INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand corner and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.

Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6" x 9" black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.

University Microfilms International A Beil & Howell Information Company 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 313/761-4700 800/521-0600

Order Number 0201633

The interaction of imidazolinone herbicides with selected adsorbents

Che, Ming-Daw, Ph.D.

The Ohio State University, 1991

UMI 300 N. ZeebRd. Ann Arbor, MI 48106

THE INTERACTION OF IMIDAZOLINONE HERBICIDES WITH SELECTED ADSORBENTS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

By

Ming-Daw Che, B.S., M.S. *****

The Ohio State University 1991

Dissertation Committee : Approved by M.M. Loux

T .J . Logan Co-Advi S.J. Traina

L.C. Brown T.J. Department of Agronomy ACKNOWLEDGEMENTS

I would like to express my appreciation to my adviser, Dr. Mark Loux, for his advice, patience, and financial assistance throughout this research. My deep appreciation is also extended to my co-adviser, Dr. Terry Logan, and to Dr. Sam Traina for their review and opinion of this dissertation, and for allowing me to use the instruments in their laboratories. The support and help of those who have worked with me in the laboratory is greatly appreciated.

I want to thank my wife, Shuling, for the sacrifices that she made over the past seven years in helping me through graduate school. Her constant support, encouragement, and understanding gave me the determination to see this work through to its completion. Finally, I thank my children, Harrison, Angela, and Jonathan, for just being there to brighten this father's days. VITA

August 16, 1957 ...... Born - Taipei, Taiwan Rep. of China 1979 ...... B.S., National Chung-Hsing University, Taichung, Taiwan 1979-1981 ...... Platoon Leader, Army in Taiwan 1981-1982 ...... Research Assistant, Taiwan Forestry Institute 1987 ...... M.S.(Environmental Chemistry) Department of Agronomy The Ohio State University Columbus, Ohio 1987-Present ...... Graduate Research Associate, Department of Agronomy The Ohio State University Columbus, Ohio

PUBLICATIONS Hsieh, Y.P., M.D. Che, and C.C. Liu. 1980. Effect of land disposal of sewage sludge on soil environmental qualities : I. Organic matter decomposition. National Science Council Monthly, vol. VIII, No. 10. p. 903-912. Che, M.D., T.J. Logan, S.J. Traina, and J.M. Bigham. 1988. Properties of water treatment lime sludges and their effectiveness as agricultural limestone substituties. J. Water Pollution Control Federation. 60: 674-680. Logan, T.J., B. Harrison, and M.D. Che. 1989. Agronomic effectiveness of cement Kiln dust-stabilized sludge. Ohio Edison Grant Report. Department of Agronomy. The Ohio State Univers ity. Columbus, Ohio. Che, M.D. and M.M. Loux. 1990. The adsorption and desorption of imazaquin and imazethapyr on soil, clay, and humic acid. Proc. North Cent. Weed Sci. Soc. 45 : 11.

FIELDS OF STUDY Major Field : Agronomy Studies in Environmental Chemistry, Soil Chemistry, and Weed Science. TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... ii

VITA ...... iii LIST OF TABLES ...... vii LIST OF FIGURES ...... ix INTRODUCTION ...... 1 CHAPTER PAGE

I. LITERATURE REVIEW ...... 4 Introduction ...... 4 Imidazolinone Herbicides ...... 5 Introduction ...... 5 Structure and Synthesis ...... 5 Physical and Chemical Properties .... 8 Mode of Action and Selectivity ..... 10 Toxicology ...... 13 Environmental Fate ...... 16 General AdsorptionConcepts ...... 16 D e f i n i t i o n ...... 16 Adsorption Isotherms ...... 19 Adsorption Constants ...... 21 Metal Cation Adsorption and Adsorption Edge ...... 25 Anion Adsorption, Negative Adsorption and dsorption Envelop ...... 27 Molecular Adsorption ...... 28 Methods for Sorption Study ...... 30 Adsorption Mechanisms for Pesticides in Soil ...... 31 Cation Exchange ...... 31 Protonation ...... 32 Anion Exchange ...... 33 Ligand Exchange ...... 34 Water Bridging ...... 35 Cation Bridging ...... 35

v Hydrogen Bonding ...... 36 Van Der Waals Interactions ...... 36 Hydrophobic Effect ...... 37 Factors Affecting the Sorption of Imidazolinone Herbicides in Soils ...... 38 Soil pH ...... 38 Soil C l a y ...... 39 Soil Organic Matter ...... 41 Soil Moisture ...... 42 Soil Temperature ...... 42 Nature of Herbicides ...... 43 Desorption ...... 45 Introduction ...... 45 Methods for Desorption Study ...... 46 Hysteresis ...... 47 References ...... 50 II. THE ADSORPTION AND DESORPTION OF IMAZAQUIN AND IMAZETHAPYR ON SOIL, CLAY, AND HUMIC ACID ...... 54 Introduction ...... 54 Materials and Methods ...... 56 Results and Discussion ...... 64 sorption ...... 64 D e s o r p t i o n ...... „ ...... 81 Conclusions ...... 95 References ...... 98 III. THE ADSORPTION AND INTERACTION BETWEEN IMIDAZOLINONE HERBICIDES AND SELECTED ADSORBENTS ...... 104 Introduction ...... 104 Materials and Methods ...... 106 Results and Discussion ...... 115 Sorption Constant Measurements ..... 115 Colloid-Solution Interface study .... 123 Conclusions ...... 134 References ...... 136

CONCLUSIONS ...... 138

APPENDICES ...... 140 A. Data Relative to Chapter II ...... 141 B. Data Relative to Chapter III ...... 149 REFERENCES ...... 162

vi LIST OF TABLES

TABLE PAGE 1. Some selected physical and chemical properties of four imidazolinone herbicides ...... 11

2. Metabolism of foliar-applied imazaquin in soybean, cocklebur and velvetleaf 3 days after application ...... 14 3. Acute toxicity of imidazolinone herbicides to mammals ...... 15 4. Selected properties of adsorbents ...... 57 5. Freundlich adsorption constants (Kd) , 1/n values, and isotherm coefficients of determination (R2) for the sorption of imazaquin and imazethapyr on various adsorbents ...... 69 6. The average sorption constants of imazaquin and imazethapyr on various adsorbents ...... 71 7. The effect of initial concentrations of herbicides on the sorption of herbicides on soils at three pH levels ...... 74

8. The experimental log K values of imazaquin and imazethapyr on soils and humic acids at various pH levels ...... 77 9. Desorption of imazaquin and imazethapyr from soils (pH = 4.5 or 4.6) ...... 82 10. The effect of soil moisture variation on herbicide desorption form soils (pH = 4.5 or 4.6) 84

11. The humic acid effect on the desorption of imidazolinone herbicides form soils ...... 86

Vll* • 12. Desorptions of imazaquin and imazethapyr from humic a c i d ...... 91 13. Desorption of imazaquin and imazethapyr from c l a y ...... 92 14. The humic acid effect on the desorption of imidazolinoneherbicides from clays ...... 94 15. The Kd and values for the interactions of three imidazolinone herbicides on humic acids ...... 119

16. Kk values for sorption of imidazolinone herbicides on two humic acids at three pH levels ...... 121

v m LIST OF FIGURES

FIGURES PAGE 1. The basic structure of imidazolinone herbicides ...... 6

2. The chemical structures of four imidazolinone herbicides used in this study ...... 7 3. The pathway of imazapyr synthesis ...... 9 4. Inhibition of branched-chain acid synthesis by imidazolinone herbicdies ...... 12

5. The three mechanisms of cation adsorption on a siloxane surface ...... 18

6. Classification of adsorption isotherms ...... 20 7. Adsorption isotherms for four selected pesticides ...... 22 8. An adsorption edge for Ca2* on an Oxisol ...... 26 9. Typical adsorption envelops on soils for fluoride, phosphate, and borate ions ...... 29

10. The sorption isotherm for imazaquin and imazethapyr on Crosby soil (pH =4.6) ...... 65 11. The sorption isotherm for imazaquin and imazethapyr on Hoytville soil (pH =4.5) ...... 66 12. The sorption isotherm for imazethapyr on Hoytville soil at three pH levels ...... 67 13. Effect of soil pH on herbicide adsorption (Initial solution concentration = 1 mg/L) ...... 75 14. The adsorption and desorption of imazaquin on Hoytville soil (pH = 4.5) ...... 88

ix 15. The adsorption and desorption of imazethapyr bn Hoytville soil (pH - 4.5) ...... 89 16. Assumed geometry of fluorescence measurement with parameters used to correct for the inner filter effect ...... 112 17. Adsrotpion of imazethapyr on Hoytville humic acid ...... 116 18. The temperature effect on the relative fluorescence intensity of imazaquin in the present of humic ac i d ...... 118 19. The fluorescence emission spectra of imazaquin solution at three pH levels ...... 124 20. The fluorescence emission spectra of imazaquin adsorbed onto Na-hectorite at three pH levels ... 125 21. The possible equilibria for the imazaquin in aqueous solution ...... 128

22. Titration - fluorescence emission spectra of imazaquin aqueous solution from pH 2.72 to 5.33 ...... 130 23. Titration - fluorescence emission spectra of imazaquin aqueous solution from pH 5.68 to 9.00 ...... 131 24. The fluorescence emission spectra of imazethapyr solution at three pH levels ...... 132 25. The fluroescence emission spectra of imazethapyr adsorbed onto Na-hectorite at three pH levels ...... 133

x INTRODUCTION

Herbicides dissipate in soil by : (1) sorption, exudation, and retention by crops and crop residues; (2) runoff in either dissolved or sorbed state; (3) sorption and desorption to organic matter, clays, and mineral surfaces; (4) vapor-phase diffusion; and (5) hydrodynamic transport, and by degradation, which includes ; (1) biological degradation, (2) chemical degradation, and (3) photochemical reactions (Weber and Miller, 1989). Sorption and desorption are major processes which can directly or indirectly influence the other dissipation processes. For instance, if the desorption rate of herbicide from soil increases, more herbicide molecules enter the soil solution, and the biological degradation, chemical degradation and volatilization rates may also increase. Therefore, sorption and desorption are the fundamental processes which should be studied to help explain herbicide behavior in soil.

Herbicide sorption in soil involves an interaction between the herbicides and sorption sites on soil components, and is regulated by the characteristics of the soil solution. There are many factors affecting the sorption equilibria of

1 2

herbicides in soil, but soil pH, soil moisture content, soil

organic matter content, soil clay characteristic and content, soil temperature, and nature of the herbicide are the six dominant ones. The imidazolinone herbicides have been developed over the past decade for crop or noncrop use. Imazaquin, imazethapyr, imazapyr, and imazamethabenz are four important imidazolinone herbicides which were studied in this research. Although imidazolinone herbicides are acidic and should be only slightly sorbed by soils, many reports have shown that crop injury could still be caused by their carryover (Curran and Knake, 1989; Gunsolus et al., 1986; Loux et al., 1989a; Renner et al., 1988a). Drought is the major factor that caused imazaquin persistence and subsequent rotational crop injury in some areas of the Corn Belt (Hackett, 1990). However, the mechanisms and role of soil moisture content in the sorption of imidazolinone herbicides in soil are still uncertain. There is also limited research on the desorption of imidazolinone herbicides from soil when soil moisture is altered. The specific objectives of this study were to :

1. Evaluate the effects of different sorbents and pH on the sorption constants (Kds) of imidazolinone herbicides, and the hysteresis in sorption-desorption isotherms. 2. Investigate the effect of soil moisture variations on herbicide sorption, and on the rate of herbicide release into the soil solution. 3. Study the sorption mechanisms and interactions between soil components, such as clays and humic acids, and imidazolinone herbicides at various pH levels. This paper is divided into three chapters. The first chapter reviews the fundamental concepts of sorption and desorption, and the basic information related to imidazolinone herbicides. The second chapter discusses the effects of pH, soil clay, soil humic acid, and soil moisture on the sorption and desorption of imidazolinone herbicides from various sorbents. The batch equilibrium method and liquid

scintillation counting technique for radiolabelled imazaquin and imazethapyr have been applied in this study. The third chapter illustrates the mechanisms and pH effect of the sorption of imidazolinone herbicides on clay using fluorescence emission spectra. The measurement of sorption constant by fluorescence quenching method is also discussed. CHAPTER I LITERATURE REVIEW

Introduction

Considerable research has been conducted on the sorption of various herbicides in soils, whereas there is limited research on the imidazolinone herbicides. The intent of this literature review then is to : 1) describe the general background information for the four imidazolinone herbicides which were used in this study, 2) illustrate the general sorption concepts, 3) present a thorough review regarding the sorption mechanisms for pesticides in soils, 4) indicate the major factors controlling the sorption of imidazolinone herbicides in soils, and 5) illustrate the desorption and hysteresis phenomena. citations are given that either were important in the elucidation of the sorption of herbicides in soils or that give results or conclusions relevant to this study.

4 5 Imidazolinone Herbicides

Introduct ion

Imidazolinone herbicides, which are derivatives of 0-(5- oxo-2-imidazolin-2-yl)aryl-carboxylates, were developed by

the American Cyanamid Company in 1983 (Los et al., 1983, Los et al., 1984). Imazaquin is a selective soybean herbicide which is applied preemergence or postemergence at low rates for the control of a broad spectrum of major broadleaf weeds (American Cyanamid Company, 1986). Imazethapyr was developed for worldwide use in soybeans and other leguminous crops, such as peas, Vicia faba and Phaseolus beans, seedling and established alfalfa (lucerne) and clover, peanut (groundnut), chickpeas, and lupines (Peoples et al., 1985). In both industrial and forestry situations, application of imazapyr to actively growing plants provides excellent control of many important brush and undesirable deciduous trees (American Cyanamid Company, 1988). Imazamethabenz is a selective postemergence herbicide for use in all varieties of fall- seeded and spring-seeded wheat and barley (American Cyanamid Company, 1989).

Structure and Synthesis

Figures 1 and 2 show the basic structure of imidazolinone herbicides, and the four imidazolinone herbicides used in this 6

X

Figure 1. The basic structure of imidazolinone herbicides. C2H, COOH . COOH

CH, CH,

HN HN

IMAZAQUINIMAZETHAPYR

COOHNH,CH(CH,):

CH

HN O

IMAZAPYR

COOCHj CH COOCH

with CH,

CH(CH,). HN O

p-isomer a-isomer

IMA2AMETHABENZ

The chemical structures of four imidazolinone herbicides used in this study. study, respectively. The highest herbicidal activity was achieved where R, was CH3 and R2 was CH(CH3)2 in Figure 1 (Los, 1983). The substituent X=H shows the highest phytotoxity, although other substituents still show good activity, with some degree of selectivity. The synthesis of the four imidazolinone herbicides is similar (Figure 3). The two isomeric products were separated by column chromatography and imazapyr was obtained by hydrolysis of the methyl ester, or by hydrogenolysis of the benzyl ester. Imazaquin was obtained by replacing the pyridine ring with quinoline and imazamethabenz was obtained by replacing the pyridine ring with a benzene ring.

Physical and Chemical Properties

All four herbicides are white to tan crystalline solids with no or slightly pungent odor (American Cyanamid Company, 1986, 1987, 1988, 1989). The melting points are between 115 to 224 °C. The order of water solubility is : imazapyr > imazamethabenz > imazethapyr > imazaquin. Solubility ranges between 0.06 and 1.5 g/100 ml at 25 °C. They are more soluble in organic solvents, such as methanol and dimethyl sulfoxide. The molecular weights are between 261.3 to 311.3 and the vapor pressures are low (< 1 x 10'7 mmHg at 60 °C) . The solution pH near each of their water solubilities is between 3.0 to 4.0. The octanol-water partition coefficients (Kw ) of imidazolinone V H;SO+

,COOMe B / N ✓ /* • T ; _

/ / r ' \ .A—A./ V •. y s I ii r \/ / S/ N COOMe ■ v y - - .

Figure 3. The pathway of imazapyr synthesis (Los et al., 1983). 10 herbicides is low (1.3 to 66). Table 1 summarizes the names, codes, molecular weights, melting points, water solubilities, and partition coefficients of the four imidazolinone herbicides.

Mode of Action and Selectivity Imidazolinone herbicides kill weeds by reducing the levels of three branched-chain aliphatic amino acids, isoleucine, leucine and valine, through the inhibition of acetohydroxyacid synthase (AHAS), an enzyme common to the biosynthetic pathway for these amino acids (Figure 4) (American Cyanamid Company, 1986). This inhibition causes a disruption in protein synthesis which, in turn, leads to an interference in DNA synthesis and cell growth. Mature tissue is not as susceptible to the imidazolinones, probably due to larger pools of amino acids as well as protein reserves (Shaner et al., 1984). All four herbicides are absorbed through the foliage and plant roots and translocated in the xylem and phloem throughout the plant. Selectivity describes a characteristic of a herbicide where some plants (weeds) are injured and other plants (crops) are not (Hackett, 1990). Selectivity is a function of the dosage of herbicide a plant is exposed to and the rate at which the herbicide is detoxified. For example, the selectivity of imazaquin and imazethapyr in soybeans is the 11 Table 1. Some selected physical and chemical properties of four imidazolinone herbicides.

Tradename, Melting Water 0ctanol-H20 Common name, molecular points solubilities Partition Code wt.(g/mole) (°C) (g/100g) Coefficient

Scepter Imazaquin 311.34 219-224 0.012 2.2 AC 252,214

Pursuit 11 - pH=5 Imazethapyr 289.34 169-173 0.14 31 - pH=7 AC 263,499 16 - pH=9

Arsenal Imazapyr 261.28 128-130 1.0-1.5 1.3 AC 252,925

Assert Imazamethabenz 288.35 115-145 0.3 p-isomer 35 AC 222,293 m-isomer 66 12

r > Leucine AHAS Pyruvate I - / - > ] ------*> - 5 s . ^ ------Valine

Acetohydroxyacid synthase (AHAS)

AHAS a-ketobutyrate > > > isoleucine

Figure 4. Inhibition of branched-chain acid synthesis by imidazolinone herbicides (American Cyanamid Co., 1986). 13 result of the rapid metabolism of the herbicide by the soybean plant (American Cyanamid Company, 1986, 1987). Susceptible weeds either cannot metabolize the herbicides or metabolize

them too slowly for detoxification. Relative plant tolerance can be measured by determining half-life of herbicide in plant tissue (Shauer and Robson, 1985). Half-life refers to the time required for one-half of the absorbed herbicide to be metabolized into inactive compounds. Table 2 shows the selectivity of imazaquin in soybeans, velvetleaf and cocklebur by comparing metabolism of imazaquin over time. Tolerant plants, such as soybeans, can rapidly metabolize imazaquin into inactive compounds (3 days), whereas susceptible plants, such as cocklebur, need longer time (30 days).

Toxicology

The toxicity of the active ingredient of imidazolinone herbicides to mammals and other nontarget species has been extensively investigated. All four imidazolinone herbicides used in this research have low mammalian toxicity partially because they act by inhibiting a biosynthetic process at a site present only in plants (American Cyanamid Company, 1986, 1987, 1988, 1989) . In addition, the herbicides are weak acids which are excreted rapidly by rats before they can accumulate in tissues or blood. Table 3 summarizes the acute toxicity of four imidazolinone herbicides to mammals. 14 Table 2. Metabolism of foliar-applied imazaquin in soybean, cocklebur and velvetleaf 3 days after application.

Half-life of Percent of CH as Species Imazaquin(days) Imazaquin Metabolite

Soybean 3 61.7

Velvetleaf 12 11.3 O H

Cocklebur 30 • 15

Table 3. Acute toxicity of imidazolinone herbicides to mammals.

Test Animal Imazaquin Imazethapyr Imazapyr Imazamethabenz

Oral LD50 Rats(Male & Female) > 5000 > 5000 > 5000 > 5000

Mise (Female) 2363 > 5000 > 2000 > 5000

Rabbits(Male & Female) ------4800 ----

Rabbits (Female) ---- > 5000 ------

Dermal LD50

Rabbits(Male & Female) > 2000 > 2000 > 2000 > 2000

s 16 Environmental Fate

Under actual field conditions, imazaquin, imazethapyr, and imazamethabenz may remain active in the soil for several weeks to several months after application depending upon environmental conditions, dosage, and method and timing of application (American Cyanamid company, 1986, 1987, 1989). On the other hand, the activity of imazapyr may persist from three months to 2 years (American Cyanamid Company, 1988). Conditions such as drought, cold weather, or limited cultivation between harvest and planting may prolong the residual activity of herbicides in the soil. There are many dissipation routes for imidazolinone herbicides in soil, but microbial degradation is the most important (Hackett, 1990). Dissipation through microbial activity would be expected to be greatest in warm, moist soils with pH levels near neutrality.

General Adsorption Concepts

Definition

Sorption is the process through which a net accumulation of a substance occurs at the common boundary of two contiguous phases (Everett, 1972). Precipitation can be defined as an accumulation of a substance to form a new bulk solid phase. 17 Both of these concepts imply a loss of material from an aqueous solution phase, but one of them is inherently two- dimensional and the other is inherently three-dimensional (Corey, 1981). When no independent data is available to distinguish between the above two conditions, the term "sorption” is used in order to avoid the implication that either adsorption or precipitation is occurring (Sposito, 1984) .

If a solid surface functional group reacts with an ion or a molecule dissolved in the soil solution to form a stable molecule unit, this formation reaction is termed "surface complexation" (Sposito, 1989). If no water molecule is interposed between the surface functional group and the ion or molecule it binds, the complex is an "inner-sphere surface complex”. If at least one water molecule is interposed between the surface functional group and the ion or molecule it binds, the complex is an "outer-sphere surface complex". If a solvated ion does not form a complex with a charged surface functional group, but instead neutralizes surface charge only in a delocalized sense, it is said to be adsorbed in the "diffuse-ion swarm". Figure 5 illstrates the three mechanisms of cation adsorption on a siloxane surface (Sposito, 1989). The diffuse-ion swarm and the outer-sphere surface complex mechanisms of adsorption involve almost exclusively electrostatic bonding, whereas inner-sphere 18

D IFFU SE ION

OUTER-SPHERE COMPLEX

INNER-SPHERE COMPLEX

Figure 5. The three mechanisms of cation adsorption on a siloxane surface (e.g. montmorillonite) (Sposito, 1989). 19 complex mechanisms are likely to involve ionic as well as covalent bonding. Inner-sphere surface complexation also is the molecular basis of "specific adsorption", whereas the diffuse-ion association and outer-sphere surface complexation is "nonspecific adsorption". The "nonspecificity" implied by this definition refers to the weak dependence on the electron configuration of the surface group and adsorbed ion to be expected for the interaction of solvated species.

Adsorption Isotherms

A graph of quantity of adsorbed species ( moles / kg of adsorbent) versus equilibrium solution concentration of the same species (moles/liter) at fixed temperature and applied pressure is called an adsorption isotherm (Sposito, 1989). A variety of isotherm shapes are possible, depending upon the affinity of the adsorbent for the adsorbate (adsorbed species). Giles et al. (1960) suggested that isotherm shape provides an indication of the adsorption mechanism operating for a given solute-solvent-adsorbent system. Figure 6 shows four basic types of adsorption isotherms. The S-type isotherm is characterized by an small initial slope that increases with the concentration of a substance in the solution. This property suggests that the relative affinity of the solid phase (adsorbent) for the substance (adsorbate) at low concentration is less than the affinity of the solution Figure 6. Classification of adsorption isotherms. (Revised (Revised isotherms. adsorption of Classification 6. Figure Am't. A dsorbed qiiru Slto Concentration C Solution Equilibrium rmGlse a. 1960) al., et Giles from 20 (Sposito, 1984), The L-type isotherm is characterized by an initial slope that does not increase with the concentration of a substance in the solution. This property is the result of a high relative affinity of the adsorbent for the substance at low concentrations coupled with a decreasing amount of adsorbing sites as the solution concentration increases. The C-type isotherm is characterized by an initial slope that remains independent of the concentration of the substance in the solution until the maximum possible adsorption. This kind of isotherm can be produced either by a constant partitioning between the interfacial region and an external solution or by a proportional increase in the amount of adsorbing surface as the solution concentration increases. The H-type isotherm is an extreme version of the L-type isotherm. Its characteristic large initial slope (in comparison with the L-curve isotherm) suggest a very high relative affinity of the adsorbent for an adsorbing substance. Figure 7 shows four types of adsorption isotherms for the sorption of some selected pesticides on soil clays (Green, 1974).

Adsorption Constants

For the purpose of comparing the sorption of numerous pesticides without showing the individual isotherms, one can obtain a distribution coefficient, Kd, for a given pesticide solution concentration. This is calculated as the ratio of Figure 7. Adsorption isotherms for four selected selected four for isotherms Adsorption 7. Figure PESTICIDE ADS.(/tmoles/kg) * * i * ' • etcds Gen 1974). (Green, pesticides OUIN OC »molti/l) o m SOLUTION CONC.(» 50 Paraquat i i t i i

100 0 1 20 10 200 Daaanit

\ 400 22 23 amount adsorbed to that in solution, i.e.,

pesticide adsorbed (umole/kg) Kd = (1.1) pesticide in solution (umole/liter)

(Green, 1974) . The higher the Kd, the more pesticide is adsorbed by adsorbent. The L-type isotherm is by far the most commonly encountered in the soil environment. The mathematical descriptions of this isotherm are conveniently given by the Freundlich and Langmuir equations (Stevenson, 1982). The Freundlich adsorption equation is :

x / m = KC1/n (1.2)

or

log (x/m) = log K + 1/n log C (1*3) where x/m is the quantity of solute adsorbed per unit weight of adsorbent, C is the equilibrium concentration of the adsorbing compound, and K and n are constants. In practical terms, a straight line is obtained when the data are plotted as log (x/m) vs log C. The intercept is equal to log K and the slope to 1/n. The constant K (i.e, K,,) provides an indication of the strength of adsorption of an adsorbate (a 24 pesticide) on an adsorbent. The Freundlich equation assumes that the quantity of solute adsorbed increases indefinitely with increasing concentration. This will occur when multilayers are formed and hetererogenous adsorption sites predominate (Stevenson, 1982). The Langmuir adsorption equation is :

x / m = KbC / (1 + KC) (1.4) or

C / (x/m) = 1/1^ + C/b (1.5) where x/m and C are the units defined above. K is a constant related to the bonding energy, and b is the adsorption maximum or total amount of solute capable of being adsorbed. In this case, a straight line is obtained when C/(x/m) is plotted against the equilibrium concentration, C.

The Langmuir equation assumes monolayer adsorption on a uniform surface with no interactions between adsorbed molecules (Stevenson, 1982).

Most herbicides exhibit L-type isotherms and the adsorption constants (Kds) for herbicides on various adsorbents can be calculated easily according to the above two equations using data collected from laboratory experiments. 25 Metal Cation Adsorption and Adsorption Edge Metal cations adsorb onto soil particle surfaces via the three mechanisms illustrated in Figure 5. The relative affinity of an adsorbent for a free metal cation will increase with the tendency of the cation to form an inner-sphere surface complex (Sposito, 1989). For a series of metal cations of a given valence, this tendency is correlated positively with the ionic radius. In respect to transition metal cations, the electron configuration plays a very important role and their relative affinities tend to follow the Irving-Williams order :

Cu2+ > Ni2* > Co2+ > Fe2+ > Mn2+.

The effect of pH on metal cation adsorption is principally the result of changes in the net proton charge on soil particles (Sposito, 1989). As pH increases, the electrostatic attraction of a soil adsorbent for a metal cation is enhanced since the net proton charge on the soil particles becomes more negative. A graph of metal cation adsorbed (q,,) versus pH will have a characteristic sigmoid shape known as an adsorption edge, shown in Figure 8 (Charlet, 1986). Figure 8. An adsorption edge for Ca2-” on an Oxisol. Oxisol. Ca2-” an for on edge adsorption An 8. Figure

qCo(mmol kg"1) Calt 1986) (Charlet, 26 27 Anion Adsorption. Negative Adsorption and Adsorption Envelop Anion sorption by soil minerals involves almost universally the two-step ligand exchange reaction :

S O H (.> + (1.6)

SOB,*,,, + L ' ^ , = SL’-',,, + HzO(1) (1.7) where S refers to the soil adsorbent, S O H (s) is one mole of inorganic surface hydroxyl groups, and L*' is an inorganic oxyanion of valence e (Sposito, 1984). If the Lewis acid site is present already, or if concentration of L is very large, the protonation in the first step is not required. If a dilute, neutral solution of KC1 is added to dry montmorillonite, the equilibrium Cl' concentration in the bulk soil solution will be greater than the Cl' concentration in the solution originally added to the clay (Bohn et al., 1979). This phenomenon is observed whenever an anion is added to a clay colloid having no adsorbing capacity for the anion at the prevailing pH. The process is called anion repulsion, or negative sorption.

The effect of pH on anion adsorption is the result of changes in the net proton charge on soil particles, if the adsorption anion does not protonate significantly (Sposito, 1989). The decrease in net proton charge with increasing pH 28 produces a repulsion of the adsorptive anion from soil particle surface. This kind of graph of anion adsorbed, vs pH, is termed an adsorption envelope, and is shown in Figure 9.

Molecular Adsorption The basic features of molecular adsorption models are perhaps best appreciated by a detailed consideration of two examples with very different foundational hypotheses : (1) The diffuse double-layer model — develops from the following assumptions (Sposito, 1989) : 1. The adsorbent surface is a uniform plane of charge density. 2. The adsorptive ions are point species that interact mutually and with the adsorbent through the coulomb force. Their only mechanism of adsorption is the diffuse-ion swarm. 3. The aqueous solution phase is a uniform continuum of dielectric constant D in which the point-ion adsorptive is immersed. (2) The constant capacitance model — develops from the following assumptions (Sposito, 1989) : 1. The adsorbent surface is a uniform plane of c h a r g e . 2. The adsorptive ions are point species that iue . yia asrto evlp o sis for soils on envelops adsorption Typical 9. Figure

MAXIMUM ADSORPTION 100 60 40 60 20 4 3 FLUORIDE fluoride, phosphate, and borate ions. (Sposito, (Sposito, ions. borate and phosphate, fluoride, 1989) PHOSPHATE 5 6 7 8 BORATE 29 30 interact with the adsorbent to form only inner- sphere surface complex.

3. The electric potential at the adsorbent surface is related linearly to the net total particle charge. The above two examples illustrate the general features of molecular adsorption models. They begin with "molecular hypotheses" about the mechanisms of sorption; they introduce

"constraint equations" that serve to relate model parameters to measurable properties, and they provide "testable predictions" of surface adsorption. The adherence of data to the model prediction does not prove that a given model is a correct molecular description of adsorption. The accuracy of a molecular model of adsorption can be ascertained only through experiments designed to verify directly the mechanisms of adsorption on which it is based (Sposito, 1989).

Methods for Sorption Study

The study of sorption in soils (or other sorbents) is characterized by three laboratory operations that define the net accumulation of a substance at the interface between the sorbent and a contiguous fluid ; (l) reaction of the sorbent with a fluid of prescribed composition for a prescribed period of time, (2) isolation of the sorbent from the reactant fluid phase, and (3) chemical analysis of the reactant fluid phase (Sposito, 1984) . Step 1 can take place either with the fluid phase at rest relative to the sorbent ("batch process") or with the fluid phase in uniform motion relative to the sorbent ("flow-through process"). The reaction time should be long enough to permit a close approach to thermodynamic equilibrium but short enough to prevent unwanted side reactions. Step 2 is usually carried out in batch processes through the application of centrifugal or gravitational force. The major chemical analyses in step 3 are ultraviolet-visible spectrophotometry, fluorescence spectrophotometry, polarography, liquid scintillation counting, and high performance liquid chromatography (HPLC). In addition, fourier transform infrared spectrometry (FTIR), X-ray diffraction, and nuclear magnetic resonance spectroscopy (NMR) are also used to provide more detailed information about sorption mechanisms and many surface-chemical phenomena. The liquid scintillation counting and fluorescence spectrophotometry methods were applied in this research for their simple and direct procedures.

Adsorption Mechanisms for Pesticides in Soil

Cation Exchange

Some cationic pesticides (such as diquat, paraquat, etc.) 32 may bind with negatively-charged soil colloids through a simple cation exchange reaction. The process involved can be illustrated by the following equation (Weed and Weber, 1974)

P+

where P+ = cationic pesticide; M+ = initially adsorbed cation; R = exchange site in soil. The adsorption of cationic pesticides in soil is principally through the interactions of amines, ring NH, or hetrocyclic N functional groups of pesticides, and carboxylic-OH and phenolic-OH groups of organic matter or/and inorganic hydroxyl groups of clay particles (Weed and Weber, 1974; Sposito, 1984). The large size of cationic pesticide molecules as compared to inorganic cations may have steric hindrance effects on their sorption in soil. Therefore, even though soil particles possess negative sites, these may not all be positionally available to large cationic pesticides.

Protonation

Less basic compounds, such as the s-triazines, may become cationic through protonation. The adsorption process can be illustrated by the following equations (Weed and Weber, 1974) 33

B (aq) + H\a q > = = = BB+(aq, (1.9) BH+<«,> + m r (8) = = = = M+{8q) + B H R {8) (1.10)

B (aq) + HR

Anion Exchange

For anionic pesticides, repulsion by the predominantly negatively charged soil surface may occur (negative sorption). Positive adsorption of anionic herbicides at low pH can also 34

be observed and illustrated by the following equation :

P ‘(aq> + ^ ( s , = = = A ‘(aq) + P R Cs> U - l * ) where P* = anionic pesticide; A' = anion; R = exchange site in soil; and AR and PR = soil exchanger saturated with A’ and P‘, respectively. This mechanism is not observed often, possibly because of the weakness of the surface complexes involved, but it should be prominent in acidic soils where the clay fraction is comprised primarily of metal oxides (Sposito, 1984).

Liaand Exchange

Ligand exchange refers specifically to inner-sphere complex formation between a carboxylate group in pesticide molecules and the incompletely chelated transition metals in organic matter or either Al(III) or Fe(III) in a soil mineral bearing inorganic hydroxyl groups (Weed and Weber, 1974; Sposito, 1984). This mechanism can be illustrated by the following equations ;

RMOH (s) + H +(aq) R M O H 2*(8) (1.13)

+ P't*,) — K M P (8> + H 20 (l) (1.14)

The chemical bonds involved are much stronger than that in 35 anion exchange. The inorganic surfaces involved are on metal oxides and the edges of phyllosilicates in soils.

Water Bridging This weak adsorption mechanism involves complexation with the proton in a water molecule solvating an exchangeable cation which can be illustrated by the following equation :

^'(aq) + (H20)nM^R === pP'd^O^R (1.15) where P, n, and m are constants. In equation 1.15, P = 0 or 1, n = 3, 4 or 6 usually, and m = 1 or 2. Water bridging is expected to occur particularly when M is a relatively hard Lewis acid (Sposito, 1984).

Cation Bridging

The mechanism of cation bridging can be illustrated by the following equation :

^'caq) + (H20)nfTR === (HjOJ^pP-lTR + H20 (1.16)

Unlike water bridging, the P**’ displaces a solvating water molecule during surface complex formation, since the M is a relative soft Lewis acid. For example, the pesticide molecule may adsorb on montmorilIonite through cation bridging when 36 monovalent exchangeable cations are present and through water bridging when bivalent exchangeable cations are present (Sposito, 1984).

Hvdroaen Bonding

The hydrogen bond is a special kind of dipole-dipole interaction in which the hydrogen atom serves as a bridge between two electronegative atoms, one being held by a covalent bond and the other by electrostatic forces (Green, 1974). The abundance of amino, carboxyl, and hydroxyl groups on organic matter suggests that hydrogen bonding will function as an important adsorption mechanism for pesticide molecules containing similar groups. The "water bridging" mechanism described above is also a hydrogen bond between the solvating water molecules surrounding the exchange cation and a polar pesticide molecule. Direct hydrogen bonding between the pesticide molecules and soil clay or organic matter is less important and may occur only under conditions of severe desiccation, and desorption would be expected when rewetting the soil.

Van Per Waals Interactions

Van der Waals interactions are produced by correlations between fluctuating induced multipole (principally dipole) moments in two nearby uncharged, nonpolar molecules (Sposito, 37 1984). This interaction between two molecules is very weak, but is additive, meaning that each atom of a molecule and of an adsorbent contributes to the total bond energy. Thus the contribution of these forces to adsorption increases with size of the molecule and also with its capacity to adapt to the adsorbent surface (Weed and Weber, 1974).

Hydrophobic Effect Some soil organic matter surfaces are hydrophobic and water molecules are not good competitors with nonpolar molecules for sorption at these sites. Such surfaces include the waxes, fats, and resins present in variable amounts, aliphatic side chains on humic and fulvic acids, and lignin- derived materials with high carbon content and few polar groups (Walker and Crawford, 1968). In addition, the adsorption of nonpolar pesticides results from a weak solute-solvent interaction. Therefore, hydrophobic sorption increases as the solute (pesticide) becomes more and more nonpolar, or as water solubility decreases (Stevenson, 1982). 38 Factors Affecting the Sorption of Imidazolinone Herbicides in Soil

Soil p H Within a group of similar soils differing mainly in their acidity or alkalinity, herbicide sorption is usually higher in the more acid soils (Calvet, 1980). Within normal soil pH range (4.5 - 7.5), slight increases in soil acidity may convert pesticides from anions to uncharged molecules or even to cations and thus dramatically increase their sorption by negatively-charged soil particle surfaces. The dissociation constant (Ka) of an acidic herbicide can be illustrated by the following equations :

HP === H+ + P' (1.17) Ka = [H+] [P ] / [HP] (1.18) pKa - pH + log [HP]/[P ] (1.19) where HP = acidic herbicide; H+ = hydrogen ion; and P‘ = anionic herbicide. The pKa value represents the pH at which half of the molecules present are in the anionic form and half are in the molecular form. When soil pH < pKa, the molecular form herbicide is predominant and more easily adsorbed by hydrogen bonding, van der Waals force, or hydrophobic interactions, etc. Since the pH at the surface of soil 39 particles may be as much as two pH units lower than that of the liquid environment, the molecular herbicide may even be protonated by soil surface acidity and become a cationic herbicide, which is more easily adsorbed by soil through cation exchange mechanisms. Many researchers have found a positive correlation between imidazolinone herbicide adsorption and soil reaction (Stougaard et al., 1985; 1990;

Goetz et al., 1986; Wehtje et al., 1987; Renner et al., 1988b; Loux et al., 1989b). These acidic herbicides have pKa values less than 4, resulting in a predominance of herbicide molecules in the anionic form at normal soil pH range, which should be less adsorbed by soil particle surfaces.

Soil Clay Clay is the most important fraction of soil mineral when considering adsorption because of its abundance and surface properties. In general, the higher the clay content in soil, the more herbicide is adsorbed.

Different types of clay possess different adsorptive properties. For example, the montmorillonitic-type clays have higher adsorption capacities due to their higher specific surface area, higher cation exchange capacities, and expanding lattices, compared to the kaolinitic-type clays with lower specific surface area, lower cation exchange capacities, and non-expanding lattices. 40 Sorption of herbicides on clay is also greatly influenced by the ionic composition of the clay surface (Calvet, 1980). Compensating cations can act in three ways : (a) by competition for adsorption sites with positively-charged organic molecules such as paraquat and atrazine; (b) by acting directly as adsorption sites through the formation of coordination bonds; (c) cations such as Al(III) and Fe(III) can form hydroxides on the clay surface which increase the adsorption capacity of the mineral. The relationship between soil clay and sorption of imidazolinone herbicides has been studied by several researchers in recent years. Dolling (1985) studied the interactions of three imidazolinone herbicides with homoionically-exchanged clays and concluded that sorption does not involve intercalation of clay by these compounds. The availability to the plant of the imidazolinones is unlikely to be significantly influenced by sorption by clays except acidic montmorillonites and illites. Goetz et al. (1986) studied the sorption of imazaquin on five Alabama soils ranging from sandy loam to clay and showed similar results; soil clay content had little effect on imazaquin sorption by the soil. The sorption of imazaquin was governed by the pH- dependent charge surfaces from aluminum and iron hydroxides and kaolinite. Hausler (1986) also found that aluminum and iron hydroxides can bind a considerable amount of the 41 imidazolinones, and that the iron hydroxides bind more herbicide than the aluminum hydroxides. These sorptive characteristics will probably be less common in temperate areas (such as corn belt in U.S.A.) due to the relative high concentration of organic compounds in the soil which complex with aluminum and iron and prevent the sorption of these herbicides on these hydroxides. On the other hand, other researchers observed increased herbicide adsorption with increasing clay content in soil (Basham et al., 1987; Loux et al., 1989b).

Soil Organic Matter Adsorption of herbicides varies greatly according to the nature of soil organic matter. Both humic substances and non­ humified organic materials may associate with herbicide by various functional groups so that it is difficult to establish general rules. The large surface area and the chemical nature of humic substances can explain their high sorptive properties. Physical sorption of herbicides on non-specific soil organic matter sites take place as a result of van der Waals forces and hydrogen bonds (Upchurch, 1966). Hausler (1986) found that the imidazolinones bind to the humic acid fraction of the soil, but the amount of binding depends on the pH of the organic matter. Basham et al. (1987) observed an increase in imazaquin sorption with organic matter 42 and clay content in soil, but could not separate out the relative importance of these two factors. Loux et al. (1989b) also found that the sorption of imazaquin was controlled by soil pH and organic matter content.

Soil Moisture The soil water content can influence herbicide sorption in two ways. It can increase or decrease the accessibility of herbicide to the soil particle surface, or can affect the physical-chemical properties of the sorbent in soil. Water molecules are themselves polar. Thus, when more water is added to a herbicide-soil complex, the water molecules begin to compete with the herbicide molecules for sorption sites on the soil colloids and force more of the herbicide into solution. As the soil dries, these herbicides molecules in soil solution are brought closer to sorptive surfaces on the soil and binding affinity increases, resulting in less herbicide in the soil solution (Hackett, 1990). Temporarily drying each soil to 25 or 50% of field capacity resulted in maximum sorption of imazaquin (Goetz et al., 1986).

Soil Temperature

As soil temperature is raised, sorption may increase, decrease or remain constant (Calvet, 1980). In general, as soil temperature increases, sorption of herbicide decreases 43 because : 1) adsorption is a exothermic reaction, and 2) increasing temperature can result in greater solubility of herbicide (McEwen and Stephenson, 1979). Increasing the soil temperature can increase the microbial degradation of herbicide and decrease the herbicide concentration in soil solution, thus indirectly influencing the sorption of herbicide on soil. There is little research concerning the effect of soil temperature on the sorption of imidazolinone herbicides in soil.

Nature of Herbicides

Two important properties of herbicides may affect their sorption on soil.

(a) Electronic structure Four structural factors which determine the chemical character of a herbicide molecule and thus influence its sorption on soil are (Bailey and White, 1970) : (1) Nature of functional group(s).

(2) Nature of substituting group(s), which may alter the

behavior of functional group(s). (3) Position of substituting groups with respect to the functional group which may enhance or hinder intramolecular bonding.

(4) Presence and magnitude of unsaturation in the 44 molecule, which affects the lyophylic-lyophobic balance. The presence of charges for herbicide molecules is an important characteristic as it permits sorption by ion exchange. This character is affected by soil pH, and the above four structural factors are in turn responsible for acidity or basicity (or pKa) of the herbicide. These factors are also resposible for electron distribution and electron mobility which can be expressed by polarity and by polarizability, respectively. These chemical properties are particularly relevant to adsorption of herbicide by charged clays.

(b) Water solubility

In general, herbicide sorption is inversely correlated with herbicide water solubility (Green, 1974). This inverse relationship is not always true and is more likely to be true within groups of chemically similar compounds. Soil pH, temperature, and electrolyte concentration can also indirectly affect the sorption of herbicides by altering their water solubility.

There are only a limited literature references comparing the sorptive capacities of different imidazolinone herbicides in soil. Dolling (1985) studied the sorption of three imidazolinone herbicides on pure clays and found 45 imazamethabenz had the highest sorption due to its basic character. He also found that sorption of imazaquin was higher than that of imazapyr, which could be explained by lower acidity and larger molecule size of imazaquin. Loux et al. (1989b) studied the sorption of imazaquin and imazethapyr in soils and found that sorption of the two herbicides was not similarly affected by changes in soil properties. The sorption of imazaquin on H/Al-montmorilIonite was greater than that of imazethapyr, whereas the results were reversed for herbicide sorption on soils and sediments using regression models, comparing for adsorption on pure clay.

Desorption

Introduction

Desorption is the reverse process of sorption. The sorption constant (Kd) is an experimental value which provides information on the degree of affinity between the herbicide and the soil, but may not predict the desorption of herbicide from the soil. The weakly acidic imidazolinone herbicides exist almost entirely as the dissociated anion in normal soil pH range, which are not strongly sorbed by the soils and are expected to be readily desorbed. There is little research concerning desorption of imidazolinone herbicides from soils 46 and other sorbents.

Approach Methods for Desorption Study The distribution between pesticide sorbed to the soil and that in solution at equilibrium for desorption is conventionally determined by the batch equilibrium method. Desorption data have been obtained by several techniques, but "decanting steps" and "repeated additions" are the two most common ones (Gamar and Mustafa, 1975? Grover, 1975; Bowman, 1979; Koskinen et al., 1979). In the decanting technique (or dilution step) , supernatant solution is decanted after sorption and separation (of solute and sorbent) steps, an equal volume of water (Gamar and Mustafa, 1975) or 0.02 M CaS04 is added (Koskinen et al., 1979), the system is equilibrated, and pesticide remaining in solution is determined. Repeating such "dilution steps" several times can generate a desorption isotherm. The repeated additions technique is conducted by adding some volume of pre-selected concentrations of pesticide solution after a maximum volume of centrifuged supernatant is removed in desorption steps. This method is specially useful for certain pesticides which are strongly sorbed by sorbents (Bowman, 1979). 47 Hysteresis In many cases, desorption is slower than sorption and appears to approach equilibrium with only a fraction of the sorbed species being removed. When the sorption isotherm is different from the desorption isotherm, it is known as hysteresis. Both organic (such as pesticide) and inorganic (such as H2P04‘) compounds may exhibit hysteresis when they are sorbed and desorbed from soil. They are six major reasons for hysteresis of organic or inorganic compounds in soil (Pignatello, 1989).

(1) Artifact due to the method. Some sorbed compounds can dissolve and form complexes with adsorbate in the solution, which increases the sorbate concentration in the aqueous phase. The supernatant is replaced or diluted by water in the desorption step, which can remove the artifact and eliminate its effect in equilibrium solution. This artifact effect on hysteresis is important for compounds which are strongly sorbed by soil particles.

(2) Failure to reach equilibrium.

The experimental solution concentration is usually greater than the true equilibrium concentration in the sorption steps, whereas the experimental solution concentration is usually 48 less than the true equilibrium concentration in the desorption steps. This is experimental error and can be corrected experimentally by increasing the equilibration time in both sorption and desorption steps.

(3) Chemical or biological transformation of the organic or inorganic compounds. During the sorption-desorption isotherm study, the compounds of interest can be degraded and transformed to other compounds which may have different sorption mechanisms than the orginal one. Biological transformation of the compounds can be inhibited by controlling the microorganisms present in the solution.

(4) Irreversible or resistant fractions of sorbed co mpound s.

Some compounds may be irreversibly sorbed by soil and sediment, and very difficult to desorb in desorption steps.

This causes a decrease of the compound concentration in the equilibrium solution during desorption and leads to hysteresis. The amounts of the irreversible fraction are dependent upon the properties of compounds and nature of the sorbe nts. 49 (5) Competitive sorption from natural (implicit) sorbates. Some natural (implicit) sorbates in the soil solution can compete for sorption sites with the compounds of interest. The sorbed natural sorbates can be desorbed in desorption steps, and some "new sites" become open and available for the compounds of interest. The equilibrium concentrations of the compounds in the supernatant will decrease, which causes hysteresis.

(6) Changes in the adsorbent such as aggregate breakdown by abrasion during the equilibrating or separating processes can also cause hysteresis. 50

References American Cyanamid Co. 1986. Technical Information Report on Imazaquin. Agric. Div., Wayne, NJ 07470. American Cyanamid Co. 1987. Technical Information Report on Imazethapyr. Agric. Res. Div., Princeton, NJ 08540.

American Cyanamid Co. 1988. Technical Information Report on Imazapyr. Agric. Res. Div., Princeton, NJ 08540. American Cyanamid Co. 1989. Technical Information Report on AC 222,293 Herbicide. Agric. Res. Div., Princeton, NJ 08540. Bailey, G.W., and J.L. White. 1970. Factors influencing the adsorption, desorption, and movement of pesticides in soil. Residue Rev. 32 : 29-92.

Basham, G.W., T.L. Lavy, L.R. Oliver, and H.D. Scott. 1987. Imazaquin persistence and mobility in three Arkansas soils. Weed Sci. 35 : 576-582. Bohn, H.L., B.L. McNeal, and G.A. O'Connor. 1979. Soil Chemistry. John Wiley & Sons, New York. 329 pp. Bowman, B.T. 1979. Method of repeated additions for generating pesticide adsorption-desorption isotherm data. Can. J. Soil Sci. 59 : 435-437.

Calvet, R. 1980. Adsorption-desorption phenomena,in R.J. Hance (ed.) In "Interactions between Herbicides and the Soil." p. 1-30. Academic Press Inc., Ner York.

Charlet, L. 1986. Adsorption of some macronutrient ions on an Oxisol. An application of the Triple Layer model, Ph.D. dissertation, Univ. of California, Riverside. Corey, R.B. 1981. Adsorption vs precipitation, in Adsorption of Inorganics at Solid-Liquid Interfaces (M.A. Anderson and A.J. Rubin, eds). Ann Arbor Science, Ann Arbor, Michigan. Dolling, A.M.1985. Studies of Interactions of Some Imidazolinone Herbicides with Clays. M.Sc. Thesis, Department of Chemistry, University of Birmingham, England. 198 pp. Everett, D.H. 1972. Manual of symbols and Terminology for Physicochemical Quantities and Units. Appendix II : Definitions, Terminology and Symbols in Colloid and Surface Chemistry. Butterworths, London. 51 Gamar, Y. and M.A. Mustafa. 1975. Adsorption and desorption of diquat2* and paraquat2* on arid-zone soils. Soil Sci. 119 : 290-295. Giles, C.H., T.H. MacEwan, S.N. Nakhwa, and D. Smith. 1960. Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurement of specific surface areas of solids. J. Chem. Soc. p. 3973-3993. Goetz, A.J., G. Wehtje, R.H. Walker, and B. Hajek. 1986. Soil solution and mobility characterization of imazaquin. Weed Sci. 34 : 788-793. Green, R.E. 1974. Pesticide-clay-water interactions, p. 3- 37 in W.D. Guenze (ed.) Pesticides in soil and water. Soil Sci. Sco. Am. Madison, WI. Grover, R. 1975. Adsorption and desorption of urea herbicides on soils. Can. J. Soil Sci. 55 : 127-135. Hackett, N. 1990. Imazaquin behavior in the soil. Technical report of American Cyanamid Company. Agri. Div., Wayne, NJ 07470. Hausler, M.J. 1986. Studies of Interactions of Some Imidazolinone Herbicides with Whole Soil, Oxyhydroxides, and with Natural and Synthetic Humic Acids. Ph.D. Thesis, Department of Chemistry, University of Birmingham, England. 228 pp.

Koskinen, W.C., G.A. O'Connor, and H.H. Cheng. 1979. Characterization of hysteresis in the desorption of 2,4,5-T from soils. Soil Sci. Soc. Am. J. 43 : 871-874. Los, M., D.R. Carliante, E.M. Ettinghouse, and P.J. Wepplo. 1983. 0-(5-oxo-2—imidazolin-2-yl)—arylcarboxylates : a new class of herbicides. American Cyanamid Co. Agric. Res. Div., Princeton, NJ 08540.

Los, M., P.J. Wepplo, R.K. Russell, B.L. Lences, and P.L. Orwick. 1984. 0-(5-oxo-2-imidazolin-2-yl)-arylcarboxylates : a new class of herbicides. 4. AC 263,499, a herbicide for use in legumes. Book of Abstracts, 188th Meet. Amer. Chem. Soc., Abstract 30 PEST.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989b. Adsorption of imazaquin and imazethapyr on soils, sediments, and selected adsorbents. Weed Sci. 37 : 712-718. 52 McEwen, F.L., and G.R. Stephenson. 1979. The Use and Significance of Pesticides in the Environment. John. Wiley & Sons, Inc., New York. Peoples, T.R., T. Wang, R.R. Fine, P.L. Orwick, S.E. Grahan, and K. Kirkland. 1985. AC 263,499 : a new broad-spectrum herbicide for use in soybeans and other legumes. Proc. Brit. Crop. Prot. Conf.-Weeds; 2 : 99-106. Pignatello, J.J. 1989. Dynamics of organic compounds in soils and sediments. In Reactions and Movements of Organic Chemicals in Soils. Sawhney & Brown (ed.) p. 45-80. SSSA Special Publication Number 22. Soil Sci. Soci. Amer., Inc., Madison, WI.

Renner, K.A., W.F. Meggitt, and D. Penner. 1988b. Effect of soil pH on imazaquin and imazethapyr adsorption to soil and phytotoxicity to corn (Zea mays). Weed Sci. 36 ; 78-83. Shaner, D.L., P.C. Anderson, and M.A. Stidham. 1984. Imidazolinones. Potent inhibitors of acetohydroxyacid synthase. Plant Physiol. 76 : 545. Shaner, D.L. and P. A. Robson. 1985. Adsorption, translocation, and metabolism of AC 252,214 in soybean (Glycine max), common cocklebur (Xantbium strumarium), and velvetleaf (AbutiIon theophrasti). Weed Sci. 33 : 469-471. Sposito, G. 1984. The Surface Chemistry of Soils, p. 113- 153. Oxford University Press, Inc. New York.

Sposito, G. 1989. The Chemistry of Soils. p. 127-169. Oxford University Press, Inc. New York. Stevenson, F.J. 1982. Humus Chemistry : Genesis, Composition, Reactions. John Wiley and Sons, New York. 443 pp. Stougaard, R.N., P.J. Shea, and A.R. Martin. 1985. The effect of soil type and pH on AC-263,499 and imazaquin. NCWSS Proc. 40 ; 15-16. Stougaard, R.N., P.J. Shea, and A.R. Martin. 1990. Effect of soil type and pH on adsorption, mobility, and efficacy of imazaquin and imazethapyr. Weed Sci. 38 ; 67-73. Upchurch, R.P. 1966. Behavior of herbicides in soil. Res. Rev. 16 : 46-85. 53 Walker, A., and D.V. Crawford. 1968. The role of organic matter in adsorption of the triazine herbicides by soils, p. 91-108. In Isotopes and radiation in soil organic matter studies. Proc. 2nd Symp. Int. Atomic Energy Agency, Vienna. Weed, S.B. and J.B. Weber. 1974. Pesticide-organic matter interactions. p. 39-66 in W.D. Guenzi, ed. Pesticides in Soil and Water, soil Sci. Soc. Am. Madison, WI.

Wehtje, G., R. Dickens, J.w. Wilcut, and B.F. Hajek. 1987. Sorption and mobility of sulfometuron and imazapyr in five Alabama soils. Weed Sci. 35 : 858-864. CHAPTER II THE SORPTION AND DESORPTION OF IMAZAQUIN AND IMAZETHAPYR ON SOIL, CLAY, AND HUMIC ACID

Introduction

Imazaquin and imazethapyr are two imidazolinone herbicides which have been developed for worldwide use in soybeans and other leguminous crops. Both herbicides are amphoteric compounds due to the presence of acidic and basic functional groups. Ionizable carboxyl groups for imazaquin and imazethapyr have pKa values of 3.8 and 3.9, respectively (Renner et al., 1988b). Therefore, both herbicides should be predominately in the anionic form at normal soil pH. Since the anionic molecule will be repulsed from negatively charged clay and organic surfaces, a very low or negative sorption of both herbicides by typical agricultural soils should be observed (Green, 1974; Need and Weber, 1974). Because of the weak sorption and relatively short environmental half-life (between a few weeks to a few months) for both herbicides, the potential for carryover effect is unlikely. However, some reports have shown that crop injury could be caused by

54 55 imazaquin and imazethapyr carryover (Curran and Knake, 1989; Gunsolus et al., 1986; Loux et al., 1989a; Renner et al., 1988a) . For example, corn and oats were significantly injured in 1986 by 0.14 and 0.28 kg/ha rates of imazethapyr and by 0.17 kg/ha rate of imazaquin applied in 1985 (Gunsolus et al., 1986). Protonation of imidazolinone N atoms may occur at lower pH (such as at the edge of a dry soil particle) , resulting in a positively charged molecule with a potential for cationic binding to soil. The sorption of herbicides can also be due to physical forces or hydrogen bonding (Loux et al., 1989b). Various soil constituents and properties have been reported to influence the magnitude of imazaquin and imazethapyr sorption (Basham et al., 1987; Goetz et al., 1986; Loux et al., 1989b; Renner et al., 1988b). However, information for the interactions between these herbicides and types of soil organic matter and clay at various pH levels is still limited. These is also limited research on the desorption of imazaquin and imazethapyr from soil when soil moisture content is altered. The specific objectives of this study were to ; 1) evaluate the effect of different sorbents and pH on the sorption constants (Kds) of imazaquin and imazethapyr, and the hysteresis in sorption-desorption isotherms, and 2) investigate the effect of soil moisture variations on the desorption of imazaquin and imazethapyr from soils. 56 Materials and Methods

Herbicides

Radiolabelled imazaquin and imazethapyr were obtained from American Cyanamid Company. The specific activities of uniformly ring-labelled 14C-imazaquin and 14C-carboxyl-labelled imazethapyr were 38 uCi/mg and 21.2 uCi/mg, respectively. Appropriate volumes of 14C- and technical grade herbicide stock solutions were diluted with distilled water to give final solution concentrations of 0.02, 0.1, 1.0, 5.0, and 10.0 mg/L (0.06, 0.32, 3.21, 16.1 and 32.1 uM) for imazaquin, and 0.04, 0.1, 1.0, 5.0, and 10.0 mg/L (0.15, 0.35, 3.46, 17.3 and 34.6 uM) for imazethapyr. Each solution contained approximately 2000 dpm of 14C-herbicide per mL.

Adsorbents

Two soils, three clays, and two humic acids were used as the sorbents in this research (characteristics of these sorbents are listed in Table 4). soils. Samples of Hoytville clay and Crosby silt loam were collected from Northwest and Western Branches of OARDC, respectively. The 0-15 cm depth of soil samples were collected from areas with pH levels of 4.5, 5.5, and 6.6 for Hoytville clay, and 4.6, 5.8, 6.8 for Crosby silt loam. 57 Table 4. Selected properties of adsorbents.

Particle size Distribution Cation Exchange Organic C (% < 2 mm) CaDacitv Adsorbents C o n t e n t (%) sand silt clay (cmolc/kg)

Crosby Silt Loam 0.80 18.0 60.7 21.3a 19

Hoytville Clay 1.84 17.7 42.0 40.3a 34 a 1 1 1 1 1 1 1 1 1 1 1 1 1 Kaolinite 0.16 H H O

Hectorite 0.10 ------89= u CM 0 0 1 1 1 1 1 1 1 1 1 1 1 Illite 0.17 t 1

Aldrich 40.88 ------Humic Acid Extracted 24.79 Humic Acid

8 Illite is the predoninant clay mineral in Crosby and Hoytville soils. b Data from Bohn et al. (1979). c Data from Jaynes and Bigham (1987). d Data from Fanning et al. (1989). 58 The soil pH levels have been maintained in the field for many years. The soil samples were air-dried, sieved through a 2 mm sieve, and stored in sealed plastic bags in the laboratory at room temperature.

Clays. Three clay samples, kaolinite (Hacon, Georgia), and illite (#35, Fithian), which were obtained from the Ward's Natural Science Establishment, Inc. at Rochester, New York, and hectorite (SHCa-1, California), which were obtained from the Source Clays Repository at the University of Missouri, were used as sorbents in this study. The hectorite was purified by three successive extractions with 1 M NaCl, then three times with distilled water. The final concentration of Na-hectorite stock solution was 0.02 g/ml. The kaolinite and illite were used without further purification.

Humic Acids. Two humic acids were used in this research. One humic acid was extracted and isolated from Hoytville soil by treatment with dilute base, acidification to pH = 1 with HC1, and purification by five successive extractions with 2% HF / 0.5% HC1 solution (Schnitzer, 1982). The second humic acid was obtained from Aldrich Co. and used without further purification. Concentrated humic acid solutions were prepared by dissolving weighed amounts in water with aid of ultrasonic agitation and temperary elevation of pH by NaOH. A working 59 pH was then obtained by acidification with HC1 when necessary. pH. Three pH levels of each soil were used in soil sorption systems. The pH levels of 3.0, 5.0, and 8.0 were maintained in clay and humic acid sorption systems, using aliquots of 1.0 M HC1 or NaOH to adjust pH.

Instrumentation

A Dohrmann DC-80 Carbon Analyzer with Horiba PIR-2000 infrared detector was used to determine the organic carbon content of each adsorbent by the dry combustion method. The organic carbon is burned in an atmosphere of pure oxygen, and the co2 produced is detected by an infrared detector. Sample pH was determined on a 1:1 sample-water suspenson with an Orion Research 701A digital ionalyzer. Particle size distributions of the two soils were determined by the hydrometer method (Day, 1965) . Cation exchange capacity (CEC) was determined by the ammonium acetate method at pH 7.0 (Chapman, 1965) . The predominant clay minerals in the two soils were determined with X-ray diffraction (Theisin and Harward, 1962) . The 14c-labelled herbicides were detected with a Beckman LS-100 liquid scintillation counter.

Sorption Study

The sorption isotherms were determined in triplicate on a 3:10 adsorbent/solution ratio for soil samples (3.0 g of adsorbent with 10.0 mL of water) or on a 1:25 adsorbent / solution ratio for clay and humic acid samples (0.4 g of adsorbent with 10.0 mL of water). Samples were equilibrated in 50-mL Teflon centrifuge tubes with Teflon-covered lids by shaking at 25 °c on a reciprocal shaker for 20 hours at 100 strokes/min. Preliminary studies had shown equilibration to be complete within 20 hours, with no degradation of herbicides occurring over this period. Sorbent and solution phases were separated by centrifugation at 10000 rpm for 15 minutes. Final concentration of herbicide in the solution phase for each sample was determined by taking 1 ml of solution phase, mixing with 15 mL of scintillation cocktail (Fisher Scientific ScintiVerse E solution), and measuring the dpm of 14C-herbicide by liquid scintillation counting for 10 minutes. The amount of herbicide sorbed by the sorbent was calculated as the difference between initial and final solution concentrations.

Data Analysis

The sorption data were fitted to the Freundlich equation

a

C8 = K X CH1/n (2.1) where C8 is the amount sorbed in nmol/g of adsorbent, and CM is the equilibrium solution concentration in nmol/mL. The constant K is a measure of the degree or strength of 61 adsorption, and the constant 1/n indicates the degree of nonlinearity between solution concentration and adsorption (Hamaker and Thompson, 1972). Freundlich K values and 1/n values were calculated from the logarithmic form of the equation 2.1 :

log Cs = log K + 1/n log Cw (2.2)

where log K is equal to the Y-intercept when 1/n log CH equals zero. The K values (also known as Freundlich sorption constants, Kds) were subjected to analysis of variance and means were separated by Fisher's Protected LSD Test at the 5% level of probability. The values for soils and humic acids were obtained from dividing the Freundlich Kd constant by percent C in the soils or humic acids (Stevenson, 1982). i.e.

Kd Koc = (2.3) % organic C in soil or humic acid

These values were compared with the theoretical values

calculated from the octanol/water partition coefficients (Kw s) of these herbicides and equation 2.4, 2.5 and 2.6 as below :

Koc - °-63 K ^ (2.4) 62 from Karickhoff et al. (1979),

log K". = 0.088 + 0.909 (log K^) (2.5)

from Hasset et al. (1983), and

log = 0.904 (log K^) - 0.779 (2.6)

from Chiou et al. (1983). The constant, K^, in equation 2.6 is the solid-phase partition coefficient normalized to the soil organic matter content. This is converted to by multiplying by 1.72 (Chiou, 1989).

Desorption Study

After sorption measurement, an additional 6.0 mL of liquid phase were removed from tubes, 7.0 mL of distilled water

added, and shaken for 24 hours. Following this equilibration period, final solution concentrations of herbicides were determined as described in the sorption study. The desorption procedure was repeated three times for soils and four times for clays and humic acids. The pH was maintained at 3.0, 5.0, and 8.0 in clay and humic acid systems throughout the entire desorption process.

In another set of samples, the effect of dissolved humic acid on the desorption of imazaquin and imazethapyr from soil 63 and clay systems was evaluated by adding a 50 mg/L humic acid solution instead of distilled water during each desorption cycle. The effect of soil moisture content on herbicide sorption was evaluated for both soils using 0.1 mg/L (0.32 and 0.35 uM for imazaquin and imazethapyr, respectively) herbicide concentration. Soils were placed under a forced-air hood and dried to a moisture content of 5% after completing the adsorption measurement. Soil water content was determined by weight. Samples were then maintained at this dry condition for 48 hours at 30 °C with Teflon-covered lids to prevent further evaporation. The soil was then rewetted using water to the 3:10 adsorbent/solution ratio and the desorption procedues was repeated as described previously. A control (pure soil without adding herbicide solution) was also maintained at 5% moisture content for 48 hours in each soil treatment. All treatments were made in triplicate and desorption data were subjected to analysis of variance and means separated by Fisher's Protected LSD Test at the 5% level of probability. 64 Results and Discussion

Sorption

Figures 10 and 11 shows that the sorption of imazaquin and imazethapyr on soils performed L-type isotherm at lower pH levels, exception of imazaquin on Hoytville soil. The sorption of imazaquin or Hoytville soil performed C-type isotherm, suggesting that the partitioning process may involved. The higher organic C content for Hoytville soil (1.84%) comparing to Crosby soil (0.80%), and the higher affinity of imazethapyr to soils than imazaquin may be the reasons why the partitioning may be the dominant process for the sorption os imazaquin on Hoytville soil. In general, imazaquin more likely performed C-type isotherm, comparing the L-type isotherm for imazethapyr in soil and clay systems at lower pH levels (Figures A-l to A-3 in Appendix A) . On the other hand, both herbicides perform C-type isotherms in two humic acid systems (Figures A-4 and A-5).

The pH effect on the shape of adsorption isotherm for imazethapyr on Hoytville soil was illustrated by Figure 12. The sorption of imazethapyr on Hoytville soil performed L-type isotherm for its relatively higher affinity to soil componant at lower pH levels. When soil pH increased to 6.8, almost all the imazethapyr molecules present as the anion species and repulsed for soil componant, the sorption of herbicide on soil Figure 10. The sorption isotherm for imazaquin and imazaquin for isotherm sorption The 10. Figure

(nm ol/g) 20 25 mztay nCob ol (pH4.6). = soil Crosby on imazethapyr 5 0 5 0 5 0 35 30 25 20 15 10 5 0 A o Imazaquin O A Imazethapyr Ceq(nmol/mL) A pH 4.6pH rsy Soil Crosby iue 1 Te opin stem o iaaun and imazaquin for isotherm sorption The 11. Figure

(nm ol/g) 20 30 40 50 imazethapyr on Hoytville soil (pH = 4.5). (pH= soil Hoytville on imazethapyr Imazaquin O I azethapyr Im A 5 10 e(mo/ ) L ol/m Ceq(nm 15 20 H 4.5 pH otil Soil_ Hoytville 530 25 Figure 12. The sorption isotherm for imazethapyr on Hoytville imazethapyr for isotherm sorption The 12. Figure

•o (nm ol/g) 20 30 40 50 ol ttrep levels. pH three at soil 5 0 • pH 6.6 pH • O pH 4.5 pH O A pH 5.5 pH A 10 e(mo/ ) L ol/m Ceq(nm 15 20 otil Soil Hoytville mazethapyr Im 25 30

35 67 6 8 performed S-type isotherm for the weak affinity of imazethapyr molecules to soil particles. The Freuhdlich sorption constants (Kd), nonlinearity constant (1/n), and isotherm coefficients of determination (R2) for the adsorption of imazaquin and imazethapyr on all sorbents at different pH levels are shown in Table 5. The coefficient of determination (R2) for equations decribing linear isotherms was 0.99 (P < 0.55) for all adsorbents except two samples at the highest pH level (R2 = 0.95 or 0.93). The low R2 values for two samples (imazethapyr adsorbed on Crosby soil and kaolinite at pH 6.8 and 8.0, respectively) should be attributed to the very low sorption in both cases (Kd value of 0.29 and 0.10). The 1/n values closed to 1.0 for imazaquin on soil and clay, indicating again that the partitioning process may involved. On the other hand, the 1/n values for imazethapyr on soil and clay (except for hectorite) were relatively less than one, indicating that the adsorption process may involved.

The order of Kd values, from highest to lowest, for imazaquin adsorption on the adsorbents was : Aldrich humic acid » extracted humic acid » hectorite > kaolinite > illite > Hoytville clay > Crosby silt loam. For imazethapyr, the order was : Aldrich humic acid » extracted humic acid » illite > kaolinite > hectorite > Hoytville clay > Crosby silt loam (Table 5). In general, the order applied to both Table 5. Freundlich adsorption constants (Kd), 1/n values, and isotherm coefficients of determination (R2) for the adsorption of imazaquin and imazethapyr on various adsorbents.

Imazaquin Imazethapyr

pH Kd 1/n R 2 Kd 1/n R 2

Hoytville 4.5 0.85 0.92 0.99 3.47 0.83 0.99 5.5 0.19 0.94 0.99 0.98 0.90 0.99 6.6 0.11 0.89 0.99 0.29 0.82 0.99

Crosby 4.6 0.33 0.85 0.99 1.69 0.77 0.99 5.8 0 0.26 0.74 0.99 6.8 0 0.10 0.82 0.95 Kaolinite 3.0 6.13 1.31 0.99 26.82 0.70 0.99 5.0 0 3.46 0.75 0.99 8.0 0 0.44 0.74 0.93 Hectorite 3.0 7.11 1.00 0.99 14.59 0.97 0.99 5.0 6.27 0.98 0.99 7.85 0.90 0.99 8.0 5.42 1.00 0.99 6.78 0.95 0.99 Illite 3.0 3.01 0.97 0.99 27.33 0.72 0.99 5.0 1.68 1.00 0.99 17.66 0.68 0.99 8.0 1.64 0.92 0.99 13.68 0.67 0.99 Aldrich Humic Acid 3.0 414.95 0.97 0.99 101.86 1.01 0.99 Extracted Humic Acid 3.0 42.76 0.95 0.99 63.24 0.92 0.99 70 herbicides (especially at low pH) was : humic acid > clay > soil. The average Kd values for both herbicides at pH = 3 were 155.7, 14.16, and 1.58 for humic acid, clay, and soil, respectively (Table 6) . Organic matter had much greater affinity for both herbicides than clay did, indicating that soil organic matter content could be the major factor controlling the sorption of imazaquin and imazethapyr in soils. The role of organic matter content will be more important when soil contain medium to high organic matter content, or when the clay particle is surrounded or mixed with soil organic matter. Dolling studied the adsorption of three imidazolinone herbicides on four clays and concluded that the availability to plants of imidazolinones was unlikely to be affected by clay adsorption because of their weak affinity to clays. Loux et al. (1989b) performed a multivariate regression analysis for the effects of soil properties on the sorption of imazaquin and imazethapyr on 28 soil and sediment samples. The results indicated that imazaquin sorption was determined by soil pH, organic matter, and clay content, whereas imazethapyr sorption was controlled only by soil pH and clay content. They explained that imazaquin may be preferentially sorbed on organic matter in soil and sediment, while imazethapyr is preferentially sorbed on clay. Our results showed that the sorption constant of imazethapyr on clay 71 Table 6. The average sorption constants of imazaquin and imazethapyr on various adsorbents (pH = 3.0).

Imazaquin Imazethapyr Ave.

Soil 0.59 2.58 1.58

Clay 5.42 22.91 14.16

Humic Acid 228.86 82.55 155.70 72 (22.91) was much higher than that of imazaquin (5.42), whereas the sorption constant of imazaquin on humic acid (228.86) was much higher than that of imazethapyr (82.55) (Table 6). Sorption of each herbicide on its preferred soil component may not totally preclude sorption on another component, especially when the preferred component (such as organic matter) is present in negligible amounts (Loux et al., 1989b). On the other hand, clay can be surrounded and coated by

organic matter, and its surface may act as an organic surface in the soil, which complicates the explanation for adsorption phenomena under field conditions.

Soils. Sorption of imidazolinone herbicides on soil was influenced by nature of herbicide, as Kd values ranged from 0.1 to 3.47, with zero adsorption on some soil samples at higher pH levels (Table 5) . Mean Kd values of imazaquin and imazethapyr on both soils, averaged over all soil pH levels, were 0.25 and 1.13, respectively. The Kd value indicated that imazethapyr had much stronger affinity for soil than imazaquin did, especially at lower soil pH ranges (Table 6) . The differences between Kd values for two herbicides decreased as soil pH increased. Previous research showed that imazethapyr had greater sorption (Renner et al., 1988; Loux et al., 1989b) and was less mobile (Stougaard et al., 1985) than imazaquin in all soil-pH combinations. The higher sorption of 73 imazethapyr in soil may result from the higher affinity of imazethapyr for clay, especially at low pH. The major clay mineral in both soils in this study is illite, which had stronger affinity (Kd = 27.33) for imazethapyr than for imazaquin (Kd = 3.01). The percent of both herbicides sorbed by soil decreased as initial herbicide concentration increased (Table 7). For example, the percentages of imazethapyr sorbed by soils decreased from 43.19 to 16.88% when initial herbicide concentration increased from 0.04 mg/L to 10.0 mg/L at pH 4.5, indicating that the sorptive capacity of each soil was approaching saturation. Both herbicides were more highly sorbed by Hoytville soil than by Crosby soil, at similar soil pH levels. This can be attributed to the higher organic matter and clay content of Hoytville soil, compared to that of the Crosby soil.

The effect of soil pH on the adsorption of both herbicides on soils is described by the results shown in Figure 13 and

Table 5. The Freundlich adsorption constant (Kd) for each soil/herbicide combination decreased as soil pH increased, but the decrease in sorption was most evident as the increase of soil pH was between 4.5 and 5.5. Renner et al. (1988b) and Loux et al. (1989b) made similar observations for imazaquin and imazethapyr adsorption on soils. Loux et al. (1989b) explained that, since pKa values both herbicides are below 4, changes in soil pH below Table 7. The effect of initial concentrations of herbicides on the sorption of herbicides on soils at three pH levels.

______Adsorption (% adsorbed) Imazaquin Concentration .. Pfl Soil fmcr/Ll 4.5 5.5 6.5 Crosby 0.02 13.45 0 0 0.1 9.48 0 0 1.0 8.15 0 0 5.0 7.42 0 0 10.0 4.90 0 0

PH 4.6 5.8 6.8

Hoytville 0.02 23.72 7.57 3.50 0.1 22.76 5.06 5.52 1.0 18.55 4.05 1.79 5.0 17.38 5.35 3.11 10.0 16.13 4.72 2.19 Imazethapyr

Concentrat ion — pH ... fmq/L). 4.5 5.5 6.5 Crosby 0.04 43.19 9.38 2.30 0.1 42.94 10.28 5.75 1.0 36.24 8.30 4.74 5.0 21.25 3.83 0 10.0 16.88 2.14 0.10

. PH..,, . 4.6 5.8 6.8 Hoytville 0.04 61.20 26.04 9.43 0.1 59.56 26.48 10.76 1.0 53.27 22.03 8.57 5.0 41.91 15.80 3.13 10.0 36.46 18.62 5.45 iue 3 Efc o si p o hriie adsorption. herbicide on pH soil of Effect 13. Figure

PERCENT ADSORPTION -10 20 0 3 0 4 0 5 0 6 70 Iiil ouincnetain= mg/). /L) g m 1 = concentration solution (Initial 4 5 pH p L I O S MZTAY O HOYTVILLE ON IMAZETHAPYR MZQI O HOYTVILLE ON IMAZAQUIN MZQI O CROSBY ON IMAZAQUIN MZTAY O CROSBY ON IMAZETHAPYR 6 7 6 should be expected to have a greater effect on sorption than changes above this pH. Increasing soil pH from 3.8 to 5.8 results in the percent of imazaquin with ionized carboxyl groups increasing from 50 to 99%, leaving very little herbicide to be ionized by further increases in pH. Loux et al. (1989b) believed that their results may have indicated some cationic binding between soils and positively charged herbicide molecules which were protonated at low pH. The pH at the surface of soil particles may be as much as two pH units lower than that of the soil solution. In this case, imidazolinone herbicide molecules may still be protonated when the soil pH is 6. Figure 13 also showed that, except for sorption on Hoytville soil, less than 10% of imazethapyr was sorbed when soil pH was higher than 6.0. This result showed that most imidazolinone herbicide molecules were in anionic form when soil pH was above 6 and were repulsed from the negatively charged soil particle surfaces. Stougaard et al. (1985) reported that increased mobility and crop injury at high pH indirectly indicated decreased herbicide sorption and an increase of herbicide in the soil solution. Table 8 shows the experimental values of both herbicides in soil and humic acid systems at various pH levels. The values are 2.00, 2.09, and 2.08 for imazaquin, calculated from log K0H and equation 2.4, 2.5, and 2.6, 77 Table 8. The experimental log values of imazaquin and imazethapyr on soils and humic acids at various pH levels.

Imazaquin Imazethapyr Sorbent pH logK^. ------

Hoytville 4.5 1.66 2.28 Soil 5.5 1.01 1.73 6.6 0.78 1.20

Crosby 4.6 1.62 2.32 Soil 5.8 1.51

6.8 1.10

Aldrich 3.0 3.01 2.40 Humic Acid

Hoytville 3.0 2.24 2.41 Humic Acid 78 respectively. For imazethapyr, the calculated log values are 10.80, 10.09, and 9.16. The apparent disagreement between the experimental and theoretical values of log clearly suggests that in addition to partitioning, other mechanisms

such as adsorption may be involved for the sorption of these herbicides on soils and humic acids. The large difference for imazethapyr between experimental and calculated values of log Km( comparing to the difference for imazaquin, suggested that partitioning is not the dominant process for the sorption of imazethapyr on soils and humic acids.

Humic acids. The high Kd values for Aldrich humic acid indicate that sorption may be partly due to partitioning between organic and aqueous phases (Table 5) . Imazethapyr had a higher affinity for humic acid extracted from Hoytville soil than imazaquin, which may be one reason for the higher sorption of imazethapyr in soil systems. The reason for the much higher adsorption of imazaquin on Aldrich humic acid is uncertain. The larger aromatic ring structure, higher molecule weight, and lower water solubility of imazaquin may have an important role in its hydrophobic affinity for humic acid. The higher affinity of both herbicides for Aldrich humic acid could be due to the higher organic C content of the Aldrich humic acid (40.88%), compared to that of the humic acid extracted from Hoytville soil (24.79%). Adsorption of 79 herbicide on humic acid at higher pH levels was not studied because humic acid will dissolve at high pH. Another method, fluorescence quenching, was used for studying herbicide sorption at higher pH.

Clays. Imazethapyr had stronger a affinity for clays than imazaquin, which could be one of the reasons why imazethapyr was more highly adsorbed on soils (Table 5). Loux et al. (1989b) also found a strong interaction between imazethapyr and clay in soils. They suggested that the interaction of imazethapyr with clay was due to a "site-specific" mechanism, and molecular size, steric effects, or electronic effects may be related to this specific sorption mechanism. The sorption of both herbicides decreased as pH increased in all three clay systems, but especially for kaolinite. Increasing solution pH would result in more herbicide molecules in the anionic form, and repulsion from the negatively-charged clay surfaces. Varying pH had less effect on the sorption of herbicides on both hectorite and illite than on kaolinite. The CEC of kaolinite is more pH-dependent than that of other clay minerals (Borchardt, 1989) . Kaolinite exhibited a greater affinity for imazethapyr than hectorite at low pH for the same reason. In general, hectorite exhibited the greatest affinity for imazaquin among the three clay systems, while adsorption of imazethapyr was greatest on 80 illite. These results were similar to those of Dolling (1985), where retention of imidazolinone herbicides was greatly influenced by the acidic montmorillonites and all illites, or probably intergrades of these two clay types. The higher sorption constant for the imazethapyr on illite (or kaolinite), which has low surface area and CEC, comparing to the lower sorption constant on hectorite, which has higher surface area and CEC, may be resulted from two reasons : (1) The illite (or kaolinite) has higher amount of impurities such as organic carbon content than hectorite, which can sorb more herbicide molecules (Table 4) . (2) Water can strongly compete with imazethapyr molecules for the adsorption sites and does not allow the herbicide molecules to enter the interlayer space of expanding clay minerals. The little hysteresis

effect, which indicated the weak sorption, and x-ray diffraction analysis (Dolling, 1985) confirmed above assumption and the imazethapyr molecules should be adsorbed on the external surface of clay such as illite. The total surface area of Wyoming montmorillonite and illite are 662 and 100 m2/g, whereas the external surface area of both clays are 31 and 100 m2/g, respectively (Greenland and Mott, 1978; van Olphen H. and Fripiat, 1979). Therefore, although the total surface area of expanding clay mineral such as hectorite is large, the sorption constant for the imazethapyr on non­ expanding illite is also large for the comparatively large 81 external surface area of illite. The total surface area of hectorite, illite, and kaolinite used in this study is 486, 44.3, and 12.1 m2/g, whereas the external surface area is 61.8, 44.0 and 12.0 m2/g, respectively. The total surface area and external surface area were determined by EGHE adsorption and N2 adsorption, respectively (Brunauer et al., 1938; Carter et al., 1965). The relatively high amounts of imazethapyr adsorption on illite is an very important information when we explain the

behavior of imidazolinone herbicides in soil. The predominant clay mineral of two soils used in this study is illite, which is also the major clay mineral fraction of many soils in Corn Belt region. Therefore, the soil containing a large amount of illite presumably has a higher adsorption capacity for imidazolinone herbicides, which should be considered when these herbicides applied in the field conditions.

Desorption

Soils. In general, the desorption rate decreased with successive desorption cycles (Table 9) . At low pH, the

greatest amount of herbicide desorbed in the first desorption cycle at each herbicide concentration level. The amounts of imazaquin desorbed from soils in the first cycle ranged from

40 to 53% at the lowest soil pH level and the two lower herbicide concentration^, and ranged from 29 to 38% for 82 Table 9. Desorption of imazaquin and imazethapyr from soils (pH = 4.5 or 4.6).

Desorption Cvcle 03 D* Total LSD Imazaauin8 % Desorbed Hovtville 44 18 10 4 76 5 Crosbv 50 17 10 4 81 5

Imazethapvrb

Hovtville 30 20 14 8 72 5 Crosbv 38 19 10 6 73 5 LSD 8 8 8 8

a Average of the 0.024 mg/L and 0.1 mg/L herbicide treatments. b Average of the 0.043 mg/L and 0.1 mg/L herbicide treatments. 83 imazethapyr. These results indicate again that imazethapyr was binding more strongly than imazaquin on soils. The amounts of both herbicides desorbed from Crosby soil were larger than that from Hoytville soil in the first desorption cycle (Table 9). There were no differences in the herbicide desorption percentages after the first desorption. These results showed that both herbicides were binding more strongly to Crosby soil than to Hoytville soil. The similar desorption percentage after the first desorption cycle indicated that the herbicide molecules still sorbed after the first desorption cycle desorbed less readily. At the highest soil pH level, the amounts of both herbicides desorbed from soils were low and irregular due to low adsorption (Table A-

1). Previous work suggested that temporary desiccation can enhance imazaquin sorption (Goetz et al., 1986). There is little research concerning the effect of moisture on desorption of imidazolinone herbicides from soils. The effect of drying the soil on herbicide desorption is shown by the results in Table 10. Imazaquin desorbed from slurry treatments at lower soil pH (4.5 or 4.6) in the first desorption cycle was significantly higher than that from temperarily drying treatments (5% moisture content for 48 hrs.; p = 0.01). The decrease in imazaquin desorption due to temporary drying of the soils may possibly be attributed to 84 Table 10. The effect of soil moisture variation on herbicide desorption from soils (pH = 4.5 or 4.6).

Desorption Cvcle Moistureb D1 D3 d4 Total LSD Imazaauin8 Treatment Desorbed -- Hovtville w/o 46 21 10 4 81 5 w 36 11 5 3 55 5 Crosbv w/o 53 18 11 8 90 5

w 10 25 11 8 54 5

Imazethapyr8

Hovtville w/o 29 19 14 8 70 5 w 31 18 11 6 66 5 Crosbv w/o 37 19 10 6 72 5

w 40 16 8 5 69 5 LSD 7 7 7 7

8 Initial concentration of herbicide solution is 0.1 mg/L. b The moisture treatment was tempararily drying the soil to 5% moisture content and maintained for 48 hours, "w/o" means "samples without the moisture treatment",

"w" means "samples with the moisture treatment". 85 the protonation of imazaquin molecules by greater clay surface acidity caused by very dry condition. These protonated molecules were strongly sorbed by clay and other soil componants and would not be easily desorbed even when soil was rewetted or extracted with water. These "trapped" molecules will be released slowly after two or three desorption cycles. The difference in desorption between wet and dry treatments would thus decrease with additional desorption cycles. There were no significant differences (p = 0.01) for the desorption of imazethapyr between slurry and 5% wt. moisture treatments (Table 10). Imazethapyr was more highly sorbed than imazaquin, and soil moisture content may thus have a less evident effect on desorption of imazethapyr from soils. The effect of dissolved humic acid on the desorption of imidazolinone herbicides from soils was evaluated by the method described previously, and the results are shown in

Table 11. Dissolved humic acid enhanced the desorption of adsorbed herbicides from Crosby soil in the first desorption cycle. The effects slowly disappeared in the following desorption cycles. The humic acid may have increased the apparent water solubility of both herbicides. This can be described by the following equation (Chiou et al., 1986) :

s.* - s. (i + ctA.) (2.7) 86

Table 11. The humic acid effect on the desorption of imidazolinone herbicides from soils®.

Desorption Cycle D, °3 Tota LSD Imazaouin - 5 % Desorbed - Hoytville water 44 13 10 3 70 3 humic acid 41 11 9 2 63 3 Crosbv water 34 14 7 4 59 3 humic acid 41 13 4 2 60 3

Imazethapvr

Hovtville water 35 23 13 9 80 3 humic acid 35 24 13 8 80 3 Crosbv water 44 18 10 6 78 3 humicacid 52 22 9 6 89 3 LSD 4 4 4 4

® pH = 4.5 and 4.6 for Crosby and Hoytville soil, respectively. where SM* and SH are the solubilities of the herbicide in the solution will dissolved humic acid and in pure water, respectively. Cha is the concentration of the humic acid (g/mL of water) and is the partition coefficient of the compound between the humic acid and water. The humic acid can also compete with the herbicide molecules for adsorption sites, and thus enhance the release of herbicide into the solution. Abdul et al. (1990) found that humic acid could enhance the removal of four less soluble aromatic hydrocarbons from a sandy material. They explained that the hydrophobic organics could partition into the hydrophobic interior of the membrances and/or micelles which were aggregated by humic acid molecules. The enhancement effect of humic acid on the desorption of imidazolinone herbicides from Hoytville soil was not observed because the herbicide molecules were strongly adsorbed, compared to the Crosby soil. Desorption of both herbicides from soils exhibited some hysteresis at high initial herbicide concentrations (Figures 14 and 15) . There were no hysteresis effects at herbicide concentrations lower than 1.0 mg/L. The hysteresis at the higher herbicide concentrations may be caused by the specific desorption method used in this study. When the initial herbicide concentrations were 5.0 or 10.0 mg/L, the adsorbed herbicide could not be easily desorbed because the equilibrium concentration of herbicide in the solution phase is so high Figure 14. The adsorption and desorption of imazaquin on imazaquin of desorption and adsorption The 14. Figure Cad(nmol/g) -10 - 20 10 5 1 5 0 5 -10 ------Hoytville soil (pH = 4.5). (pH = soil Hoytville 0 10 q( lmL) L ol/m m (n eq C 20 30 Adsorption Desorption 050 40 88 Figure 15. The adsorption and desorption of imazethapyr on imazethapyr of desorption and adsorption The 15. Figure Cad(nmol/g) 0 2 0 3 0 4 10 50 - 0 -10 - - - - Hoytville soil (pH = 4.5). (pH = soil Hoytville 0 10 q( lmL) L ol/m m (n eq C 20 050 30 Adsorption Desorption 40 89 90 that the higher concentration gradient did not allow the adsorbed herbicide molecules released into the solution. Less hysteresis at the lower herbicide concentrations indicated that both herbicides were weakly adsorbed by the soils.

Humic acids. Both herbicides had similar desorption rates from humic acids (Table 12). The desorption rates of both herbicides from Hoytville humic acid were higher than those from Aldrich humic acid during the first and second desorption cycles. The desorption rates became similar between the two humic acids after the third desorption cycle, because only more strongly bound herbicide molecules remained. The lower desorption rates from Aldrich humic acid reflected a greater affinity for both herbicides, compared to the humic acid extracted from Hoytville soil. This greater affinity may be due to the higher organic C content of Aldrich humic acid. Data in Table 12 also shows that the desorption rate increased as pH increased. Since the sorption of these herbicides on soil increased as pH decreased, the herbicide desorption rate should increase when more herbicide molecules are anionic and repulsed by humic acid as soil pH increases.

Clays. In general, the desorption rates from clays decreased with each desorption cycle as in other sorbent systems (Table 13). The desorption rate of imazethapyr from clays was 91

Table 12. Desorptions of imazaquin and imazethapyr from humic acid.

______Desorption Cvcle______D1 D2 Dj D4 D5 Total LSD Imazaquin % Desorbed------Extracted humic acid pH 3.0 20 9 7 4 2 42 9 pH 5.0 35 23 13 6 2 79 9 pH 8.0 46 25 6 4 2 83 9 LSD 12 12 12 12 12 Aldrich humic acid pH 3.0 2 2 1 4 2 11 1 pH 5.0 1 1 2 2 2 8 1 pH 8.0 2 1 1 2 1 7 1 LSD 1 1 1 1 1

Imazethapyr Extracted humic acid pH 3.0 22 25 22 8 7 84 8 pH 5.0 35 26 19 18 2 100 8 pH 8.0 41 31 18 7 3 100 8 LSD 11 11 11 11 11 Aldrich humic acid pH 3.0 12 8 5 7 7 39 0.38 pH 5.0 12 8 6 7 7 40 0.38 pH 8.0 12 8 6 7 7 40 0.38 LSD 0.5 0.5 0.5 0.5 0.5 92

Table 13. Desorption of imazaquin and imazethapyr from c l a y s .

______*££sawi. i w y v j . 9 D1 DZ °3 Dc Total LSD Imazaauin - % 1Desorbed ------Kaolinite pH 3.0 3 0 0 0 — 3 6 pH 5.0 0 4 5 7 — 16 6 pH 8.0 0 6 6 3 — 15 6 LSD 7 7 7 7 Hectorite pH 3.0 40 17 9 4 — 70 3 pH 5.0 31 15 9 4 — 59 3 pH 8.0 40 18 9 4 — 71 3 LSD 3 3 3 3

ImazethaDvr Kaolinite pH 3.0 59 23 10 5 3 100 6 pH 5.0 50 23 10 6 3 92 6 pH 8.0 46 23 13 7 5 94 6 LSD 8 8 8 8 8 Hectorite pH 3.0 54 24 13 6 3 100 2 pH 5.0 56 24 11 6 3 100 2 pH 8.0 55 24 13 5 3 100 2 LSD 3 3 3 3 3 greater than that of imazaquin. Since the amounts of imazaquin sorbed by clays were very low compared to that of imazethapyr, the desorption rate of imazaquin was slow due to the much higher herbicide concentration in solution phase, compared to the much lower herbicide concentration sorbed in sorbent phase, which inhibit the adsorbed imazaquin molecules easily to release into the solution. The desorption rates of imazaquin from hectorite were more rapid than from kaolinite for the same reason, whereas desorption rates from the two clays were similar for imazethapyr. There was no significant effect of pH on the desorption of herbicides from clays, except both herbicides had higher desorption rates in first desorption cycle from kaolinite at low pH (pH = 3.0), compared to the higher pH ranges (pH = 5.0 or 8.0). As mentioned previously, the CEC of kaolinite is more pH-dependent than other clay minerals, and varying pH would have a greater

effect on the desorption of herbicides from kaolinite than from hectorite. The effect of dissolved humic acid on the desorption of imazaquin and imazethapyr from clays was not significant (p = 0.05) as shown in Table 14. Since the clays had large surface areas and could bind with humic acid, the effectiveness of the dissolved humic acid would become significant only after the sorption capacity of the clays for the humic acid is reached. 94

Table 14. The humic acid effect on the desorption of imidazolinone herbicides from clays®.

Desorption Cvcle Di °2 d 4 d 5 Total LSD Imazaauin6 - ? > Desorbed - Kaolinite water 3 0 0 0 3 6 humic acid 7 7 0 1 15 6 Hectorite water 40 17 9 4 70 6 humic acid 43 17 9 4 73 6

Imazethapvrb Kaolinite water 59 23 10 5 3 100 6 humic acid 61 23 8 3 2 97 6 Hectorite water 54 24 13 6 3 100 6

humic acid 56 17 15 6 3 97 6

LSD 10 10 10 10 10

8 pH = 3 for all systems.

b Herbicide concentration is 0.1 mg/L. 95 Conclusions

The results of this study indicate that the adsorption of imazaquin and imazethapyr is affected by soil pH, the nature of herbicide, and the type and amount of organic matter and clay in soils. Goetz et al. (1986) reported that clay and organic matter content had minimal effect on imazaquin adsorption by soils, but both fractions appeared to be important in this experiment. The soils which were used in their study contain large amounts of aluminum and iron hydroxides and kaolinite, whereas the predominant dry fraction of soils used in this study was illite, which is a highly adsorption soil component especially for imazethapyr. Humic acid had much greater affinity for both herbicides than clay did, indicating that organic matter content could be the major factor controlling sorption of imazaquin and imazethapyr on soils especially in soil containing high organic matter content. Illite and kaolinite, which exhibit more pH- dependent charge characteristics and less total surface area than hectorite, can impart important sorptive character to soil, especially with regard to imazethapyr at lower pH levels. The sorption constant of both herbicides in soils decreased as soil pH increased, but the increase in sorption was most evident as soil pH decreased from 5.5 to 4.5. This pH effect may be due to hydrogen bonding and protonation 96 mechanisms. Both sorption and desorption data suggested that imazethapyr binds more strongly than imazaquin on soils and clays; but imazaquin binds more strongly to Aldrich humic acid. The sorption of imazethapyr on soils and clays was more

likely to be controlled by adsorption process at lower pH levels, whereas the sorption of imazaquin seemed to be controlled by partitioning process. The desorption of imazaquin from soil decreased when soil moisture decreased and maintained for only 48 hours, indicating rainfall and soil moisture status after herbicide application may affect the desorption of imazaquin under field conditions. Dissolved humic acid can enhance the desorption of sorbed herbicide from Crosby soil. There were no significant hysteresis effects at herbicide concentration lower than 1.0 mg/L, indicating that both herbicides were weakly sorbed by the soils. Both herbicides showed similar desorption rates from humic acids, whereas the desorption rate of imazethapyr from clays was greater than that of imazaquin. The desorption rates of both herbicides from Hoytville humic acid were higher than those from Aldrich humic acid. The desorption rates of imazaquin from hectorite were greater than from kaolinite, whereas desorption rates from two clays were similar for imazethapyr. However, in the batch equilibrium method used in this study, the sorbent-solution ratios were not the typical moisture contents under field conditions. Therefore, use of the desorption data from this study to explain field wetting and drying effects should be carefully evaluated, and a more accurate desorption method is recommended for future experiments. 98

REFERENCES Abdul, A.S., T.L. Gibson, and D.N. Rai. 1990. Use of humic acid solution to remove organic contaminants from hydrogeologic systems. Environ. Sci. Technol. 24 : 328-333. American Cyanamid Co. 1986. Technical Information Report on Imazaquin. Agric. Div., Wayne, NJ 07470. American Cyanamid Co. 1987. Technical Information Report on Imazethapyr. Agric. Res. Div., Princeton, NJ 08540. American Cyanamid Co. 1988. Technical Information Report on Imazapyr. Agric. Res. Div., Princeton, NJ 08540. American Cyanamid Co. 1989. Technical Information Report on AC 222,293 Herbicide. Agric. Res. Div., Princeton, NJ 08540. Bailey, G.W., and J.L. White. 1970. Factors influencing the adsorption, desorption, and movement of pesticides in soil. Residue Rev. 32 : 29-92.

Basham, G.W., T.L. Lavy, L.R. Oliver, and H.D. Scott. 1987. Imazaquin persistence and mobility in three Arkansas soils. Weed Sci. 35 : 576-582. Bohn, H.L., B.L. McNeal, and G.A. O'Connor. 1979. Soil Chemistry. John Wiley & Sons, New York. 329 pp. Borchardt, G. 1989. Smectites, p. 675-728. In J.B. Dixon and S.B. Weed (ed.) Minerals in Soil Environments. 2nd ed. Soil Sci. Soc. Am. Madison, WI.

Bowman, B.T. 1979. Method of repeated additions for generating pesticide adsorption-desorption isotherm data. Can. J. Soil Sci. 59 : 435-437.

Brunauer,S., P. H. Emmett, and E. Teller. 1938. Adsorption of gasses in multimolecular layers. J. Amer. Chem. Soc. 60 : 309- 319.

Calvet, R. 1980. Adsorption-desorption phenomena, in R.J. Hance (ed.) In "Interactions between Herbicides and the Soil." p. 1-30. Academic Press Inc., Ner York. Carter, D. L., M. D. Heilman, C. L. Gonzalez. 1965. Ethylene glycol monoethyl ether for determing surface area of silicate minerals. Soil Sci. 100 : 356-360. 99 Chapman, H.D. 1965. Cation exchange capacity, p. 891-901. In C.A. Black (ed.) Methods of Soil Analysis. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI. Charlet, L. 1986. Adsorption of some macronutrient ions on an Oxisol. An application of the Triple Layer model, Ph.D. dissertation, Univ. of California, Riverside. Chiou, C.T. 1989. Theoretical considerations of the partition uptake of nonionic compounds by soil organic matter. In : B.L. Sawhney and K. Brown (ed.) Reactions and Movement of Organic Chemicals in Soils. Soil Sci. Soc. Am., Madison, WI., Spec. Publ. No. 22, pp. 1-30. Chiou, C.T., R.L. Malcolm, T.I. Brinton, and D.E. Kile. 1986. Water solubility enhancement of some organic pollutants and pesticides by dissolved humic and fulvic acids. Environ. Sci. Technol. 20 : 502. Chiou, C.T., P.E. Porter, and D.W. Schmedding. 1983. Partition equilibria of nonionic organic compounds between soil organic matter and water. Environ. Sci. Technol., 17 : 227-231.

Corey, R.B. 1981. Adsorption vs precipitation, in Adsorption of Inorganics at Solid-Liquid Interfaces (M.A. Anderson and A.J. Rubin, eds). Ann Arbor Science, Ann Arbor, Michigan. Curran, W.S. and E.L. Knake. 1989. Effect of herbicide combinations on rotational corn. NCWSS Research Report. 46 : 274.

Day, P.R. 1965. Particle fractionation and particle size analysis, p. 545-567. In C.A. Black (ed.) Method of Soil Analysis. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI. Dolling, A.M. 1985. Studies of Interactions of Some Imidazolinone Herbicides with Clays. M.Sc. Thesis, Department of Chemistry, University of Birmingham, England. 198 pp. Edward J.T. and I. Lantos. 1972. Solvent effects on the tautomeric equilibria of 4,4/5,5-disubstituted 4H/5H-imidazol- 5/4-ones. J. Heterocycle Chem. 9 : 363-369. Everett, D.H. 1972. Manual of Symbols and Terminology for Physicochemical Quantities and Units. Appendix II : Definitions, Terminology and Symbols in Colloid and Surface Chemistry. Butterworths, London. 100 Panning, D.S., V.Z. Keramidas, and M.A. El-Desoky. 1989. Micas. p. 551-634 in J.B. Dixon and S.B. Weed (eds.) Minerals in Soil Environments. 2nd ed. Soil Sci. Soc. Amer. Madison, WI. Gamar, Y. and M.A. Mustafa. 1975. Adsorption and desorption of diquat2+ and paraquat2* on arid-zone soils. Soil Sci. 119 : 290-295. Gauthier, T.D., E.C. Shane, W.F. Guerin, W.R. Seitz, and C.L. Grant. 1986. Fluorescence quenching method for determining equilibrium constants for polycyclic aromatic hydrocarbons binding to dissolved humic materials. Environ. Sci. Technol. 20 : 1162-1166.

Giles, C.H., T.H. MacEwan, S.N. Nakhwa, and D. Smith. 1960. Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurement of specific surface areas of solids. J. Chem. Sco. p. 3973-3993. Goetz, A.J., G. Wehtje, R.H. Walker, and B. Hajek. 1986. Soil solution and mobility characterization of imazaquin. Weed Sci. 34 : 788-793.

Green, R.E. 1974. Pesticide-clay-water interactions, p. 3- 37 in W.D. Guenze (ed.) Pesticides in soil and water. Soil Sci. Sco. Am. Madison, WI. Green, R.W. and H.K. Tong. 1956. The constitution of the pyridine monocarboxylic acids in their isoelectric forms. J. Cm. Chem. Soc. 78 : 4896-4900.

Greenland, D.J. and C.J.B. Mott. 1978. Surface of soil particle. P.32 in D.J. Greenland and M.H.B. Hayes (eds.), The Chemistry of Soil Constituents. John Wiley & Sons, Inc., Chichester. Grover, R. 1975. Adsorption and desorption of urea herbicides on soils. Can. J. Soil Sci. 55 : 127-135. Gunsolus, J.L., R.H. Behrens, W.E. Lueschen, D.D. Warnes, and J.V. wiersma. 1986. Carryover potential of AC-263,499, DPX- F6025, FMC57020 and imazaquin in Minnesota. North Cent. Weed Control Conf. Proc. 41 : 52.

Hackett, N. 1990. Imazaquin behavior in the soil. Technical report of American Cyanamid Company. Agri. Div., Wayne, NJ 07470. 101 Hamaker, J.W. and J.M. Thompson. 1972. Adsorption, p. 49- 143. In C.A. Goring and J.W. Hamaker, (ed.) Organic Chemicals in the Soil Environment. Marcel-Dekker, New York. Hasset, J.J., W.L. Banwart, and R.A. Griffin. 1983. Correlations of compound properties with sorption characterisitcs of nonpolar compounds by soils and sediments : concepts and limitations. In : C.W. Francis and S. I. Auerbach (ed.), Environmental and Solid Wastes. Butterworths, Boston, MA. pp. 161-178. Hausler, M.J. 1986. Studies of Interactions of Some Imidazolinone Herbicides with Whole Soil, Oxyhydroxides, and with Natural and Synthetic Humic Acids. Ph.D. Thesis, Department of Chemistry, University of Birmingham, England. 228 pp. Jaynes, W.F. and J.M. Bigham. 1987. Charge reduction, octahedral charge, and lithium retention in heated, Li- saturated smectites. Clays Clay Miner. 35 : 440-448. Karickhoff, S.W., D.S. Brown, and T.A. Scott. 1979. Sorption of hydrophobic organic pollutants on natural sediments. Water Res. 13 : 241-248.

Koskinen, W.C., G.A. O'Connor, and H.H. Cheng. 1979. Characterization of hysteresis in the desorption of 2,4,5-T from soils. Soil Sci. Soc. Am. J. 43 : 871-874. Los, M., D.R. Carliante, E.M. Ettinghouse, and P.J. Wepplo. 1983. 0-(5-oxo-2-imidazolin-2-yl)-arylcarboxylates : a new class of herbicides. American Cyanamid Co. Agric. Res. Div., Princeton, NJ 08540.

Los, M., P.J. Wepplo, R.K. Russell, B.L. Lences, and P.L. Orwick. 1984. 0-(5-oxo-2-imidazolin-2-yl)-arylcarboxylates : a new class of herbicides. 4. AC 263,499, a herbicide for use in legumes. Book of Abstracts, 188th Meet. Amer. Chem. Soc., Abstract 30 PEST.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989a. Availability and persistence of imazaquin, imazethapyr, and clomazone in soil. Weed Sci. 37 : 259-267.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989b. Adsorption of imazaquin and imazethapyr on soils, sediments, and selected adsorbents. Weed Sci. 37 : 712-718. McEwen, F.L., and G.R. Stephenson. 1979. The Use and Significance of Pesticides in the Environment. John. Wiley & Sons, Inc., New York. 102 van Olphen, H. and J.J. Fripiat (eds.)* 1979. Data Handbook for Clay Materials and other Non-metallic Minerals. Pergamon Press, Oxford. Parker, C.A. 1968. Photoluminescence of Solutions. Elsevier, New York. pp. 220-232. Peoples, T.R., T. Wang, R.R. Fine, P.L. Orwick, S.E. Grahan, and K. Kirkland. 1985. AC 263,499 : a new broad-spectrum herbicide for use in soybeans and other legumes. Proc. Brit. Crop. Prot. Conf.-Weeds; 2 : 99-106. Pignatello, J.J. 1989. Dynamics of organic compounds in soils and sediments. In Reactions and Movement of Organic Chemicals in Soils. Sawhney & Brown (ed.) p. 45-80. SSSA Special Publication Number 22. Soil Sci. Soci. Amer., Inc., Madison, WI. Renner, K.A., W.F. Meggitt, and R.A. Leavitt. 1988a. Influence of r^te, method of application and tillage on imazaquin persistence in soil. Weed Sci. 36 : 90-95. Renner, K.A., W.F. Meggitt, and D. Penner. 1988b. Effect of soil pH on imazaquin and imazethapyr adsorption to soil and phytotoxicity to corn (Zea mays). Weed Sci. 36 : 78-83. Schnitzer, M. 1982. Organic matter characterization, p. 581- 594. In A.L. Page et al. (ed.) Methods of Soil Analysis. Part 2. Chemical and microbial properties. 2nd ed. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI.

Shaner, D.L., P.C. Anderson, and M.A. Stidham. 1984. Imidazolinones. Potent inhibitors of acetohydroxyacid synthase. Plant Physiol. 76 ; 545.

Shaner, D.L. and P.A. Robson. 1985. Adsorption, translocation, and metabolism of AC 252,214 in soybean (Glycine max), common cocklebur (Xantbium strumarium), and velvetleaf (AbutiIon theophrasti). Weed Sci. 33 ; 469-471. Sposito, G. 1984. The Surface Chemistry of Soils, p. 113- 153. Oxford University Press, Inc. New York. Sposito, G. 1989. The Chemistry of Soils. p. 127-169. Oxford University Press, Inc. New York.

Stevenson, F.J. 1982. Humus Chemistry ; Genesis, Composition, Reactions. John Wiley and Sons, New York. 443 pp. 103 Stougaard, R.N., P.J. Shea, and A.R. Martin. 1985. The effect of soil type and pH on AC-263,499 and imazaquin. NCWSS Proc. 40 : 15-16. Stougaard, R.N., P.J. Shea, and A.R. Martin. 1990. Effect of soil type and pH on adsorption, mobility, and efficacy of imazaquin and imazethapyr. Weed Sci. 38 : 67-73. Theisen, A. and M.E. Harvard. 1962. A paste method for preparation of slides for clay mineral identification by x- ray diffraction. Soil Sci. Soc. Am. Proc. 26 : 90-91. Traina, S.J., D.A. Spontak, and T.J. Logan. 1989. Effects of cations on complexation of naphthalene by water soluble organic carbon. J. Environ. Qual. 18 : 221-227. Upchurch, R.P. 1966. Behavior of herbicides in soil. Res. Rev. 16 : 46-85.

Walker, A., and D.V. Crawford. 1968. The role of organic matter in adsorption of the triazine herbicides by soils, p. 91-108. In Isotopes and radiation in soil organic matter studies. Proc. 2nd Symp. Int. Atomic Energy Agency, Vienna. Weber, J.B. and C.T. Miller. 1989. Organic chemical movement over and through soil. p. 305-334. In B.L. Sawhney and K. Brown (ed.) Reactions and movement of organic chemicals in soils. Soil Sci. Soc. Am., Inc. Madison, WI. Weed, S.B. and J.B. Weber. 1974. Pesticide-organic matter interactions. p. 39-66 in W.D. Guenzi, ed. Pesticides in Soil and Water. Soil Sci. Soc. Am. Madison, WI.

Wehtje, G., R. Dickens, J.W. Wilcut, and B.F. Hajek. 1987. Sorption and mobility of sulfometuron and imazapyr in five Alabama soils. Weed Sci. 35 : 858-864.

Willard, H.H., L.L. Merritt, Jr., J.A. Dean, and F.A. Settle, Jr. 1988. Fluorescence and phosphorescence spectrophotometry. In Instrumental Methods of Analysis. 7th ed. p. 197-223. Wadsworth Publishing Co., Belmont, California. CHAPTER III THE SORPTION AND INTERACTION BETWEEN IMIDAZOLINONE HERBICIDES AND SELECTED ADSORBENTS

Introduction

The effects of various soil constituents and properties on the adsorption of imidazolinone herbicides on soils have been reported by some researchers in recently years (Basham et al., 1987; Goetz et al., 1986; Loux et al., 1989b; Renner et al., 1988b). Among these factors, soil pH, organic matter content, clay types and content, and soil moisture content appear to be dominant. In general, adsorption of imidazolinone herbicides increases as soil pH increases, soil organic matter and clay content increase, and soil moisture content decreases (Basham et al., 1987; Goetz et al., 1986; Loux et al., 1986b). There is limited information on the mechanisms of adsorption of imidazolinone herbicide on specific clay minerals or organic matter over a range of soil pH. The relative affinitives of these herbicides for various clays and organic matter are also unknown. Dolling (1985) investigated the interactions between three

104 105 imidazolinone herbicides and four homoionically-exchanged clays and concluded that mechanisms such as van der Waal forces, hydrogen bonding, coordination complex formation, and association or bridging complexes might be involved in the adsorption of these herbicides on clays. Hausler (1986) drew similar conclusions for the adsorption mechanisms of imidazolinone herbicide on soils, oxyhydroxides, and humic acids. Fluorescence spectroscopy has assumed a major role in analysis, particularly in the determination of trace contaminants in the environment, industries, and bodies, because for applicable compounds fluorescence gives high

sensitivity (in the low parts per trillion) and high specificity (Willard et al., 1988). Fluorescence quenching can be used to study the adsorption of imidazolinone herbicides on humic acid. The method is based upon the observation that imidazolinone herbicides fluoresce in aqueous solution but not when associated with humic acid (Gauthier et al., 1986). As a consequence, the fraction of herbicides associated with various adsorbents may be determined directly from the fractional decrease in fluorescence upon addition of adsorbents. This method has been previously used to determine the equilibrium constants for polycyclic aromatic hydrocarbons (PAHs) binding to dissolved humic materials or water soluble organic carbon (Gauthier et al., 1986; Traina et al., 1989). 106 For compounds that fluoresce, this method offers advantages over other methods for its application without separation of adsorbent and solution phases. Fluorescence is expected in molecules that are aromatic, contain multiple- conjugated double bonds with a high degree of resonance stability, or contain a nonbonding pair of valence electrons - - such as an amine with a lone pair on its nitrogen atom (Willard et al., 1988). The imidazolinone herbicides, which have aromatic character for pyridine, quinoline, or benzene ring structure on imidazole and a lone pair of electrons on its nitrogen atom, should also fluoresce. The objectives of this study are to : 1) determine adsorption constants for the association of imidazolinone herbicides with two humic acids by using the fluorescence quenching, and 2) evaluate the effect of pH on the mechanisms of imidazolinone herbicide adsorption on clay using fluorescence spectroscopy.

Materials and Methods

Materials

The four imidazolinone herbicides used in this study were imazaquin (96.0% pure), imazethapyr (99.2% pure), imazapyr

(97.8% pure) and imazamethabenz (90.0% pure). All were obtained from American Cyanamid Company, and were used without 107 further purification. The concentrations of solutions of the herbicides were 0.10, 1.00, 5.20, and 1.00 g/L for imazaquin, imazethapyr, imazapyr, and imazamethabenz, respectively. These solutions were prepared by dissolving weighed amounts in water and stirring for 24 hours. The herbicide solutions were stored in a refrigerator in glass volumetric flasks in the dark. A small amount of adsorption of herbicides to the glass walls of containers made it impractical to know the exact concentrations of herbicides in solutions, but accurate herbicide concentrations are not required in the fluorescence quenching method (Gauthier et al., 1986). Two humic acids were used as the sorbents in this study. The characteristics of the humic acids were given in Table 4 (the organic c content of Hoytville humic acid was increased to 35.67% by additional purification). The 500 mg/L stock solutions for both humic acids were prepared and used as adsorbents. The herbicide solutions were maintained at pH

3.0, 5.0, and 8.0 with HC1, acetic buffer (0.1 M sodium acetate + acetic acid), and 0.1 M phosphate buffer, respectively.

Instrumentation

Fluorescence measurements were made on a Perkin-Elmer LS- 5 Spectrofluorometer with slit widths set for band widths of 5 nm on both excitation and emission monochromators. The 108 excitation/emission wavelengths used for the herbicide fluorescence measurements were 320/495 nm for imazaquin, 339/392 nm for imazethapyr, 359/413 nm for imazapyr, and 325/432 nm for imazamethabenz. Light adsorption measurements were made with a Beckman DU6, UV-VIS spectrophotometer.

Sorption Constant Measurement

Fluorescence was measured as a function of added humic acids at a fixed wavelength. A 2.25-mL herbicide solution and 0.25 mL of the appropriate aqueous buffer solution were equilibrated in a 1x1x4 cm quartz fluorescence cuvette for 30 minutes using a micro stir bar, then the initial fluorescence intensity (F0) was measured. A 10-50 uL (depending on herbicides and adsorbents) aliquot of stock humic acid solution was added to the cuvette. The solution was stirred for 3 to 4 minutes, and then allowed to stand for one minute without stirring, and the second fluorescence intensity (F) was measured. This step was repeated several times, with more humic acid suspension added until the initial fluorescence intensity decreased by approximately 50% (Gauthier et al., 1986).

The sorption of herbicides on humic acid can be represented by the following equations (Gauthier et al., 1986; Traina et al., 1989) : 109 Herb + Hu = Herb-Hu (3.1)

where Herb = imidazolinone herbicide, Hu = humic acid, and Herb-Hu = humic-associated herbicide. A conditional constant cKd for reaction 3.1 can be defined as :

cKd = [Herb-Hu] / [Herb][Hu] (3.2)

The mass balance on herbicide is given as :

HerbT - [Herb] + [Herb-Hu] (3.3)

where HerbT = total aqueous concentration of all forms of herbicide. By combining eq. 3.2 and 3.3 one obtains :

HerbT / [Herb] - 1 + cKd[Hu] (3.4)

Assuming that the fluorescence intensity of the herbicide is proportional to the concentration of uncomplexed herbicide, then from equation 3.4, one can obtain :

F0 / F = l + cKd[Hu] (3.5) where F0 and F are the fluorescence intensities in the absence and presence of humic material, respectively. Equation 3.5 110 is in the form of the Stern-Volmer equation. It implies the formation of a ground state complex between herbicide and humic acid, resulting in static quenching. The slope of a plot of F^F vs [Hu] will yield a value for cKd. Dividing cKd by the fraction of organic carbon in the adsorbent yields a K^. value. This KM value was compared with the theoretical K^. values calculated from the octanol/water partition coefficients (KOHs) of these herbicides and equation 2.4, 2.5, and 2.6, described in Chapter II.

The background fluorescence from the humic acid was corrected for by subtracting the fluorescence intensity measured for adsorbent only from the total fluorescence intensity measured for herbicides in the presence of adsorbent. Both fluorescence intensities were measured at the same adsorbent concentrations and under the same instrumental conditions. To minimize background fluorescence relative to herbicide fluorescence, the herbicide concentrations used were as close to their solubility limits as possible in this experiment. In general, there are three different types of quenching processes that can occur : static, dynamic, and apparent (Gauthier, et al., 1986). Humic acid can potentially quench imidazolinone herbicide fluorescence by the formation of nonradiant complexes with ground-state herbicide molecules (static quenching) and/or by diffusion controlled (and thus Ill temperature dependent) collisions with excited state herbicide molecules (dynamic quenching). The apparent quenching is not a quenching process at all, but is rather due to an attenuation of the excitation beam and/or absorption of emitted radiation by an excess concentration of fluorophore (humic acid or clay) or by the presence of an additional absorbing species in solution. This phenomenon is known as the "inner filter effect", and can be corrected by the method of

Parker (1968). All the fluorescence intensity measurements were corrected by the following equation as described by Parker :

F 2.3 d Aex 2.3 s Aem = 108Aeffl (3.6) Fobsd l - 1 0 ' d Aex 1 - 1 0 ' 8 Aem

where is the observed intensity, Fcor is the corrected intensity, and Aex and Aem are the absorbences per centimeter at the excitation and emission wavelengths, respectively. The d, g, and s depend upon the geometry of the measurement and are defined in Figure 16.

The data analysis used in this study assumed that the attenuation of imidazolinone herbicide fluorescence by humic acid resulted from static quenching, not dynamic quenching.

This assumption was tested by measuring the temperature dependence of the fluorescence intensity measurement, in 112

o

C uvet

F

Figure 16. Assumed geometry of fluorescence measurement

with parameters used to correct for the inner filter effect. I0 represents the excitation beam with thickness s = 0.10 cm. The distance from the

edge of the sample beam to the edge of the cuvet g = 0.40 cm. The F represents the observed fluorescence beam with width 1.00 cm.(Gauthier, et al., 1986) 113 herbicide solutions with and without humic acid (Traina et al., 1989).

The herbicide solutions were equilibrated with two humic acids for 2 hours at 298 °K (The concentration of Aldrich humic acid and Hoytville humic acid was 50.0 and 30.0 mg/L, respectively). A control (herbicide solution without adding adsorbent) for each herbicide solution was also measured for its fluorescence intensity. Quartz cuvettes were filled completely with aliquots of the above equilibrated solutions, stoppered, and placed into the sample chamber of the spectrofluorometer. The temperature of the cuvette holder was then varied over a range of 274 to 343 °K with a circulating water bath. All fluorescence intensity measurements were made 15 minutes after the circulating water bath had stabilized at the desired temperature. There were 28 fluorescence intensity measurements for each sample through two heating and two cooling cycles. The excitation shutter was kept closed between measurements to minimize photochemical alteration of the experimental solutions.

Colloid-Solution Interface Study

Fluorophores which exhibit pH-dependent fluorescence spectra can be used to study colloid-solution interface phenomena. Two imidazolinone herbicides, imazaquin and imazethapyr, under the concentrations described above, were 114 equilibrated with Na-hectorite for 2 hours. The total concentration of Na-hectorite in the equilibrated solution was 1 g/L. The emission spectra of herbicides adsorbed onto

Na-hectorite were measured through a fixed excitation wavelength (320 nm and 339 nm for imazaquin and imazethapyr, respectively). Spectra for two controls, herbicide solution only or hectorite solution only, were also measured as for the herbicide-clay systems. Three pH levels, 3.0, 5.0, and 8.0, were maintained for all systems by adding appropriate amounts of 1 M HCl or NaOH. The titration-emission spectra of pure solutions of the two herbicides were also recorded. The herbicide solutions were titrated slowly by 0.025 M HCl or NaOH to reach the desired solution pH, and equilibrated for two minutes before pouring into the cuvet for fluorescence measurements. There were about 20 fluorescence measurements for each herbicide over a solution pH range of from 2.6 to 9.0.

All fluorescence emission data in this study were smoothed by "FFT" function in computer program "Lab Calc". This function smooths the "active" data file by the Fourier transforming the data file, applying a user selectable filter function, and reverse Fourier transforming the data. The chosen filter factor in this study was 8.0. There are no significant peak position and height changes by smoothing the data. 115 Results and Discussion

Sorption Constant Measurements In all cases, after correction for background fluorescence (maximum background was 7% of total) and inner filter effects, the sorption of imazaquin and imazethapyr on humic acid

resulted in linear Stern-Volmer plots. Figure 17 shows the Stern-Volmer plot of F^F values vs [Hu] for adsorption of imazethapyr on Hoytville humic acid at three pH levels. The slopes and intercepts at pH 3.0, 5.0 and 8.0 are 0.014, 0.018, 0.017 L/mg and 0.95, 0.97, 0.99, respectively. The average of R2 values for the plots is 0.99. The values of F^F were greater than 1 for all experimental solutions shown in Figure 17, which indicates that some herbicide molecules were complexed by humic acid in all solutions. The maximum background fluorescence for imazapyr was over 100% of the total fluorescence in both humic acid systems, except at pH 3.0 in the Hoytville humic acid system (between 16 and 32%; Table B-l in Appendix B). The sorption constants of imazapyr on humic acid can not be measured by this method because the fluorescence is quenched by the presence of humic acid at the specific exitation/emission wavelength. The background fluorescences for imazamethabenz were between 0.2 and 18% at various humic acid concentrations and pH levels with the exception of that at pH 8.0 for both humic acid 116

2.5 O pH=3.0

• pH=5.0

A pH=8.0 2.0

0 20 40 60 80 100

[HU] (mg/L)

Figure 17. Adsorption of imazethapyr on Hoytville humic acid. 117

systems (Table B-2). The humic acids became more soluble and dissolved at higher pH, which resulted in higher background fluorescence. The effect of solution temperature on the relative fluorescence intensity of imazaquin in the presence of Aldrich humic acid or Hoytville humic acid is shown in Figure 18. There was no apparent temperature effect on the values of F^F occurred between 274 and 348 °K. Since dynamic quenching is a diffusion controlled process, it should become more prevalent with increasing solution temperatures, and the magnitude of the relative fluorescence intensity should be directly proportional to the temperature of the solution (Traina et al., 1989). Figure 18 does not show this trend, which indicates that the attenuation of imazaquin fluorescence resulted from the formation of ground-state complexes with humic acids, and that the Stern-Volmer plots provide accurate values of adsorption constants for these complexes. Table 15 shows the Kd and values for the interactions of three herbicides with the humic acids. Each Kd or is the average of three pH levels. In general, the order of herbicide affinity for humic acids at each pH level was : imazamethabenz > imazethapyr > imazaquin. The reported octanol-water partition coefficients (Kw ) are 2.2, 11-31, and 35-66 for imazaquin, imazethapyr, and imazamethabenz, respectively (Table 1). The adsorption of imidazolinone 118

2.0

0 Aldrich HA

8 £ HojrtvUl* H A

6

4

2 «— 260 280 300 320 340 360

Temperature (°K)

Figure 18. The temperature effect on the relative fluorescence intensity of imazaquin in the

presence of humic acids. 119

Table 15. The K.,8a and K oc8 valuesb for the interactions of three imidazolinone herbicides on humic acids.

Herbicide Aldrich HA Hovtville HA Kd K , Kd Koc imazaquin 0.82 2.0 0.92 2.6 imazethapyr 3.20 7.9 1.60 4.5 imazamethabenz 26.00 64.0 3.00 8.4

8 Kd and values are expressed x 10'4 mL/g. b All Kd and K^. are the average of three pH levels. 120

herbicides on humic acid seems to follow the order of OH value of each herbicide. However, estimation of K^. for the adsorption of these herbicides on humic acid from KM values may not always be valid and should be carefully applied because different mechanisms may be involved for the adsorption of various herbicides on various humic acids. The water solubilities of imazaquin, imazethapyr, and imazamethabenz are 0.012, 0.14, and 0.3 g/lOOg HzO, respectively (Table 1). The K^. values for these herbicides are also proportional to their water solubilities. The imidazolinone herbicides are weakly adsorbed by soil and may show some hydrophobic adsorption at lower pH. Compared to the other herbicides, imazamethabenz showed much higher affinity for both humic acids. The imazamethabenz contain ester functional group, whereas the other two herbicides have acid characteristicss (See Chemical Structures in Figure 2). Imazamethabenz adsorption may involve more hydrophobic properties as indicated by its higher Kw value and was strongly adsorbed by humic acids. The larger K^. values for the adsorption of imazethapyr and imazamethabenz on Aldrich humic acid, compared to Hoytville humic acid, may be simply due to the higher organic c content of Aldrich humic acid. The effect of solution pH on sorption of imidazolinone herbicides on two humic acids is shown in Table 16. The Kd values for imazaquin sorption on two humic acids decreased 121 Table 16. values8 for sorption of imidazolinone herbicides on two humic acids at three pH levels.

Herbicide p H Aldrich HA Hovtville HA K„oc log Kk K o c log K0 Imazaquin 3.0 2.2 4.34 3.1 4.49 5.0 2.2 4.34 2.5 4.40 8.0 1.7 4.23 2.2 4.34

Imazethapyr 3.0 9.0 4.95 3.9 4.59 5.0 5.1 4.71 5.0 4.70 8.0 9.5 4.98 4.8 4.68

Imazamethabenz 3.0 12.0 5.08 5.0 4.70 5.0 22.0 5.34 11.0 5.04 8.0 150.0 6.18 9.2 4.96

8 values are expressed x 10'4 mL/g. 122 when pH increased front 3.0 to 8.0. The effect of pH on the sorption of imazaquin and imazethapyr also showed a similar trends. Many researchers have found similar correlations

between sorption of imazaquin and imazethapyr and soil reaction (Stougaard et al., 1985; Goetz et al.r 1986; Wehtje et al., 1987; Renner et al., 1988b; Loux et al., 1989b). These acidic herbicides have pKa values less than 4, resulting in a predominance of herbicide molecules in the anionic form at higher solution pH, forms which should be less sorbed by humic acid.

The effect of pH on the sorption of imazethapyr and imazamethabenz on humic acids was irregular and varied with herbicides and adsorbents. This irregular pH effect may be because both herbicides were much strongly bound by the humic acids, compared to other herbicide and sorbent combinations. By comparing the log K^. values of both herbicides for two

humic acids at pH 3.0 in Table 8 and Table 16, it is obviously

that all log Kk values which was obtained by using fluorescence quenching method were higher than that by using batch equilibrium method. However, these larger log Kov values still significant less than the theoretical values calculated from equation K0H, indicating again that the sorption of imazaquin and imazethapyr on humic acid is not just a simple partitioning process, but involved other mechanisms such as adsorption process. 123

Colloid-Solution Interface Study Figures 19 and 20 show the fluorescence emission spectra of imazaquin only and imazaquin adsorbed onto Na-hectorite at three pH levels, respectively. The data indicate that imazaquin molecules were protonated and adsorbed on the surface of Na-hectorite at pH 3.0. The fluorescence intensity in the imazaquin-clay system was enhanced to about 120% of that in the imazaquin-only system. The peak position changed when pH increased from 3.0 to 5.0, which indicated that a "new species" was present in the system. Moreover, the old peak which was present at pH 3.0 reappeared at pH 8.0.

The tautomerism of 4,4/5,5-disubstituted 4H/5H-imidazol- 5/4-ones had been investigated by Edward and Lantos (1972), and they concluded that these compounds may exist as the imidazolin-4-one (conjuated) or imidazolin-5-one (unconjugated) tautomers. The equilibrium of two tautomers can be described by the following equation :

(3.7)

Conjugated form Unconjugated form

They found that the unconjugated tautomer predominated in nonpolar solvents, and small proportions of the conjugated tautomer became apparent in more polar solvents. The iue 9 Te loecne msin pcr o imazaquin of spectra emission fluorescence The 19. Figure

Intensity (arbitrary units) 200 300 400 500 600 100 solution at three pH levels. pH three at solution 0 300 A pH=*8. 0 C B B --- -- pH*5.0 p pH*3.0 400 aeegh (nm) Wavelength 500 600 124 iue 0 Te loecne msin pta f imazaquin of spetra emission fluorescence The 20. Figure

Intensity (arbitrary units) 1200 1500 300 900 600 adsorbed onto Na-hectorite at three pH levels. pH three at ontoNa-hectorite adsorbed 300 350 pH-8.0 pH"5.0 pH"3.0 aeegh (ran) Wavelength 0 5 0 550 500 450 400 600 125 126 proportion of conjugated tautomer was increased by the presence of electron-releasing groups at the 2-position. The acidity constants of these compounds were determined by UV- Spectrophotometry as following :

compound pKa (conjugate acid) pKa (acid) R = H 2.13 ± 0.10 9.6 R = Me 2.73 ± 0.10 R = Et 2.80 ± 0.10

Therefore, protonation may occur on the imidazolone group of imazaquin at similar pH value, which was showed by the fluorescence emission spectra of imazaquin at pH 3.0 (Figure 19 and 20).

The situation is more complicated for imazaquin, where a heterocyclic system bearing a carboxyl group is attached to the imidazolone group. Green and Tong (1956) studied the constitution of the pyridine-monocarboxylic acids in their isoelectric form and found that less than 10% of the isoelectric form is present as uncharged molecules. Similarly, the quinoline-3 -carboxyl ic acid in imazaquin should have the following equilibrium :

N,

(3.8) C00H ■COO"

Zwitterionic form 127 and the zwitterionic form is predominant in aqueous solution. In summary, the possible equilibria for imazaquin in aqueous solution can be illustrated by Figure 21. Notice that forms II, III, IV, V, and VI, are zwitterionic, and forms II and IV are the dominant in each group, respectively. If we assume that the first peak which appears at 475 nm represents the protonated species, it is by no means that this peak reappears again at pH 8.0 because all imazaquin molecules should be presents as anionic form in such high pH condition. Therefore, by comparing the fluorescence spectra in Figure 19 and possible equilibria in Figure 21, an intramolecular H- bonding mechanism for imazaquin in aqueous solution is proposed. Imazaquin is present in a protonated form as "free cation" in aqueous solution at pH 3.0, and only one peak appears at 475 nm. When solution pH increases from 3.0 to 5.0, imazaquin becomes neutral or even negative, and a proposed "polymer" gradually becomes the dominant species, which shifts the peak to 400 nm. This polymer is connected by H-bonding of N-H or O-H bonds between imazaquin molecules which form "bulk" structure in aqueous solution. When solution pH becomes alkaline, most protons will dissociate from imazaquin molecules, which increases the total negative charge and breaks the intramolecular structure. As more and more imazaquin is present as the "free" form in solution, the peak at 475 nm reappears and two peaks are present at pH 8.0 128

,COOH ^ C O O H COOH

‘N H* HN — ^

I II III

\

2 0 H * r2 H * OH* H * OH* H \ COO COOH CCC< HN HN HN

IV V VI

O* OH' H

COO*

HN

VII

Figure 21. The possible equilibria for the imazaquin in

aqueous solution. 129 (Figure 19) . This proposed mechanism assumes that both "free" cationic and anionic forms of imazaquin molecules show only one peak (475 nm) in fluorescence emission spectra, whereas the proposed polymer structure species show the other one (400 nm) . Figures 22 and 23 show the titration-fluorescence emission spectra for imazaquin aqueous solution from pH 2.72 to 5.33, and from pH 5.68 to 9.00, respectively. The fluorescence spectra in Figure 19 indicate that the second peak (400 nm) appeared at pH 3.78, which is the pKa value of imazaquin. Figure 20 indicates that the "free cation" peak (475 nm) reappears again when solution pH is above neutral. Figures 24 and 25 show the fluorescence emission spectra of imazethapyr only and imazethapyr adsorbed onto Na-hectorite at three pH levels, respectively. Unlike imazaquin, the imazethapyr molecules were only slightly protonated and adsorbed on the surface of Na-hectorite at pH 3.0. The fluorescence intensity of imazethapyr was enhanced only about 20% by Na-hectorite. The peaks were shifted to the right and only one "species" appeared (410 nm) at pH 8.0. Two peaks appeared when solution pH was 3.0 (Figure 24). The peak at

365 nm may represent the protonated form and that at 335 nm is the uncharged molecular form. As solution pH increases from 3.0 to 5.0, a new peak appears at 410 nm, suggestory an increase in the concentration of the uncharged molecular species is changing to the anionic form because the pKa value iue 2 Ttain fursec eiso seta of spectra emission fluorescence - Titration 22. Figure

Intensity (arbitrary units) 200 300 100 400 500 600 mzqi auos ouin rmp 27 t 5.33. to 2.72 pH from solution aqueous imazaquin 300 350 5.33 4.38 3.78 3.25 2.72 400 (nm) h t g n e l e v a W 450 0 550 500 600 130 Figure 23. Titration - fluorescence emission spectra of of spectra emission fluorescence - Titration 23. Figure

Intensity (arbitrary units) mzqi auos ouin rmp 56 t 9.00. to 5.68 pH from solution aqueous imazaquin 200 300 400 100 100 500 600 0 0 30 0 40 0 50 600 550 500 450 400 350 300

9.00 8.23 7.76 7.12 6.76 6.44 5.68 aeegh (nm) Wavelength 131 iue2. h fursec msinsetao imazethapyr of spectra emission fluorescence The 24. Figure

Intensity (arbitrary units) 200 250 100 300 150 solution at three pH levels. pH three at solution 50 0 300 aeegh (nm) Wavelength 0 500 400 A H 8 * pH— 0 C B B ------p—5.0 pH— p—3.0 pH— 132 Figure 25. The fluorescence emission spectra of imazethapyr of spectra emission fluorescence The 25. Figure

Intensity (arbitrary units) 200 250 300 350 100 150 50 adsorbed onto Na-hectorite at three pH levels. pH three at Na-hectorite onto adsorbed 300 aeegh (nm) Wavelength 400 pH-8.0 500 133 134 of imazethapyr is 3.9. The different behaviors of imazaquin and imazethapyr in aqueous solution may be due to chemical structureal differences. Imazaquin displays more aromatic character due to its quinoline ring, compared to the pyridine ring of imazethapyr, which allows delocalization of the positive charge that stabilized the zwitterion and formed a "polymer- like" structure by hydrogen bonding. However, the proposed mechanism must be further investigated using other techniques such as FTIR or NMR spectroscopy.

Conclusions

In general, the order of affinity of imidazolinone herbicide for humic acids at all three pH levels was : imazamethabenz > imazethapyr > imazaquin, which follows the order of herbicide values. These herbicides may interact hydrophobically with humic acids at lower soil pH.

The fluorescence quenching method is a quick and easy method to determine the sorption constants of imidazolinone herbicides on humic acid. This technique does not involve the time-comsuming separation steps which usually cause error, and the initial concentration of herbicides need not be determined accurately. The high analytical sensitivity and low detection 135 limit of fluorescence spectroscopy allows study of the adsorption of herbicides on colloidal surfaces, especially those with low water solubilities such as the imidazolinone herbicides. However, this technique is not applicable in systems where fluorescence is quenched by the presence of adsorbents. For example, sorption of imazapyr can not be studied with this technique because the fluorescence is quenched by humic acids at its specific exitation/emission wavelength. The effect of pH on the sorption of these herbicides may be due to hydrogen bonding and protonation mechanisms, which were observed using fluorescence emission spectra. The imazaquin may perform differently comparing to imazethapyr in aqueous solution at various pH levels. However, further investigation is required using other techniques such as FTIR or NMR spectroscopy. 136

References Basham, G.W., T.L. Lavy, L.R. Oliver, and H.D. Scott. 1987. Imazaquin persistence and mobility in three Arkansas soils. Weed Sci. 35 : 576-582. Dolling, A.M. 1985. Studies of Interactions of Some Imidazolinone Herbicides with Clays. M.Sc. Thesis, Department of Chemistry, University of Birmingham, England. 198 pp. Edward J.T. and I. Lantos. 1972. Solvent effects on the tautomeric equilibria of 4,4/5,5-disubstituted 4H/5H-imidazol- 5/4-ones. J. Heterocycle Chem. 9 : 363-369. Everett, D.H. 1972. Manual of Symbols and Terminology for Physicochemical Quantities and Units. Appendix II : Definitions, Terminology and Symbols in Colloid and Surface Chemistry. Butterworths, London. Gauthier, T.D., E.C. Shane, W.F. Guerin, W.R. Seitz, and C.L. Grant. 1986. Fluorescence quenching method for determining equilibrium constants for polycyclic aromatic hydrocarbons binding to dissolved humic materials. Environ. Sci. Technol. 20 : 1162-1166. Goetz, A.J., G. Wehtje, R.H. Walker, and B. Hajek. 1986. Soil solution and mobility characterization of imazaquin. Weed Sci. 34 : 788-793.

Green, R.W. and H.K. Tong. 1956. The constitution of the pyridine monocarboxylic acids in their isoelectric forms. J. Cm. Chem. Soc. 78 : 4896-4900. Hausler, M.J. 1986. Studies of Interactions of Some Imidazolinone Herbicides with Whole Soil, Oxyhydroxides, and with Natural and Synthetic Humic Acids. Ph.D. Thesis, Department of Chemistry, University of Birmingham, England. 228 pp.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989b. Adsorption of imazaquin and imazethapyr on soils, sediments, and selected adsorbents. Weed Sci. 37 : 712-718.

Parker, C. A. 1968. Photoluminescence of Solutions. Elsevier, New York. pp. 220-232.

Traina, S.J., D.A. Spontak, and T.J. Logan. 1989. Effects of cations on complexation of naphthalene by water soluble organic carbon. J. Environ. Qual. 18 : 221-227. 137 Willard, H.H., L.L. Merritt, Jr., J.A. Dean, and F.A. Settle, Jr. 1988. Fluorescence and phosphorescence spectrophotometry, in Instrumental Methods of Analysis. 7th ed. p. 197-223. Wadsworth Publishing Co., Belmont, California. CONCLUSIONS

The adsorption of imidazolinone herbicides is affected by soil pH, the nature of herbicide, and the type and amount of organic matter and clay in soils. The sorption of imidazolinone herbicides in soils and selected adsorbents decreased as pH increased, but the decrease in sorption was most evident as soil pH increased from 4.5 to 5.5. The sorption of these herbicides on clay was more dependent upon solution pH, compared to the sorption on humic acids. The sorption of imazethapyr on soils and clays was more likely to be controlled by adsorption process, especially at lower pH levels, whereas the sorption of imazaquin seemed to be controlled by partitioning process. Humic acid had much greater affinity for imidazolinone herbicides than clay did, indicating that organic matter content could be the major factor controlling the sorption of these herbicides in soils, especially in soils containing high organic matter content. Illite and kaolinite, which exhibit more pH-dependent charge characteristics and less total surface area than hectorite, can impart important adsorptive character to soil, especially with regard to imazethapyr at lower pH levels.

138 139 The desorption of ixnazaquin from soil decreased as soil moisture decreased and maintained for only 48 hours, indicating rainfall and soil moisture status after herbicide application may affect the desorption of imazaquin under field conditions. There were no significant hysteresis effects at herbicide concentration lower than 1*0 mg/L, indicating that imidazolinone herbicides were weakly sorbed by the soils, and leaching may take place under wet conditions. By comparing the batch equilibrium method, the fluorescence quenching method can be used to accurately and quickly measure the sorption constants for imidazolinone herbicides on humic acid, without the separation of adsorbent and solution phases necessary in the batch equilibrium method. The effect of pH on the sorption of imidazolinone herbicides may be due to hydrogen bonding and protonation mechanism, which were observed by fluorescence emission spectra. Comparing to imazethapyr, the imazaquin may perform differently in aqueous solution at various pH levels. However, further investigation is required using other techniques such as FTIR or NMR Spectroscopy. APPENDICES

140 Appendix A

Data Relative to Chapter II

141 142 Table A-l. Desorption of imazaquin and imazethapyr from soil.

Imazaquin Desorption Cvcle Soil Concentrat ion Moisture8 D1 D2 D3 D4 (mq/L)____ Treatment % Desorbed ---- Hovtville pH 4.5 0.024 w/o 42 16 11 3 0.1 w/o 46 21 10 4 1.0 w/o 45 20 10 3 5.0 w/o 41 16 8 4 10.0 w/o 40 23 10 5 0.1 w 36 11 5 3

pH 5.5 0.024 w/o 0 0 2 2 0.1 w/o 0 22 5 0 1.0 w/o 0 9 6 1 5.0 w/o 0 16 5 0 10.0 w/o 0 10 5 2 0.1 w 0 0 11 4 pH 6.6 0.024 w/o 0 0 7 7 0.1 w/o 0 9 7 5 1.0 w/o 0 19 35 3 5.0 w/o 0 11 22 9 10.0 w/o 0 39 34 18 0.1 w 18 0 1 1 Crosby pH 4.6 0. 024 w/o 46 16 9 --- 0.1 w/o 53 18 11 8 1.0 w/o 41 18 7 --- 5.0 w/o 40 24 8 --- 10.0 w/o 49 22 16 --- 0.1 w 10 25 11 8 143 Table A-l (Continued) Imazethapyr Soil Concentrat i on Moisture8 D1 D2 D3 D4 (ma/L) Treatment ---- % Desorbed --- Hovtville pH 4.5 0.043 w/o 30 20 13 7 0.1 w/o 29 19 14 8 1.0 w/o 32 20 13 8 5.0 w/o 37 21 11 7 10.0 w/o 36 21 12 7 0.1 w 31 18 11 6 pH 55 0.043 w/o 38 14 8 6 0.1 w/o 39 17 10 6 1.0 w/o 36 14 9 5 5.0 w/o 40 13 9 7 10.0 w/o 25 11 7 4 0.1 w 34 15 11 4 pH 6.6 0.043 w/o 16 11 11 7 0.1 w/o 25 14 13 6 1.0 w/o 12 9 11 8 5.0 w/o 0 17 24 21 10.0 w/o 3 12 15 11 0.1 w 13 14 7 4 Crosbv pH 4.6 0.043 w/o 38 19 10 5 0.1 w/o 37 19 10 6 1.0 w/o 38 17 10 6 5.0 w/o 38 18 12 6 10.0 w/o 34 17 11 6 0.1 w 40 16 8 5 pH 5.8 0.043 w/o 20 1 8 3 0.1 w/o 21 5 9 4 1.0 w/o 3 5 6 5 5.0 w/o 0 0 10 6 10.0 w/o 0 18 22 15 0.1 w 11 14 14 7 pH 6.8 0.043 w/o 0 0 30 5 0.1 w/o 0 2 17 6 1.0 w/o 0 7 4 15 5.0 w/o 0 0 0 0 10.0 w/o 0 0 0 0 0.1 w 0 45 23 13

8 The moisture treatment was tempararily drying the soil to 5% moisture content and maintained for 48 hours, "w/o" means "samples without the moisture treatment", "w" means "samples with the moisture treatment". 35 30 pH 3.0 Kaolinite 25 20 15 Ceq(nmol/mL) 10 Imazethapyr 5 A O O Imazaquin imazethapyr on kaolinite = (pH 3.0). 0 50 150 100 200

(3/louiu) pB0 Figure A-l. The sorption isotherm for imazaquin and iueA2 Te opiniohr fr mzqi and imazaquin for isotherm sorption The A-2. Figure

CJ (nmol/g) 200 100 250 150 300 50 imazethapyr on hectorite (pH = 3.0). (pH= hectorite on imazethapyr Imazaquin O A 5 Imazethapyr 10 Ce^(nmol/mL) 15 H 3.0 pH Hectorite 20 530 25 Figure A-3. The sorption isotherm for imazaquin and imazaquin for isotherm sorption The A-3. Figure

u (nm ol/g) 1000 200 400 800 600 mztay nilt (pH3.0). » illite on imazethapyr Imazaquin O Imazethapyr A 20 06 80 60 40 e (mo/ L) ol/m Ceq(nm H 3.0 pH Illite 100 iueA4 Te opin stem o iaaun and imazaquin for isotherm sorption The A-4. Figure

(nm ol/g) 200 400 600 800 2 6 10 8 6 4 2 0 mztay nAdihhmcai (pH3.0).= acid humic on Aldrich imazethapyr O A o A g jn lmL) Cg(j(nm ol/m A Imazaquin O lrc HA Aldrich H 3.0 pH Imazethapyr 147 iue -. h srto iohr fr mzqi and imazaquin for isotherm sorption The A-5. Figure o

T3 (nm ol/g) (8 20 30 40 50 60 70 0.0 mztay nHyvlehmcai (H= 3.0). (pH= acid humic onHoytville imazethapyr Imazaquin O A 0.2 Imazethapyr 0.4 mo/ ) L ol/m nm ^ C 0.6 0.8 H 3.0 pH otil HA- Hoytville 1.0 1.2 148 Appendix B

Data Relative to Chapter III

149 150 Table B-l. The fluorescence intensity for the adsorption of imazapyr on humic acids at three pH levels.

Humic Acid Aldrich HA Concentration PH 3 OH 5 d H 8 (mg/L) A 8 Bd A B A B

0 99.9 13.2 100.3 18.2 100.3 16.8 9.8 108.2 62.5 165.1 129.0 101.0 18.2 19.2 133.5 49.8 170.7 132.4 110.5 30.2 28.3 124.8 62.6 127.0 130.1 146.9 53.2 37.0 102.9 69.2 149.5 136.1 144.5 110.6 45.5 97.0 71.2 125.6 123.4 142.6 110.9 53.6 92.8 62.0 109.5 111.9 176.9 109.9 61.4 113.7 77.1 126.0 104.2 156.9 109.1

Hovtville HA d H 3 d H 5 d H 8 A B A B A B

0 101.7 16.2 100.5 15.8 100.4 4.2 4.0 132.0 19.3 112.5 43.4 104.3 43.9 7.9 113.2 19.2 122.8 44.0 88.8 44.1 11.7 102.7 26.6 94.6 45.6 87.1 51.4 15.5 113.0 31.0 99.8 55.5 105.9 59.2 19.2 102.8 33.2 157.2 53.1 119.4 63.8 22.9 120.5 32.0 117.6 74.6 128.7 66.3 26.5 107.3 35.8 96.9 57.2 117.0 69.2 37.0 129.8 44.4 117.9 66.1 121.0 71.5

8 The fluorescence intensity of imazapyr-humic acid system. b The fluorescence intensity of humic acid only (background fluorescence). 151 Table B-2. The fluorescence intensity for the adsorption of imazamethabenz on humic acids at three pH levels.

Humic Acid Aldrich HA Concentration DH 3 DH 5 DH 8 (mg/L) A a BA BA B pH=3 or pH=5 pH=8 only 0 0 100.0 0.3 100.0 1.6 100.0 2.5 4.0 4.9 92.5 1.4 91.0 1.2 105.9 80.0 7.9 9.8 76.8 3.0 86.1 7.3 133.8 81.0 11.7 19.2 67.2 4.5 57.0 8.6 166.2 89.8 15.5 28.3 61.8 4.3 46.5 8.5 141.9 89.8 19.2 37.0 55.3 4.1 37.9 8.4 127.6 80.0 22.9 45.5 49.7 4.0 33.0 8.5 26.5 42.3 4.1

Hovtville HA d H 3 DH 5 d H 8 A B A B A B

0 99.9 0.7 100.1 1.2 99.6 1.3 4.0 81.8 1.8 97.0 3.0 81.5 2.5 7.9 80.1 2.0 94.8 5.8 84.2 4.5 11.7 81.0 2.6 93.7 7.3 90.1 7.6 15.5 77.2 3.1 90.4 9.2 81.9 8.3 19.2 74.6 3.7 87.3 9.8 80.6 9.6 22.9 73.0 4.1 79.8 10.3 81.2 10.9 26.5 66.5 4.3 70.5 11.2 76.9 12.8 30.1 66.2 4.3 60.2 10.9 70.5 14.6 68.2 13.2

a The fluorescence intensity of imazamethabenz-humic acid s y s t e m .

b The fluorescence intensity of humic acid only (background fluorescence). 152 Table B-3. The fluorescence intensity for the adsorption of imazaquin and imazethapyr on Na-hectorite at three pH levels.

Na-Hectorite imazacmin Concentration p H 3 p H 5 p H 8 (mg/L) A a Bd A B A B

0 99.6 1.3 100.2 1.4 100.0 1.5 75.8 76.5 1.3 86.4 1.4 98.3 1.5 228 61.6 1.3 81.6 1.4 87.7 1.5 567 68.9 1.5 78.9 1.6 83.7 1.9 900 69.2 2.2 72.9 2.3 84.0 2.4 1218 54.5 2.8 77.9 3.0 81.7 2.5 1526 58.9 3.1 70.6 3.5 79.0 3.6 2113 50.6 3.4 70.6 3.5 2926 71.9 4.0 3180 71.8 5.1 3428 63. 3 5.9

Imazethapyr PH 3 PH 5 P H 8 A B A B A B

0 100.1 0.3 100.1 0.4 100.5 0.4 392 92.8 0.3 88.0 0.4 97.2 0.4 769 71.5 0.4 83.8 0.6 84.6 0.6 1481 62.0 0.5 82.5 0.8 83.5 0.8 3333 52.4 0.6 82.1 0.9 82.4 0.9 4375 74.9 0.9 77.7 0.9 5294 69.9 1.0 80.1 0.9 6111 65.8 1.1 75.2 1.1 6842 65.7 1.1 73.6 1.1 7500 58.0 1.2 71.0 1.1 8095 57.1 1.5 69.7 1.2 8636 55.8 2.4

a The fluorescence intensity of herbicide-clay system. b The fluorescence intensity of clay only (background fluorescence). 153

Table B-4. The fluorescence intensity for the adsorption of imazapyr and imazamethabenz on Na-hectorite at three pH levels.

Na-Hectorite XmazaDvr Concentration t>H 3 d H 5 d H 8 (mg/L) Aa B° A B A B

0 100.1 10.2 97.8 11.7 98.7 17.4 392 92.5 11.2 113.7 12.8 121.1 21.9 769 152.8 12.5 117.5 14.0 120.1 24.4 1132 165.8 14.1 128.0 15.8 132.5 30.6 2143 209.0 22.2 155.0 21.0 169.9 54.6 3333 263.7 37.0 134.8 28.4 158.2 71.4

Imazamethabenz DH 3 d H 5 OH 8 A B A B A B

0 100.0 0.5 100.4 1.3 101.2 1.9 237 104.0 0.6 138.0 1.7 113.6 4.6 469 108.9 0.6 129.0 2.3 121.1 4.6 695 118.0 0.7 134.0 2.0 125.5 4.3 1343 128.0 0.9 147.5 2.8 131.8 6.4 2143 139.8 2.2 157.5 4.4 136.0 9.4 3051 152.4 3.3 178.8 6.1 137.2 11.0 3871 166.5 6.2 207.8 8.8 139.8 14.2

a The fluorescence intensity of herbicide-clay system. b The fluorescnece intensity of clay only (background fluorescence). 154 Table B-5. The effect of solution temperature on the fluorescence intensity of imazaquin and imazethapyr in the presence of Aldrich humic acid, Hoytville humic acid, or Na-hectorite.®

Temoerature f°lO Treatment 274 285 298 308 323 336 348 -- Fluorescence Intensity (arbitrary units) — Imazaquin 105.6 95.1 87.3 77.9 71.1 64.0 60.1

Imazaquin + 69.0 61.4 57.4 51.9 46.2 41.0 40.1 Aldrich HA

Imazaquin + 72.8 66.5 59.4 54.1 49.0 43.5 42.9 Hoytville HA Imazaquin + 58.7 53.1 49.0 44.5 40.4 36.7 35.1 Na-hectorite

Imazethapyr 87.5 83.4 84.0 86.3 85.0 88.2 90.2 Imazethapyr + 37.8 36.9 36.8 37.8 37.8 39.2 40.4 Aldrich HA imazethapyr *4- 58.3 55.2 56.8 59.5 58.6 62.1 63.5 Hoytville HA Imazethapyr + 70.0 67.2 67.7 70.7 69.7 72.9 77.1 Na-hectorite

8 The concentration of Aldrich HA, Hoytville HA, and Na- hectorite were 50.0, 30.0, and 3330 mg/L, respectively. 155 Table B-6 . The titration data of fluorescence emission spectra for imazaquin and imazethapyr.

Imazacmin______Imazethapyr ml Of NaOH ml of NaOH pH (0.05 N) PH (0.025 N)

2.72 2.64 3.01 0.364 2.80 0.500 3.25 0.168 3.01 0.500 3.49 0.110 ml of NaOH 3.78 0.058 ...PH f0.05 N> 4.05 0.041 3.25 0.240 4.38 0.025 3.49 0.149 ml of NaOH 3.77 0.098 PH f0.025 m 4.03 0.069 4.77 0.036 4.24 0.040 5.33 0.027 _1,4? . 0.020 5.68 0.009 ml of NaOH 6.11 0.015 PH , f 0.025 N) 6.44 0.013 4.74 0.031 6.76 0.017 5.19 0.021 7.12 0.018 5.66 0.029 7.45 0.015 6.10 0.014 7.76 0.019 6.60 0.015 8.00 0.017 6.93 0.021 8.23 0.016 7.30 0.017 8.49 0.021 7.68 0.016 9.00 0.041 8.06 0.017 8.30 0.020 8.50 0.017 9.00 0.038 iue -. h fursec eiso seta f imazaquin of spectra emission fluorescence The B-l. Figure

Intensity (arbitrary units) 1200 1500 300 600 900 300 — Na-hectorite at three pH levels. pH atthree Na-hectorite — A C C B — pH"5.0 — B 350 -- -- pH-8.0 pH-3.0 0 0 550 500 400 aeegh (nm) Wavelength 450 600 156 iueB2 The fluorescence emission spectra of Na-hectorite Figure B-2.

Intensity (arbitrary units) 200 300 400 500 100 300 imazaquin) at three pH levels. pH three at imazaquin) ueectto/msinwvlnt for wavelength (useexcitation/emission 350 400 aeegh (nm) Wavelength 450 pH-5.0 .0 3 - H p 500 550 600 157 Figure B-3. The fluorescence emission spectra of imazethapyr of spectra emission fluorescence The B-3. Figure

Intensity (arbitrary units) 200 250 300 350 400 100 150 50 — Na-hectorite at three pH levels. pH three at Na-hectorite — 300 aeegh (nm) Wavelength 400 pH"3.0 pH-5.0 pH"8.0 500 158 Figure B-4. The The fluorescence emission Figure spectra B-4.of Na-hectorite

Intensity (arbitrary units) 100 200 150 250 50 0 L_ 0 300 imazethapyr) at three pH levels. atpH three imazethapyr) ueectto/msinwvlnt for wavelength (useexcitation/emission aeegh (nm) Wavelength 400 .0 5 - H p pH-3.0 500 159 iue -. irto-loecne msin pcr of spectra emission Titration-fluorescence B-5. Figure

Intensity (arbitrary units) 200 100 250 150 300 350 50 5.19. imazethapyr aqueous solution from pH 2.64 to to 2.64 pH from solution aqueous imazethapyr 300 350 avelength (nm) h t g n e l e v Wa 400 5 500 450 5.19 4.03 3.77 3.49 2.64 550 160 iue -. irto-loecne msin pcr of spectra emission Titration-fluorescence B-6. Figure

Intensity (arbitrary units) 200 250 300 100 150 50 0 0 300 9.00. imazethapyr aqueous solution from pH 5.66 to to 5.66 pH from solution aqueous imazethapyr 5 0 450 400 350 (nm) h t g n e l e v a W 3 2 1 500 9.00 8.06 7.30 6.93 6.60 6.10 S.66 PH

550 161 REFERENCES

Abdul, A.S., T.L. Gibson, and D.N. Rai, 1990. Use of humic acid solution to remove organic contaminants from hydrogeologic systems. Environ. Sci. Technol. 24 : 328-333. American Cyanamid Co. 1986. Technical Information Report on Imazaquin. Agric. Div., Wayne, NJ 07470. American Cyanamid Co. 1987. Technical Information Report on Imazethapyr. Agric. Res. Div., Princeton, NJ 08540. American Cyanamid Co. 1988. Technical Information Report on Imazapyr. Agric. Res. Div., Princeton, NJ 08540. American Cyanamid Co. 1989. Technical Information Report on AC 222,293 Herbicide. Agric. Res. Div., Princeton, NJ 08540. Bailey, G.W., and J.L. White. 1970. Factors influencing the adsorption, desorption, and movement of pesticides in soil. Residue Rev. 32 : 29-92.

Basham, G.W., T.L. Lavy, L.R. Oliver, and H.D. Scott. 1987. Imazaquin persistence and mobility in three Arkansas soils. Weed Sci. 35 : 576-582.

Bohn, H.L., B.L. McNeal, and G.A. O'Connor. 1979. Soil Chemistry. John Wiley & Sons, New York. 329 pp. Borchardt, G. 1989. Smectites, p. 675-728. In J.B. Dixon and S.B. Weed (ed.) Minerals in Soil Environments. 2nd ed. Soil Sci. Soc. Am. Madison, WI.

Bowman, B.T. 1979. Method of repeated additions for generating pesticide adsorption-desorption isotherm data. Can. J. Soil Sci. 59 : 435-437.

Calvet, R. 1980. Adsorption-desorption phenomena, in R.J. Hance (ed.) In "Interactions between Herbicides and the Soil." p. 1-30. Academic Press Inc., Ner York.

Chapman, H.D. 1965. Cation exchange capacity, p. 891-901. In C.A. Black (ed.) Methods of Soil Analysis. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI.

162 163 Charlet, L. 1986. Adsorption of some macronutrient ions on an Oxisol. An application of the Triple Layer model, Ph.D. dissertation, Univ. of California, Riverside. Chiou, C.T. 1989. Theoretical considerations of the partition uptake of nonionic compounds by soil organic matter. In : B.L. Sawhney and K. Brown (ed.) Reactions and Movement of Organic Chemicals in Soils. Soil Sci. Soc. Am., Madison, WI., Spec. Publ. No. 22, pp. 1-30. Chiou, C.T., R.L. Malcolm, T.I. Brinton, and D.E. Kile. 1986. Water solubility enhancement of some organic pollutants and pesticides by dissolved humic and fulvic acids. Environ. Sci. Technol. 20 : 502. Chiou, C.T., P.E. Porter, and D.W. Schmedding. 1983. Partition equilibria of nonionic organic compounds between soil organic matter and water. Environ. Sci. Technol., 17 : 227-231. Corey, R.B. 1981. Adsorption vs precipitation, in Adsorption of Inorganics at Solid-Liquid Interfaces (M.A. Anderson and A.J. Rubin, eds). Ann Arbor Science, Ann Arbor, Michigan. Curran, W.S. and E.L. Knake. 1989. Effect of herbicide combinations on rotational corn. NCWSS Research Report. 46 : 274.

Day, P.R. 1965. Particle fractionation and particle size analysis, p. 545-567. In C.A. Black (ed.) Method of Soil Analysis. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI. Dolling, A.M. 1985. Studies of Interactions of Some Imidazolinone Herbicides with Clays. M.Sc. Thesis, Department of Chemistry, University of Birmingham, England. 198 pp. Edward J.T. and I. Lantos. 1972. Solvent effects on the tautomeric equilibria of 4,4/5,5-disubstituted 4H/5H-imidazol- 5/4-ones. J. Heterocycle Chem. 9 : 363-369. Everett, D.H. 1972. Manual of Symbols and Terminology for Physicochemical Quantities and Units. Appendix II : Definitions, Terminology and Symbols in Colloid and Surface Chemistry. Butterworths, London.

Fanning, D.S., V.Z. Keramidas, and M.A. El-Desoky. 1989. Micas. p. 551-634 in J.B. Dixon and S.B. Weed (eds.) Minerals in Soil Environments. 2nd ed. Soil Sci. Soc. Amer. Madison, WI. 164 Gamar, Y. and M.A. Mustafa. 1975. Adsorption and desorption of diquat2* and paraquat2* on arid-zone soils. Soil Sci. 119 : 290-295. Gauthier, T.O., E.C. Shane, W.F. Guerin, W.R. Seitz, and C.L. Grant. 1986. Fluorescence quenching method for determining equilibrium constants for polycyclic aromatic hydrocarbons binding to dissolved humic materials. Environ. Sci. Technol. 20 : 1162-1166. Giles, C.H., T.H. MacEwan, S.N. Nakhwa, and D. Smith. 1960. Studies in adsorption. Part XI. A system of classification of solution adsorption isotherms, and its use in diagnosis of adsorption mechanisms and in measurement of specific surface areas of solids. J. Chem. Sco. p. 3973-3993. Goetz, A.J., G. Wehtje, R.H. Walker, and B. Hajek. 1986. Soil solution and mobility characterization of imazaquin. Weed Sci. 34 : 788-793. Green, R.E. 1974. Pesticide-clay-water interactions, p. 3- 37 in W.D. Guenze (ed.) Pesticides in soil and water. Soil Sci. Sco. Am. Madison, WI.

Green, R.W. and H.K. Tong. 1956. The constitution of the pyridine monocarboxylic acids in their isoelectric forms. J. Cm. Chem. Soc. 78 : 4896-4900. Greenland, D.J. and C.J.B. Mott. 1978. Surface of soil particle. P.32 in D.J. Greenland and M.H.B. Hayes (eds.), The Chemistry of Soil Constituents. John Wiley & Sons, Inc., Chichester.

Grover, R. 1975. Adsorption and desorption of urea herbicides on soils. Can. J. Soil Sci. 55 : 127-135. Gunsolus, J.L., R.H. Behrens, W.E. Lueschen, D.D. Warnes, and J.V. Wiersma. 1986. Carryover potential of AC-263,499, DPX- F6025, FMC5702 0 and imazaquin in Minnesota. North Cent. Weed Control Conf. Proc. 41 : 52.

Hackett, N. 1990. Imazaquin behavior in the soil. Technical report of American Cyanamid Company. Agri. Div., Wayne, NJ 07470.

Hamaker, J.W. and J.M. Thompson. 1972. Adsorption, p. 49- 143. In C.A. Goring and J.W. Hamaker, (ed.) Organic Chemicals in the Soil Environment. Marcel-Dekker, New York. 165 Hasset, J.J., W.L. Banwart, and R.A. Griffin. 1983. Correlations of compound properties with sorption characterisitcs of nonpolar compounds by soils and sediments : concepts and limitations. In : C.W. Francis and S.I. Auerbach (ed.), Environmental and Solid Wastes. Butterworths, Boston, MA. pp. 161-178. Hausler, M.J. 1986. Studies of Interactions of Some Imidazolinone Herbicides with Whole Soil, Oxyhydroxides, and with Natural and Synthetic Humic Acids. Ph.D. Thesis, Department of Chemistry, University of Birmingham, England. 228 pp. Jaynes, W.F. and J.M. Bigham. 1987. Charge reduction, octahedral charge, and lithium retention in heated, Li- saturated smectites. Clays Clay Miner. 35 : 440-448. Karickhoff, S.W., D.S. Brown, and T.A. Scott. 1979. Sorption of hydrophobic organic pollutants on natural sediments. Water Res. 13 : 241-248.

Koskinen, W.C., G.A. O'Connor, and H.H. Cheng. 1979. Characterization of hysteresis in the desorption of 2,4,5-T from soils. Soil Sci. Soc. Am. J. 43 : 871-874. Los, M., D.R. Carliante, E.M. Ettinghouse, and P.J. Wepplo. 1983. O-(5-oxo-2-imidazolin-2-yl)-arylcarboxylates : a new class of herbicides. American Cyanamid Co. Agric. Res. Div., Princeton, NJ 08540.

Los, M., P.J. Wepplo, R.K. Russell, B.L. Lences, and P.L. Orwick. 1984. O-(5-oxo-2-imidazolin-2-yl)-arylcarboxylates : a new class of herbicides. 4. AC 263,499, a herbicide for use in legumes. Book of Abstracts, 188th Meet. Aroer. Chem. Soc., Abstract 30 PEST.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989a. Availability and persistence of imazaquin, imazethapyr, and clomazone in soil. Weed Sci. 37 : 259-267.

Loux, M.M., R.A. Liebl, and F.W. Slife. 1989b. Adsorption of imazaquin and imazethapyr on soils, sediments, and selected adsorbents. Weed Sci. 37 : 712-718.

McEwen, F.L., and G.R. Stephenson. 1979. The Use and Significance of Pesticides in the Environment. John. Wiley & Sons, Inc., New York. van Olphen, H. and J.J. Fripiat (eds.). 1979. Data Handbook for Clay Materials and Other Non-metallic Minerals. Pergamon Press, Oxford. 166 Parker# C.A. 1968. Photoluminescence of Solutions. Elsevier, New York. pp. 220-232. Peoples, T.R., T. Wang, R.R. Fine, P.L. Orwick, S.E. Grahan, and K. Kirkland. 1985. AC 263,499 : a new broad-spectrum herbicide for use in soybeans and other legumes. Proc. Brit. Crop. Prot. Conf.-Weeds; 2 : 99-106. Pignatello, J.J. 1989. Dynamics of organic compounds in soils and sediments. In Reactions and Movement of Organic Chemicals in Soils. Sawhney & Brown (ed.) p. 45-80. SSSA Special Publication Number 22. Soil Sci. Soci. Amer., Inc., Madison, WI. Renner, K.A., W.F. Meggitt, and R.A. Leavitt. 1988a. Influence of rate, method of application and tillage on imazaquin persistence in soil. Weed Sci. 36 : 90-95. Renner, K.A., W.F. Meggitt, and D. Penner. 1988b. Effect of soil pH on imazaquin and imazethapyr adsorption to soil and phytotoxicity to corn (Zea mays). Weed Sci. 36 : 78-83. Schnitzer, M. 1982. Organic matter characterization, p. 581- 594. In A.L. Page et al. (ed.) Methods of Soil Analysis. Part 2. Chemical and microbial properties. 2nd ed. Agron. Ser. #9 : Am. Soc. Agron. Madison, WI.

Shaner, D.L., P.C. Anderson, and M.A. Stidham. 1984. Imidazolinones. Potent inhibitors of acetohydroxyacid synthase. Plant Physiol. 76 : 545. Shaner, D.L. and P. A. Robson. 1985. Adsorption, translocation, and metabolism of AC 252,214 in soybean (Glycine max), common cocklebur (Xantbium strumarium), and velvetleaf (AbutiIon theophrasti). Weed Sci. 33 : 469-471. Sposito, G. 1984. The Surface Chemistry of Soils, p. 113- 153. Oxford University Press, Inc. New York. Sposito, G. 1989. The Chemistry of Soils. p. 127-169. Oxford University Press, Inc. New York. Stevenson, F.J. 1982. Humus Chemistry : Genesis, Composition, Reactions. John Wiley and Sons, New York. 443 pp.

Stougaard, R.N., P.J. Shea, and A.R. Martin. 1985. The effect of soil type and pH on AC-263,499 and imazaquin. NCWSS Proc. 40 : 15-16. 167 Stougaard, R.N., P.J. Shea, and A.R. Martin. 1990. Effect of soil type and pH on adsorption, mobility, and efficacy of imazaquin and imazethapyr. Weed Sci. 38 : 67-73. Theisen, A. and M.E. Harward. 1962. A paste method for preparation of slides for clay mineral identification by x- ray diffraction. Soil Sci. Soc. Am. Proc. 26 : 90-91. Traina, S.J., D.A. Spontak, and T.J. Logan. 1989. Effects of cations on complexation of naphthalene by water soluble organic carbon. J. Environ. Qual. 18 : 221-227. Upchurch, R.P. 1966. Behavior of herbicides in soil. Res. Rev. 16 : 46-85. Walker, A., and D.V. Crawford. 1968. The role of organic matter in adsorption of the triazine herbicides by soils, p. 91-108. In Isotopes and radiation in soil organic matter studies. Proc. 2nd Symp. Int. Atomic Energy Agency, Vienna. Weber, J.B. and C.T. Miller. 1989. Organic chemical movement over and through soil. p. 305-334. In B.L. Sawhney and K. Brown (ed.) Reactions and movement of organic chemicals in soils. Soil Sci. Soc. Am., Inc. Madison, WI. Weed, S.B. and J.B. Weber. 1974. Pesticide-organic matter interactions. p. 39-66 in W.D. Guenzi, ed. Pesticides in Soil and Water. Soil Sci. Soc. Am. Madison, WI. Wehtje, G., R. Dickens, J.W. Wilcut, and B.F. Hajek. 1987. Sorption and mobility of sulfometuron and imazapyr in five Alabama soils. Weed Sci. 35 : 858-864. Willard, H.H., L.L. Merritt, Jr., J.A. Dean, and F.A. Settle, Jr. 1988. Fluorescence and phosphorescence spectrophotometry. In Instrumental Methods of Analysis. 7th ed. p. 197-223. Wadsworth Publishing Co., Belmont, California.