20 µM, [Fe(II)]i = 5.0 mM, [NaCl] = 50.0 mM, Fe(III) oxide loading = 10 g/L, pH 7.0. Error bars represent 95% confidence intervals.

3.0E-05 BN FZ

2.0E-05 ) 1 - (s obs k 1.0E-05

0.0E+00 0.0 50.0 100.0 150.0 200.0 Concentration of NaCl (mM)

Figure 3.5. Observed reduction rate constant (kobs) for benoxacor (BN) and furilazole (FZ) with changing concentration of NaCl. Reaction conditions: [BN]i = 20 µM, [FZ]i = 20 µM, [Fe(II)]i = 5.0 mM, [MOPS] = 30.0 mM, Fe(III) oxide loading = 10 g/L, pH 7.0. Error bars represent 95% confidence intervals.

CHAPTER FOUR

Effects of Iron Oxide and Manganese Oxide Mineral Systems on the Reductive Dechlorination of Dichloroacetamide Safeners

35 4.1 Oxidation of Fe(II) by Mn(IV) Oxide

The oxidation of Fe(II) by Mn(IV) oxide is thermodynamically favorable under both environmentally-relevant pH/pE conditions and the anaerobic conditions examined herein.29,35 To quantify the extent of Fe(II) oxidation in systems containing Mn(IV) oxide, reactions were performed in which oxidation of Fe(II) was monitored over approximately 3.5 hours and an observed rate constant was determined. In the presence of 1 g/L Mn(IV) oxide and 6.0 mM Fe(II), approximately 40% of the added Fe(II) was oxidized over approximately 4 hours (Figure 4.1).

1.0

0.8

0.6 Fe(II)

0.4 Fraction of Fraction

0.2

0.0 0 75 150 225 Time (min) Figure 4.1. Oxidation of Fe(II) by Mn(IV) oxide over an approximately 3.5-hour period. Reactor conditions: Mn(IV) oxide loading = 1 g/L, [Fe(II)]I = 6.0 mM, [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0.

36 Because the oxidation of Fe(II) by Mn(IV) results in the reduction of Mn(IV) to

Mn(II), the competitive adsorption between Fe(II) and Mn(II) was examined. Fe(II) adsorption was monitored in hematite systems under conditions where the molar ratio of

Mn(II) to Fe(II) was 0, 0.5, 1, and 2 (Figure 4.2). The presence of Mn(II) at molar concentrations less than or equal to that of Fe(II) did not have an appreciable effect on the extent of Fe(II) sorption, with approximately 40% of the Fe(II) recovered from the aqueous phase. In systems with a Mn(II)-to-Fe(II) molar ratio of 2, only 17% of the

Fe(II) was recovered, indicating that the recovery of Fe(II) decreased compared to systems with Mn(II) at concentrations less than or equal to Fe(II).

1.4

1.2

1.0

0.8

0.6

Fraction Fraction Recovered 0.4

0.2 No solids Hematite 0.0 0.0 0.5 1.0 1.5 2.0 Mn(II)/Fe(II) Molar Ratio Figure 4.2. Fraction of Fe(II) recovered (i.e., Fe(II) dissolved in the aqueous phase), calculated as [Fe(II)]aq/[Fe(II)]i after 3 hours. Reactor conditions: [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0. In hematite systems, the Fe(III) oxide loading was 10 g/L. In the absence of solids, [Fe(II)]i = 6.0 mM. In hematite systems, [Fe(II)]i and [Mn(II)]i were either 3.0 or 1.5 mM in order to achieve the desired Mn(II)/Fe(II) molar ratio. Error bars represent 95% confidence intervals.

37 4.2 Effects of Mixed-Mineral Systems on Dichloroacetamide Reduction

The effect of Mn(IV) oxides and the combined effects of Mn(IV) oxides and

Fe(III) oxides on the reduction of dichloroacetamide safeners have not been studied previously. To determine the influence of Mn(IV) oxide on the transformation rate, the safeners benoxacor, dichlormid, and furilazole were monitored in the presence of

Fe(II)-amended birnessite, Fe(II)-amended hematite, and Fe(II)-amended mixed-mineral systems (mixture of birnessite and hematite solid phases). The effect on the addition order was also determined by comparing the effect of adding the safener, Fe(II), or birnessite at time zero.

4.2.1 Effect of Mn(IV) Oxide

In the presence of Fe(II)-amended birnessite with no added hematite, reduction rates of dichlormid, benoxacor, or furilazole were too slow to quantify within a 6-hour sampling period (data not shown). This suggests that Fe(II)-amended Mn(IV) oxide, alone, is insufficient to transform dichloroacetamides. A control reaction containing

Fe(II) in the absence of a mineral phase did not show any appreciable loss of Fe(II) over a 14 day period (data not shown), indicating that trace levels of oxygen in the anaerobic chamber were not sufficient to oxidize Fe(II). In all of the reactions monitored, the reductive dechlorination of dichlormid did not occur within a 6-hour period. This is consistent with previous work, which has shown that the transformation of dichlormid is approximately an order of magnitude slower than that of benoxacor in Fe(II)-amended hematite and goethite.17

38 In Fe(II)-amended mixed-mineral systems, the reduction rate constant of both benoxacor and furilazole changed as the molar ratio of Fe(II)-to-Mn(IV) oxide changed.

As the ratio increased from 0.2 to 2.0, the observed reaction rate constant for benoxacor increased by a factor of 12 (Figure 4.3). The observed reaction rate constant for furilazole increased as the Fe(II)-to-Mn(IV) oxide ratio increased, up until a molar ratio of 1. Above a molar ratio of 1, the reaction rate constant for furilazole did not change appreciably as the ratio of Fe(II) to Mn(IV) oxide increased (Figure 4.3, Appendix F).

A further experiment was performed with benoxacor to confirm that the change in observed rate constant was due to the oxidation of Fe(II) by Mn(IV) oxide. Duplicate reactions were performed for benoxacor in Fe(II)-amended hematite systems. After approximately 4 hours, 1 g/L of birnessite was added to one of the reactors, and the reduction of benoxacor slowed (Figure 4.4). At the time birnessite was added, the fraction of benoxacor remaining in both reactors was approximately 82%. In the absence of birnessite, after 50 hours, 56% of the remaining benoxacor had transformed. In the reactor with added birnessite, after 50 hours, only 30% of the remaining benoxacor had transformed. Natural agricultural soils contain between 0.1 and 0.74% total manganese and between 1.4 and 2.8% total iron.61 On average, manganese makes up 10 – 30% by mass of the total iron + manganese loading.61 In this system, the added birnessite amounted to approximately 10% of the total solids loading (by mass); a larger fraction of birnessite would likely be sufficient to effectively stop the reduction of benoxacor.

A previous study examining the oxidation of Fe(II) by birnessite at varying pH conditions concluded that the oxidation of Fe(II) above pH 4 is controlled by reactions at the surface of birnessite.62 The study found that as Fe(II) was oxidized, surface sites on

39 birnessite minerals were filled, slowing the oxidation rate over time.62 Our results are consistent with this, and indicate that as Mn(IV) oxide loading increases relative to the

a. 5.0E-05 4.5E-05 4.0E-05 3.5E-05

) 3.0E-05 1 -

(s 2.5E-05 obs k 2.0E-05 1.5E-05 1.0E-05 5.0E-06 0.0E+00 0 0.4 0.8 1.2 1.6 2 Fe(II):Mn(IV) Oxide Molar Ratio

b. 3.0E-05

2.5E-05

2.0E-05 ) 1 -

(s 1.5E-05 obs k 1.0E-05

5.0E-06

0.0E+00 0 0.4 0.8 1.2 1.6 2 Fe(II):Mn(IV) Oxide Molar Ratio

Figure 4.3. Observed reduction rate constant (kobs) for a. benoxacor (BN) and b. furilazole (FZ) with changing molar ratio of Fe(II) to Mn(IV) oxide. Reaction conditions: [BN]i = 20 µM or [FZ]i = 20 µM, Fe(III) oxide loading = 9 g/L, Mn(IV) oxide loading = 1 g/L, [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0. Error bars represent 95% confidence intervals. Dotted line represents an Fe(II) to Mn(IV) oxide molar ratio of 1.0.

40 concentration of Fe(II), the reduction rate of dichloroacetamide safeners slows. This is likely due to the increased number of surface sites on the birnessite mineral that must be blocked by Fe(III) prior to exhausting the oxidative capacity of birnessite.

1.2

0.8 * *

0.6

Fraction Fraction of Analyte 0.4

0.2 Birnessite Present Hematite Only 0 0 10 20 30 40 50 Time (h) Figure 4.4. Comparison of the reduction of benoxacor (BN) in the absence of and after the addition of birnessite. Reactor conditions: Fe(III) oxide loading = 10 g/L, [Fe(II)]i = 3.0 mM, [BN]i = 20 µM, [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0. Dotted line indicates the addition of 1.2 g/L birnessite (filled symbols). Error bars represent 95% confidence intervals. Hematite only data are significantly different (at the 95% confidence level) than birnessite data at 47 and 50 hours (marked with asterisks).

41 4.2.2 Effect of Addition Order

To determine effects of addition order on dichloroacetamide reduction reactions, the reduction of benoxacor was monitored in two sets of paired reactions: (1) the addition of Fe(II) at time zero compared to benoxacor addition at time zero, and (2) the addition of birnessite at time zero compared to benoxacor addition at time zero. These comparisons are designed to model environmental systems in which the safener is present prior to anaerobic conditions being established and prior to Mn(IV) oxide formation, respectively.

In both systems, the addition order did not significantly affect the transformation rate of benoxacor (Figure 4.5 and 4.6). The comparison of Fe(II) and benoxacor addition at time zero yielded nearly overlapping data for the reduction of benoxacor (Figure 4.5).

The addition of Mn(IV) oxide at time zero yielded results that were not statistically different at the 95% confidence level from addition of the safener at time zero

(Figure 4.6).

These results are consistent with the assumption that surface reaction is rate-limiting in our Fe(II)-amended hematite systems. If, for example, the adsorption reaction (eq 1) were rate-limiting, we would expect to observe a slower loss of the parent safener (benoxacor) when the safener (rather than Fe(II)) was added at time zero.

0.8

0.6

0.4 Fraction Fraction of Analyte

0.2 Safener as t=0 Fe(II) as t=0 0 0 20 40 60 80 100 Time (h) Figure 4.5. Comparison of benoxacor (BN) reduction with safener and Fe(II) addition as time zero. Reactor conditions: Fe(III) oxide loading = 10 g/L, [Fe(II)]i = 3.0 mM, [BN]i = 20 µM, [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0. Error bars represent 95% confidence intervals.

0.8

0.6

0.4 Fraction Fraction of Analyte

0.2 Safener as t=0 Mn(IV) oxide as t=0 0 0 20 40 60 80 100 Time (h)

Figure 4.6. Comparison of benoxacor (BN) reduction with safener and Mn(IV) oxide addition as time zero. Reactor conditions: Fe(III) oxide loading = 9.8 g/L, Mn(IV) oxide loading = 0.2 g/L, [Fe(II)]i = 3.0 mM, [BN]i = 20 µM, [MOPS] = 30.0 mM, [NaCl] = 50.0 mM, pH 7.0. Error bars represent 95% confidence intervals.

43 4.3 Broader Impacts

From reactions performed in the presence of agrochemical co-formulants, we conclude that the presence of the herbicide S-metolachlor and the presence of the surfactants SDS, MyTAB, and Triton X-100® does not appreciably affect the reductive dechlorination of the dichloroacetamide safener benoxacor in homogeneous solutions containing CrII as a model reductant. The ratios of safener-to-co-formulants used herein are consistent with those present in commercial agrochemical formulations. Accordingly, our results suggest that rates of homogeneous, abiotic reductive dechlorination of benoxacor are unlikely to be affected by co-formulants.

Due to the extensive use of MOPS as a pH buffer and NaCl as an ionic strength adjuster in laboratory systems containing dissolved metals, it is necessary to examine the potential influence of buffer concentration and ionic strength to ensure these reactor components do not contribute to experimental artifacts. Reductive dechlorination rate constants for benoxacor and furilazole did not change appreciably as a function of MOPS or NaCl concentration. Although previous research has shown that MOPS can affect reduction rates of CCl4 in Fe(II)-amended goethite systems, our results demonstrate that

MOPS and NaCl do not affect reduction rates of dichloroacetamide safeners in

Fe(II)-amended hematite systems.

Although hematite is present in soils in abundance, birnessite and other redox-active mineral phases are also present. In Fe(II)-amended hematite systems at pH

7.0, the presence of birnessite had an appreciable impact on reductive dechlorination rate constants of benoxacor and furilazole. As the molar ratio of Fe(II)-to-birnessite loading increased, the observed reduction rate constant for benoxacor and furilazole increased

44 12-fold and 2-fold, respectively. Most previous laboratory studies include only one mineral at a time when examining mineral-mediated transformations of environmental contaminants.17,43-48,50,51,53,56,57 The results of our mixed-mineral systems highlight the utility of studying more realistic model environmental systems (i.e., those containing multiple mineral phases) when seeking to better represent field conditions in a laboratory setting.

Previous work investigated the reductive dechlorination of benoxacor and dichlormid in Fe(II)-amended hematite and Fe(II)-amended goethite systems, and concluded that, while the behavior of the safeners in both systems was similar, the goethite system had somewhat higher reactivity.17 In both systems, the observed reduction rate constant for benoxacor was approximately an order of magnitude greater than that of dichlormid.17 Our results indicate that transformations of dichloroacetamide safeners will happen more slowly in the presence of birnessite. Overall, when assessing risks associated with using dichloroacetamide safeners, it is important to consider the possible formation of monochlorinated products possessing increased bioactivity relative to the parent compounds.

To understand the complete picture of dichloroacetamide fate and transformation, it will be necessary to examine these compounds in the presence of natural organic matter and in complex soil samples. As previously demonstrated for chloroacetamide herbicides,12-15 dichloroacetamide safeners may conceivably enter drinking water sources and transform during drinking water treatment processes (e.g., upon reactions with strong oxidizing agents, Appendix G).

APPENDICES

APPENDIX A

Reagents

Aldrich Aldrich Aldrich Aldrich Aldrich Aldrich ------Organics Organics Organics Organics Organics Vendor Alfa Aesar Alfa Aesar Alfa Aesar Alfa Chem Service Chem Service Chem Sigma Sigma Sigma Sigma Sigma Sigma Acros Organics Acros Organics Acros Organics Acros Organics Acros Organics Acros Organics Acros Acros Acros Acros Acros Acros Fisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher ScientificFisher Tokyo Chemical IndustryChemical Tokyo Ricca Chemical Company Chemical Ricca Cambridge Isotope Laboratories Isotope Cambridge 6% – 99% 99% 99% 99% 70% 99% 20% 50% 98+% >97% ≥99% ≥99% >99% 99+% Purity 96.8% 99.9% 99.8% 99.9% 99.9% 99.5% 99.9% 99.9% 96.5% 98.4% 99.5% 98.4% ≥99.5% 99.7+% ≥98.0% ≥95.0% ≥99.5% 99.99+% 99.999% 5.65 electrophoresis grade electrophoresis endor. V CN) 3 dibasic urity, and and urity, disulfonic acid, disodium salt hydrate salt disodium acid, disulfonic P - 100 sulfonic acid (MOPS) acid sulfonic - - X D3 (CD D3 1 - - p,p' ® - chloroacetamide bicarbonate - heir 2 toluene Reagent - T furilazole acid nitric methanol vernolate metolachlor acetochlor benoxacor dichlormid acetonitrile - acid sulfuric Triton iron(III) oxide iron(III) S sodium nitrate sodium sodium chloridesodium chlorobenzonitrile bromidesodium triazine - - sodium thiosulfate sodium sodium diallyl 2 manganese(IV) oxide manganese(IV) chromium(II)chloride - acetonitrile iron powder, <10 micron <10 powder, iron sodium phosphate, sodium iron(II) chloride, anhydrous chloride, iron(II) anhydrous acetate, sodium 1,2,4 sodium hydroxide (aqueous) hydroxide sodium sodium chloride (high purity) (high chloride sodium - N,N sodium hypochlorite (aqueous) hypochlorite sodium eagents, eagents, sodium dodecyl sulfate (aqueous) sulfate dodecyl sodium manganese(II) chloride tetrahydrate chloride manganese(II) R myristyltrimethylammonium bromide myristyltrimethylammonium morpholinopropane - of 3 diphenyl - List List 5,6

- pyridyl) - (2 - 3 Table A.1. Table

APPENDIX B

Preparation and Standardization of Aqueous Fe(II) and Cr(II)

49 B.1 Preparation and Standardization of Aqueous Fe(II)

Fe(II) spiking solutions were prepared in an anaerobic chamber (3% H2, 97% N2) by dissolving FeCl2 solid into 40 mL of 30.0 mM MOPS buffer with 50.0 mM NaCl.

Fe(II) concentration in solution was standardized prior to each use via UV-vis spectrophotometry. UV-vis standardization followed Stookey59 and used the disodium salt of 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine-p,p'-disulfonic acid (ferrozine) as an indicator for Fe(II). The FeCl2 solution was filtered through a 0.2 µM syringe filter and

1.7 mL was combined with 0.6 mL of acetate buffer (1.0 M, pH 5.5) and 1.9 mL of ferrozine solution (2.0 M). The mixture was analyzed via UV-vis spectrophotometry at

562 nm (ε 20,735 M-1cm-1)17 and used within 24 hours of standardization.

B.2 Reduction and Standardization of Aqueous Cr(II)

Cr(II) stock solutions were prepared following Sivey and Roberts.17 Under anaerobic conditions (3% H2, 97% N2), in an Erlenmeyer flask, CrCl2 and Fe(0) powder

(<10 µm) were combined in 18 MΩ•cm water at a CrCl2-to-Fe(0) molar ratio of 1:3. The solution was magnetically mixed at room temperature for approximately 12 hours to reduce any oxidized Cr species to Cr(II). After mixing, the solution was filtered through a

0.2 µm nylon syringe filter and the concentration of Cr(II) was determined using UV-vis spectrophotometry (714 nm, ε 5.6 M-1cm-1).58 Stock solutions were prepared by diluting the concentrated Cr(II) to 500 µM in 5.0 mM H2SO4. Stock solutions were used within 24 hours of standardization. Control experiments determined that the Fe(0) particles, in the absence of Cr(II), did not transform the dichloroacetamide safeners examined herein.

APPENDIX C

Synthesis and Characterization of Monochlorinated Analogue of Benoxacor

51 C.1 Synthesis of Monochlorinated Analogue of Benoxacor

Synthesis reactions followed procedures outlined in Sivey and Roberts17 with some modifications noted below. Acetonitrile (6 mL) was added to 18 MΩ•cm water

(200 mL) in a 250-mL Erlenmeyer flask. Benoxacor (0.10 mmol) was added to the solution as a pure solid and dissolved via sonication. The solution was subsequently sparged with high purity N2 and transferred to an anaerobic chamber (3% H2, 97% N2).

After further sparging with chamber atmosphere, 1.0 mmol of CrCl2 was added to the solution to serve as a reductant. The solution was magnetically mixed at room temperature for approximately 1.5 hours to obtain the maximum amount of monochlorinated analogue, without appreciable formation of the deschlorinated analogue

(as determined by GC/MS). To quench the reaction, the solution was removed from the anaerobic chamber and magnetically mixed to oxygenate the solution and oxidize any remaining Cr(II). Products were isolated via extraction into toluene and volatilization of the solvent under forced air.

Synthesis results were confirmed via qualitative GC/MS analysis (SIM mode, see

Appendix D) and further characterized via proton nuclear magnetic resonance

(1H NMR), as described below. The product was reconstituted in toluene for preparation of calibration standards.

52 C.2 Characterization of Monochlorinated Analogue of Benoxacor

Characterization via 1H NMR was carried out on a JEOL 400SS at 290 K and

399.78 MHz, using CD3CN solvent. Chemical shifts for the monochlorinated analogue were compared to those reported for benoxacor in the literature.9 Chemical shifts for benoxacor are: δ 1.22 (C9, d, 3H), 4.24 (C2, d, 2H), 4.69 (C3, b, 1H), 6.78 (C11, s, 1H),

6.9-7.0 (C7-8, c, 2H), 7.17 (C6, t, 1H), and 7.61 (C5, b, 1H).9 For consistency, carbon atom numbering in the monochlorinated analogue matches those used by Miller, et al.9

Chemical shifts for monochlorinated product are as follows (Figure C.1): δ 1.15 (C9, b,

3H), 4.17 (C2, d, 2H), 4.33-4.40 (C3, 2d, 1H), 6.76 (C11, s, 2H), 6.89 (C6-8, t, 3H), and

7.07 (C5, t, 1H). Shifts at δ 1.90 and 2.14 correspond to CD3CN and toluene, respectively; toluene is likely residual from the extraction step of the synthesis.

Analysis via GC/MS provided further information on the purity of the product.

No unidentifiable peaks were present on the chromatogram (Figure C.2) and the presence of deschlorinated product was negligible. GC/MS analysis also allowed for quantification of the amount of parent compound present and calculation of the monochlorinated analogue concentration for preparation of calibration standards.

toluene

5.0

11 4.0 10 5 4 3 9 6

7 1 2 8 CD3CN 3.0 Abundance

2.0 C6-8

1.0 C9 C11 C5

0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 Chemical Shift (δ 1H) Figure C.1. 1H NMR spectrum for the monochlorinated analogue of benoxacor. For consistency, carbon numbering follows that of Miller, et al.9 Chemical shifts at δ 1.90 and 2.14 correspond to CD3CN and toluene, respectively.

1.9 e+7 1.8 e+7 A 1.7 e+7 1.6 e+7 1.5 e+7 1.4 e+7 1.3 e+7 1.2 e+7 1.1 e+7 1.0 e+7 9.0 e+6 Abundance 8.0 e+6 7.0 e+6 6.0 e+6 5.0 e+6 4.0 e+6 3.0 e+6 B 2.0 e+6 1.0 e+6 C 0.0 5.00 10.00 15.00 Time (min)

Figure C.2. GC/MS chromatogram for the monochlorinated benoxacor synthesis product. Peak A corresponds to the monochlorinated analogue, peak B corresponds to benoxacor, and peak C corresponds to the deschlorinated analogue.

APPENDIX D

GC Method Details

56 D.1 Instrument Parameters

GC analyses were performed on an Agilent 7890A GC with µECD and an Agilent

5975C MSD. Both µECD and MS detection used a single, splitless injection of 1 µL.

Inlet temperature was 250 ⁰C for µECD and 280 ⁰C for MS detection. MS transfer line temperature was 280 ⁰C. An Agilent DB-5MS+DG column (30 m, 250 µm i.d., 0.25 µm) was used with both detectors. Carrier gas for both systems was He, with a flow rate of 2.0 and 1.0 mL/min for µECD and MS detection, respectively. Total run time for both the

µECD and MS methods was 18.00 minutes. The temperature ramp program for both

µECD and MS detection is given in Table D.1. The retention times for µECD and the retention times and quantitation ions for MS detection are given in Table D.2.

Table D.1. GC Oven Temperature Program for Both µECD and MS Detection. Total Run Time Was 18.00 Minutes.

Rate (⁰C/min) T (⁰C) Hold Time (min)

Initial --- 85 0.5

Ramp 12 280 1.25

Table D.2. GC Retention Times for All Analytes with Both µECD and MS Detection, and Quantitation Ions Used with Selected Ion Monitoring (SIM) for MS Detection. µECD Retention Time MSD Retention Time Quantitation Ions Analyte (min) (min) (m/z) a

Internal Standard (CBN) b 5.41 5.12 137, 139

CDAA 6.58 6.31 173, 132

Dichlormid 7.24 7.05 207, 172

Vernolate c n.a. e 7.95 203, 128

Monochlorinated Furilazole d 10.69 n.a. e n.a. e

Furilazole d 11.02 n.a. e n.a. e

Monochlorinated Benoxacor 11.39 11.27 225, 134 Benoxacor 11.99 11.87 259, 120

S-Metolachlor 13.03 12.96 283, 162 a Ions are listed with the monoisotopic molecular ion mass first, followed by the most abundant fragment ion with m/z greater than 100 b CBN = 2-chlorobenzonitrile c Vernolate was not detected using µECD due to the lack of halogens d Furilazole and the monochlorinated analogue were not analyzed via GC/MS e Not available

58 D.2 Calibration

A linear calibration range of 0.3 – 16 µM was employed for all analytes. Peak separation with baseline resolution was observed for all analytes with both µECD and

MSD (Figures D.1 and D.2, respectively). For calibration curves, analyte detector response was normalized to that of the internal standard (2-chlorobenzonitrile, CBN). A typical calibration curve is shown in Figure D.3, with corresponding limits of detection

(LODs) and R2 values given in Table D.3. For the GC measurements performed herein,

LODs were calculated using eq D163:

2ts 1 1 x̅2 OD y √ [D1] m I 2 ∑ (xi-x̅ where I is the number of calibration standards; J is the number of replicates of each standard; is the number of replicates of the unknown; t is the Student’s t, obtained from a one-tailed t-distribution with (I×J) – 2 degrees of freedom; m is the slope of the calibration curve; sy is the standard error of y for the calibration curve; and x̅ is the mean concentration of the calibration standards.

5.5 e+8 Benoxacor 5 e+8

4.5 e+8

4 e+8 Furilazole

3.5 e+8 Dichlormid 3 e+8

Abundance 2.5 e+8

2 e+8

1.5 e+8 1 e+8 S-Metolachlor 5 e+7 CBN CDAA

5.00 10.00 15.00 Time (min) Figure D.1. GC/µECD chromatogram of a calibration standard. All analytes are at approximately 16 µM, except for the internal standard (CBN, 10.2 µM). Peak separation and retention times are typical for all standards. The monochlorinated benoxacor analogue is not shown as it was calibrated with a separate set of standards; the monochlorinated furilazole analogue is also not shown as reference material was not available for this analyte (see Table D.2 for retention times). Vernolate lacks halogen atoms and was not detected with µECD.

S-Metolachlor 8 e+6 7.5 e+6 7 e+6 Benoxacor 6.5 e+6 6 e+6 5.5 e+6

5 e+6 Vernolate

4.5 e+6 CBN 4 e+6

Abundance3.5 e+6 3 e+6 2.5 e+6 2 e+6

1.5 e+6 Dichlormid 1 e+6 CDAA 5 e+5

5.00 10.00 15.00 Time (min) Figure D.2. GC/MS chromatogram of a calibration standard. All analytes are at approximately 16 µM, except for the internal standard (CBN, 10.2 µM). Peak separation and retention times are typical for all standards. The monochlorinated analogue of benoxacor is not shown as it was calibrated with a separate set of standards (see Table D.2 for retention time). Furilazole and its monochlorinated analogue are not shown as they were not analyzed using MS detection.

a. 225 BN FZ DL MB SM 150 CDAA

75 Corrected DetectorCorrectedResponse

0 0 3 6 9 12 15 18 Concentration (μM)

b. 40 SM BN VN DL 30 CDAA

10 Corrected DetectorCorrectedResponse

0 0 3 6 9 12 15 18 Concentration (μM) Figure D.3. Calibration curves for all analytes with a. µECD and b. MS detection. Detector response was normalized to the internal standard response for all analytes. Analytes are abbreviated as follows: benoxacor (BN), dichlormid (DL), furilazole (FZ), S-metolachlor (SM), monochlorinated benoxacor (MB), and vernolate (VN). Limits of detection (LODs) and R2 values are given in Table D.3.

2 Table D.3. Limits of Detection (LODs) and R Values for All Compounds, Corresponding to Calibration Curves in Figure D.3. µECD MSD Analyte LOD (µM) R2 LOD (µM) R2 Benoxacor 1.18 0.9993 2.19 0.9975

CDAA 0.943 0.9995 0.717 0.9997

Dichlormid 1.41 0.9990 1.46 0.9989 Furilazole b 1.28 0.9991 n.a. a n.a. a

S-Metolachlor 2.50 0.9966 1.43 0.9989 Monochlorinated Benoxacor b 4.26 0.9911 n.a. a n.a. a

Vernolate c n.a. a n.a. a 1.80 0.9982 a Not available b Furilazole and the monochlorinated analogue of benoxacor were not analyzed with MSD c Vernolate is not detected with µECD due to the lack of halogens on this compound

APPENDIX E

Quality Assurance and Quality Control

64 E.1 Extraction Efficiencies

To quantify analyte recovery following extraction into toluene, extraction

efficiencies were experimentally determined for benoxacor, dichlormid, and furilazole,

both alone and in the presence of surfactants. For extraction efficiencies in the absence of

surfactants, aqueous solutions of analytes were prepared in 30.0 mM MOPS buffer at

pH 7.0 with 50.0 mM NaCl. Analytes were added to the solution as a methanolic spike

(7–14 µM). At room temperature in 4-mL amber glass vials with Teflon-lined caps,

1.0 mL of aqueous analyte solution was combined with either 1.0 or 2.0 mL of toluene

and shaken vigorously by hand for approximately 20 seconds. After phase separation, the

toluene layer was removed for analysis via GC. GC results were compared with the initial

concentrations to obtain extraction efficiencies as a percentage of the amount of analyte

in aqueous solution (Table E.1). Recovery for each analyte was comparable between

extractions using 1:1 and 2:1 toluene-to-aqueous volume ratios. Extraction efficiencies

greater than 100% likely result from temporal variability in the relative detector response

between the internal standard and analytes.

Table E.1. Extraction Efficiency for Safeners in the Absence and Presence of Surfactants. Values are Given as Percentages ± 95% Confidence Intervals.

Safener Without Surfactant SDS a MyTAB b Triton® X-100

Benoxacor 126 ± 8 140 ± 15 138 ± 12 159 ± 13

Dichlormid 121 ± 4 139 ± 11 135 ± 9 161 ± 9

Furilazole 100 ± 20 88 ± 12 85 ± 13 86 ± 13 a Sodium dodecyl sulfate (SDS) b Myristryltrimethylammonium bromide (MyTAB)

65 For extraction efficiencies in the presence of surfactants, aqueous solutions were prepared in the same manner as in the absence of surfactants, with either SDS, MyTAB, or Triton® X-100 present at 5 µM. Extractions with surfactants were performed following the same methods as described above, using only a 1:1 toluene-to-aqueous volume ratio and an aqueous safener concentration of 10 µM (Table E.1).

E.2 Recoveries Following Centrifugation

Recovery of analytes after centrifugation was quantified experimentally to determine potential loss to the surfaces of plastic centrifuge tubes. Aqueous solutions of the safeners benoxacor, dichlormid, and furilazole were prepared alone and in the presence of surfactants, following the same methods as described above for determination of extraction efficiencies. At room temperature, 1.0 mL of aqueous solution was placed into plastic centrifuge tubes and centrifuged for 10 minutes at 13,000 rpm. After centrifugation, samples were extracted into 2.0 mL of toluene, following the method described above. A sample of aqueous solution, subjected to all steps except centrifugation, was analyzed as the control sample. Results from GC analyses were compared to the control sample to determine centrifugation recovery as a percentage of the amount of analyte originally present in the aqueous solution (Table E.2).

Table E.2. Centrifugation Recovery for Safeners in the Absence and Presence of Surfactants. Values are Given as Percentages ± 95% Confidence Intervals. Safener Without Surfactant SDS a MyTAB b Triton® X-100

Benoxacor 104 ± 9 93 ± 10 106 ± 11 101 ± 9 Dichlormid 102 ± 8 96 ± 8 104 ± 8 97 ± 6 Furilazole 90 ± 20 86 ± 19 90 ± 20 100 ± 20 a Sodium dodecyl sulfate (SDS) b Myristryltrimethylammonium bromide (MyTAB)

E.3 Adsorption to Fe and Mn Oxides

Adsorption of the safeners benoxacor, dichlormid, and furilazole to Fe(III) oxide and Mn(IV) oxide was experimentally determined. Adsorption reactors were prepared at room temperature in 40-mL amber glass vials with Teflon-lined caps. Reactors contained

15 mL of aqueous slurry containing 10 g/L solids as either Fe(III) oxide or Mn(IV) oxide in 30.0 mM MOPS buffer at pH 7.0 with 50.0 mM NaCl. Benoxacor, dichlormid, or furilazole was added as a methanolic spike at 20 µM. Adsorption reactors were mixed continuously on a vial rotator for three days before aliquots were centrifuged

(13,000 rpm, 10 min) and extracted into toluene for analysis via GC. Figure E.1 shows the adsorption of analytes onto Fe(III) oxide and Mn(IV) oxide. Aliquots taken from adsorption reactors after 4 days confirmed that equilibrium between the aqueous and solid phases was reached within 3 days. Experiments to determine the adsorption of safeners in the presence of surfactants were also performed. Reactor conditions were the same as in the absence of surfactants, except that reactors also contained SDS, MyTAB, or Triton® X-100 at 4-20 µM. Results of experiments in the presence of surfactants

67 showed that none of the examined surfactants had an appreciable effect on the extent of adsorption of benoxacor, dichlormid, or furilazole (data not shown).

0.5 Fe(III) Oxide Mn(IV) Oxide 0.4

0.3

0.2 Fraction Fraction Adsorbed 0.1

0 BN DL FZ Safener

Figure E.1. Adsorption of safeners benoxacor (BN), dichlormid (DL), and furilazole (FZ) to Fe(III) oxide and Mn(IV) oxide. Furilazole and dichlormid adsorption to both solids is not significantly different from zero. Adsorption of benoxacor is not significantly different between Fe(III) oxide and Mn(IV) oxide. Error bars represent 95% confidence intervals.

APPENDIX F

Summary of Observed Reduction Rate Constants

Table F.1. Reaction Conditions and Observed Rate Constants for the Reduction of Benoxacor in the a Presence of the Herbicide S-Metolachlor. b [S-metolachlor] [Cr(II)]i Benoxacor kobs (μM) (μM) (s-1) 0 500 (4.3 ± 0.4) x 10-4

8.0 500 (4.4 ± 0.2) x 10-4

16 500 (4.3 ± 0.3) x 10-4 24 500 (4.4 ± 0.3) x 10-4 32 500 (4.3 ± 0.3) x 10-4 40 500 (3.8 ± 0.1) x 10-4 a All reactions were performed in a 5.0 mM H2SO4 solution under anaerobic conditions b Observed rate constants are shown ± 95% confidence intervals

Table F.2. Reaction Conditions and Observed Rate Constants for the Reduction of Benoxacor in the Presence of the Surfactant SDS. a b c [SDS] [Cr(II)]i Benoxacor kobs (μM) (μM) (s-1) 0 500 (4.3 ± 0.4) x 10-4 2.0 500 (3.6 ± 0.2) x 10-4 4.0 500 (3.6 ± 0.1) x 10-4 6.0 500 (3.5 ± 0.2) x 10-4 8.0 500 (3.3 ± 0.1) x 10-4 10 500 (3.6 ± 0.2) x 10-4 a All reactions were performed in a 5.0 mM H2SO4 solution under anaerobic conditions b Sodium dodecyl sulfate c Observed rate constants are shown ± 95% confidence intervals

Table F.3. Reaction Conditions and Observed Rate Constants for the Reduction of Benoxacor in the a Presence of the Surfactant MyTAB. b c [MyTAB] [Cr(II)]i Benoxacor kobs (μM) (μM) (s-1) 0 500 (4.3 ± 0.4) x 10-4 2.0 500 (3.9 ± 0.3) x 10-4 4.0 500 (3.9 ± 0.4) x 10-4 6.0 500 (3.8 ± 0.3) x 10-4 8.0 500 (3.9 ± 0.3) x 10-4 10 500 (4.0 ± 0.2) x 10-4 a All reactions were performed in a 5.0 mM H2SO4 solution under anaerobic conditions b Myristryltrimethylammonium bromide c Observed rate constants are shown ± 95% confidence intervals

Table F.4. Reaction Conditions and Observed Rate Constants for the Reduction of Benoxacor in the Presence of the Surfactant Triton X-100®. a

b [Triton X-100®] [Cr(II)]i Benoxacor kobs (μM) (μM) (s-1) 0 500 (4.3 ± 0.4) x 10-4 2.0 500 (4.1 ± 0.4) x 10-4 4.0 500 (4.3 ± 0.5) x 10-4 6.0 500 (4.1 ± 0.3) x 10-4 8.0 500 (4.4 ± 0.3) x 10-4 10 500 (4.1 ± 0.7) x 10-4 a All reactions were performed in a 5.0 mM H2SO4 solution under anaerobic conditions b Observed rate constants are shown ± 95% confidence intervals

5 5 5 5 5 5 5 5 5 5 5 ------b,c 10 10 10 10 10 10 10 10 10 10 10 obs x x x x x x x x x x x k ) 1 - 0.3) 0.3) 0.3) 0.4) 0.2) 0.3) 0.3) 0.4) 0.4) 0.4) 0.4) 0.3) (s ± ± ± ± ± ± ± ± ± ± ± (1.9 (2.0 (2.1 (1.9 (2.2 (1.7 (2.1 (2.1 (1.5 (2.0 (1.8 Furilazole Furilazole

5 5 5 5 5 5 5 5 5 5 5 b,c ------10 10 10 10 10 10 10 10 10 10 10 obs k x x x x x x x x x x x ) 1 - 0.3) 0.3) 0.3) 0.3) 0.3) 0.3) 0.3) 0.2) 0.3) 0.2) 0.3) 0.3) (s of Benoxacor and Furilazole and Furilazole of Benoxacor

± ± ± ± ± ± ± ± ± ± ± (2.0 (2.1 (2.3 (2.3 (2.4 (1.7 (2.1 (2.3 (1.7 (2.3 (2.0 Benoxacor Reduction

Loading Loading 10 10 10 10 10 10 10 10 10 10 10 (g/L) Hematite

. i ) )] 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 (mM [Fe(II 95% confidence intervals 95% confidence

± Concentrations ] mM) 50.0 50.0 50.0 50.0 50.0 10.0 25.0 50.0 ormed at pH 7.0 under anaerobic conditions pHanaerobic under 7.0 ormed at ( 100.0 150.0 200.0 [NaCl ] Reaction Conditions and Observed Rate Constants for the for Constants Rate and Observed Conditions Reaction

mM) 10.0 20.0 30.0 40.0 50.0 30.0 30.0 30.0 30.0 30.0 30.0 ( MOPS [ Changing MOPS and NaCl MOPSNaCl Changing and

Observed rate constants are shown shown are constants Observed rate All reactions were perf were All reactions System contained benxoacor and furilazole in a single reactor for each conditions set of for reactor in a andsingle furilazole benxoacor contained System Table F.5. Table with a b c

5 5 5 5 5 5 5 5 5 ------b,c 6 - 10 10 10 10 10 10 10 10 10 obs x x x x x x x x x k 10 d ) . x 1 - a . 0.2) 0.2) 0.7) 0.7) 0.5) 0.9) 0.6) 0.7) 0.5) 0.4) (s 2) n ± ± ± ± ± ± ± ± ± ±

(6 (1.2 (1.5 (1.4 (1.7 (1.7 (1.7 (1.6 (1.5 (1.3 Furilazole Furilazole 5 5 5 5 5 5 5 5 b,c ------6 6 5 - - - 10 10 10 10 10 10 10 10 obs k x x x x x x x x 10 10 10 ) x x x 1 - 0.7) 0.7) 0.9) 1.4) 1.0) 1.2) 1.7) 1.2) 1.3) (s 3) 5) 2) ± ± ± ± ± ± ± ± ± ± ± (2 (6 (3 (1.5 (1.6 (1.9 (1.9 (1.9 (2.4 (3.3 (3.0 Benoxacor of Benoxacor and Furilazole in in and Furilazole of Benoxacor

) Loading 9 9 9 9 9 9 9 9 9 9 9 Reduction (g/L

Hematite at pH 7.0 and NaCl (50.0 mM) under anaerobic conditions underNaCl7.0 mM) (50.0 pH and anaerobic at ) Loading 1 1 1 1 1 1 1 1 1 1 1 (g/L

Birnessite 95% confidence intervals 95% confidence

± i ) was observed was )]

10 12 14 16 18 20 22 2.0 4.0 6.0 8.0 (mM [Fe(II

Oxide Ratio 0.2 0.3 0.5 0.7 0.9 1.0 1.2 1.4 1.6 1.7 1.9 Reaction Conditions and Observed Rate Constants for the for Constants Rate and Observed Conditions Reaction

Mineral Systems. Mineral Molar Molar - Fe(II):Mn(IV) Observed rate constants are shown shown are constants Observed rate transformation No appreciable All reactions were performed with MOPS buffer (30.0 mM) mM) (30.0 MOPSbuffer with performed were All reactions Reactions with benoxacor and furilazole were performed in separate reactors, under identical conditions under identical reactors, separate in performed were furilazole and benoxacor with Reactions a b c d Table F.6. Table Mixed

APPENDIX G

Transformation of Dichloroacetamide Safeners in the Presence of Free Chlorine and Free Bromine

74 G.1 An HPLC Method for Dichloroacetamide Safeners

Although a method for GC analysis of dichloroacetamides has been developed

(see Appendix D), some studies may warrant analysis via high performance liquid chromatography (HPLC). For the preliminary experiments with dichloroacetamide safeners under model drinking water treatment conditions described in this Appendix, the following HPLC method was developed.

HPLC analyses were performed on an Agilent 1260 Infinity HPLC with diode array detector (DAD) and an Agilent Poroshell 120 EC-C18 column (2.1 mm × 50 mm ×

2.7 µm). Temperature was fixed at 20 °C in the column and autosampler compartments.

Total run time for the method was 3.00 minutes, with a 5 µL injection volume and a mobile phase of 50% acetonitrile and 50% 18 MΩ•cm water (by volume) at a flow rate of

0.55 mL/min. Absorbance data were collected at 254 nm. Wavelengths of maximum absorbance and retention times for all analytes are given in Table G.1.

A calibration range of 1 – 100 µM was used for analyses of benoxacor and furilazole and 2 – 200 µM for dichlormid. Peak separation with baseline resolution was observed for all analytes (Figure G.1). Figure G.2 shows a typical calibration curve with limits of detection (LODs) and R2 values.

Table G.1. Wavelengths of Maximum Absorbance (λmax) and HPLC Retention Times for All Analytes. Retention Time Analyte λ (nm) a max (min) Benoxacor 256 1.15

Dichlormid b 216 0.70

Furilazole b 219 0.89 a Determined experimentally b Dichlormid and furilazole were not monitored at their

λmax due to increased noise from the mobile phase at those wavelengths

Benoxacor

Detector Detector Response Dichlormid

10 Furilazole

0.5 1.0 1.5 2.0 2.5 Time (min) Figure G.1. HPLC chromatogram of a calibration standard. Concentrations of analytes are as follows: [Dichlormid] = 104 µM, [Furilazole] = 52.8 µM, [Benoxacor] = 51.8 µM. Peak separation and retention times are typical for all standards (see Table G.1).

a. 350 R2 = 0.99998 300 LOD = 0.0539 µM 250 • min) • 200

mAU 150 100 50 Peak Peak Area( 0 0 20 40 60 80 100 Concentration of BN (uM)

b. 50 R2 = 0.9999 40 LOD = 1.44 µM • min) • 30 mAU 20

10 Peak Peak Area( 0 0 50 100 150 200 Concentration of DL (uM)

c. 30 R2 = 0.9999 25 LOD = 0.657 µM

• min) • 20

mAU 15

5 Peak Peak Area( 0 0 20 40 60 80 100 Concentration of FZ (uM) Figure G.2. Calibration curves and limits of detection (LODs) for: a. benoxacor (BN), b. dichlormid (DL), and c. furilazole (FZ) using HPLC.

77 G.2 Chlorination and Bromination Rates of Dichloroacetamide Safeners

Observed chlorination rate constants for benoxacor, dichlormid, and furilazole were determined following the experimental method used by Sivey et al.,64 with modifications noted below. Reaction solutions were buffered at pH 7.0 with phosphate buffer (10.0 mM) that also contained NaNO3 (50.0 mM) and high purity NaCl

(10.0 mM). The initial concentration of free chlorine was 500 µM for all reactions, and the initial concentration of safener was 20 µM. Chlorination rate data were analyzed via

GC/MS (see Appendix D). The observed chlorination for benoxacor was an order of magnitude faster than dichlormid and furilazole (Table G.2).

Table G.2. Observed Pseudo-First-Order Rate Constants (± 95% Confidence Intervals) for Chlorination and Bromination of Safeners at 20.0 °C. Observed Chlorination Rate Observed Bromination Rate Analyte -1 -1 Constant, kobs,Cl (s ) Constant, kobs,Br (s ) Benoxacor (4.0 ± 0.7) x 10 –4 (9.8 ± 0.8) x 10 –6

Dichlormid (1.4 ± 0.2) x 10 –5 (4.8 ± 0.5) x 10 –3

Furilazole (7 ± 3) x 10 –5 (5.7 ± 0.4) x 10 –3

Observed rate constants for bromination of benoxacor, dichlormid, and furilazole were also determined following the method reported by Sivey et al.64 For bromination reactors, pH was buffered using bicarbonate (20.0 mM) at pH 7.0, with

NaNO3 (90.0 mM) and high purity NaCl (10.0 mM). The initial concentrations of free chlorine and bromide were 600 µM and 500 µM, respectively, and the initial concentration of safener was 40 µM. Bromination rate data were obtained via HPLC. The bromination rates increased in the following order: benoxacor < dichlormid < furilazole

(Table G.2).

78 G.3 Phosphate Catalysis of Benoxacor Bromination

In order to determine the possible effect of phosphate buffer on the bromination of safeners, benoxacor bromination was monitored as a function of the concentration of phosphate at pH 6.0, 6.5, and 7.0. Reactor conditions and analyses followed those described above for bromination. Buffer solutions contained [phosphate] = 5.00 –

50.0 mM, [NaNO3] = 55.0 mM, [NaCl] = 5.00 mM, and were adjusted to the desired pH using HNO3 (1.6 M) or NaOH (1 M). For most pH values, the bromination rate constant for benoxacor increased as the concentration of phosphate increased (Figure G.3). These results suggest that the bromination of benoxacor is catalyzed by phosphate buffer. The increasing dependency on phosphate concentration as pH decreases also suggests that

– protonated phosphate species (e.g., H2PO4 ) play a larger role in this catalysis relative to

2– 3– species such as HPO4 and PO4 .

1.6E-03 pH 6.0

pH 6.5 y = (0.006 ± 0.004)x + (0.0010 ± 0.0001) R² = 0.8912 1.2E-03 pH 7.0 ) 1 -

(s 8.0E-04

obs y = (0.006 ± 0.003)x + (0.00029 ± 0.00007) k R² = 0.9510

4.0E-04 y = (0.0027 ± 0.0006)x + (0.00010 ± 0.00002) R² = 0.9847

0.0E+00 0 0.01 0.02 0.03 0.04 0.05

[Phosphate]tot, M Figure G.3. Observed pseudo-first-order rate constants for benoxacor bromination as a function of the concentration of phosphate buffer at varying pH. Reaction conditions: [benoxacor]i = 40 µM, [HOCl]i = 600 µM, [NaBr]i = 500 µM,

[NaNO3] = 55.0 mM, [NaCl] = 5.00 mM, [Phosphate]tot = 5.00 – 50.0 mM. Error bars represent 95% confidence intervals. Error bars that are not visible are smaller than the symbol.

79 REFERENCES

1. Abu-Qare, A. W.; Duncan, H. J. Herbicide safeners: uses, limitations, metabolism, and

mechanisms of action. Chemosphere 2002, 48, 965-974.

2. Cobb, A. H.; Reade, J. P. H. Herbicide Selectivity and Metabolism. In Herbicides and

Plant Physiology, 2nd ed.; Wiley-Blackwell: West Sussex, U.K., 2010.

3. Davies, J. Herbicide safeners - commercial products and tools for agrochemical

research. Pestic. Outlook 2001, 12, 10-15.

4. Davies, J.; Caseley, J. C. Herbicide safeners: a review. Pestic. Sci. 1999, 55, 1043-

1058.

5. Rosinger, C.; Kocher, H. Safener for Herbicides. In Modern Crop Protection

Compounds: Volume 1, 1st ed.; Kramer, W.; Schirmer, U., Eds.; Wiley-VCH:

Weinheim, Germany, 2007.

6. Komives, T.; Hatzios, K. K. Chemistry and structure-activity relationships of herbicide

safeners. Z. Naturforsch. 1991, 46c, 798-804.

7. Newman, M. C. Organic compound detoxification. In Fundamentals of Ecotoxicology,

3rd ed.; CRC Press: Boca Raton, FL, 2010.

8. Tomlin, C. D. S., Ed. The Pesticide Manual: A World Compendium, 15th ed.; British

Crop Protection Council: Hampshire, U.K., 2009.

9. Miller, K. D.; Irzyk, G. P.; Fuerst, E. P.; McFarland, J. E.; Barringer, M.; Cruz, S.;

Eberle, W. J.; Fory, W. Time course of benoxacor metabolism and identification of

benoxacor metabolites isolated from suspension-cultured Zea mays cells 1 h after

treatment. J. Agric. Food Chem. 1996, 44, 3326-3334.

80 10. Abu-Qare, A. W.; Duncan, H. J. Photodegradation of the herbicide EPTC and the

safener dichlormid, alone and in combination. Chemosphere 2002, 46, 1183-1189.

11. Aga, D. S.; Thurman, E. M.; Yockel, M. E.; Zimmerman, L. R.; Williams, T. D.

Identification of a new sulfonic acid metabolite of metolachlor in soil. Environ. Sci.

Technol. 1996, 30, 592-597.

12. Hladik, M. L.; Bouwer, E. J.; Roberts, A. L. Neutral degradates of chloroacetamide

herbicides: occurrence in drinking water and removal during conventional water

treatment. Water Res. 2008, 42, 4905-4914.

13. Hladik, M. L.; Hsiao, J. J.; Roberts, A. L. Are neutral chloroacetamide herbicide

degradates of potential environmental concern? Analysis and occurrence in the upper

Chesapeake Bay. Environ. Sci. Technol. 2005, 39, 6561-6574.

14. Hladik, M. L.; Roberts, A. L.; Bouwer, E. J. Removal of neutral chloroacetamide

herbicide degradates during simulated unit processes for drinking water treatment.

Water Res. 2005, 39, 5033-5044.

15. Kolpin, D. W.; Barbash, J. E.; Gilliom, R. J. Occurrence of pesticides in shallow

groundwater of the United States: initial results from the National Water-Quality

Assessment Program. Environ. Sci. Technol. 1998, 32, 558-566.

16. Loch, A. R.; Lippa, K. A.; Carlson, D. L.; Chin, Y. P.; Traina, S. J.; Roberts, A. L.

Nucleophilic aliphatic substitution reactions of propachlor, alachlor, and metolachlor

- 2- with bisulfide (HS ) and polysulfides (Sn ). Environ. Sci. Technol. 2002, 36, 4065-

4073.

81 17. Sivey, J. D.; Roberts, A. L. Abiotic reduction reactions of dichloroacetamide

safeners: transformations of "inert" agrochemical constituents. Environ. Sci.

Technol. 2012, 46, 2187-2195.

18. Schwarzenbach, R. P.; Escher, B. I.; Fenner, K.; Hofstetter, T. B.; Johnson, C. A.;

von Gunten, U.; Wehrli, B. The challenge of micropollutants in aquatic systems.

Science 2006, 313, 1072-1077.

19. Joly, P.; Bonnemoy, F.; Charvy, J.; Bohatier, J.; Mallet, C. Toxicity assessment of the

maize herbicides S-metolachlor, benoxacor, mesotrione and nicosulfuron, and their

corresponding commercial formulations, alone and in mixtures, using the

Microtox® test. Chemosphere 2013, 93, 2444-2450.

20. Castro, M. J. L.; Ojeda, C.; Cirelli, A. F. Advances in surfactants for agrochemicals.

Environ. Chem. Lett. 2014, 12, 85-95.

21. Katagi, T. Surfactant effects on environmental behavior of pesticides. Rev. Environ.

Contam. Toxicol. 2008, 194, 71-177.

22. Knowles, A. Recent developments of safer formulations of agrochemicals.

Environmentalist 2008, 28, 35-44.

23. Liu, Z. Effects of surfactants on foliar uptake of herbicides - a complex scenario.

Colloids Surf., B 2004, 35, 149-153.

24. Pena, A.; Rodriguez-Liebana, J. A.; Mingorance, M. D. Persistence of two

neonicotinoid insecticides in wastewater, and in aqueous solutions of surfactants and

dissolved organic matter. Chemosphere 2011, 84, 464-470.

82 25. Tanaka, F. S.; Wien, R. G.; Mansager, E. R. Survey for surfactant effects on the

photodegradation of herbicides in aqueous media. J. Agr. Food Chem. 1981, 29,

227-230.

26. Hess, F. D.; Foy, C. L. Interaction of surfactants with plant cuticles. Weed Technol.

2000, 14, 807-813.

27. Davies, J. T. A Quantitative Kinetic Theory of Emulsion Type. I. Physical Chemistry

of the Emulsifying Agent. In Proceedings of the 2nd International Congress of

Surface Activity, Shulaman, J. H., Ed.; Butterworths: London, 1957.

28. Maurice, P. A. Environmental Surfaces and Interfaces from the Nanoscale to the

Global Scale, John Wiley & Sons, Inc.: Hoboken, NJ, 2009.

29. Sposito, G. The Chemistry of Soils, 2nd ed.; Oxford University Press: New York,

2008.

30. Learman, D. R.; Voelker, B. M.; Vazquez-Rodriguez, A. I.; Hansel, C. M. Formation

of manganese oxides by bacterially generated superoxide. Nat. Geosci. 2011, 4, 95-

98.

31. Herndon, E. M.; Jin, L.; Andrews, D. M.; Eissenstat, D. M.; Brantley, S. L.

Importance of vegetation for manganese cycling in temperate forested watersheds.

Global Biogeochem. Cycles 2015, 29, 160-174.

32. Herndon, E. M.; Martinez, C. E.; Brantley, S. L. Spectroscopic (XANES/XRF)

characterization of contaminant manganese cycling in a temperate watershed.

Biogeochemistry 2014, 121, 505-517.

33. Herndon, E. M.; Brantley, S. L. Movement of manganese contamination through the

Critical Zone. Appl. Geochem. 2011, 26, S40-S43.

83 34. Herndon, E. M.; Jin, L.; Brantley, S. L. Soils reveal widespread manganese

enrichment from industrial inputs. Environ. Sci. Technol. 2011, 45, 241-247.

35. Langmuir, D. Aqueous Environmental Geochemistry, Prentice-Hall: Upper Saddle

River, NJ, 1997.

36. Thelin, G. P.; Gianessi, L. P. Method for estimating pesticide use for county areas of

the conterminous United States; USGS Open-File Report 00-250; U.S. Department

of the Interior: Sacramento, CA, 2000.

37. Hofstetter, T. B.; Schwarzenbach, R. P.; Haderlein, S. B. Reactivity of Fe(II) species

associated with clay minerals. Environ. Sci. Technol. 2003, 37, 519-528.

38. Nano, G. V.; Strathmann, T. J. Ferrous iron sorption by hydrous metal oxides. J.

Colloid Interface Sci. 2006, 297, 443-454.

39. Jeon, B.; Dempsey, B. A.; Burgos, W. D. Kinetics and mechanisms for reactions of

Fe(II) with iron(III) oxides. Environ. Sci. Technol. 2003, 37, 3309-3315.

40. Klein, A. R.; Silvester, E.; Hogan, C. F. Mediated electron transfer between FeII

adsorbed onto hydrous ferric oxide and a working electrode. Environ. Sci. Technol.

2014, 48, 10835-10842.

41. Soltermann, D.; Fernandes, M. M.; Baeyens, B.; Dahn, R.; Miehe-Brendle, J.; Wehrli,

B.; Bradbury, M. H. Fe(II) sorption on a synthetic montmorillonite. A combined

macroscopic and spectroscopic study. Environ. Sci. Technol. 2012, 47, 6978-6986.

42. Mohatt, J. L.; Hu, L.; Finneran, K. T.; Strathmann, T. J. Microbially mediated abiotic

transformation of the antimicrobial agent sulfamethoxazole under iron-reducing soil

conditions. Environ. Sci. Technol. 2011, 45, 4793-4801.

84 43. Klausen, J.; Trober, S. P.; Haderlein, S. B.; Schwarzenbach, R. P. Reduction of

substituted nitrobenzenes by Fe(II) in aqueous mineral suspensions. Environ. Sci.

Technol. 1995, 29, 2396-2404.

44. Pecher, K.; Haderlein, S. B.; Schwarzenbach, R. P. Reduction of polyhalogenated

methanes by surface-bound Fe(II) in aqueous suspensions of iron oxides. Environ.

Sci. Technol. 2002, 36, 1734-1741.

45. Kim, D.; Strathmann, T. J. Role of organically complexed iron(II) species in the

reductive transformation of RDX in anoxic environments. Environ. Sci. Technol.

2007, 41, 1257-1264.

46. Strathmann, T. J.; Stone, A. T. Reduction of the carbamate pesticides oxamyl and

methomyl by dissolved FeII and CuI. Environ. Sci. Technol. 2001, 35, 2461-2469.

47. Strathmann, T. J.; Stone, A. T. Reduction of oxamyl and related pesticides by FeII:

influence of organic ligands and natural organic matter. Environ. Sci. Technol. 2002,

36, 5172-5183.

48. Strathmann, T. J.; Stone, A. T. Mineral surface catalysis of reactions between FeII and

oxime carbamate pesticides. Geochim. Cosmochim. Acta 2003, 67, 2775-2791.

49. Yanina, S. A.; Rosso, K. M. Linked reactivity at mineral-water interfaces through

bulk crystal conduction. Science 2008, 320, 218-222.

50. Kohn, T.; Arnold, W. A.; Roberts, A. L. Reactivity of substituted benzotrichlorides

toward granular iron, Cr(II), and an iron(II) porphyrin: a correlation analysis.

Environ. Sci. Technol. 2006, 40, 4253-5260.

85 51. Nano, G. V.; Strathmann, T. J. Application of surface complexation modeling to the

reactivity of iron(II) with nitroaromatic and oxime carbamate contaminants in

aqueous TiO2 suspensions. J. Colloid Interface Sci. 2008, 321, 350-359.

52. Zwank, L.; Elsner, M.; Aeberhard, A.; Schwarzenbach, R. P. Carbon isotope

fractionation in the reductive dehalogenation of carbon tetrachloride at iron

(hydr)oxide and iron sulfide minerals. Environ. Sci. Technol. 2005, 39, 5634-5641.

53. Postma, D.; Appelo, C. A. J. Reduction of Mn-oxides by ferrous iron in a flow

system: column experiment and reactive transport modeling. Geochim. Cosmochim.

Acta 2000, 64, 1237-1247.

54. Villinski, J. E.; Saiers, J. E.; Conklin, M. H. The effects of reaction-product formation

on the reductive dissolution of MnO2 by Fe(II). Environ. Sci. Technol. 2003, 37,

5589-5596.

55. Villinski, J. E.; O'Day, P. A.; Corley, T. L.; Conklin, M. H. In situ spectroscopic and

solution analyses of the reductive dissolution of MnO2 by Fe(II). Environ. Sci.

Technol. 2001, 35, 1157-1163.

56. Stone, A. T. Reductive dissolution of manganese(III/IV) oxides by substituted

phenols. Environ. Sci. Technol. 1987, 21, 979-988.

57. Nowack, B.; Stone, A. T. Homogeneous and heterogeneous oxidation of

nitrilotrismethylenephosphonic acid (NTMP) in the presence of manganese(II, III)

and molecular oxygen. J. Phys. Chem. B 2002, 106, 6227-6233.

58. Kelsall, G. H.; House, C. I.; Gudyanga, F. P. Chemical and electrochemical equilibria

and kinetics in aqueous Cr(III)/Cr(II) chloride solutions. J. Electroanal. Chem. 1988,

244, 179-201.

86 59. Stookey, L. L. Ferrozine - a new spectrophotometric reagent for iron. Anal. Chem.

1970, 42, 779-781.

60. Buchholz, A.; Laskov, C.; Haderlein, S. B. Effects of zwitterionic buffers on sorption

of ferrous iron at goethite and its oxidation by CCl4. Environ. Sci. Technol. 2011, 45,

3355-3360.

61. Siegel, F. R. Environmental Geochemistry of Potentially Toxic Metals, Springer: New

York, 2002.

62. Postma, D. Concentration of Mn and separation from Fe in sediments - I. Kinetics

and stoichiometry of the reaction between birnessite and dissolved Fe(II) at 10oC.

Geochim. Cosmochim. Acta 1985, 49, 1023-1033.

63. Harris, D. C. Quantitative Chemical Analysis, 8th ed.; W.H. Freeman and Company:

New York, 2010.

64. Sivey, J. D.; Bickley, M. A.; Victor, D. A. Contributions of BrCl, Br2, BrOCl, Br2O,

and HOBr to regiospecific bromination rates of anisole and bromoanisoles in

aqueous solution. Environ. Sci. Technol. 2015, 49, 4937-4945.

CURRICULUM VITA

NAME: Allison N. Ricko

PROGRAM OF STUDY: Environmental Science

DEGREE AND DATE TO BE CONFERRED: Master of Science, August 2015

Secondary Education:

Patapsco High School and Center for Arts, Dundalk, MD 21222, 2008

Collegiate Institutions Attended:

Institution Dates Degree Date of Degree

Towson University 2009 – 2012 Bachelor of Science May 2012 Major: Geology

Towson University 2013 – 2015 Master of Science August 2015 Major: Environmental Science

Professional Publications:

Wunsch, A.; Navarre-Sitchler, A. K.; Moore, J.; Ricko, A.; McCray, J. E. Metal release from dolomites at high partial-pressures of CO2. Appl. Geochem. 2013, 38, 33-47.

Professional Positions Held:

Graduate Research and Teaching Assistant September 2013 – May 2015 Towson University, Towson, MD 21252

Laboratory Assistant June 2013 – August 2013 Towson University, Towson, MD 21252

Adjunct Faculty, Physical Geology Laboratory September 2012 – May 2013 Department of Physics, Astronomy, and Geoscience Towson University, Towson, MD 21252