IAEA-TECDOC-768

Research and development of controlled release formulations of

VolumeI Development and evaluation of controlled release formulations of pesticides

Proceedings seminara of organized by the Joint FAO/IAEA Division Nuclearof Techniques Food in Agriculture and held Vienna,and in September6-9 1993

INTERNATIONAL ATOMIC ENERGY AGENCY

October 1994 The IAEA does not normally maintain stocks of reports in this series. However, microfiche copies of these reports can be obtained from

INIS Clearinghouse International Atomic Energy Agency Wagramerstrasse5 0 10 x P.OBo . A-1400 Vienna, Austria

Orders should be accompanied by prepayment of Austrian Schillings 100, for e for e chequa th f m th IAEf m o n i o n i r Aeo microfiche service coupons orderee whicb y hdma separately fro INIe mth S Clearinghouse. The originating Section of this document in the IAEA was: Residued san s Section International Atomic Energy Agency Wagramerstrasse5 P.O. Box 100 A-1400 Vienna, Austria

RESEARC DEVELOPMEND HAN F TO CONTROLLED RELEASE FORMULATIONS OF PESTICIDES VOLUME I DEVELOPMENT AND EVALUATION OF CONTROLLED RELEASE FORMULATION PESTICIDEF SO S IAEA, VIENNA, 1994 IAEA-TECDOC-768 ISSN 1011-4289 Printed by the IAEA in Austria October 1994 FOREWORD

Pesticides are commonly used for the control of pests of agriculture and vectors of human and animal diseases. However, their use may involve risks to human health and the environment . Mosth f o t e released intenvironmene oth loss i t t before reachins git target. The loss occurs due to physical, chemical or biological factors. Some pesticides are more persistent than others e morTh .e persistent pesticide n accumulatca s e th i h e environment and, thereby, prove hazardou non-targeo st t species. Therefore, ther greates ei r emphasis on phasing those persistent pesticides out and on developing less persistent but more selective pesticides.

Because of the cost and other limitations in the development of new pesticides increasing attentio beins ni gdesige paith f o improvedt no d deliver pesticidese th f yo n A . effective way to improve the performance of the pesticides is to improve formulations. Controlled release technology offers the opportunity to develop formulations which can improv performance eth f pesticideo e increasiny sb g their efficac safetd makind an y an y g them environmentally less harmful. Recognizing these potential advantages in the use of controlled release technolog Joine yth t FAO/IAEA Divisio Nucleaf no r Technique i Foosh d and Agriculture in 1988 initiated the Co-ordinated Research Programme on Research and Developmen f Controlleo t d Release Formulation f Pesticideso s s designewa t I . assiso dt t scientist developinf so g Member State conduco st t researc developmene th n hi controllef o t d release formulation pesticidef o s s utilizing nuclear techniques programme Th . divides ewa d into two sub-programmes: (i) Development of controlled release formulations of e controfoth r f e rice-fisweedo lth ricn d i s an eh ecosystem d (iian ), Developmenf o t improved formulation f insecticideo s contro e tsetse th th r f programme sfo o l flyTh . s ewa completed in 1988 and an international seminar was held to review the status of research in these areas.

IAEe Th Agratefus i mane th yo t l expert havo sewh contribute thio dt s document either as participants in the co-ordinated research programme or as participants in the seminar. The IAEA officer responsibl finae th lr compilatioefo f thino s documen . HussainM e s th wa tf o , Joint FAO/IAEA Divisio f Nucleano r Technique Foon si Agricultured dan . EDITORIAL NOTE

In preparing this document for press, staff of the IAEA have made up the pages from the original manuscripts as submitted by the authors. The views expressed do not necessarily reflect those governmentsofthe nominatingthe of Member nominating the States of or organizations. Theof use particular designations countriesof territoriesor does imply judgementnot any by publisher,the legalthe IAEA,to the status as of such countries territories,or of their authoritiesand institutions delimitationthe of or of their boundaries. The mention of names of specific companies or products (whether or not indicated as registered) does not imply any intention to infringe proprietary rights, nor should it be construed as an endorsement or recommendation on the part of the IAEA. The authors are responsible for having obtained the necessary permission for the IAEA to reproduce, translate materialuse or from sources already protected copyrights.by CONTENTS

SUMMAR SEMINAE TH F YO R ...... 7 .

Controlled release delivery of : Looking back and looking forward ...... 13 G.G. Allan, J.P. Carroll UNIDO's international activities in pesticides and controlled-release formulations ...... 23 B. Sugavanam Molecular encapsulation of pesticides with cyclodextrins ...... 35 J. Szejtli, L. Szente Novel clay carriers for the controlled release of organic agrochemicals ...... 47 Z. Gerstl, U. Mingelgrin, A. Nasser Infrare FT-Ramad dan n studie pesticide-organoclaf so y interaction relation si n with fixation-release processes ...... 59 R. Prost, A. Vimond-Laboudigue Polyethylene (PE) film formulation incorporated with metolachlor for controlled-release ...... 9 6 . Byung-Youl Jin-HwaOh, Kirn Radiochemical preparatio polyacrylamidf no e hydrogel agriculturn i thei d e san rus e .....1 8 . P. Yousefzadeh, M.S. Khadjavi, M. Sohrabpour Novel controlled-release (CR) agrochemical formulations: Development and evaluations .... 91 Rajagopalan,N. Bhaskar,C. P.G. Shukla, AmarnaihN. Preparation of controlled-release starch encapsulated pesticides: Advantages and opportunitie extrusiof so n processing ...... l Il . U.E. Can Process studies, engineering feasibility and cost analysis for starch encapsulation of herbicide co-rotatine th n so g twin screw extruder ...... 5 12 . Grillo,J. M.S. Starer, M.E. Carr, PapazoglouE. Environmental factors affecting starch encapsulated herbicide rates of release ...... 139 T.J. Gish, B.J. Wienhold, ShirmohamadiA. Mobility, transpor environmentad an t l impac f starco t h encapsulated formulations of herbicides ...... 5 15 . M.V. Hickman, MM. Schreiber Efficacy of starch-encapsulated formulations of herbicides ...... 169 MM. Schreiber, M. V. Hickman Starch encapsulatio f microbiano l pesticide r sustainesfo d activity ...... 3 18 . M.R. McGuire, B.S. Shasha The use of controlled-release technology to improve the performance and reduce the environmental impac herbicidef o t s ...... 7 19 . A. Flynn, P. Stork, J. Gittins, M. Finlay, K. Williams Kinetic studie f controlled-releasso e formulation f Diuroo s n containing pall moi mill effluent ...... 5 20 . B.M. Yamin, A.Sahali Mardi, R.B. Mohamad Efficac alginate th f o y e controlled-release formulation (CRF f thiobencar)o b herbicidr efo weed control in transplanted rice ...... 215 H. R. Soltan Efficacy evaluation of controlled-release formulations of thiobencarb in rice ...... 225 CM. Bajet, I.E. Fabro Controlled-release herbicides in rice-fish culture: Evaluation of thiobencarb and butachlor formulations ...... 9 23 . M. Soerjani, S. Warnoutomo Controlled-release herbicide formulations in rice and rice-fish culture ...... 249 M. Soerjani, S. Warnoutomo Field plot test of efficacy of thiobencarb formulations for weed control in direct seeded rice ...... 257 Omar,D. R.B. Mohamad Evaluatio f controlled-releasno e formulation f thiobencaro s rice-fisn bi h ecosystems using radiochemical techniques ...... 1 26 . Bajet,CM. L.C. Araez, E.D. Magallona Potted plants greenhouse efficace studth n thiobencaryf o y o b formulations against Echinochloa crusgalli in direct seeded rice (Abstract) ...... 277 D. Omar, R.B. Mohamad Studieefficace th n herbicidef so o y ricn si e cultures (Abstract) ...... 9 27 . /. Dombovari, M. Oncsik, L. Szilvassy Study on the dynamics of release of carbon-14 labelled herbicides from controlled-release formulation waten i s r (Abstract) ...... 1 28 . Fujun Wang, Mengwen HuaguoQi, Wang Studies on the efficacy of the controlled-release formulations of herbicides against weeds in transplanted rice (Abstract) ...... 283 Fujun Wang, Mengwen Qi, Genhai Yang, Huaguo Wang, Qianhua Guo Study on the fate of controlled-release formulation of carbon-14 thiobencarb in a model paddy ecosystem ...... 285 J.H. Sun, X.M. Li, Q.Z. Zhang, Z.Y. Chen Studies on the controlled release pesticide formulations for in cotton and maize using isotope techniques ...... 287 F.F. Jamil, JamilM. Qureshi, Haq,A. S.H. Mujtaba Naqvi f controlleo e Us d release formulation f insecticideo s controe th r f termitesfo o l pests sa f so crops and forestry ...... 299 J.W.M. Logan Control of insect pests using slow release pheromone-containing devices ...... 311 F.S. Rankin Use of ionizing radiation in the manufacture of matrix materials for drug controlled-release systems selected an , d applications (Abstract) ...... 9 31 . E.E. Smolko

List of Participants ...... 321 SUMMARY OF THE SEMINAR

1. INTRODUCTION

Insect pests, disease organism competitiod san n from weeds cause enormous lossen si food production. About 30% of the food grown in the world goes to feed insect pests rather than people. Weeds generally do not damage crop plants directly, but by competing with them they restrict crop o reducgrowts d ean h yield. Contro f weedo l n contributca s e significantly to increased food production. In addition to causing losses to agricultural crops and stored food, insects which infest animals and man carry and transmit major debilitating diseases r exampleFo . , African animal trypanosomiasis, also know humann ni sleepins sa g sickness, which occurs in almost 9 million km of the savannah and forest areas of Africa, is transmitte severay db l specie tsetsf so e fly t limit.I productioe sth f livestocno 2 moskn i f o t these area thao ss t milk yiel mead dan t productio countrie7 3 e th n si affecte tsetsy db y efl are among the lowest in the world. As a result, they have to import large amounts of food.

Pesticides are commonly used to control agricultural pests and vectors of human and animal diseases. The impact of these chemicals on agricultural production in developed countries has been revolutionary. According to an estimate by the PAO, a decision to stop f croo pe protectious e th nwoul A chemicalUS d e reducth n si totae eth l outpu cropf to d san livestock by 30% and would increase the price of farm products by between 50 and 70%. In tropical countries potential yield losses from weed competition alone range from 20 to 100%. In addition to increasing yield, pesticides can reduce labour costs and energy consumption as well as protect man against insect-born diseases and so increase labour efficiency. For these reasons the use of pesticides has been increasing, especially in developing countries.

Pesticides have been deliberately developed to be toxic to some living organisms and because of this their use involves risks to human health and the environment. Excessive residue producn o s e could have deleterious, acut r lono e g term healte effectth f n ho o s consumers. Pesticides may be hazardous to non-target species such as beneficial insects including bees, to wild animals, birds and fish.

Most of the pesticide in an application fails to reach the site of action in the target organism. Losses occu resula s f a physicalr o t , chemica biologicad an l l factors. Physical processes include dispersion with wind, evaporation, washoff with rainwater, leaching into soirunofe d th lan f with irrigation water. Chemical transformatio resuly nma t fro effece mth t of light (photolysis), reaction with moisture (hydrolysis), oxidatio r reductiono n . Some pesticide degradee sar d within environmene hourth i sh t while other persisy sma r man fo t y monthsmore th s ei persistent I . t one mose sth thate likeltransportear e t b o yt d away from the site of application and cause possible detrimental side effects and for this reason there is pressure to restrict or even prohibit their use. Consequently the emphasis is on the development of less persistent and more selective pesticides. However, the less persistent pesticides may require more frequent application and higher standards of crop management. The alse yar o usually more expensive developinn I . g countries many older, more persistent pesticide stile usesn ar i l , primarily because the cheapere yar . Considerable effort beine ar s g f pesticideo mad e mako us et e seth mor othed e an efficien rn non-targesafed ma an o tt r t species.

Recognizing the cost and limitations in the design of new pesticides, pesticide scientists in the 1960s began to turn to improving the delivery of pesticides. This approach resulted f controlled-releaso e us e inth e (CR) technolog pesticidn yi e formulations controlleda n I . - release formulatio e pesticidth n s trappei e s releasea matrii n d i d an x d over tima t a e predetermined rate. In a conventional formulation, a pesticide is applied at a higher rate than is necessary initially in order to compensate for dissipation losses and so maintain activity fo reasonabla r eformulatio R periodC a n I . amoune nth t tha loss i treplace s i t d continually so the initial high dose is not needed, and harmful effects to human health and the environment are reduced. In addition, pesticide trapped in the matrix of a CR formulation protectes i d against losse physicaly sb , chemica microbiad an l l processes resulta s s i A . t i , effectiv longea r efo r perio timef do . Thi particularls si y usefu non-persistenr fo l t pesticides. CR formulations thus offer the following advantages over conventional formulations:

Prolonged effective duratio f non-persistenno t pesticides; Much less pesticid same th use r e dfo perio f activitydo , resultin lesn gi s wastd an e fewer applications; Reduced environmental contamination, particularly reduced nonpoint source contaminatio f surfacno e wate groundwaterd an r ; Reduced losses due to environmental factors (evaporation, photolysis, leaching with wate degradatiod an r chemicao t e nmicrobiadu d an l l factors), resultin savingn gi n si active costh e f th eo t ingredient; Reduced toxicit o non-target y t specie f plantso s , mammals, birds, d othefisan h r organisms; Improved efficac f pesticideo y betteo t e rsdu targeting; Greate e ruser d th thossafeto com an r s wh n contacefo i ye t wite pesticidth h e formulations.

The development of CR formulations for agrochemicals began about 25 years ago. A professional group Controlled-release Th , t heli s d dit an e A SocietyUS forme s e wa th , n i d first symposiu m197n i almosd 4an papere t th hal f o fs presented dealt with agrochemicals. Since then considerable research in this field has been done by scientists in industry, universities and government laboratories and the efficacy of the CR technology has been demonstrate n multipli d e way n croi s p protection, forestry, animal healt d aquatian h c systems. Several types of CR formulations have been prepared using a variety of matrices including synthetic as well as natural polymers. A few formulations have been developed commercially, but these are mostly microencapsulated formulations prepared in synthetic naturaf polymerso e us l e polymerTh . s suc starchs ha , cellulose, lignin, alginate othed san r polysaccharides in commercial products has been rare although considerable research with natural polymer based CR formulations has been done.

Researc formulationR C n ho bees sha n done mostl Australian yi , Germany, Japane th , United Kingdo e Uniteth d dman State f Americo s t verabu y little wor bees kha n donn ei developing countries.

Research on pesticide product development by the chemical industry, which is mostly located in the developed countries, is driven primarily by market forces so if a product does not have a substantial market it will not be developed. Development and registration of CR product costla s si y proces industrye chemicath o n r sd fo an , l compan likels yi develoo yt p formulatioR C a competo nt e wit existins hit g conventional product othee th n r handO . R C , technology offers the potential to solve some of the problems associated with the use of pesticides. Farmer developinn si g countries frequentl t takno e o adequatyd e precautionn si handling pesticides ,f protectiv o particularl e us e eth clothingn yi . Therefore, illness among farmers from exposur pesticideo et quits si e common, resultin increasen gi d medical costs and low labour productivity. In addition, a considerable number of deaths result from the accidenta r deliberato l e ingestio advantagee pesticidesf th no f formulationo R e C f On so . s is that the acute toxicity of the pesticide is usually reduced. Thus, in developing countries the use of CR technology would not only make pesticide use more efficient but also safer to farm worker otherd san s potentiall t riskya .

2. OBJECTIVES OF THE MEETING

Becaus e potentia th pesticide R f C o e f o l advantages comparee a s us e th dn i s with conventional pesticides Joine th , t FAO/IAEA Divisio f Nucleano r Technique Foon si d dan Agriculture in 1983 initiated an internationally co-ordinated research programme (CRP) on developmene th controlled-releasf to e pesticides using nuclear techniques maie Th .n objective of the programmes, which lasted 5 years, was to transfer this promising emerging technology developino t g countries 1988 n completioI e . th t ,a thif no s programm seconea d 5-year CRP, funded by the IAEA, was initiated. This programme was designed, not only to transfer the technolog otheo yt r developing countries t als seebu o ,t k solution specifio st c problems.

Rice is cultivated in many Asian countries, and frequently fish, shrimp and/or azolla are grown as mixed cultures in rice. In many countries the use of herbicides in rice is increasing. Some of the herbicides commonly used are toxic to one or more of the organisms involved in such ecosystems. In order to reduce the phytotoxicity to rice of one of the herbicides applie chemicada l calle dsoftenea useds i r , addin t anothegye r chemicae th o t l ecosystem. Controlled-release formulations of herbicides have the potential to reduce the phytotoxicit ricazollao yd t ean eliminatino s , neee r safenersgth dfo alsd reduco an t , e eth toxicit fisho yt , shrim othed pan r non-target specie ecosysteme th n i s . They would alse ob safer to the applicators. Because of these potential advantages the development of CR herbicides was one of the objects of the CRP.

Insecticide-impregnated cloth target screend san commonle sar controye useth r f dfo o l n i Africatsets y fl e. Synthetic pyrethroid , especially deltamethrid an n cypermemrin, are the most commonly used insecticides. However, the compounds degrade quickl exposurn yo environmente th o et clote th ho s ,mus e frequentlb t y retreated. This increase t onl environmentae sno yth l burde pesticidef no t alscoste sbu oth . Stabilizatiof no insecticidese th , especiall effecte th f sunligho yt o s raind an t , woul expectee db prolono dt g targetse th lif e f th eo , makin tsetse g th contro y efl l operation more efficient, economicad an l environmentally more acceptable. Therefore, the development of improved formulations of tsetsr s fo contro y efl l comprise secone dth d componen CRPe th f o t.

Radioisotope e commonlar s y use n studyini d e environmentath g l transformation, transpor residued tan pesticidef so s because the vere yar y sensitive, reliabl accuratd ean d ean require minimum sample processing. In these studies the pesticide molecule is tagged with a radioactive tracer sucr s o a hp C .S , Some studies, suc s determinatioa h f o n unextractable residues 3214, can35 be performed only by using radiotracers. The rate of release of a radiolabelled pesticide from a controlled-release matrix and the amount remaining in the matrix can be accurately determined by using radiotracer techniques.

Scientists from China, Hungary, India, Indonesia, Egypt, Ghana, Kenya, Malaysia, Nigeria, Pakistan, Philippines Unitee th , d Kingdom Unitee th , d Republi f Tanzanico d aan e Uniteth d State f Americo s a participate e seconth n di d 5-yea whicP s entitleCR r hwa d "Researc developmend han f controlled-releaso t e formulation f pesticideo s s using nuclear 1993techniques"n i d Internationan ,en a s it t A . l Semina organizes rwa revieo dt statue wth s pesticideR oC f s formulation technology.

. CONCLUSION3 P CR E TH F SO

e rat Studf th releaso e f o yf butachloo e d thiobencaran r b herbicides froR mC formulations, prepared by several programme participants, indicated that it was feasible to control the release of the active ingredient by manipulating the composition of the formulation. In alginate formulations, the size and composition of the granule and in corn cob formulations the proportion of polyvinyl acetate and polyoxethylene glycol determined the rate of release.

Field plo tn alginata test f o formulatioR s C e f thiobencaro n a commercia d an b l formulation (SaturnR) conducted in several countries indicated that the CR formulation provided better weed control and resulted in higher yield of grain as compared with the commercial formulation, especiall lowea t ya r rat applicationf eo examplr Fo . direcn ei t seeded rice, when the herbicides was applied at a rate of 0.75 kg/ha, the commercial formulatio ineffectives nwa , whereas formulatioR C e th , n increase yiele d th f grai do n by 37%. Similarly transplanten i , formulatioR C f o e dus n ric e resulteeth % 16 n di more rice grain as compared with the commercial formulation.

weee Th d contro formulationR l efficacC e th f yo s depende timine th n applicatiof dgo o n e herbicidoth f transplanteo et vardy riceyma t froI .e are manothero on t a n i t bu , general applicatio formulatione th f no day6 o t s afte4 s r transplanting provided better weed control than wheaftey da r n e befor transplantingy applieon da r e o e don .

formulationR C f butachloo s thiobencard an r b were showe safeb d fiso t o rnt an h shrimp than the commercial formulations.

4. CONCLUSIONS OF THE SEMINAR

Controlled-release technolog economin startes a ys wa a year0 d3 o csag measure from companiese th mors i t i ' e poin w importan f viewno o t t bu , t fro environmentae mth l protection poin f viewo t .

Almost all major corporations dealing in pesticides are or have been involved in researc developmend han pesticidesR C f o t . Interes mucs wa t h greate yearw o fe sag ra thanows i t ni .

Developmen registratiod an t formulationR C f no s woul costla e db y process, whics hi the main reason why so few CR formulations are commercially available. However, increasing concern for environmental contamination from pesticides may force the industr looo yt k more seriously inttechnologyR oC .

The United States Department of Agriculture, Agriculture Research Service (USDA- ARS) has done considerable research over the past 15 years on starch encapsulated CR formulations of pesticides. Their data indicate that ground and surface water contamination from pesticides can be reduced by using CR formulations, and crop yield resulting from eithes i thei e rus r comparabl conventionae th o et l formulationn i , or s some cases, better.

10 The FAO/UNEP Programme for the Operation of Prior Informed Consent (PIC) has decide tako dt e accoun e typ th f formulatio o ef o t n when considering pesticider fo s listC inclusioPI . Thie alsy th osma n n i forc industre moreth y pa eo y t attentio R C o nt formulations.

In developing countries, CR formulations would be preferable for two special reasons. Firstly, the safe e would yar an r d reduc numbee eth f illnessero fatalitied san s resulting from exposur accidentad ean r deliberato l e ingestio conventionaf no l pesticides, which are quite common in developing countries. Secondly, they will reduce the loss of pesticide volatility sb degradationd yan , whic fastes hi tropican i r l tha temperatn ni e climates.

Criteria for the development of CR formulations in developing and developed countries may be different. In developing countries patent laws and registration procedures may make it easier to develop and use CR formulations. International Organizations such s FAOa , IAE UNIDd Aan O could platransfee th yrola n ei f thi o r s technologd yan pilot scale product development. Safety considerations may become the driving force behind the use of CR formulations in developing countries.

. RECOMMENDATION5 S indicateP CR e dTh that ther considerabls ewa e interes need thir an d fo s n i ttechnolog y in the developing countries. The results of research contract holders suggested that it was feasible to develop CR formulations of thiobencarb and butachlor herbicides which reduce phytotoxicity to rice seedlings and toxicity to fish and shrimp and increase yield of rice grain. Pilot scale developmen thesf to e formulation possible b y projeca sma n e i t under consideratio UNIDOy nb .

Institutes participating in the CRP should make the results generated under this programme availabl appropriate th o et e national authorities responsibl r pesticidefo e formulation development, registration, use and regulation.

Becaus potentiae th f eo l advantage formulationsR C f so , participating institutes should continue research on the development of these formulations, particularly where local inert ingredients can be utilized.

Further collaboration among participating institutes, i.e. exchange of information and chemicals shoul encouragede db .

Next page(s) left blank CONTROLLED RELEASE DELIVERF YO AGROCHEMICAL SLOOKIN: G BACK AND LOOKING FORWARD

G.G. ALLAN, J.P. CARROLL Departmen Chemicaf o t l Engineering, Colleg Foresf eo t Resources, Universit f Washingtonyo , Seattle, Washington, United State f Americso a

Abstract A brief histor evolutioe th f yo growtd science nan th f h o f controlle eo d release delivery technology is given with emphasis on its application to plants and crops. The various underlying physical and chemical concepts are reviewed and their advantages and disadvantages compared and contrasted. Present worldwide trend basin si applied can d researc controllen hi d release technology are identified and some future patterns of utilization predicted.

1. HISTOR CONTROLLEF YO D RELEASE AGROCHEMICAL FORMULATIONS

science Th controllef eo d release delivery technolog agrochemicalr yfo s began rigorously aboua t quarter of a century ago [1]. Previous to that there had been some scattered papers and patents dealing with bioactive formulations for various practical purposes [2] but no theoretical underpining beed sha n developed untit . Indeedno ls severawa t i , l years later tha professionaa t l group was formed, The Controlled Release Society, to provide an international annual forum where scientists in this field could present their work. Over the 20 years that these fora have taken place the emphasis on medical aspects of controlled release has steadily increased. Thus, at the 1993 Symposiu ] helm [3 Washington n di publishee , th D.C. n i 2 , donl 25 paper 0 f y1 o t sou proceedings dealt with agrochemicals contrastn I . firse th tt symposiuma , , whic hels hn wa di 1974 [4]numbee th , f articlero s dealing with agricultural application closs swa hal o e t thosf fo e presented. This trend is certainly due in part to the fact that the constraints of permissible cost are much less confinin arene th healtf n ago i h care tha agriculturen ni , where every penny countso T . further understand these trends and to forecast the future of controlled release agrochemicals it is instructive to look back at some of the earliest formulations.

2. CLASSIFICATION OF CONTROLLED RELEASE SYSTEMS; PHYSICAL Controlled release delivery system agrochemicalr sfo broadle b n sca y classifie physicas da r o l chemical. In the former category the bioactive substance is simply admixed with some entity which will retar loss dvolatilizatioit y sb leachingd nan chemicay b , l breakdow actioe th o t n e ndu of water, air and sunlight, or by microbial degradation. The long known use of petroleum oils with herbicides dissolved therein is a very primitive form of a controlled release delivery system. Later, the use of polymers as solid equivalents of the oil ushered into the marketplace a new type of product. The plastic insecticidal strip for the control of flying insects was an early commercial success produce Th . t consiste flaa f td o piec polyvinyf eo l) chloridcm 0 3 x e 3 (abou x 1 0. t containing therei dissolvena d organophosphorus insecticide (2,2-dichlorovinyl dimethyl phosphate strie ,mountes DDVP)th p wa e us dn I .diagonall y withi nrectangulaa r cardboard tube ) whic huns cm 0 hwa 3 g (aboulikx 3 lanter ea x somewha3 ta n ni t confined space, suca s ha stable. The cardboard container had several large open windows cut in the walls to permit insects to ente DDVd ran P vapo exito rt .

13 3. UNDERLYING PRINCIPLES OF PHYSICAL CONTROLLED RELEASE SYSTEMS

There are a number of important scientific principles of controlled release which are embodied in this deceptively simple device. The first of these was the choice of insecticide. The pure chemical itself is a liquid (b.p. 35 C at 0.05 torr) which has a significant vapor pressure (0.012 torr root a ) m temperature. This selection meant tha levea t f vapoo l r toxi incomino ct g insects coul readile db y maintaine atmosphere th n di e withi semi-opee nth n cardboard tube. The selection of insecticide then determined the choice of the polymer needed to make the plastic strip t rooA . m temperature this musgooa e b td solveninsecticidee th r fo t wert i f .I e not, plastie th c strip would exude visible liquidrope th f dso insecticide fact n I insecticid.e th , e must gooa e b d plasticize polymere th r rfo everyda n A . y exampl thif eo s plasticization phenomenos ni provide vinye th y l db upholster automobilef yo s where phthalat ecounterparte esterth e sar f so the insecticide. Initially, of course, the concentration of insecticide is uniform throughout the plastic strip. However, most of the is within the body of the strip where it can be regarded as being inside a molecular bottle created by the long chains of the polymer of the plastic. This fraction of the insecticide cannot volatilize or wash off and the degradative effects of water, air and sunlight are certainly reduced. Any possible attack by micro-organisms is also frustrated because these are too large to penetrate into the plastic strip. Thus, only a small fraction of the bioactive agent is actually on the surface where it can physically contact a landing insect or volatilize. This is the material whic actualls hi y workin givey an t nga momen timen i t . materiae surfacth e s th A n consumeds o lei parn i s ti replace t i ,biocid w ne y db e which diffuses fro interiore mth , just belo surface wplastice th th vitas f i eo appreciatt o I .t l e that this a , s occurs, the concentration of insecticide throughout the plastic strip is no longer uniform. At the center of the strip the concentration will probably be very close to what it was when the strip was brand new, surfacwhile th t e a wilt e i considerabl e b l y depleted. This nonuniformit becauss yi e th f eo low diffusivity of the insecticide through the polymer matrix. The nonuniformity can be exacerbated by the selection of other possible geometric shapes for the plastic. Thus exampler fo , sphericaa , l preferreplastie b flae o t th t tc slab-liko baldno t s i l e geometry.

4. DEFICIENCIES OF PHYSICAL CONTROLLED RELEASE SYSTEMS Notwithstanding the many instructive features of this initial design there are some significant shortcomings which are not so obvious. The most notable among these is the relatively large amount of insecticide still left in the plastic when the strip first becomes ineffective. This can be originae thire mucs th a f on do s ha l loading. Thi serioua s si s economic proble controllea r mfo d release agrochemical. To circumvent this, attention was turned to formulations where the pure active chemical was contained within some reservoir and released through a thin film of plastic. simplese Th t manifestatio thif no s concep microencapulsations twa producte .Th s were tiny hollow spheres usually consisting of expensive polyamides or polyurethanes. The first commercially important pesticide so packaged was, surprisingly, the very inexpensive organophosphate, parathion-methyl [5]. Unfortunately varioue th , s processe f encapsulatioso n available are all rather demanding technically, and therefore impose a substantial economic penalty, particularly when the pesticide is low in cost. lookino S g earlies e bacth o kt t days numbea , simplf ro criticat ebu l guiding principler sfo physical controlled release delivery systems could even then be clearly stated as follows; i. The agrochemical should be soluble in the dissolving polymer. pesticide-plastie Th ii. c combination shoul flae db t flakes rather than spherical granules.

14 iii. A substantial amount of the original biocide will remain entrapped within the plastic and thereby wasted. Despit simplicite eth thesf yo e requirements, their implementatio fraughs ni t with practical difficulties. Thus, a. if there are several candidate agrochemicals that can control the target pest, how should the selection be made ? therf numbea i e ear possiblf ro . b e polymeric solvents, wha tfinae shoulbasie th r lth choicsfo e db e? c. how can a flat flake be made in quantity ? d. how can flat flakes be applied to the crops ? e. how can the wasteful entrapment of pesticide within the plastic matrix be avoided ? Answers to some of these questions can also be found by looking back to some of the early research results.

DESIGE TH PHYSICAF . NO 5 L CONTROLLED RELEASE SYSTEMS

rationae Th l metho choosinf do g from among several candidate biocide wels si l illustratea y db comparison of the effectiveness of a series of insecticides for the control of the shootborer which prevents the establishment of Spanish cedar plantations in Central and South America [6]. By setting up a well defined experimental protocol the single best performing pesticide will often emerge. However wely thae scientificalle ma b l t th ti , y based choice will subservienhave b o et t to other considerations, suc availabilitys ha willingnese ,th cosd manufacturee tan th f so o t r cooperate by supplying pure unformulated chemical. Usually, it will be prudent to proceed with more than a single candidate. This emphasizes one of the key inherent difficulties of controlled release delivery research which is the tendency for the number of required experiments to mushroom unmanageably.

6. SELECTION OF THE POLYMERIC MATRIX irresistible Th e expansio experimentae th f no l program will als promotee ob procese th y db f so selection of the polymer solvent. There are usually, but not always, several polymers which can be considered as candidate solid solvent matrices. This number should be reduced to one by careful consideration of the solvent power of the polymer, its melting point as well as availability, cosbiodegradabilityd an t mid-sixtiee th n I .show s wa t si n that these requirements could me met by both the polyamides and the polyurethanes. These are therefore extremely attractive general options for most [7], particularly also because these are commercially available in great diversity from different suppliers at a moderate cost. Nonetheless, there is customarily some other pesticide-specific polymer which can substitute and probably at lower cost.

7. PREPARATION OF THE AGROCHEMICAL PLASTIC COMBINATION The blending of the agrochemical, if thermally stable, and the polymer can best be accomplished by addition of the former to the latter in a molten state. This solventless system avoids the later complication of differential removal of solvents, the presence of which, however, may be inevitabl pesticider efo s which canno heatede b t . Whil relativels i t ei y eas convero yt moltee th t n

15 pesticide-polymer blend into flakes on a laboratory scale by casting in trays, or by pressing, for example, it is much more challenging to move to the larger scale needed for field trials or production e continuouTh . s extrusio thif no n flat ribbon conventionan si l commercial extruders is always a possibility if access to such high value machinery can be gained. Of course, a device to coole'e breath p ku d ribbon into flake alss i s o required. The flakes thus obtained are really only suitable for soil application as it is difficult to see how this geometric form can interact with and be retained by the surfaces of the plant foliage. Furthermore, since there are no existing commercial products having this shape, application equipment capable of dispensing such materials is not currently readily available. This need not b probleea less-developee th mn i d countries wher controllee eth d release flakes coule db scattered by hand. Of course, afte certaia r n perio timef do , these flakes wil longeo n l capable rb maintaininf eo e gth minimum effective level of pesticide in the surrounding environment. Analysis of the depleted flakes will then disclos presence eth substantiaf eo l amount entombef so d agrochemicalo .T prevent this wastage, attention focuse f biodegradablo de upous e nth e polymer matricee th n si expectation that the depleted surfaces would be consumed by soil microrganisms and thus expose the essentially non-depleted core with its original complement of agrochemical.

. 8 BIODEGRADABILIT PLASTIF YO C MATRICES

Although all polymers are ultimately biodegradable, only seldom does the rate of deterioration correspon growte th o dt h cycl agrochemicad ean l need f agriculturaso l crops, especialle th f yi plasti layins ci g upo soie nth l surface. Amon polymee gth r classes mose th , t biodegradabln ei order of decreasing susceptibility are probably proteins, polysaccharides, natural polyphenolics, polyamides, polyurethanes, polyesters, polyethers, polyvinyls and polyolefms. However, neither the protein polysaccharidee th r sno vere sar y good solvent mosr sfo t agrochemicals. Moreover, both are non-thermoplastic and cannot be melted.

9. DISSOLUTION OF POLYMERS IN PESTICIDES

This obstacle was overcome by a careful consideration of the process of mixing molten plastic" and agrochemical. Thus, if lOg of a pesticide is dissolved in 90g of a molten polymer and cooled without separation then the mixture can be described as a 10% w/w solution of the biocide in the plastic matrix. Likewise, if 20g of the pesticide is dissolved in 80g of polymer, the result is a 20% w/w solution. As the percentage of pesticide solute is increased similar descriptors can be conventionally applie 50%o t p thit du A .s becomepointw no t i , s uncertain whethe pesticide rth e is dissolved in the polymer or the polymer is dissolved in the pesticide. Beyond 50% of agrochemica blene cleas th i t n di i l r thamixture th t e mus describee b t solutioa s e da th f no polymer in the pesticide. Although at first, this distinction might seem to be only of philosophical interesprofouna s ha t i t d effect upo release bioactive nth th f eo e componene th f o t controlled release blend. When the pesticide fraction is above 50%, the pesticide and not the plastic is the continuous matrix. When the pesticide is lost from the surface the remaining polymer chains are so far apart that they can no longer maintain intimate contact with one another and the surface cracks and disintegrates exposin virgie gth n blend beneath . casslabe Thua pesticidth e f eo n th , si e leaving causes the slab to become progressively thinner and thinner, but the concentration of the agrochemical withi slae nth b remains constant. This means tha rate releasf th te o e remains constant. It also means that ultimately the slab disappears and no pesticide is permanently entombed [8].

16 Of course, sometimes it is not practical to have more than 50% pesticide in the composition; perhap blenstickye o sth to examples r di fo , thesn I . e situations pesticide , parth f o t replaces ei d b ynonbioactiva e material, typifie benzoly db e acid, whic alss hi o capabl leavinf eo blene gth d with the departing agrochemical. The family of polyoxyethyleneglycols are particularly suitable for this role because of their broad range of molecular weights and compatibilities as well as industrial availability at moderate cost.

. CONTROLLE10 D RELEASE FORMULATIONS CREATE FARMEE TH Y DB R singulae Th r propertie thesf so e polyoxyethyleneglycols have also been extremely helpfun i l devising controlled release delivery systems which are compatible with the spray application methods and equipment currently used by farmers worldwide [9]. This technique is based on the admixture of a polymer latex with the commercial, registered and approved, water dispersible pesticide formulations, by the fanner, in the field. An immediate and obvious benefit of this approach is that the active cooperation of the pesticide manufacturer in providing unformulated pesticide is not mandatory. A polymer latex is a water suspension of tiny balls of polymer, usually prepared directly and inexpensively from the parent vinyl monomers. Latices base vinyn do l acetat suitable ear wida r efo rang agrochemicaisf eo . Water-based household paint commonla e sar y encountered for f thimo s latex polyvinye .Th l acetate balls are about 1500 A in diameter, do not settle out under gravity and constitute about weighy latexe b th f % .o t Whe50 agrochemican na l tha organis i t c solvent solubl stirres ei d into the latex it will preferentially locate within the tiny plastic spheres, and not in the aqueous phase. When the latex modified with the agrochemical is sprayed onto the crop an aqueous layer containing the little spheres forms on the foliage. As the water evaporates the tiny plastic balls with their contained pesticide coalesc foro et mcoherena t flat plastic-pesticide film. Thisf ,o course, is the preferred geometric shape for a controlled release delivery system having a constant rat releasef eo rate f releasadjustee .Th o b n manipulatioeca y db amoune th f no f to pesticide and polyoxyethylene glycol added to the polymer latex. This latex-based system clearly advantagee th s ha easf so f manufactureeo , low-cost, ready availability worldwidd ean compatibility with existing commercial agrochemicais and farm application equipment. It should be the technology given the first consideration in tackling a new controlled release design problem.

. CLASSIFICATIO11 CONTROLLEF NO D RELEASE SYSTEMS, CHEMICAL

However, looking bac tim n kapparens i i t ei t that ther s anotheewa r approac developmene th o ht t of controlled release agrochemicais. This was founded upon the idea that an agrochemical could chemicalle b y linke substrata o dt discreta y e b identifiabl d ean e chemical bond. This would completely deactivate the agrochemical until it was liberated from its chemical prison by some process, suc hydrolysiss ha , which would soiloccue th .n ri The idea was early expressed in a very limited way in a U.S. Patent awarded to Baltazzi in 1967 [10]. In this document is described the addition of a herbicidal carboxylic acid, such as 2,4- dichlorophenoxyacetic typica e acidth o t , l mixtur diacidf eo diold san s use preparo dt w lo ea molecular weight styrène soluble polyester. Becaus herbicide eth monofunctionas ei must i l e b t located at the ends of the short polyester chains. When the styrène is copolymerized with the olefinic double bonds withi polyestee nth r chai three-dimensionana l crosslinked networs ki obtained. A remarkable feature of this type of polymerization is that the molecular weight change vera n si y short interva f reactioo l n time from quit level infinitw n a elo o st e value [11], determined by the size of the container. This transformation occurs at the so-called "gel point". actuae Th l controlled release performanc thif eo s typ polymef eo neves rha r been reported. Becaus remarkable th f eo e propertie three-dimensionaf so e b l l networkpoinn ge ca t e i t th t sa anticipated tha t soma t e degre hydrolysif eo infinite sth e network would suddenly disintegrate

17 into a mass of very Jow molecular fragments. This would correspond to reaching the gel point from the infinite network side and should provide a sudden burst of herbicidal action. A much broader expression of the idea of chemically attaching pesticides to polymers was put forward by Allan in 1963 [12] and Later studied in great variety [13] by the University of Washington Fibe Polymer& r Group. Their attentio largels nwa y focuse vern do y long-lasting formulation temperaturn i e us r tropicad fo s ean l forest plantations temperature th n I . e regionsa , main thrust was the study of controlled release herbicides for the suppression of competitive vegetation around seedling conifen si r plantations [14] formulatione Th . s were preparee th y db heterogeneous esterificatio hydroxye th f no l group wastn si e Dougla barr sfi k with 2,4-dichlorophenoxybutyric acid [15]. In the soil, these ester linkages progressively hydrolysed to release the original acid. The rate of release could be modified by the extent of esterification. Higher degrees of substitution reduce the hydrophilicity of the combination and the rate of release choic e substrate .Th th f eo onl d e ha ymino a r effect bare Th k. itself coursef o , alss o,wa subjec microbiao t t l breakdow soithie d th san l n tendeni increaso dt rate f releaseth e o e wite hth passage of time.

12. THE PROBLEM OF REGISTRATION

Although excellent practical performanc exhibiteds ewa , attempt registeo st r these materialr sfo commercia frustrates governmentawa e e th y us l db l authorities positioe Th . n take thas e nth wa t bark esters of known herbicides would be regarded as new compositions and would therefore throughavo g o e t entir e hth e approval process fro beginning.e mth . Since thitasa s f si ko immense proportions in the U.S.A. further progress was effectively stymied. Consequently, new chemically-bonded compositions based on old, tried and true pesticides should probably only be investigate countrien i e us r sdfo that would assuredly grant registration approval e baseth n do pesticide actually liberated.

13. NEWER CONTROLLED RELEASE SYSTEMS

A revie controllee th f wo d release literature ove pase rth t decade doe indicatt sno incipiene eth t emergence of any strikingly new technologies for agrochemicals. Perhaps the only partial exception thio st s generalizatio vero tw y e simplnar practicad ean l procedures e basee th on ; n do limited solubilit 2,4,6-triamino-l,3,5-triazinf yo waten ei microporou e othere th [16] th n d ro ,an s structur cellulose th f eo e fiber woodf so y plants [17].

14. CONTROLLED RELEASE FRO MCRYSTALLINA E MATRIX

In the former, the starting material is a commercially available heterocyclic substance which is produced in enormous quantities at many locations around the world and is sold under the trivial nam melaminef eo startine .Th g materia manufacturs it r fo l low-coss ei t urea whic simpls hi y heated to cause trimerization by the elimination of ammonia, which is recycled into the synthesis of urea. Melamine is used mainly, after reaction with formaldehyde, to prepare the tough, wear- resistant plastic laminates found on kitchen tabletops, counters and the like. On its own, without the formaldehyde, melamine can function as a controlled release nitrogen fertilizer [18] by virtue solubilits o it fnitroges it d f onlan yo n, ycontenC wate n 0.00 i 2 t L 366.6%f ra g/ o tsoile th ,n I . melamine breaks down so slowly that it is most effective on crops with long growing seasons that flourish in higher temperature locations. Thus, melamine has been shown to be an excellent fertilizer for rice [19]. The main obstacle to its widespread adoption seems to be the cost, which is about $1.50/k .nitrogenf go .

18 Since melamine is sold as a fine white crystal it can be readily converted into fertilizer-type granules in conventional rotatory pan granulation equipment. When a solid pesticide is admixed with the melamine before granulation the granule obtained has the pesticide distributed throughout and entombed between the melamine crystals. Since the pesticide is not soluble in, cannot dissolve in, and cannot diffuse through, the highly crystalline melamine matrix the agrochemica onln ca ly escap melamine th s ea e slowly dissolve soin si l moisture this i t sI . moisture-controlled, slow dissolutio melamine th f no e which constitute mode sth f actioeo f no this very simple controlled release delivery system. Of course, the granules are inevitably spherical as a consequence of the means of manufacture. This shape requires that the rate of release of the entombed pesticide decrease with time as melamine dissolves away and the spheres theid an , r associated surface area, become smaller difficuls i t I . imagino t tflakea w -eho shaped formulation could be made economically to remedy this declining rate characteristic. Nonetheless sphericae th , l melamine granule merisystee th extremf s to mha e simplicity, could be practiced anywhere and should not pose insuperable registration hurdles.

15. CONTROLLED RELEASE FROM MICROPOROUS CELLULOSE FIBERS The latter of these newer controlled release systems starts from angiosperm woody plants. Their worldwide occurrence can be an important consideration for less-developed countries where some of the other polymer alternatives discussed may be unavailable. The cellulose fibers can be separated fro othee mth r component woode th f so y plant vern si y crude equipmen boiliny b t g with a 1:1 mixture of ethanol and water containing a trace of acid catalyst [20]. The ethanol can be obtained locally by fermentation and recovered after the treatment of the wood. This new pulping process has been the subject of intense research and has already been commercialized in Canada [21]. The cellulose fibers obtained are similar to those produced by the old kraft or sulfite pulping processes wooe th pulpes s di A . removae d th lignine th f lo , hemicellulosed san extractives creates voids in the fibers which, after washing, are filled with water. The volume of thes mucs a rnL/2 e e s voidhb cellulosa f g o n sca e fibers. This materia manufactures i l n di various gradethin i sd conditiosan knows ni "never-drieds na pulp". After dryin sols i t gdi worldwide in multimillion ton quantities as the raw material for papermaking. Never-dried pulp itself is not an article of commerce although samples can readily be obtained in any pulp mill. The water-filled voids or pores, range in width from 5 to 300 A, and are enclosed by lamellae of cellulos abouthicke A b e 5 thay 3 t. Thesma t e pore substantialle sar y slit-lik shapn ei e becausf eo the lamellar morphology of the fiber cell wall. During the drying process, surface tension forces caus collapse eth porouf muce eo th f ho s structur holloe th s ea w circular fiber becomesa flattened ribbon. If the liquid within the pores contains an agrochemical, then when the fiber is dried the active ingredient will be trapped between the layers of cellulose within the cell wall. Thus, to make a controlled release formulation, the never-dried fibers need only be immersed in a solutio agrochemicale th f no , presse removo dt solutioe eth n outsid fibere theed th san n dried [22]. Thus, this controlled release delivery meritsystee th localla s f mso ha r internationallyo y available, low-cost, biodegradable matrix, together with a simple manufacturing procedure for the desired flat geometry.

. FUTUR16 E RESEARCH TRENDS From all of the foregoing it seems clear that it is fair to state that as of now almost any agrochemical can be delivered in a controlled fashion over a period of time by any one of a variet techniquesf yo future th r e Fo .thi s means thaincentive th t develoo et controllew pne d release delivery system lows si . However, there will alway researce sb h along these lined san nowaday everr polymesfo w yne r invente shibbolete dth attacheds hi havy , thama e t promisi t n ei controlled release. In contrast, the incentive to develop new formulations of old agrochemicals is high. There are several reasons for this and two are particularly significant.

19 Firstly mose th , t commercially successful biocide reachine sar where ag n gea their patent protectio expirings ni originae th f I . l manufacture firse d introducth o t t an e b w n ne rca e a improved version of the now patent-free pesticide then their market position may be protected fo furthea r r perio f yearsdo , combinationespeciallw ne e th f yi themselvee sar s patentable. Accordingly woult i , expectee db d that there woul continuoua e db s strea mcontrollef o d release versions of old agrochemicals reaching the marketplace. However, this is not the case, partly because ther vera s ei y strong force acting agains realizatioe tth f thino s expectation. Thuss ,a designatioe sooth s na n "controlled release attaches "i producta o dt U.Se th , . regulatory agencies require specific dat supporo at t this descriptor momentw fe A . s contemplation shoul enouge db h to appreciate that such experimental data can be exceeding expensive and difficult to obtain. It therefore makes better business sense to not attempt to claim publicly any controlled release characteristics. Second, in the U.S., the re-registration of existing commercial pesticides is underway and environmental concern beinw no g e givesar n much more weight than whe originae nth l approvals were granted. This will mean that agrochemicals sol relativeln di y small amountr sfo specialized markets will vanis U.Se th n .hi sinc coste ere-registratior th sfo higno r wilto hfo e b l the biocide to bear. These economics notwithstanding, if pesticide manufacture takes place outsid U.Se eth . borders these chemical stily available b lsma countrien ei s t followhicno o whd the stricture Environmentae th f so l Protection Agenc U.S.Ae th f yo . futuree Inth , there will continuoualsa e ob s driv reformulato et e existing agrochemicale us r sfo in countries with special requirements. The work on Tsetse fly control which will be presented late thin ri s symposiu exampln a ms i thisf eo . Obviously, ther othee ear r extremely serious pest problems around the world but these will not attract the attention of the agrochemical industry unles lucrativsa e marketing opportunit apparents yi . Such financial justification difficule sar o t t find in the less-developed countries. developee Inth d countries majoe generatio e efforth ,w th n ne ri t f controlleno d release products is likel arecome o y t planth f ao n ei t growth stimulants [23]. Thi becauss si e these agrochemicals are effective in such small doses that even their physical application is a testing challenge. Moreover, after application prolongea , d perio effectivenesf do s will ofte requirede nb . Increasingly, publications dealing with the delivery of plant growth stimulants are appearing [24, wel y , tha25]be lma tt I becaus. minute th f eo e amount chemicaf so l needed appropriate th , e delivery systems will be applied to the seed before planting. Some remarkable increments of crop yields are now being obtained, on rice [19] for example, by the use of the marine polysaccharide, chitosan, as its nonphytotoxic salt [26].

17. CONCLUSIONS

Thus, looking bac apparens i t ki t tha powerfua t l arra controllef yo d release technologies sha been created that can be used to deliver almost any agrochemical. Looking forward, it seems clear that the future major thrust of research will be to identify those agricultural needs which can be satisfied by existing agrochemicals when these are supplied at an appropriate rate over the necessary period of time by a simple and economical controlled release formulation.

REFERENCES ] [1 ALLAN, G.G., CHOPRA, C.S., NEOGI, A.N., WILKINS, R.M., Desig synthesid nan s of controlled release pesticide-polymer combinations, Nature 234 5328 (1971) 349. [2] ALLAN, G.G., CHOPRA, C.S., FRIEDHOFF, J.F., GARA, R.I., MAGGI, M.W., NEOGI, A.N., ROBERTS, S.C., WILKINS, R.M., Pesticides, pollutio polymersd nan , CHEMTEC (19733 H3 ) 171 referenced ,an s therein cited.

20 [3] Proceedings 20th. International Symposiu Controllen mo d Releas Bioactivf eo e Materials, July 25-30, 1993, Washington, D.C., The Controlled Release Society, Inc., Deerfield, IL60015, U.S.A. [4] Proceedings 1st. Controlled Release Pesticide Symposium, Sept. 16-18, 1974, Akron, OH, The University of Akron, Akron, OH 44325, U.S.A. [5] LOWELL jr., J.R., CULVER, W.H., de SA VIGNY, C.B., "Effects of wall parameters on the release of active ingredients from microencapsulated insecticides", in Controlled Release Pesticides (SCHER, H.B., Ed.) ACS Symposium Series 53, American Chemical Society, Washington, D.C. (1977 . 145)p . [6] ALLAN, G.G., GARA, R.I., WELKINS, R.M., Studies on the shootborer Hypsipyla grandella, Zeller. III. The evaluation of some systemic insecticides for the control of larvae in Cedrela odorata L., Turrialba 20 (1971) 478. ] [7 ALLAN, G.G., "Controlled release pesticides", Canadian Patent 846,785, Jul, 1970y14 . [8] ALLAN, G.G., NEOGI, A.N., THOMPSON, H.E., The control of pine tip moths by using sustained release systemic insecticides, International Pest Contro (19811 . 3 2 l 10 ) [9] ALLAN G.G., ALLAN, R.G., CARROLL, J.P., NEOGI, A.N., "Simple creation of controlled release formulations of insecticides by a polymeric spray tank additive" in Proceeding Seminae th f so Africr rfo Animan ao l Trypanosomiasis: Tsetse Control, Diagnosis and Chemotherapy using Nuclear Techniques, Nairobi, Kenya, Feb. 11-15, 1991. Food and Agricultural Organization of the United Nations International Atomic Energy Agency ( 1992) p. 197. [10] BALTAZZI , "HerbicidallE. , y active alkyd resins", U.S. Patent 3,318,769, Sept. 26, 1967. [11] FLORY, P.J., "Principles of Polymer Chemistry", Cornell University Press, Ithaca (1953Y N , 377. )p .

[12] ALLAN, G.G., "Pesticid elibératioà n retardé procédn so t ee é d'obtention", Belgian Patent 706,809, Dec , 1967.15 .

[13] ALLAN, G.G., BEER, J.W., COUSIN, M.J., MIKELS, R.A., "The biodegradative controlled releas pesticidef eo s from polymeric substrates"n i , Controlled Release Technologies; Methods, Theor Applicationsd yan , Vol2 . (KYDONIEOUS, A.F., Ed.), CRC Press, Inc. Boca Raton, FL (1980) p. 7. [14] ALLAN, G.G., CHOPRA, C.S., RUSSELL, R.M., Controlled release pesticides. III. Selective suppression of weeds and deciduous brush in the presence of conifers, International Pest Control 14 2 (1972) 15. [ 15] ALLAN, G.G., CHOPRA, C.S., NEOGI, A.N., WILKINS, R.M., Controlled release pesticides. Part II. synthesie Th herbicide-foresf so t solid waste combinations, Tapp (19714 5 i ) 1293. [16] SMOLIN, E.M., RAPOPORT, L., VTriazines and Derivatives" in "The Chemistr Heterocycliyof c Compounds", Interscience Publishers Inc.Yorkw (1957Y Ne , N , . 309)p . [17] ALLAN, G.G., KO, Y.C., RITZENTHALER, P., The microporosity of pulp. nature pore th Th ef esizo e distribution ,(199 3 Tapp 4 7 1. 205) J i . [18] ALLAN, G.G., FREEPONS, D.E., CREWS, G.M., "Fertilizer compositions, processef so making them and processes of using them", U.S. Patent 4,560,400, Dec. 24, 1985.

21 [ 19] ANON, "California rice growers - a commercial fertilizer program for increased yields", Technical brochure, Melamine Chemicals Inc., 748x P.OBo , . Donaldsonville 70346A L , , 1986. [20] SARKANEN, K.V., Chemistry of solvent pulping, Tappi J. 73 10 (1990) 215. [21 ] GOYAL, G.C., LORA, J.H., , AutocatalyzePYEK. , d organosoiv pulpinf go hardwoods; effect of pulping conditions on pulp properties and characteristics of soluble and residual lignin, Tappi J. 75 2 (1992) 110.

[22] ALLAN, G.G., BALABAN, C, DUTKffiWICZ, J., LEE, A.W.W., STRUSZCZYK, H., "A new simple controlled release delivery system", in "Macromolecules as drugs and as carriers for biologically active materials", TIRRELL, D.A., DONARUMA, L.G., TUREK, A.B., Eds. Yorw , kNe Annal e Academth f so Sciencesf yo , Vol. 466, Yorw kNe Academe Th f Sciencesyo Yorkw (1985. Y Ne , N 14 , . )p [23] McLAREN, J.S., "Chemical manipulatio crof no p growt development"d han , Butterworth Scientific, London (1982)K U , . [24] ALLAN, G.G., COUSIN, M.J., MIKELS, R.A., "Controlled release growth stimulants for plants", in "Biomédical polymers, polymeric material pharmaceuticald san biomédicar sfo l use", (GOLDBERG, E.P.. NAKIJIMA , Eds.A. , ) Academic Press Yorkw (1980Y Ne , N , . 381)p .

[25] KORNAKOV, ML, TSATSAKIS, A., SHTILMAN, M., ZALUKAEVA, T., VLAHAKIS, J., ASSITHIANAKIS, P., KURUSHINA, N., "Polymeric forms of 1-aminocyclopropane-l-carboxylic acid: structure and properties relation", Proceedings 20th. International Symposium on Controlled Releas Bioactivf eo e Materials, July 25-30, 1993, Washington, D.C., Controllee Th d Release Society, Inc., Deerfield 60015L I , , U.S.A. . 400p , . [26] FREEPONS, D.E., "Plant growth regulators derived from chitin", U.S. Patent 4,812,159, March 14, 1989.

22 UNIDO INTERNATIONA1S L ACTIVITIES IN PESTICIDES AND CONTROLLED- RELEASE FORMULATIONS

B. SUGAVANAM United Nations Industrial Development Organization, Vienna

Abstract The United Nations Industrial Development Organizations (UNIDO) mandate o assisit s e industriath t l developmen n developini t g countries n thisI . , transfe f appropriato r e maieth n technolog f o component e on s i yf technica o s l co. operation programme of UNIDO. The Agrochemical Industries Unit of the Chemical Industries Branch of UNIDO has been providing assistance to developing countries in the safe development of pesticides. Technology transfer, among other things has been in the manufacture of active ingredients, formulation technology, development of new environment friendly pesticides and their formulations and in disposa f toxio l c wastes e papeTh . r describes selected UNIDO projects dealing with technology transfer in the area of pesticides , development of integrated safety guidelines and in data collection and dissemination of information.

1. INTRODUCTION

Toda pesticide yth e industr goins yi g through both reactionar d radicayan l change reactin s thai n s i t ti pressureo gt s brought abou extremy tb e e viewth n so pesticidesf o e us . While accepting many drawbacks associated with excessive eus of pesticides and resisting; non.scientific facts that are mainlv based on isolated and unproven cases the industry during the last two decades has undergone radical changes thisn I . , newer, safe d highlan r y active pesticides, their formulations and application technologies have been introduced. In addition, strict registration and re-registration schemes and verv low residue limits in food and other environmental matrices the industry is becoming one of the highly scrutinized industry sub.sectors. This has also culminated in accepting integrated pest management (IPM) as a way to achieve environmentally sustainable agricultural production in which pesticides will play an important role. Whil e benefitth e f modero s n pesticide technolog s reacheyha d developed countries e developinth , g countrie a grea o t st exten e oldeus t r generation pesticides, use outdated technologies and suffer from lack of expertise and facilities for proper quality control, waste management and safety aspects during production, distributio d usean n .n thiI s context UNIDO's activitien i s pesticides cover a wide range to induce the catalytic effect to promote safe developmen f pesticido t e industr developine th n i y g world.

2. CLASSIFICATIO F DEVELOPINNO G COUNTRIES

In UNIDO's technical assistance programm n pesticidei e e developinth s g countries are broadly divided into four categories.

Categor - Countrie 1 y s wit o manufacturhn d inadeouatean e usae pesticidesf o e .

Categor - Countrie 2 y s with substantial pesticide demand an d some distribution o locasysten t lbu m uroduction

Category 3 - Countries wifh sizeable cesticide markets and local formulation o productioplantn t sbu f no basic active ingredients.

23 Category 4 - Countries with capabilities of manufacture of pesticides with potential for export and also capable of doing R&D work in exploitation of local raw materials.

3. NATURE OF ACTIVITIES

s UNIDA s involvei O n assistini d g developing countrie n industriai s l development its activities cover a wide area in each industry sub. sector. In the pesticide sector they cover the whole spectrum from development to monitoring the fat f theso e e chemical environmente th n i s . These activities include:

--Opportunity studies --Pre d fulan . l feasibility studies --Acquirin assessmend gan technologf o t pesticidn yi e production and formulation --Establishment and management of pilot/demonstration plants locallf o e us ye availabl--R&th n i D materiaw n ra ei d an l developing newer/safer --Bio.assay, toxicology evaluation --Bio/botanical pesticides --Ecotoxicology --Integrated safety aspect pesticiden i s s productiod nan formulation(SHE aspects) --Publications

Technical assistanc provides ei nationaln do , sub.regional, regionad lan global basis. This paper highlights selected project d frovere an sth m y nature of UNIDO's assistance the controlled release formulations are covered within the context of overall pesticide formulation projects.

4. NATIONAL PROJECTS

4.l Myanmar (Burma)

Myanmar during the '50s and the '60s had been one of the economically prosperous countries in the region and was a prime exporter of rice. Subsequently, the country went through a process of economic decline to the exten beinf tleaso e th g tf declare o develope e on s da d countrie U.Ne e th .Th y s b country was mainly dependant on imported finished formulations for use in agriculture orden locaf I .o mako e t rl eus resource d savsan e foreign exchange UNIDO proposed during late '70s establishmen pesticida f to e formulation complex in Myanmar.

Base n markeo d t demands decidewa t o establisi ,t d C formulatioE n a h n plant that could initially produce formulations according to FAO specifications using importe w materialra d d slowlan s y phas n localli e y available solvents. Therefore a formulation plant was designed and established during 1985-1989 and in parallel experiments were carrie witt ou dh locally available solvents. Table I give e ssolvent th som f o e s available locall d theiyan r physical properties. pesticidee Th s chosen wer elocae baseth ln o ddemand . Tests were carried out with different local solvents regarding solubility of selected pesticides, corrosive nature against packing material rubbee th d rs an washer sprayen si r used by the farmers. Storage properties and phytotoxicity tests were carried out accordin standaro gt d methods. studie e Basefouns th wa n do t ds i tha localle tth y available superio rn combinatio i kerosen r o n eow ncouls wit it e use b dn h o d xylen variour efo s formulations meetin specificationsO gFA plane .Th t itsels fwa constructed accordin internationao gt l standards wit extractorsr hai , scrubber.

24 TABLE I . PHYSICAL PROPERTIES OF SOLVENTS AVAILABLE IN MYANMAR

Myanmar Moisture B.P F.P Density Aromatic Acidity Solvent s/i 1 g/ Content

Maphtha 0.05 205 3 .75 Nil

SBP 62/82 0.05 62-82 % 7 3 .71 Nil

SBP 50/135 0.04 50-135 3 -- --

Superior Kerosene 0.058 300 64 .82 Nil

% 20 Mann 2 Reformat.8 e3 0.1 - - 2

Xylene 7 .8 138-14 7 1002 l 4 Ni % (imported)

drum crusher, carbon adsorption unit and a fully equipped analytical laboratory. A small high temperature incinerator is being installed to take care of toxic wastes including organo-chlorines plane dea.n Th tca l with practicall wastel yal s d disposan onlf crushee o yth burned dan d drums. Therpiloa s i et scale mixing tank for small scale trial formulations. The project gave the country a capabilit formulato yt e pesticide st exis whicno td hdi before futurn s I .i t i e expected that a granular and a plant would be established subject to availability of funds.

2 4. Egypt

Egypt is one of the biggest importers of pesticides in the Arab region and in 198 governmene 3th t requested UNID assistechnologe o Ot th n ti y transfer rfo the productio organf no o phosphorus pesticides dimethoat malathiond an e . UNIDO assisted in getting know-how from Spain and the plant was established at Kafr el Dawar near Alexandria The plant was integrated with the existing utility services and effluent treatment plant at the site for treatment of effluents. The project was operationally completed in 1990. The dimethoate production, around 0 tonnes/annum30 directls i , y formulatioa use y b d n plan Kaffat a t Zayatl re . This e firsprojecth tr fo timt e gave countrth e y capabilit o product y e active ingredients and is catalyzing further investment and possibly joint ventures.

3 4. India

India is one of the unique countries in the sense that it is one of the countries with r capit verw pe leve lo yf ao l consumptio f pesticideo n d an s fertilizers. Mainly conventional pesticides suc s organo-chlorinea h d an s organo-phosphorus compounds dominate the local market and traditional formulations such as EC and dusts are produced. Yet, at the same time India competes with multinationals in the export of pesticides. In fact it is already in the list of major exporters of pesticides(Table II.). There are more than 50 basic pesticide producer d abou0 an sformulators50 t . Somf thee o ear m multinationals and others are medium and small scale formulators making their own formulation contracd san t formulation thirr sfo d parties. Recent economic reforms in India could revolutionize the pesticide industry and bring it in line with western standards.

25 TABLE IL MAJOR PESTICIDE EXPORTING COUNTRIES

Country 989 1989 1990 1990 '000 value '000 value Tonnes SMillion Tonnes SMillion

Germany 137 1,195 139 1,480

U.S.A. 185 995 195 1,025

France 128 812 148 1,066

U.K. 98 692 92 839

Netherlands 65 357 69 431

Switzerland 56 530 52 593

Japan 25 245 28 275

India - 41 - 88

Source : Wood McKinzie, RENPAP Gazette, India Report.

In 1980 India requested assistanc establiso et D facilitiehR& promoto st e environmentally friendly formulations. A research station was established near Delhi with full facilities for developing experimental formulations such as SC, WDG, micro encapsulation d biocidan W ,E e formulations e secon.Th de phasth f o e project is now ongoing and will be completed by end 1994.

Some of the typical formulations developed for local market include: isoproturo Coppe, SC 0 rn5 oxychloride flowable, Isoproturo sulphu, WG 5 n7 r 60w/w SC. Various types of B.thuringiencis and B.sphaericus formulations were prepared. A novel self spreading bio.pesticide formulation was developed but could not be commercialized due to lack of reliable biological activity. The project is in the process of establishing a pilot plant for WDG using spray drying process.

centee Th r provides high level training cours formulation eo n technology d analyticaan l method r industriefo s d governmenan s t organizations whics i h normally not available in developing countries. Formulation technology courses are also conducted for the Asia region. We also hope that this project would catalyze the process of modernization of pesticide industry in India. The second phase of the project will be completed in 1994.

4.4. South Korea South Korea is one of the highly advanced developing country in the Asia regiolacket i d dnan facilitie toxicologr sfo y evaluatio chemicalf no s introduced in the country and also in the evaluation of locally invented chemicals for various outlets includin s pesticidesa g n 198I .e Ministr 3Th f Scienco y d an e Technology(MOST) requested UNDP/ UNIDO assistance for establishing a toxicology research center. Korea Research Institute of Chemical Technology (KRICT) located at Daejeon near Seou s e chosecenterwa th l r settine maif fo no Th . n p u g objective of the project was to establish a capacity for the country to carry out

26 TABLE III. CURRENT STAT TOXICITF EO Y TESTIN KRICT GA T

GENERAL TOXICITY TEST STANDARDS

Acute Toxicity Test Mouse/Rat IS

Subacute Toxicity(l month) Mouse/Rat IS

Subchronic Toxicity Tes months3 ( t S I )

Chronic Toxicity 12months& Tes 6 ( t ) KS

Carcinogenicty Test (Direc d Gavagetan ) KS

Inhalation toxicity/Chronic Toxicity/ Subacute toxicity/Acute Toxicity (all dogs) ID

REPRODUCTIVE TOXICOLOGY Mouse/Rat KS

SPECIAL TOXICITY TESTS

Immunogenicity Test KS Neurotoxicity tests KS Drug dependence Toxicity KS Skin Irritation tests KS Eye irritation tests KS Repeated dose dermal toxicity KS

MUTAGENICITY TESTS IS KS

AQUATIC TOXICITY TESTS

S K Acute Toxicit r H Fis6 9 hy Acute Toxicity 96 Hr Daphnia KS Acute Toxicity 96 hr Algae KS S EnvironmentaK l Chemistry Subacute Toxicity(Fish) Reproduction(Daphnia) ID

IS-International Standard, KS-Best in Korea ID-In development

27 a systematic toxicity testin f chemicalo ge internationath t a s l levef o l expertise. This involved the following stages:

-preparatory assistanc asseso et s requirementd san exposure to facilities abroad -Preparatio facilitf no y -Recruitmen trainind tan personney ke f go l -Sending the staff abroad for training -Recruitmen foreigf to n experts -Procurement and installation of equipment -carryin toxicitt gou y testing -setting up the operations system

Toxicologe Th y Research Cente locates ri buildina n di g occupying 6002 0m of floor wel 9 spac5 ld traineean leadin a w no d s gstafi Cented fan toxicit n ri y testing in the Republic of Korea. The second phase of the project started in 1990 with the idea of developing capability for doing long term toxicology testing and aiming to introduce Korean and international glp standards. The second phase under Government executio s e completei Institutnth d an de carriet ou s toxicological evaluatio Korear nfo n industrie Governmend san t institutions. These tests include general toxicity, reproductive toxicology, mutagenicity tests, aquatic toxicity tests and special toxicity tests.(Table III )

4.5. China

China and India are almost in a similar level of development in pesticide productio d use nan e Governmen .Th Chinf to a wante strengtheo t d country'e nth s capacity to carry out toxicological evaluation and in bio-assay facilities at Shenyang Research Institute in North Eastern part of China. Supported by UNDP and British funds UNIDO strengthened the capability of Shenyang Research Institute to carr t fulou yl toxicological evaluatio pesticidf no d theian e r formulations for registratio n Chinani . Unde ra secon d phas a ebio.assa y cente s builwa r t which coul l dscreenin al carr t ou yf pesticido g e formulation r potentiafo s l development n 199I . 0 durin e commemoratioth g e centeth f n internationao a rn l semina recenn ro t development pesticidn i s e technolog d theiyan r applicatioo t n the Asia region was organized at Shenyang. Following this project, a new programme has been approved by UNDP in 1992 to assist Nanshen Chemical Corporation in establishing R&D facilities for development of state of the art formulations for use in China. This includes R&D on controlled release formulations

4.6. Hungary

In order to Assist Hungarian Academy of Science in the invention of non.conventional environmental friendly pesticides UNIDO provided technical assistance to develop novel juvenile hormone analogues as insect growth regulators. Unde able rth e guidanc Prof eo f .Matolcs unfortunatelo ywh y died under tragic circumstances the project invented a number of new juvenile hormone analogues, UNIDO assisted extensive biological testing of the compounds. This came up with two compounds (NKI-35120, and NKI-43049) - with better or comparable activity to standard insect growth regulator such as fenoxycarb. An European company (Phillips Duphar) was very much interested in developing the compounds. In their tests these compound larvae th yellon f so eo w fever mosquit 3-1s wa 0t oi times more active thae standarth n d fenoxycarb n additioI . n these compounds showed excellent activity against housefly, desert locust (Schistocerca gregaria) superio fenoxycarbo rt . Unfortunatel organizationao t e ydu l change compane sth y could not pursue the matter. The project is completed and the Hungarian Academy of Science has given additional funding to develop new environment friendly pesticides.

28 5. SUB-REGIONAL AND REGIONAL PROJECTS

5.1. Africa Sub-Region

It is well recognized that seed dressing is one of the most effective and controllinf o saf y ewa g certain economic pest Whil. s e majorit developine th f yo g countries in Asia, Latin America have a very good net work of supply of treated hybrid seeds, Africa region still suffers from lack of infrastructure and facilities for the supply and distribution of treated seeds. UNIDO's studies in four least developed countries suc Tanzanias ha , Zambia, Malaw Rwandd e ian th f ao Preferential Trade Area (PTA) revealed that they have thei f barelo % r 0 seedy2 s treated while many smal mediud lan m scale farmer untreatee sus d farm saved seeds. As an experimental project supported by the German Government UNIDO has recently started developmen protf to o type mobile seed treatment machines which coule db used by African farmers or co.operatives to treat their own seeds in a safe and efficient manner to increase their food production. Based on existing machines we are providing assistance to these countries to design, fabricate and test prototype machines. Both Engineering and pesticide research institutions are involved in the project. Emphasis is placed on mobility of the machine in African terrain, flexibilit operato yt e both manuall electriy b d yan c power suitablo et African conditions, safety during application and easy cleaning, reasonably priced usind an , g only least toxic pesticides (mainly ). Traininn i g seed dressing and in carrying out biological trials is being provided to project counterparts. Base performancn do economid ean c viabilit hopes i attraco t dyt i t private investors to look into commercialization. We are expecting to display one proto type model at a seed treatment conference scheduled in the United Kingdom during earl projece yth 1994 f successfui s ti . wilt li l have spi effectf nof r sfo post harvest treatment of food grains and other produce..

5.2. Asia

One of the very successful regional projects has been the Regional Network on Pesticides for Asia and the Pacific called RENPAP executed by UNIDO in association with FAO, WHO and ESCAP.. This was originally started as an experiment to bring together countries of the Asia region having similar crops, climatic condition facind san g similar problem productionn si , distributiod nan use of pesticides. The project started in 1982 with nine member countries. The initial objectives were:

i. -to survey pesticide demand and supply in the region ii. -to improve pesticide formulation technology in member countries through technical training and consultancy encourago ii -t ilocaf o materialw e lra e us n si pesticide production

iv. -to develop common criteria for regional: -harmonizatio datf no a required for registration -standardization of analytical method qualitr sfo y control and residue analysis colleco v-t . disseminatd tan e information through publications of periodical bulletins and other materials.

Durin firse gth t phase projece ,th t conducted workshop experd san t group meetings on each topic and established a National Coordinator Unit in each

29 member country in.deptn .A h evaluatio projece th f no t carrieUNDP/UNIDy b t dou O experts in 1988 observed that:

-the projec mads tha e significant progrese th n si harmonizatio registratiof no n requirements. Sri Lanka and the Philippines have already implemente recommendationde th som f eo otherd san s arprocesn ei adoptinf so g them

-the member countries have agreed to adopt the standardization procedure qualitr sfo y controd lan uniform methodologies for pesticide formulation and residues analyses. Some countrie alreade sar y participating in CIPAC collaborative studies. FAO and WHO specifications are being broadly accepted in the region.

-the countrie regioe th f sno have already become aware of use of locally available raw materials for pesticide formulation.

-pesticide data collectio regioe n th a s r ni n fo important activity and should be strengthened.

The experts strongly recommended extensio projece th f strengthed no tan n activitie improvo st e quality control, safety aspects, greater participatiof no industrie d bettesan r technical cooperation among member countries.

e favorablBaseth n o d e recommendation e in.deptth f o s h evaluation team the regiona t wors extendedne lwa k . World Bank joine e projecn th a d s a t associated agenc r sometimfo y d includean e s PEST(Pesticidit d e Efficacd an y Safety Testing )e benefitprogrammeth o t se accrueDu e .projec th y b dt more

Global Safety in / ESCA O FA P/ O WH Pesticide Formulatio- 4 n UNDP / UNIDO (supported by Finland) WORLD BANK

Regional Co-ordination Unit India

National Co-ordinator Units (14)

INDIA Formulation Technolog Qualitd yan y Control (supporte UNDPy db ) INDONESIA Industrial Safet y/ Wast e Management (supporte UNDPy db ) PHILIPPINES Occupational Health and Safety REPUBLIC OF KOREA Impuritie Technican si l Pesticides PAKISTAN Ecotoxicology (supported by Denmark) THAILAND Bio. Botanical Pesticides MALAYSIA Application Technology

FIG. 1. Institutional network for RENPAP.

30 countries joine 4 membee 1 projecty therth dda e rar eo T .countrie z vi s Afghanistan,Bangladesh,India, Indonesia, Iran, Malaysia, Myanmar, Pakistan, People's Republic of China, Republic of Korea, the Philippines, Sri Lanka, Thailan d Vietnamdan , coverin fifta g earth'f o h s surfac d almosean t hale th f world's population which depend agriculturn o s livelihoods it r fo e .

From 1988 RENPAP grew from strengt o strengtt h d countriean h s with sufficient facilities too responsibilite kth y establishing technical coordinator unit closd san e links were maintained with national project provido st e support to the member countries. A permanent Secretariat supported by the Government of Indi UNDd bees aan Pha n establishe whole th Indin i ded ainstitutionaan p u t lse is give Figurn ni . 1 e

The fourth phase of the project has recently been approved and the project would put emphasis on safety, quality control, establishing a regional data base, bio-botanical pesticides, application technology, eco.toxicolog d supporan y t Integrated Pest Management.

6. GLOBAL

Durin lase decadeo gth ttw s ther s beeeha greana t increas settinn i e g f upesticido p e formulation plant developinn i s g countries. Whil O CodFA f eo e Conduc r Pesticidfo t e Distributio e tooUs k d car nan f safet o e y aspecte th t a s user end UNIDO felt that an integrated safety guidelines covering plant safety, occupational healt environmentad han l safety (SHE aspects) woul vere db y valuable to smal d mediuan l m scale formulator n developini s g countries. Supportey b d Finland a series of expert group meetings were conducted and in a final meeting in Brussels experts from developed and developing countries jointly developed integrated guideline r pesticidsfo e formulatio developinn ni g countriess i t I . proposed to implement the guidelines on a global basis as shown in figure 2. This

WHO FA/ O UNE/ P f UNIDO US / EPA

Database for International Safety Advisory Committee dissemination (around 6 experts)

Regional Focal Points (Technical Committee*)

Asia (RENPAP) Africa Arab Europe Latin America Philippines / Zimbabwe Egypt Hungary Mexico Indonesia

National Focal Point (Safety Stewards Committee**) Tech. Committee (4 members): Pesticide Formulation, Overall Safety, Safer Pesticides and Formulations, Waste Management. Safety Stewards Committee (6 members): Plant Maintenance Safety, Warehouse Safety, Waste Management, Occupational Health and Safety, Legal Consultant.

FIG. 2. Global network for safety in pesticide formulation in developing countries (proposed).

31 would creat responsibilite eth y withi e industrnth developinn i y g countrieo st adopt measures similar to responsible care voluntarily accepted in developed countries by the chemical industries.

7. ACTIVITIE CONTROLLEN I S D RELEASE FORMULATIONS.

As mentioned before UNIDO deals this topic within the context of major projects mentioned above. Obviously countries belongin categoro gt y e 4(sear ) e2 capable of developing such formulations. These formulations are always considered safer, economica environmentalld lan y friendly within certain limitations.The advantages are : -reductio mammaliaf no n toxicity -reduction of dermal toxicity -reduction of losses of a.i by evaporation -reductio degradatiof no a.if no . (physical/chemical;) -reductio rat n volumnd i ean applicatiof eo n -prolongation of activity

Obviously one has to balance the advantage against: -large variation between laboratory and field studies -cos preparatiof to involvet i f ni sitn si u polymerization or other reactions.

TABLE IV. PESTICIDES RECOMMENDED FOR CRF STUDIES

Pest ic tde Actiod nan Foraualting Typf o e Area of Appl. Agent Formulation

) BPM1 C non-systemiR C c . Ca + A I. rice possible combination with carriers CE -

2) Butachlor systeoic H, rice a C + A 3) Cartap non-systesnic I, rice - Ncros s - linke H G d

4) Diazinon non-systeaic I A t Ca SL - CR upland

5) Fenitrothion contact I. vegetables A + Ca SL - CR

P 6IB ) systemi . I ricc e A + Ca CR

7) Isoprothiolane systemiR cG F a C + A

8) htelalochlor uplan , vegetablH d e N - cross linked CE

9) Napropamid uplan , fruiH d t treeR G s Aa C t

10) Pendiosethalin pre-emergence H. A + Ca GR upland

11) Probena2o . ricF e l R G a C t A

12) Tricyclazol F, rice - crosN s linkeR C d

= insecticide I = F , = herbicide H , sodiu= A , m alginate, a ionC C= sa e.g. fron» CaC12 natroso= N , l (hydroxyethyl-cellulose)

32 -longer development time compare o conventionat d l formulations -slow releas againss ea t controlled release.

In India, UNIDO under the guidance of Dr.Gardarelli, initiated studies on controlled release formulations as early as 1983. Research projects started to develop controlled release formulations of temephos for mosquito larva control, carbofuran for the control of rice stem borer, dursban for soil nematode contro d nee an ls anitfeednta m l caseal n sI .natura l polymers suc s rubbea h r latex and starch were used as monolithic polymer matrix. Many of theses formulations were successfully developed and performed well in laboratory condition t repeatabilitbu s e maie resultth th n s f problemo ywa s . While they coul acceptee db slos da w releas undet definitioee bu rth controllef no d release there were many short comings. Still in the Institute for Pesticide Formulation (IPFT) in India experiments are being carried out to use starch as encapsulation. Duo development e f otheo t r formulations, controlled release formulations development has been rather slow. Under the regional project RENPAP assistance was given to South Korea to improve capability in developing CRF. This is actually linked to the Pesticide Formulations Unit at the Agricultural Chemical Research Institute. A number of fungicides, insecticides and herbicides for use ricn vegetabled o ean s selectedswa . Polysacharides suc alginates ha lesr sfo s water soluble pesticide d cellulosan s morr fo ee water soluble pesticides were used. Tabl giveV eI controllee s th som f eo d release formulations being trien i d South Korea. This is actually in the same institution where Joint FAO/IAEA has a project. Hopefully Dr.Oh in his paper will give more recent developments. At the moment UNIDO in collaboration with the FAO/IAEA Joint Division is a regionat fund tryinr ge fo so t gl projec o develot t p controlled release formulation Asie th a n regionsi . Subjec availabilito t fundf yo s India, South Korea, Pakistan and Indonesia will participate in the project making use of the existing facilities.

REFERENCES

Holly,K., Copping, L.G., Brooks, G.T.,(Eds), Recent Development Fiele th Pesticidef do n i s d theisan r Applicatio Peso nt t Control, UNIDO/RENPAP Delhiw ,Ne . Matolcsy, G.,Practical Developmen Non-toxif o t c Anti-Insect Agents, Terminal Report, DP/ID/SER.B/707, UNIDO, Vienna, (1992) Toxicology Research Center, Phase II, Project Terminal Report, Korea Research Institute of Chemical Industry,(1993). Mandava,N.B., Donnez,M.P., Regional network on Pesticides for Asia and the Pacific- Report of the Evaluation Mission, DP/ID/SER.C/20 UNIDO, Vienna (1988). Dhua, S.P., Regional Network on Pesticides for Asia and the Pacific, Interim Report, UNIDO, Vena(1992) Integrated International Safety Guidelines for Pesticide formulation in Developing countries. Cardarelli,N.F.,Bioscienc Engineering-Technicad ean l Report, DP/TD/SER.A/422, UNIDO, Vienna,1983. Shasha,B., Strengthening of Pesticide Development Center,DP/ID/SER.A/1372, UNIDO,Vienna, 1990. Vollner.L., Regional Network on Pesticides on Pesticides for Asia and the Pacific, DP/ID/SER.A/1418, UNIDO, Vienna,1990.

Next page(s) left blank 33 MOLECULAR ENCAPSULATIOF NO PESTICIDES WITH CYCLODEXTREVS

J. SZEJTLI, L. SZENTE Cyclodextrin Researc Developmend han t Laboratory Ltd, Budapest, Hungary

Abstract

The majority of organic pesticide substances can be complexed with cyclodextrins. The complex formation (molecular encapsulation) of pesticides in most cases results in the improvement of the physical, chemical stability, wettability and aqueous solubility of those rather lipophylic molecules. These novel formulations improv physicae eth chemicad an l l propertie knowe th f sno pesticides without the formation of any chemical bonds i.e. new molecule is not formed. The cyclodextrin complexation thus improve the bioavailability of pesticides which in general leads to the possible reduction of applied doses to reac e requireth h d biological response advantagee Th . e d limitth an f so s utilization of cyclodextrin pesticide inclusion complexes in novel formulations will be presented on selected examples of different pesticide cyclodextrin complexes.

. STRUCTUR1 TYPED ECYCLODEXTRINAN F SO AGRICHEMICAF SO L SIGNIFICANCE cyclodextrine Th producee sar enzymiy db c degradatio f starcno h thus they belon groue th enzyme-modifief go pt o d starch derivatives. Three different parent cyclodextrins are known: the alpha-cyclodextrin (CD) which consists of 6 glucose units, the beta-CD {7 glucose units) and gamma-CD (8 glucose units) (Fig.1.). All of the crystallinee mar , chemically uniform, non-hygroscopic substancese Th . doughnut-shaped cyclic molecules hav n internaa e l cavit f welo y l defined dimensions. These molecular cavitie rathee sar r hydrophobic, whil outee eth r surface of the cyclodextrin molecules are of hydrophilic character.

i CD yCO

l,37nm 1,69nm ^

•* 0,95-'--.» rub 0,78

Fig. 1. Structure and dimensions of «-, ß- and -)~cydodextrin

35 Molecule r certaiso n functional group f moleculeso s thalese ar ts hydrophilic tha nincludee b wate n ca dr inte cavit cyclodextrioa th f o y n molecule th n ei presence of water. In an aqueous solution the slightiy apolar cydodextrin cavity is occupie watey db r molecules thaenergeticalle ar t y unfavoure thesd dan e water molecules can be replaced by an appropriate guest molecule present in the system. This substitution process can be considered as essence of the molecular entrapment (encapsulation on molecular level) of the potential guest molecule (Fig.2.). The formed generally biner adduct is an inclusion complex, in which the guest substanc reversibts ei y boun apolar-apolay db r interactio parn i y b td nan hydrogen bond chemicay t nevean sbu y b r l interaction formee Th . d inclusion

Fig 2 Illustration of the mechamism of inclusion complex formation

complexes readily dissociat presencn ei f watereo , particularl presencn yi a f eo third componen potentiaa ( t l competitive guest substance) agrichemicae Th . l utilizatio f cydodextrino n s alsf o chemicallo e involveus e y th smodifie d cyclodextrins (especially substituted beta-cydodextrins) usually of higher water solubility. Beside non-modrfiee sth d cyclodextrins particularly four type f derivativeo s s can be taken into consideration: /1 / • alkylated cyclodextrins hydroxyaikylated cydodextrins (hydroxypropylate hydroxyethylated dan d ones) • branched cyclodextrins (enzyme produced glycosyl d maltosyan - ! derivatives). • crosslinked cyclodextrins

E UTILIZATIOTH . 2 CYCLODEXTRINF NO THEID AN SR DERIVATIVE PESTICIDN I S E FORMULATIONS

The utilization of the molecular encapsulation of biologically active substances by cydodextrins usually aims at: • modificatio f physicochemicano l propertie f pesticideo s s (wettability, solubility, volatility, smell, etc.) • improvement of stability in physical and chemical sense • enhancement of wettability and bioavailability of poorly soluble and absorbable pesticides ensuro t • e homogeneit contend yan t uniformit finan yi l formulationd san finished products • reduction of the environmental pollution caused by overdosing pesticides in fields All of these imporvements are achieved by cyclodextrin complexabon without any chemical modificatio active th f neo substance whic f practicao hs i l importance in terms of registration and approval of new formulations The cyclodextrins themselves - being non toxic, biodegradable carbohydrates - are not noxious for the environment.

36 3.EFFECTS ATTAINABLE WITH CYCLODEXTRIN COMPLEXATION 3.1. Phase transformation

The oily liquids and greasy or waxy solids can easily be transformed into freee flowing microcrystafline. easy to handle powders via inclusion complex formation. The crystalline character of these solids have been investigated in details using X-ray powder diffractometry. On the example of the oily liquid insecticide MGR- 264/jS-cyclodextrin comple shows i x n thaX-rae th t y dtffractogram indicatesa novel crystalline structur rathef eo r high crystallinit complee th r yfo x comparedo t that of the empty 0-cydodextrin hydrate.(Figure 3.) 12.1 The crystalline appearance of the inclusion complex of liquid synthetic pyrethroid, Sumithrin has been found by powder diffractometry to exhibit lower crystallinity, as shown in Figur. e4

a n n H n K u n n » n a n » u K u t! u o t i >< i u

Fig . .3 X-ray diffractogra mMGR-264/0-cyciodextrif o n inclusion complex

Fig . .4 X-ray dtffractogra mSumJthrin/0-cyckxiextrif o n inclusion complex

The solid state and good flowing properties of the pesticide cyclodextrin complexes further provide remarkably homogeneous distributio iipophilie th f no c pesticide carbohydrate th n i s e matrix, thus ensurin acceptabln ga e content uniformity in the finished products. The solid pesticide complexes exhibit outstanding moisture resistance their hygroscopicity is negligible under normal and elevated humidities (R.H. 60-80 %). The results of the clumping tendencies are listed in Table 1.following a storage of the adsorbed and cyclodextrin complexed solid Dursban, Malathion and DDVP formulations under Rel.humidity of 95% at 25 °C for two days.

37 Table 1. dumping tendency by screening test of adsorbed and complexed pesticides after a two-day storage under rel. humidity of 95% at 25 °C .The results are weight percentages of passed and retained fractions of samples. /3/

Sample pesticide content passed fraction retained fraction malathton/ lactose 20% 11% 89% malathion/0CD 19% 78% 22% DDVP/iactose 16% 9% 91% DDVP/0CD 16% 86% 14% Dursban/lactose 15% 6% 94% Dursban//3CD 14% 81% 19%

moleculae Th r encapsulatio pesticidee th f no s also provide pronouncea s d micronizing effect, sinc complexee e th regardee b n sca establishins da e gth possible highest dispersity. the molecular dispersrty of the pesticide. The cydodextrin pesticide complexes generally do not show any electric charging and o theit e r du hydrophilic character theeasile ar y y suspendabl n aqueoui e s systems.

3.2. Stabilizing effect of complexation 3.2.1. Stabilizing in physical sense

Majority of the practically used pesticides possesses considerable vapour pressure even at ambient temperature, that will result in loss of active substances froe formulationmth s applied e moleculaTh . r entrapmen f theso t e volatile agrichemlcals effectively decreases or even eliminates their vapor tension in solid dry form under normal conditions. This is manifested in physically stable, rather odourless solid products. This stabilizing power of the cycle-dextrin complexation illustrates i Thermay db l Evolution Analysis (TEAfreee th f , o )adsorbe d dan complexe de volatilth form f o es insecticide, Malathion.(Figure Th ) 5. e compfexation of a series of organophosphorous, tjoester type insecticides by cyclodextrins therefore result considerablsn i e masking their unplesent smell, until the complesolidy dr a . This xi s effect play importann sa developmene t rolth en i t of more acceptable insecticide formulations particularl r indooryfo , household uses. Accelerated heat stability tests performed at elevated temperatures for weeks provide usefu reductioe l th dat n ao f volatilitno f insecticideyo s (Fenithrothion, Malathion, and N.N.-diethyl-toJuamide.DEET) by cydodextrin complexation.(Figures 6,7,8) These stress data are in a good agreement with thos f normaeo l long term storage tes yearsr fo t . Sumithio Malathiod nan d nan Dursban /3-cydodextrin complexes have been stored for nine years under normal laboratory conditions and their insecticide content followed. Figure 9. illustrates the change f insecticidso e conten solin i t d formulations (complexe pysicad san l mixtures) upon a nine-year storage.

38 CHBK-KS5 HaMhion-fCDTU wiAiHta<-«cn TU-MW.TSIS Mm » utn

Malayan

BCD-control

»cowicu. KIXI

analysiA TE freef so . , 5 adsorbeFig. comptexed dan d ftxm Malathtof so n

Accelerated storage stability test of Fenitrothion formulations Intact Fenllrothlon («)

6 8 W 12 14 t6

( t-60°C ; RH-58% )

. 6 Fig. Accelerated storage tes adsorbef to comptexed dan d Fenithrothion

39 Accelerated storage stability test of Malathion formulations

Intact Malathion <*)

024

( t-60°C ; RH-58% )

Fig. 7. Accelerated storage test of adsorbed and complexed Malathion

Accelerated storage stability test of DEBT formulations

Intact Deot (%)

SOW Time { days ) ( t-60'C ; RH-58% )

Fig . 8 . Accelerated storage tes adsorbef o t comptexed dan d N.N.-diethyl- toluamide, DEET

term storage test of insecticides formulations

Fig. 9. Loog term storage stability of adsorbed and complexed insecticides under normet conditions

40 3.2.2.Stabilization in chemical sense The molecular encapsulation has long been known to provide effective protection for the chemically unstable, sensitive substances (drugs, flavours, fragrances.colours and pesticides.) /1J The majorit naturaf yo synthetid an l c pyrethroid lighe sar t sensitive insecticides that obviously need protection against photochemical degradation to maintain their biological effectivity. Among pesticide pyrethroide th se mos th te ar s thoroughly studied ones in terms of the cydodextrin complexatk>n74.,5/ Light stability test under UV irradiation of the adsorbed and complexed Allethrin, Resmethri Permethrid nan n showed that complexing these insecticides wit~ hß cydodextrin their photochemical degradation was strongly decreased.(Figures 10..11..12.) Methylparathio s alswa no succesfully stabilized against chemical decomposition by cydodextrins767

Light stability test of solid Allethrin formulations

Remnant Allethrin (%) 110

0123 Time ( week } ultraviolem n 4 25 V t U (ligh ) t

^«9. 10 -Ligh t stability tes adsorbef to complexéed dan ! Allethnn

Light stability tesf o t solid Resmethrin formulations

Resmathrln content (*(

0 16 0 16 0 14 O 13 0 12 0 11 0 10 0 9 0 8 0 7 0 6 0 5 0 4 0 3 0 2 0 1 0 Time (hours) ; t«30" m n C) 4 25 V U (

Fig 1 1 . Light stability tes adsorbef to complexed dan d Resmethrin

41 Light stability test of solid Permethrin formulations

Permethrin content (*>)

0 4 0 13 0 0 2 5O 60 Time (days) ( UV 254 nm ; t-SO'C )

Fig. 12. Ught stability test of adsorbed and comptexed Pennethrin

Solubility isotherm of Difluron

Cone, of Dllluron (ug/ml)

4 6 10 ) (% ConeD C f o ,

. 13 SolubilitFig. y isotherm f Diflubenzuroso aqueounn i ßC, d sa- D an RAMEB solutions

3.3.Solubilizing effec cyclodextrif to n complexation

For the powerful solubilization of poorly water soluble pesticides methylated ß- cyclodextrins have bee mose nth founte b suitabl o dt e oomplexing agents./?./ methylatee Th d /3-cyclodextrins wit averagn ha e degre f substitutioeo f DS=1,5no - 2,0 showed excellent solubilizing effect on a series of lipophilic pesticides. The commercially available randomly methylated /3-cyclodextrin called also RAMEB (produce Wackey db r Chemie, Munnich, Germany) solubilized very effectivele yth almost insoluble chitin synthesis inhibiting insecticide Difiubenzuron, while neither o-nor /3-cyclodextrin showed practically useful solubility enhancing effect. (Figure 13.) Also RAMEB was found to exert remarkable solubilizing potency for natural and sythetic pyrethroids, as shown in Figure 14. on the example of Tetramethrin substrate.

42 McBCDlmgftnJ]

-RAMEB - RAMEB - RAMES --o- — RAMES 05-a< DS-LO rHARM WACX

- RAMEB - OIME - ——B — RAMEB CHIN CS_2J> FEN

. Fig14 Solubilit. y isotherm Tetramethrif so aqueounn i s methyt-0- cycfodextrin solutions

3.4. Wettabilit d Dissolutioan y n propertie f pesticido s e cycJodextrin complexes

The pesticide cydodextrin complexe y solidr d n i sstat e preserve their entrapped active ingredien r lonfo t g time. However, upon contacting with water (e.g. dissolving thecasn i r mf rain eo immediato e th ) e releas idudee th f eo d pesticides takes place extene ratd .Th releasf e to an entrappe e th f eo d pesticides depends primaril associatioe th n yo n constan givee th f no t inclusion complexn o , the mola rpesticidercydodextrie ratith f oo extene th f dilutionn o to d e nan Th . presence of a third component especially of a competitive potential guest substance further improve release sth pesticideee th rat f eo s from complexes. Employing a suitable competitive complexant even a programmed release formulation can be designed based on a pesticide cydodextrin complex. The very first step of the dissolution process of a pesticide cydodextrin complex is its wetting by the dissolving liquid. Since the comptexed lipophilic pesticides are surrounde hydrophilia y db c CD-rin solie gth d powde fairls i r y well wettable. Almost all pesticide cydodextrin complexes were found to posses improved wettability in water. This is illustrated on the example of adsorbed and complexed MGR-26 Resmethrid 4an n formulations followin change gcontace th th f eo t angle value functioa s sa f timeno . (Figure 15.)

»•

100

Resmetrm complex 50 MGmis4 R26 t

- MGR 26t Com

10 20 30 M) SO 60timo(s)

Fig. 15. Wettability of adsorbed and complexed MGR-264 and Resmethrin formulations, measure ooncaty db c angle changes

43 The wettability profiles of free, adsorbed and /3-cyc!odextrin complexée! Propoxur determined by migration technique show similar tendencies to those of e contacth t angle measurement e possibilitd pointth an so t s o improvt y e wettabilit simoly yb v adding cydodextri pesticidese th no t . (Figure 16. gooe )Th d wettability of the pesticide complexes are usually followed by a significantly enhanced dissolution rate. On the example of free and complexed plant growth regulators (Gibberellic acid and N-6-benzyl adenine) is illustrated the improved dissolution inclusioe profilth f o e n complexes. (Figure 17,18.) Anticoagulant of very poor aqueous solubility in a 0-cyclodextrin complexed form also provide remarkably enhanced dissolution rates, whic presentehs i Figurdn i e 19. on the example of Chlorophacinone//3CD formulations. In certain cases the simple addition of /3-cydodextrin to the pesticides can significantly improve their dissolution properties in aqueous systems. This phenomenon is illustrated on the example of Baygon/ßcydodextrin formulations. As Figure 20 shows the mechanical mixtures of Baygon with 0-cydodextrin give almost the same dissolution profil compares a e true th e o dt indusio n complexe f identicao s l composition.

Wettability of solid Propoxur formulations WattabMMy (%) 110

_ _.L...__|___;_____ ._. Propoxur-BCD mlxtür« K

1 1 0 1 9 8

Fig. .16 Weoabifit y profile f freeso , adsorbe complexed dan d Propoxuy b r migration technique

Rate of dissolution of solid Gibberellic acid A4+7 formulations

cone ol Gibberellic acid (ma/ml) 1 2.——^————————————————————

CO-complflnBd Glbbcfalltc «eld

0 10 20 30 40 50 60 70 t (mm) ( <-22*C .in water)

Fig 7 1 . Dissolution profil complexel freed o ean d Gibberellic ao watedn i r

44 Rat f dissolutioeo f solino d N6-benzy1-adenine formulations in water

cone of N6-benzyl-adenine (mg/ml) 250,————

200 -

150 -

100 -

0 7 0 6 0 5 0 4 0 3 0 2 10

( t-22'C)

. Fig18 .Dissolutio n profil complexéefref d eo ean ! N-6-benzyi adenine

Rate of dissolution of solid Chlorophacinone formulations

cone. (mfl/100mO

0 4 8 12 16 20 24 28 32 36 40 44 48

Fig. .19 Dissolutio n profil complexe fref d eo ean d Chlorophacinone

Rate of dissolution of Baygon formulations in water c (mg/ml) 56

S •8*ygon-CÛ comp 46 iaygon-CD mlxl .nygon-CD comp 4 ygon-CO mUl 3.G

3

2.5

2

1.6

1

5 0

0 8 5 7 0 7 5 6 0 6 5 5 0 5 5 4 0 4 5 3 0 3 5 2 0 2 5 1 0 1 5 0

t (mm)

Fig. .20 Dissolutio n profil freef eo , adsorbe complexed dan d Baygon

45 5. LIMITS OF THE UTILIZATION OF CYCLODEXTRINS IN PESTICIDES

Due to the present price, availability and approval status to date only ß- cyclodextri consideree b n nfeasiblca a s da e additiv f agrochemicalseo e th n I . near future particularl r solubilizafofo y n purposes technical grade RAMEB (randomly methylated /3-cyclodextrin practicaf also n e o)ca b l importance. Cyclodextri- n complexatio wortns i h usin gcase onl th expensivef eo yn i , highly potent pesticides, since the pesticide content of such a complex is between 5- 25% by weight. When the aim of complexation is not stabilization then the raoo of cydodextri considerable b n nca y lower. - Only organic molecules, of appropriate size, shape, and polarity can be complexed with cyclodextrins.

REFERENCES

1. SZEJTLI, J-, Cydodextrin Technology. Kluwer Publ. Co., Dordrecht. (1988) 2. SZENTE, L., MAGISZTRAK, H., SZEJTLI, J., Formulation of Insect Controlling Agents with j3-cydodextrin, Pestic, Sd. 28 7 (1990) 3. CYCLOLAB unpublished data 4. YAMAMOTO, I., UNAI. T., SUZUKf, Y., KATSUDA, Y., Preparation. Stabilization and Insecticidal Activity of Cydodextrin-lndusion Compounds of PyrethrokJs, J.Pesticide Sd. (Nippon Noyaku Gakkaishi), 1. 41 (1976) 5. MIFUNE. A., KATSUDA, Y.. YONEDA, T., insectiddal and Acaricidal Compositio Procesd nan Controllinr sfo g Pests PatentS U , , 8463 , , 551, Nov. 5 . (1974) 6. YAMAMOTO, I., OHSAWA, K., PLAPP, F.W., Effect of the Indusion Compounds of Pyrethroids and Methyl Parathion on Certain Cotton Insects, J.Pesticide Sd., (Nippon Noyaku Gakkaishi), 2. 41 (1977)

46 NOVEL CLAY CARRIERS FOR THE CONTROLLED RELEASE OF ORGANIC AGROCHEMICALS

. GERSTLZ . MINGELGRINU , . NASSEA , R Department of Soil Physical & Environmental Chemistry, Institute of Soils & Water, ARO Volcane Th , i Center, Bet Dagan, Israel

Abstract Controlled-release (CR) formulation alachlof so atrazind ran e herbicides were prepared using sodium alginate and pectin as natural, biodegradable matrices and clay minerals as inert fillers. The release of the two herbicides from differen tformulation R typC f eo studies wa s statin di c water e releasTh . f eo alachlor from alginate based formulation a commercia d an s l formulation i n sandy loam soil and it movement in a soil column was also studied. The rate of release was affected by the type of clay and the size of the formulation beads. The addition of Fisher bentonite to the alginate reduced the rat releasf eo herbicidese th f eo release Th slowes . ewa r fro largee mth r beads and from those prepared using high viscosity alginate. The release of alachlor s movemenit e d soith an l n i t throug e soith hl colum s fastewa n r froe mth commercial formulation than the CR formulation.

1. INTRODUCTION

An ideal delivery system is one that delivers or releases the active ingredient (a.i.) at a constant rate directly to the site of desired action. Tremendous efforts have gone int developmene oth reliablf o t e formulation controllee th r fo s d (slow) releasf eo chemical botn si h agricultur medicind ean e [1-3]. Controlled release enables efficien economicad an t l applicatio e activth f neo ingredient while at the same time reducing the danger of undesirable side-effects such as environmental pollution. The controlled release of pesticides and other organic agrochemicals, can, in many cases, permit safer, more efficient and more economical crop production. The results present herein represent our efforts to prepare a controlled release formulatio agriculturar nfo froe lus m natural polymer clad san y minerals studyiny B . g interactione th s betwee active nth e ingredien formulante th isolatn d ca an e t e eth w s factors which contro release th l e proces thud san s achieve almos desirey an t d ratf eo releas activn a f eo e ingredient.

2. MATERIALS AND METHODS

Materials: The polymers studied were alginate and pectin. Alginic acid is an unbranched, hydrophilic colloid consisting of D-mannuronic and L-glucuronic acid

47 residues. Various alginates differin thein i g r viscosities were used generaA . l grade Na-alginate (Fluka, Buchs, Switzerland) was used for all studies unless otherwise indicated. In addition, Na-alginates of low, medium and high viscosity (250, 3,500 and solutions% 2 r 14,00fo s ) cp 0wer e obtained from Sigma Chemica . (Stco l . Louis, U.S.A.) e seconTh . d polymer a studiecommercia s wa d l grade pectin (partially methoxylated poly-D-galacturonic acid) supplie Israeln a y db i food manufacturer. e clayTh s studied were Na-montmorillonite (Fisher Scientific, U.S.A.), kaolinite (Supreme, St. Austell, U.K.), Edasil (a commercial product with « 50% bentonite, Germany) and several attapulgites (Diluex and Minugel, Floridon, GA, U.S.a sampl d Aan e from Spain). The pesticides studied were alachlor and atrazine. Technical grade material was used for all the experiments. Experimental: The formulations prepared are based on the gelling properties of alginate and pectin in the presence of divalent cations. Suspensions of the polymer, clapesticidd yan e were prepare shows da ratie Tablth n n i o n i 3:4:e1 2 (3:0: thosr 2fo e préparâte t containinsno g clay slowld an ) y drippe molad5 0. int n roa solutio CaClf no 2 with the aid of a peristaltic pump. The resulting beads were allowed to gel in the CaCltotaminutea 4 r f o lfo 2 s after which they were filteret a allowe d y dan dr o dt room temperature controlle e beade b sizth orifice tubine n e f th esth ca o Th . y f edb o g used for dripping the suspension into the solution. One preparate was prepared with a smaller orifice to study the effect of bead diameter on release. The actual concentration of the active ingredient in the beads was determined by dissolvin beade Na-citratga th n si e solutio extractiod nan n into ethyl acetatee Th . organic phas thes enwa analyze (seC eG below)y db . e releas active Th th f eo ingredients fro differene mth t préparâtes into water s studie wa closea n di d system beade (FigurTh . s 1) were e suspende stainlesa n i d s steel mesh basket in doubly distilled water. The water circulated through a variable wavelengt HPLV hU C HPLdetecton a f Co d pumr ai returnewitd e e pan hth th o dt glass beaker. The detector was set at a wavelength suitable for detection of the a.i. atrazine)r fo m detectoe n alachlor 0 Th . fo (2326 rm d 0n respons ran founs e b e wa o dt lineaentire th n eri rang concentrationf eo s encountered detectoe Th .connecte s rwa o dt a strip chart recorder and a trace of the solution concentration with time was monitored continuously. release Th f alachloeo r from selected préparâtes into s determinedsoiwa l A . 100 g sample of a sandy loam soil (10% clay, 0.3% OC) was packed into a glass beaker, brought to field capacity and a predetermined number of beads was placed ont slightld soie oth an l y pushed into place, after which s f soi, o anothe wa lg 0 10 r beadse beakerth e f o Th applie. p sto wern do e covered with parafil weighed man d dan incubated at 25°C. Several small holes were made in the parafilm to allow gas exchange beakere .Th s were weighed twic losy weighn san weei ea d kan t mady b p eu applying water. At different times three beakers were sampled by removing the beads and thoroughly mixing the soil. The soil was extracted with water:ethyl acetate (1:1 ratio) and the organic phase analyzed for alachlor by GC. The beads were also analyzed for the remaining alachlor content as described above. As a control, commercially formulated alachlor was applied to the same soil, incubated under identical condition sampled san d periodicall alachlorr yfo .

48 magnetic stirrer 0

Figur : Flo1 e w through syste r direcmfo t measuremen f rato t f releaseo f organico e s from clay-polymer beads. Detector attenuatio recorded nan r setting o givt t e se s about 0.8 FS for 100% release.

The movement of alachlor in soil columns was studied in the same sandy loam soil. Air dry soil was packed into lucite columns (20 cm i.d., 50 cm length) to within 10 cm of the top. Beads or a commercial formulation were placed on top of the soil and another 5 cm of soil applied. The columns were irrigated with tap water to field capacit thed yan n onc weeeweeka 5 r kfo s they were irrigate wate e mako dt th p reu lost by evaporation. At the end of the experiment the bottom of the column was still dry. After 5 weeks the soil was extruded from the column and the alachlor content in different layers determined beade Th . s were also recovere theid dan r alachlor content determined.

3. RESULTS

3.1 Release into water

compositioe Th varioue th f no s controlled release formulation f alachloo s d an r atrazine is presented in Table 1. The beads were generally spherical in shape and in all case beade sth s werradiua f eo s wit of-1.m exceptioe hm th 2 5- preparatf no 8 e# where the radius of the beads was considerably smaller, about 0.4 - 0.6 mm. The average loadin l alachloal r gfo r préparâtes (excludin s 12.5±1.7wa r ) fo g#8 d %an atrazine 11.4± 1.1%. The results, summarized in Figures 2-6, are presented as Mt/Mo, the relative amount of alachlor released, where Mt is the mass of alachlor released up to time t and originae Mth Os i l mas alachlof so r containe beadse th n di . release Th f alachloeo r fro differene mth t préparâtes depend greaa o st t degree on their composition release Th . alachlof eo r from beads prepared with different clays as fillers with alginat presentes ei seee b nn Figurthan dca i onle t I tth y. e 2 cla havo yt e

49 Tabl . eI Compositio loadind nan f slogo w release formulations containing alachlor and atrazine.

Preparate #______Composition______% Active Ingredient

Alachlo. a r

1. Alginat eFishe- r Bentonite 14.9 2. Alginat eEdasi- l (Germany) 11.5 3. Alginate only 11.7 4. Pecti nFishe- r Bentonite 11.6 . 5 Pectin only 13.9 6. Alginate - Pectin - Fisher Bentonite 9.4 7 9. . 7 Alginat ePecti- n 8. Alginate - Fisher Bentonite (small beads) 6.6 9. Low viscosity alginate - F. Bentonite 12.6 10. Low viscosity alginate only 16.6 11. High viscosity alginate - F. Bentonite 12.0 12. High viscosity alginate only 13.0

b. Atrazine

13. Alginate - Fisher Bentonite 11.8 14. Pectin - Fisher Bentonite 12.4 15. Alginate - Pectin - Fisher Bentonite 10.0

any impact on release was Fisher bentonite. The addition of Fisher Bentonite to alginate based beads exampler fo , , reduce rate releasf sth e o (Figur 2 factoa « y ef b o r e 3). The effect of Fisher bentonite on a pectin based preparate is also quite evident but relatively less pronounced Edasilf o e us montmorillonitia , e Th . c based materiald ha , releas e effeco th n n alachlof o teo systee th n ri m under study (dat t shown)ano . 'This may be due to the fact that Edasil is only 50% clay and contains a fairly high percentag coarsf eo e material compare Fishee th do t r Bentonite. Combining alginate and pectin for bead preparation resulted in release rates greater than with either polymer alone (Figur . Afte e3) hour0 e 4 r th sf o nearl % y90 alachlo s releasewa r d fro alginate mth epecti- n based beads d whilan 8 e 4 onl , y72 alachloe th f o r % containe28 pectine th n di , alginat alginatd ean e- Fishe r bentonite préparâtes, respectively, were released. Addition of Fisher Bentonite to the alginate - pectin mixture had a dramatic effect in reducing the rate of release when compared to polymee th r mixture alone t thi,bu s tri-component mixture still released alachlor faster than the alginate - Fisher bentonite preparate. Although the actual mechanism of release has yet to be fully elucidated, it is postulated that the different behavior of alginate and pectin, as well as the effect of the clay, are the results of differences in the 3-dimensional network poref so s forme eacn di h system. Further wor thin ko s topis ci in progress.

50 0.40

0 10 15 20 25 Time (hours)

Montmorillonite Kaolinite Minugel Spanish Attapulgite Diluex

Figure 2: Release of alachlor from alginate-clay beads.

0 4 10 0 3 20 50 60 Time (hours)

- Alginate-F-* B —t— Alginate-Pectin -$K- Alginate -B- Pectin-FB -X- Pectin -jar Alginate-Pectin-FB

Figure 3: Release of alachlor from polymer-clay based beads

51 10 20 30 40 50 60 Time (hours)

AJginate-FB Alginate-FB, small

Figure 4: Effect of bead size on the release of alachlor.

0 5 0 4 0 3 0 2 0 1 0 60 Time (hours)

V L - -t B LF V+ HV + FB -&- HV

Figur Effec: e5 f alginato t e viscosity on the release of alachlor.

52 0.25

S Alginate Pectin Alginat Pectie+ n

0.00 10 20 30 40 50 Time (hours)

Figure 6. Release of atrazine from polymer Fisher bentonite préparâtes.

Figur showe4 effece s th beaf o t rate d releasef th sizeo n eo expecteds A . e th , smalle beae th r d diamete fastee ratrth e releasef th reo . Accordingly t tooi , hour2 k1 s alachloe th releasee f b o o rt % fod50 r fro smallee mth r beads during which time only active 12th %f eo ingredien releases twa d fro largee mth r beads. Alginate is available in several different forms which differ primarily in the degre f polymerizatioo e e alginatth f consequentld no an e e viscositth e n th i y f o y resulting suspensions. Beads containing alachlor, both with and without fisher bentonite as an additive, were prepared with both a high and a low viscosity grade alginate (Sigma) and the rate of release of the active ingredient was determined. These préparâtes were compared to an alginate of an intermediate viscosity (Fluka) which is e standarth use s a d d polyme e presenth n i r t study (Tabl . 1) Froe e resultmth s presente thae difference se th t n rate releasf ca Figurth en o di n e i e w f alachlo eo 5 r from the high and low viscosity alginate beads was small but both these alginates responded differently to the addition of the clay. Furthermore, the release of alachlor fro standare mth d alginate beads (Fig considerabls i ) .3 y faster than fro hige md th han low viscosity préparâtes. The standard alginate (Fluka) is polydisperse, encompassing wida e rang polymerizationf eo , wherea viscositw lo hige d sth h an y alginates (Sigma) were prepared to give a narrow range of molecular weights. This difference in composition results, we believe, in different internal networks of pores in the various alginates and hence, in different rates of release of the active ingredient. The release of atrazine to water from 3 different polymer - Fisher Bentonite préparâte shows si Figursemblance n i Th . e6 releasn ei préparâte3 ee rateth r sfo s si in contrast to the results for alachlor (Figure 3) in which the alginate - Fisher Bentonite preparate released alachlo considerabla t ra y slower rat othee e th tha d r npréparâtesdi .

53 A second difference between the release rates for the two compounds studied is the much slower rate of release for atrazine. After 30 hours only 17.5% of the atrazine containe beade releases th n di swa compares da 22-42o dt case th f alachlor eo %n i . This is not suprising given the difference in the solubilities of the two compounds. Alachlo nearls ri timey8 solubls sa atrazins ea e (242 |ig/m(J,g/ml0 3 s v l , respectively [4]). Pfister et al. [5] observed that the rate of release of several pesticides from alginate beads (without clay fillers) corresponded to the aqueous solubilities of the active ingredients.

Releas2 3. soin ei l

e releasTh f alachloo e r into soil under static condition s subsequenit d an s t degradatio e degradatios presentei nTh . 8 n Figurei f dd commerciallo n an 7 s y formulated alachlor in the sandy loam soil is fairly rapid, with a half-life of about 4 expectn ca days e ,W . therefore, that alachlo thin i se soius r l e highlwouldb t yno effective l casesal standare n I .th , d size beads base alginatn do e (Figur exhibite) e7 da slow and relatively steady release of the material over the course of over 61 days, resultin averagn gi e soil concentration mg/kg9 0. - 2 rang.e Throughou0. th f n eo si t most of the incubation period, the alginate - Fisher Bentonite beads gave the slowest release. The smaller beads exhibited a higher soil concentration («1.1 mg/kg) which is in line with their faster rate of release to water. The pectin based préparâtes (Figure 8) behaved similarly to the alginate based beads, exhibiting a slow but steady release active ofth e ingredient notabl.A e exceptio pure th ee pectinar n beads which displayed a sharp peak in the release of alachlor between 40 and 55 days. This may have resulted

2.0 •&• Alginate + FB "^"Alginate + Edasil Comm. Formulation J8P ^ Alginate only ® Small beads

o. 20 30 40 50 60 70 Time (days) Figur : Alachloe7 r concentratio sanda n ni y loam soil from alginate based préparâte frod san ma commercial formulation.

54 2.0- ,Sf ^Alginate + FB ^Pectin only "5l S Comm. Formulation -*• PectiB F + n

10 20 30 40 50 60 70 Time (days) Figure 8: Alachlor concentration in a sandy loam soil from pectin based beads e commerciath , l formulation and from the standard alginate preparate.

from the observed swelling and disintegration of the pectin beads. The amount of material remainin e differenth n gi t beads, M(f)/M(0), show a slossteadd wan y decrease with time (Figur . Afte e9) day 1 6 rincubatiof so fine nw d betwee% 75 - 5 n3 of the alachlor still in the beads. The lowest fraction of alachlor was left at the end of the incubation in the small diameter beads, in agreement with the higher release into soil from these beads (Figure 7). Nègre et al. [6] studied the persistence of commercial formulations of alachlor and microencapsulated alachlor. Their results were similar to ours in that the half life of alachlor in the commercial formulation was 9-10 days compared to > 56 days for the microencapsulated material.

Transpor3 3. soin ti l

Figurmovemene n I th 0 e1 t throug soiha l colum f alachlono r released from alginate - Fisher bentonite beads is compared to the movement of alachlor applied as e commerciath l e thae peaformulationse th t kn concentratioca e W . f alachloo n r released fro e clamth y- polyme r based preparat s stilei l centere e deptth f t ho da application, wherea alachloe sth r applie commerciae th s da l formulation moved into the soil. Furthermore, a mass balance for alachlor in the columns (Table 2) shows that while alachloe onlth yf o abou stils i r % l 50 tpresen day9 2 t s after applicatioe th s na commercial formulation, about 90% is still present when alachlor was added as the controlled release formulation ,beadse most stil th i bead e n f i lth t o t . f Thussno o e us , only retards leaching of the active ingredient, but reduces degradation of the compound as well.

55 1.2

0.2- "^Alginate + FB "^"Alginate + Edasil ^Alginate only -o- Small diam. Pectin + FB Pectin only

0.0 \ I \ 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Time (days) Figure 9: The amount of alachlor remaining in beads after incubation in a sandy loam soil at field capacity.

Alachlor content (mg/kg) 0 5 10 15 20 25 30

Application c Depth 5

10

u 15 Polymer-clay beads

20 * Commercial product

25

30 FigurelO: Distributio f alachlono columnn ri a f o s sandy loam soil when applie commerciaa s da l formulatio alginatn i d nan e based beads.

56 Tabl . AlachloeII r mass balanc soie th l n columnei s afte days9 2 r .

Column System Alachlor Alachlor Alachlor left Applied Applied (mg) in soil (%) in beads (%)

1 Beads 5930 9.7 77.7 2 Beads 6124 15.3 77.4 3 Commercial 7104 54.6 4 Commercial 7104 45.7

4. DISCUSSION

Controlled release formulations for agricultural chemicals serve two main purposes. The prevenn yca t leachin f mobilgo r semi-mobileo e chemical grouno st d water and they can retard degradation of chemicals thus extending their effective life time in the soil. resulte Th dato st e indicat investigatee th e f thao d t ai wit de hth formulations , fine e tuninreleasth f o ge rat f activo e e ingredient s possibli s y matchinb e e th g compositio preparate th f no e wit propertiee hth active th f eso ingredient.

ACKNOWLEDGEMENT

This work was funded by a grant from the German-Israeli Foundation (GIF) for Scientific Researc Developmenth& .

REFERENCES

[1] CAOUDRY, B.M., PRABHU PRASAD, B. and LAKSHMI, M. New intralameller pesticide-metal-montmorillonite complexes: A novel technique for controlled release. J. Agric. Food Chem. 37 (1989) 1422-1426. [2] DEASY, P.B. "Microencapsulation and Related Drug Processes", Marcell Dekker, New York, (1984). ] [3 KYDONIEUS, A.F. "Fundamental Concept Slof so w Release" Controlle: In . d Release Technologies: Methods, Theor Applicationsd yan , Vol (A.F. .1 . Kydonieu s, ed.C )CR Press, Florida (1980) 1-19. ] [4 WORTHING, C.RHANGED .AN , R.J. (Eds. Pesticide )Th e Manual ednh e 9t , ,Th British Crop Protection Council, (1991). ] [5 PFISTER , BAHADIRG. , KÖRTEd an . M ,Releas. F , e characteristic herbicidef so s from Ca alginate gel formulations. J. Controlled Release 3 (1986) 229-233. [6] Negre, M. Alachlor dissipation in soil as influenced by formulation and soil moisture. J. Agric. Food Chem. 40 (1992) 1071-1075.

Next page(s) left blank 57 INFRARE FT-RAMAD DAN N STUDIEF SO PESTICIDE-ORGANOCLAY INTERACTIONS IN RELATION WITH FIXATION-RELEASE PROCESSES

. PROSTR . VEMOND-LABOUDIGUA , E Station de science du sol, Institut national de la recherche agronomique, Versailles, France

Abstract

e mechanise studth Th f o y f pesticidmo e fixation-releas organoclayn o e s wa s undertaken to get a better understanding of the remanence phenomenon and the controlled release processes approace Th . h included fixation isotherms, X-Ray determinations andn i , addition to these classical methods, the study of pesticide-substrate interactions at a molecular level using spectroscopic methods, in order to obtain more information on the physico- chemical state and the location of the xenobiotics, and to determine the nature and the strengt f bondho s involve fixatioe th n di n process. Thinecessara s i s y e ablb ste o et o pt predic fate pesticidef th teo soiln si . Vermiculite hectorite-decylammoniud -an m were chosen as models of organoclays. Isotherm and X-Ray data show that fixation was much higher for vermiculite than for hectorite and that intercalation occured in the case of vermiculite. Infrare d Ramaan d n spectroscopies showed that pesticide-organoclay interactions induced change substrate th n si e itselfixee t alsth dfbu on i pesticide molecule whic e longeo hth n d rha structur cristallinee th f dissolvee eo th r o d forms. This specific physico-chemical stats ewa analysed.

1. INTRODUCTION Seed coatin relativela s gi y common practice treatmenth r fo e f seedo t s with fungicide insecticidesd an s controllee Th . d releas f suceo h xenobiotics incorporate seen di d coating encapsulater so formulatioa n di necessara s ni y approac increaso ht e their efficiency and to protect our environment by reducing the amounts of pesticides being used. Several studies have concerne e slow-releasdth f pesticideo e s whic s mainlhi y dependen biologicae th n o t physico-chemicad an l l characteristic soie th l f systeso mwhico t h they are applied. This aspect is of importance since the flow of pesticide release is part of their efficiency, but it cannot be considered as a control of the release. Only a better understanding of the mechanisms involved in the fixation of molecules on substrates used in coatings or in formulations will allow us to act on the release of these molecules. The knowledge of the interactions involved in the adsorption process at a molecular level should choice resulth n f suitabli teo e formulation substrate bettea r fo sr controrelease th f f o leo active molecules and a reduction of the amounts spread in the environment. e classicaTh l method adsorption i sr fa use o ds n studies consis determininn i t g adsorption isotherms. This approach makes it possible to determine the extent of adsorption of a compound and its affinity for different substrates; desorption isotherms provide informatio release th n f adsorbeno eo d molecules. However, different interpretatione th f o s results show that thes t sufficiene no dat e aar explai o t t n mechanisms involve adsorption di n/ desorption processes [1,2]. An investigatio e molecularth t na leve s needei l o studdt y these mechanisms: characterizatio f adsorbeno d molecules (locatio physico-chemicad nan l state) should leao dt information about adsorbate-surface interactions. Spectroscopic method amone sar mose gth t interesting approaches for this type of study since analyses can be made on the adsorbed t necessitatmoleculeno o d d s extractioneit an , . Example sutilizatioe r existh o f R o tI f no R spectroscopieNM e studth f adsorptioo r y fo s f neutrao n l organic molecule n soio s l constituant formulatiod an s n substrates [3,4]. f spectrometriHowevero e us e th , c methods si often limited by the low level of adsorption, and the lack of knowledge about the nature of the substrate.

59 The purpose here was to develop spectroscopic methods in order to characterize the physico-chemical state and the location of adsorbed molecules and to determine interactions involved in adsorption / desorption mechanisms; they are complementary data to the former approach, useful to explain results obtained at a macroscopic level coming for example from adsorption isotherms n thaI . t way, organoclays were selecte s modea d l substrates able to fix xenobiotics with very low solubility in water. Classical adsorption isotherms were performed wit neutraha l pesticid clay-alkylammoniuo tw n eo m substrates. Results were studied together with those obtained at the molecular level, by means of spectroscopies. Diffuse reflectanc e UV-Visiblth n i e e region permitte o follot e s u dwth modification e electoth f o s n e adsorbe th stat f o e d molecule absorptioR I ; d Ramaan n n diffusion were found very useful to characterize the substrates and to study the arrangement of adsorbed molecules. e knowledgTh e gaine shouly thin di wa s d contribut bettea o et r controe th f o l releas f formulateo e d pesticides. This approach could als e applieb o o describt d e th e behaviour of pesticides already introduced in soil, since their mobility and bioavailability will greatly depen adsorption do interactiond nan s with soil constituents.

. 2 MATERIAL METHODD SAN S

2.1. The substrates Two clay-decylammonium substrates were used in this investigation. Organoclays were obtained with Llano vermiculit d hectoritan e e from Hector which differ froe mon another in the location of the deficit of positive charge (for vermiculite and hectorite, the deficit is due to isomorphic substitutions in tetrahedral and octahedral sheets of the layers respectively) e densitth n f i chargeo y d e catioan ,th : n exchange capacit s equai y o t l 210meq/100 r vermiculitgfo 76meq/100d an e r hectoritegfo . This result differenn i s t interactions betwee hydratee nth d interlayer siliccatione th ad surfacsan vermiculitn ei d ean in hectorite. It was shown that hydration water molecules of compensating cations are involved in hydrogen bonds with oxygen atoms of the surface only when isomorphic substitutions are located in the tetrahedral sheet of the layer. Clay-alkylammonium complexes were prepare exchanginy db g mineral interlayer cations with decylammonium cations [5]. Basal spacing clay-decylammoniuf so m complexes were obtained by examining the d(001) X-Ray reflections. Infrared and Raman spectra were used to determine the arrangements of organic cations in the interlayer spaces of both phyllosilicates (models of arrangements are presented on figure 1). The type of interactions between alkylammonium clae cationth y d surfacsan e wer espectrR I deduce e th a n frodo m variatione th absorptioe th f so n frequencie botf o s h clacationd yan ; Raman spectroscops ywa specially usefu determino t l e intra inter-moleculad -an r conformation hydrocarbof o s n chains and their lateral packin e complexesth n s showgi wa t ne casI . th f vermiculitetha eo n i t - decylarnmonium, hydrocarbon chains have their axes inclined 55° to the silicate planes and generate a 20.8 A basal spacing; their arrangement in interdigitated bilayers is close to that of cristallized decylammonium chloride. Terminal NH3+ groups penetrate deeply into the hexagona lgroup H surface holeN th involvee d f sar o ean hydrogen di n bonds with oxygen atoms of the silica sheet. In the case of hectorite, hydrocarbon chains lie on the silicate surface without enhancing the spacing (13.5 A), interactions between chains are comparable to thos decylammoniuf eo liquie th mn di phase[5].

2.2. Pesticide herbicide Th e use 2-sec-butyl-4,6-dinitrophenos dwa whicf o l commoe hth n name is dinoseb; figure 2 gives its structural formula. This nitro compound was chosen for its UV- Visible absorption characteristics. Dinoseb is sparingly soluble in water: 52 mg/1 at 20°C.

60 (b)

Figur Model. e1 hydrocarbor sfo n chain arrangement vermiculitee th f so hectorite d an ) -(a - decylamrnoniu substrate) m(b s

Figure 2. Structural formula of dinoseb (2-jec-butyl-4,6-dinitrophenol)

2.3. Methods The adsorption isotherms of dinoseb on the clay-alkylammonium complexes were f aqueoucarrieequilibratiny o b l f t substrato m ou dg 0 s 2 m solutio n 0 i e5 g f dinosebo n . Samples were 4 shakehour2 t r 20°Ca sfo n . After equilibration e suspensionth , s were centrifuge supernatante th d an d s analysed spectrophotometricall UV-Visibly yb e absorption absorptioe th t a n maxim dinosebf ao amounte Th . dinosef so b adsorbed were calculated from difference th concentration ei n befor afted ean r equilibration. The interplanar spacing for the organoclays and for their respective dinoseb complexes were determined by X-Ray diffraction, using several orders of 001 reflections. The measurements were mad orienten eo d deposit fluoridn so e plates. At the same time, organoclay-pesticide complex samples were examined by spectroscopic methods. UV-Visible diffuse reflectance and FT-IR absorption spectra were obtained with the oriented deposits. For FT-Raman diffusion examination (the wavelength of the laser source was 1.06 fi), samples were gently pressed in an appropriate cell.

61 3. RESULTS AND DISCUSSION

3.1. Adsorption of dinoseb on vermiculite-decylammonium

3.1.1.___Location of the adsorbed molecules: compared analysis of adsorption isotherms and X-Ray data. figurn O , curve3 show) e(a amounte sth dinosef so b adsorbe V-decya n do s a ) (X l function of the equilibrium concentration Ce, and curve (b) shows corresponding evolution of basae th l spacin organoclay-pesticide th f go e complex. It should be noted that the amount of dinoseb adsorbed on vermiculite- decylammonium is very high: the X value reaches 90 mg / g of substrate for the higher value of CP.

100 T X d(001) (mg/g) 90 • * (Ä) 80' (0) . 30 70' 60- • •28 50' * * * •26 40' * (b) . 30- 24 * 20- • .. . 22 10- 'fr * * 2n 0 2 5 1 10 25 equilibrium concentration (mg/1)

Figur . Curve3 : amountea dinosef so b adsorbe mg/n (i vermiculite-decylammoniuf gdX o m substrate)

functioa equilibrius e a th f no m concentratio mg/1)n (i nC ; curv : evolutioeb basae th f lno e spacing of corresponding organoclay-dinoseb complex d(001) (in A) as a function of Ce.

Simultaneous analysis of both curves shows that adsorption occurs in several steps. At the beginning of the process, adsorption reaches a first plateau, and then remains invariant with increasing concentrations, while the basal spacing of the substrate does not vary from its initial value. This indicates that, at this point of adsorption, only sorption on external surfaces takes place. Then, both curves sho wshara stepw increaspne takea d sean place where adsorption gently increases with increasing concentration, while basal spacing of the complex remains at its new value of 26 A. Several 001 reflections corresponding to the 26 A value on this stage shows that layer spacing is regular throughout the organoclay- pesticide complex. This phenomenon could be explained by the fact that hydrocarbon chains of the organoclay substrate regularly slided along each other, creating a vacant space at the f eaco hd en chain, neae silicatth r e layer (figur . Thi4) e s slip could result froe mth intercalation of a few pesticide molecules, which are located in these sites, acting as wedges holding the silicate layers apart, and allowing more molecules to penetrate between the layers.

62 Figur . Hydrocarboe4 n chain arrangemen vermiculite-decylammoniuA 6 2 e th n ti m dinoseb- complex.

Onestimatn eca e from vermiculite characteristics numbee th , f sitero s that could createe b sucy d b expansio n ha interlayee th f no r spaces calculatd an , filline eth g ratio from adsorbed amount pesticidesf so . This rati durino% varielas0 e 2 g th t o t stepd fro% t :i 0 m1 seeme e sufficienb o t d o maintait t e layerth n st explainei e adsorbe apard th an y t wh dd amounts kept on increasing beyond the last points of the curve. Molecular modelling applied dinosee th o t b molecule indicate basa compatiblw s dne wa thal e spacinA th t 6 2 ef g o wit e hth volume occupie herbicidy db e moleculehydrocarboe th f o d sen locatene chainon t da , near the silicate surface.

3.1.2. Pesticide / substrate interactions : spectroscopic data Diffuse reflectance spectr f organoclay-dinoseao b complexe e UV-Visiblth n i s e region showed a dramatic shift of the main absorption band to lower energies, compared to spectra of both pure dinoseb and the deprotonated form of the molecule. This must be the resulincreasen a f o t d electron delocalizatio aromatie th n i c ring. Figur 5 showe e R spectr1700-115I sth n i a 0 cm-1 regio f vermiculiteo n - decylammonium (curve a), dinoseb adsorbed on this substrate (corresponding to Ce = 20 mg /1 on the isotherm curve) (curve b), and pure dinoseb in the solid phase (curve c). This regiospectrue th f mosstude e no th th mf dinoser s tyo i usefufo e bon l functional groups vibrations. The strongest bands observed on these spectra are listed in table 1. Adsorption modes of dinoseb were attributed with the help of deuteration experiments, by comparing spectra obtained in several solvants and by studying the deprotonated form of the molecule. mose Th t significant features observe spectrue th organoclay-dinosee n di th f mo b complex, compared wit pure hth e dinoseb spectrum wer strone eth g absorptioe shiftth f o s n bands characteristi f bot co e aromati hth e functionacth rind gan l groupmoleculee th f o s . modificatiopara n i Thes o e t du intra-moleculae ee th effectb f y no ma s r hydrogen bond that concern grouppurH e sO th NO ed n i smolecule an 2 creatioe resula th s f a , o tf specifi no c interactions betwee molecule nsubstratee th th d ean . Increased electron delocalizatioe b y nma due to a spatial rearrangement of the aromatic ring substituents or to charge transfer interactions with the surface. Further modifications observed on the organoclay-dinoseb spectrum, compared with the original organoclay spectrum, indicated a decrease of clay/alkylammonium cation interactions, which resulted from a slight withdrawal of ammonium groups from the

63 1700 1600 1500 1400 1301200 0 Wavenumbe) 1 - m c ( r

Figure 5. Infrared spectra: a, vermiculite-decylammonium; b, adsorbed dinoseb (for Ce=20 mg/1); c, pure dinoseb.

Table 1. Vibration bands ( in cm"* ) in the 1700-1150 cur* region of the IR spectra. va, vg: asymmetric and symmetric stretching vibrations; 8: bending vibration , s: shoulder.

pure dinoseb vermiculite-decylammonium adsorbed dinoseb (KBr) substrate

1632s vaNH 1632 e 1613 vC-C 1604 vaNH 1606 1566 vsNH C C- v 1554 1565 1529 vaNO2 1556 1506 1465 5CH2,CH3 1467 2 CH S 1454 1458s 1458s 1435 v C-C 1420 1383 8 CH3 CH8 3 1394 1378 1378 8 CH3 1365 1341 vs NO2 1342 1310s VSNO2 1306 1259 ß OH C-Ov H 124 s 5 1230 1203 1212 1167 1164 1177

64 hexagonal silicate holeth f so e layer, together wit diminutioha strengte th f n o hydrogef ho n bonds between NH groups and surface oxygen atoms. Raman spectroscop uses ywa d previousl determino yt e intra inter-moleculad -an r conformation f o hydrocarbos n e clay-alkylammoniuchainth n i s m substrates. When comparing spectr vermiculite-decylammoniuf ao correspondine th d man g dinoseb complen xi stretchinH C- e th g region (figurobservee w , e6) d dramatic shifts whic characteristie har a f co diminution of lateral interactions between the hydrocarbon chains of the organoclay substrate [6]. These modifications showed that rearrangements occured in the interlayer space of the organoclay pesticido t e du , e adsorption.

3100 3050 3000 2950 2900 2850 2800 2750 2700 Wavenumber ( cm -1 )

Figure 6. Raman spectra: a, vermiculite-decylammonium; b, vermiculite-decylammonium - dinoseb complex.

3.2. Adsorptio othef no r nitrocompound vermiculite-decylammoniun so m In orde determino rt role hydrophobif eth eo hydrophilid can c dinosee poleth f so b molecul adsorptioe th n ei n process, adsorption experiments were carrie t witotheo dou htw r nitro compounds: 2,4-dinitropheno- l differs from e dinoseabsence sec-butyth th y f b o be l hydrophobic ring substituent. This compound is much more soluble in water than dinoseb, - 2,4-dinitroanisole is the methyl ether corresponding to 2,4-dinitrophenol. The absence of hydroxyl group implies that neither intra-molecular hydrogen bond (which involves bot group phenol)H e h O th NO d n sr i inter-molecula 2an no , r hydrogen bonds (with the substrate existn )ca .

3.2.1. Adsorptio 2.4-dinitrophenof no I Simultaneous analysis of isotherms and X-ray results show that adsorbed amounts of 2,4-dinitrophenol were comparable to those of dinoseb, and that interlayer adsorption occured. Vibrational and electronic variations observed on spectrometric data were characteristic of specific interactions between the molecules and the substrate.

3.2.2. Adsorption of 2.4-dinitroanisole No significant adsorptio observes nwa d with 2,4-dinitroanisol spectrometriy eb c methods X-rad an , y spectrt sho changey no wbasaan d e adi th n lsi spacin substratee th f go .

65 From thes e deduce b resultsn ca dt i , thae alkyth t l chai f dinosebno , which contribute solubilitw lo molecule e e th th f yo o s t , doe plat maie sno y th nadsorptioe rolth n ei n process. However, the hydrophilic pole constituted by the hydroxyl group plays a fondamenta processle rolth n ei .

3.3. Adsorption of dinoseb on Hectorite-decylammonium From the first points of the adsorption isotherm (figure 7, curve a), it can be seen that hectorite-decylammonium adsorbs much lower amount f dinoseo s b than vermiculite- decylammonium. X-ray spectra of the dinoseb complexes give a basal spacing of 13.5 A for all the points of the curve (figure 7, curve b), which corresponds to the initial d(001) value of the substate. Prolonged contacts of the substrate with highly concentrated solutions of dinoseb did not give different results.

5r X d(001) (mg/g) (A) 4,. (a)

3-- * * •• 13.5 (b)

1 ••

4 6 8 10 12 14 equilibrium concentration (mg/1)

Figure !.. Curve a: amounts of dinoseb adsorbed X (in mg/g of hectorite-decylammonium substrate) as a functio equilibriue th f no m concentratio mg/1)n (i e n;C curv : evolutioeb basae th f ln o spacin g of corresponding organoclay-dinoseb complex d(001) (in A) as a function of Ce.

IR and Raman spectra of the organoclay-dinoseb complex showed new bands of low intensity, which were characteristic of the pesticide molecule; neither substrate nor pesticide specific bands showe shifty dan . deducee W d from these results thaadsorbee th t thad an td onlw amounlo y s wa t external surfaces were concerne case hectorite-decylammoniumth f eo n di .

4. CONCLUSION addition I classicao nt l adsorption isotherms, spectroscopic methods were usen di this study to determine the location and the physical state of a pesticide interacting with two different organoclays: vermiculite-decylammoniu hectorite-decylammoniumd man . Adsorptio e s mucfounb wa no t hd more intense cas f th vermiculito e n i e e substrate than in that of hectorite. For the latter, adsorption occured only on external surfaces and adsorbed amounts were rather e cas lowth f vermiculite-decylammoniumo en I . ,

66 adsorption occured in different steps including a primary adsorption on the external surfaces and a secondary adsorption where molecules were intercalated between the silicate layers. The resulting organoclay-pesticide complex had a highly regular structure. Adsorbed molecule electronid sha c modifications which were characteristie th f co formation of strong interactions with the substrate. Adsorption experiments carried out with 2,4-dinitrophenol and 2,4-dinitroanisole molecules showed that the hydrophilic OH group of dinoseb playe maia d ne adsorptio th rol n i e n processe hydrophobith t bu , c sec-butyl ring greaa substituen f o tt see e importanceb no m o t d di t . These results suggested that the main interactions existing between adsorbed molecule substrate th d an s e were specific. They were probabl hydrogea f yo n bond nature involving both the polar groups of the molecule (OH and NO2) and the equivalents on the silica sheet (oxygen atom f substituteso d tetrahedron case vermiculitf th eo n si NHe th 3d ean group f decylammoniuo s m cations). Dramatic shifts observe+ vibrationan do electronid an l c spectr f adsorbeo a d molecule e cas f th adsorptioo e n i s f dinoseo n n vermiculiteo b - decylammonium indicated tha charga t e transfer phenomenon delocalizef !o d electrone th f so molecul adsorptiooccue y th ema n i r n process silicate Th . e surfac f hectorit eo t hav no ed edi hydrogen bond acceptor sites, and no pesticide intercalation was observed in this case. Hydrophobic r forceWaalde n ss Va sucinteraction s a h e involvear se th n i d intercalatio neutraf no l pesticide interlayere th n si phyllosilicatef so s since adsorptio thesf no e molecules on hydrated clays such as vermiculite or hectorite saturated with Na+ cations is not significant. However lace f adsorptioth ,k o dinosef no hectorite-decylammoniun bo r mthao t of 2,4-dinitroanisol n vermiculite-decylammoniumeo e stronth d g an ,interaction s observed between intercalated molecules and vermiculite-decylammonium substrate show that hydrophobic interaction t sufficienno e sar induco t t intercalatioe eth f moleculesno thad an ,t specific interactions with hydrophilic poles of the adsorbed molecules are necessary in the process.

REFERENCES

[1] CHIOU, C.T., PETERS, L.J., FREED, V.H., A physical concept of soil-water equilibria for nonionic organic compounds, Science, 206 (1979) 831. [2] MINGELGREN, U., GERSTL, Z., Réévaluation of partitioning as a mechanism of nonionic chemicals adsorptio soilsn i Environ. J , . Qual. (19831 2 1 ,. 1 ) [3] SENESI, N., TESTINI, C, Physico-chemical investigations of interaction mechanisms between s-triazine herbicide soid san l humic acids, Geoderma (19828 2 , ) 129. ] [4 GARBOW , GAEDER. . J , methode analysiw th ,ne r B.J. f A fo granulao s , r pesticide formulations, J. Agric. Food Chem., 38 (1990), 996. [5] VIMOND-LABOUDIGUE, A., PROST, R., Etude comparée des complexes hectorite- et vermiculite-decylammonium à l'aide des spectrométries Infrarouge et Raman, submitted to Clay Miner. (1993). ] [6 LEVIN, I.W., "Vibrational spectroscop f membranyo e assemblies", Advancen si Infrare Ramad an d n spectroscopy, (CLARK,1 vol1 . R.J.H., HESTER, R.E., Eds), WileyHeyden( 1984) 1-48.

Next page(s7 )6 left blank POLYETHYLENE (PE) FILM FORMULATION INCORPORATED WITH HERBICIDE METOLACHLOR FOR CONTROLLED RELEASE

Byung-Youl OH, Jin-Hwa KIM Pesticide Formulation Laboratory, Agricultural Chemicals Research Institute, Rural Development Administration, Suweon, Republic of Korea

Abstract

Modern agricultural practice r somfo es cashe th crop n i s Republic of Korea accompany with plastic sheet mulching to keep soil moisture, to elevate soil temperature and to increase crop productivity E P film formulatio. n incorporated with metolachlor was investigated as a controlled-release pesticide to eliminat operation a e pesticidr fo n e application uniforn .A m film was able to be formulated by thermal extrusion of low density polyethylene (LDPE ethylend LDPr )o an E e vinyl acetate e physica(EVA)Th .l propertie e filmt domestith me s f o sc industrial standards of PE, even with little lower values in tensile, elongation and tear strength than LDPE alone. Incorporated rat metolachlof eo r inte filth om amended witA hEV was to some extent higher than those with LDPE alone. On the other hand, release chemicae profilth f o le e frofilth m prepareintA EV o y wateb LDPd dan Er systems showe littla d e slower than that prepared by LDPE alone. Metolachlor residue in the top soil of red pepper field mulched the films was maintained around one ppm up to 90 days after application. Weed contro l y annua e filmke th effecn o s l f e o weedtth n i s s fielequivalenwa d o conventionat t l practices; spraying metolachlor EC followed by mulching PE film. No phytotoxicity s observewa d durin e wholth g e growth e cropperioth .f o d Retained amoun metolachlof o t e filmth n si r after crop harvest, 90 days after mulching, was 0.28ug to 0.3ug per sq. cm., which s equawa o 1.6t l originae %o 1.7th t f %o l amount.

1. INTRODUCTION Modern agricultural practices require high input of pesticides in crop environment to decrease the crop damage by harmful organism upheavo t e d croth san e p productivity, which leo concentratiot d sociaf no l theio concert e rdu nbiologica l activity. These adverse impact originatee ar s d froe facth mt that the major portion of pesticides applied enters to the environments without contributin e targeth to t gorganis m control [1]. Controlled-release formulation technology of pesticide mose th tproves e sha b tolerabl o nt e counter-measure o solvt e sideth e impact pesticidesf o s e technolog.Th y imparts various merit pesticido st e active ingredient(a.i.); reduction of acute toxicity and phytotoxicity, extension of biological activity, protection from environmental degradation, controf o l release pattern, masking of noxious material, and etc. [2].

69 Environmental advantages gained by using controlled-release products in agriculture will become of increasing significance in accordance with sustainable agriculture which involves less loads of agrochemicals. Various dispensers sucs synthetia h c polymers [3,4,5], natural polymers [6,7,8], or inorganics [9] have been utilized s controlled-releasa e devicesdispensere th f O . s thermoplastic polyme s use rwa o t dformulat mulchine th e g film incorporated with herbicides [1]. Upland cas Republie h th crop n i f o cs Korea require mulco st soie th hl witplastie th h c shee o keet p soil moisture, to elevate soil temperature and in consequent to increase crop harvests. Weed control practices for the crops e entirelar y depended upon spraying conventional formulation, ECs or WPs, of herbicides followed by mulching the polymer sheet. e authorTh s prepared polyethylene film formulations incorporated with metolachlo y thermab r l extrusion method. e behavioTh metolachlof ro e formulatioth n i r s studienwa n i d submerged water, by condensed water or by artificial rainfall. The stability of the chemical in the film or in the mulched s soiinvestigatewa l d under laborator r o fiely d condition. Weeding effect of the film was also evaluated in red pepper fielde reporW .t her resulte seriae eth th f lo s experiments.

2. MATERIAL METHODD SAN S 2.1. Chemical d instrumentan s s Metolachlor [2-chloro-6'-ethyl-N-(2-methoxy-l-methylethyl) aceto-0-toluidide] with 96.3% of purity was kindly provided by Ciba-Geigy(Basel, Switzerland) densitw Lo . y polyethylene(LDPE) with 0,921g/m densitf o l ethylend yan e vinyl acetate(EVA) with 0.943g/ml of density were used as dispenser. All the solvents used for metolachlor behavior study in the film and soil were reagent grade or HPLC grade. Equipments involved to formulate the plastic film were Banbury kneader(Moriyama MFC Co., Japan), Cutting mill(Alpine Co.,Germany), Film extruder(Yoojin ENG Co., Korea), Micromill(Thomes Co.,U.S.A.), Pelleter(Brabender Co., Germany) d Rol,an l mill(Bongsi Co.,Korea)C nMF . Metolachlon i r e dissolve th mulchee th n i d e filwaterr s soio th wa mln i , analysed by GLC(Varian Vista 6000), HPLC(SP 8100) or GC/MS (Finnigan MAT 4510B).

2.2. Productio plastif no c film formulation An intermediate with high content of metolachlor (3%, w/w) in LDPE or LDPE and EVA was prepared to formulate the film uniformly o typeTw polymef .so r dispenser were involved; LDPE alone and LDPE mixed with EVA (9+1, w/w). The mixture of metolachlor and polymers was thermally blended at 110°C in the Banbury kneader, extruded through Roll mill, pulverized by Cutting mill, and pelletized. Each intermediate was subject to ten fold dilution with LDPE, mixing and extrusion under 160°C o 180°t manufacturo t C e 0.3% metolachlor plastic film with 75cm of widt d 0.03man h thicknessf o m .

70 2.3. Physico-chemical characteristic e filth m f so Metolachlor distribution in and on the film was measured by washing it with distilled water and by extracting it with acetonitrile. Storage stabilit e fil s testea.if th o y mwa n i .d under different temperatures. Physical properties including tensile strength, elongation tead an r, strength were measured in accordance with the Korean Industrial Standards for polyethylene products. 2.4. Release experiments Metolachlor release study was done in the static submerged watecondensee t 10°Cth a r y b , d 22° r waterd o 50°Can C . All the experiments were executed under dark condition in order to avoid photodegradation effect. Metolachlor dissipatioy nb rain was tested by using specially designed artificial rainfall facility. Rainfall intensit s adjusteywa 60mo t dr 120m o mr pe m o tw houe raie on r perio ro e s kepth r n th wa f fo to dhoud an r hours. Evaporation s measureratewa metolachlo f sr o ai de th n i r under closed conditio y intb r trappinn ai o e wateth g r with constan velocityr tai . Release entrapper o d d metolachlon i r wate s extracterwa d with n-hexane twice, dehydrate hexane th d e laye y b anhydrour s sodium sulfate, concentrate adequato t d e volume HPLCr o d analyse ,an C . GL y b d 2.5. Field experiments Weed control effect was tested in red pepper field with sandy loa soif o m l texture, 1.6 organif %o c matteH 5.8p d .ran Plot size was 4 sq. meter with 0.4m in width and 10m in length of the ridge. Each plot was covered with the film and made holes with 30cm interval r transplantatiofo s d peppere f o nr seedlings. Referenced plos sprayetwa d with diluted solutiof no metolachlor 40EC at the dose of 800g a.i./ha and covered with polyethylene film. Weeding effect was investigated by collectin e weedth sl al geace fro th hm ploday0 4 ts after mulching and calculated efficacy from the fresh weight of the weeds. 2.6. Metolachlor residue Metolachlor residue in the mulched soil or in the film afte croe rth p harves s analysetwa y samplinsoip b d n to i l e th g sequence after mulching or collecting the film. The residue s extractewa y shakinb d air-driee th g d sample with acetonitrile for one hour. The extracted solution was partitioned with n-hexane, saturated NaCl solutio d distillean n d water. e hexanTh e layes subjecwa r o dehydrationt , concentration, clean-u y b Florisip l column chromatography d analysian , y sb GLC/ECD.

3. RESULT DISCUSSIOD SAN N 3.1. Formulation aspects PE film formulation incorporated with pesticide requiresa certain period of heat treatment to form the film; therefore

71 heat essentian stabilita a.ie th s i . f o ly requisite. Mass spectru f metolachloo m r incorporate e filth m showen i d n a d identical one with the pure chemical; molecular ion 283.8, major fragments 162.0 and 238.1, which means no degradation of a.i. during the process. Metolachlor distribution in/o plastie th n c film showed that e a.ith mos.f o (ovet r 87% s incorporate)wa e filth m n i d e retaine e surfac Th e filth (Tabl th n m. o f d o eI) e t i rat f o e formulated with mixed polymer(PFF-II) was a little bit lower than that with PE alone(PFF-I). This effect seems to be originated from the fact that EVA contains acetate group with a relatively high polarity, thus retains more chemicals.

Table I. Distribution and incorporation of metolachlor in/on the plastic films

Percent of a. i. Formulation ————————————————————— Percent of incorporation e filth m n i e filonth m

PFF - I 11.66 ± 0.34* 75.64 ± 0.09 87.30 ± 0.22

PFF - II 10.69 ± 0.05 78.61 ± 0.04 89.30 ± 0.05

• Standard error from the values of three replications

Metolachlor stability in / on the film showed a great difference to the storage temperatures; degradation rate of the chemical after 12 weeks storage at 17°C, 30°C and 50°C was 1.1%, 5.0% and 15.6%, respectively. These rates are relatively higher value than those in conventional pesticide formulation, whic y resultema h d e froexpandeth m d surfacee th are f o a formulation. Physical e strengt filmth s givewa sTablf n o i . hn II e Tensile or tear strength of the films was not greatly influenced as compared to those of PE alone. Meanwhile the elongation rate of the film was changed in both direction; increase n film-emittei d d directio decreased nan n verticai d l direction. Howeverphysicae valuee th th n l i s,al l parameters met the Korean Industrial Standards of PE film.

3.2. Release profiles Metolachlor release pattern from the film was investigated at different temperature of distilled water in static submerged condition. Conventional granular(GR) formulation of metolachlor was also included in the release profile study, which was formulated by coating method with sand as a carrier. Fig. 1 show percene sth f o cumulativt e metolachlor release froe th m film prepared with PE alone (PFF-I) in a close flask under dark condition. The fastest release rate was observed in the GR

72 Tabl . PhysicaII e l propertie f plastiso c films incorporated with metolachlor

Tensile strength Tear Strength Elongation (kg/en/) (kg/at) ( % ) Formulation VD* FED** VD FED VD FED PFI F- 189 200 95 98 558 355

1 PF1 F- 193 180 89 101 543 374

LDPE 210 200 97 99 400 500

EVA 300 200 105 110 400 550

LDPE + EVA*** 320 210 90 95 400 450

Korean Industrial > 160 > 55 > 200 Standard

• Vertical direction «* Rim - emitted direction

•«• The mixing ratio of LDPE and EVA was 90% and 10% (w/w)

formulation which recorded 95% of the metolachlor content during the first one day. e filTh m release e chemicath d t a l a n i C 0 5 t a % 54 22° t a d Can % muc42 h, C slo0 1 w t ratea % ;39 same day. Similarly, during the next three days an additional 4.5% metolachlor was released from GR and 8% to 12% was released froe filmth m . After this perio release th d e rate from e filth m slow downdifferencee Th . releas e th n i se rate from the fil t 10° a md 22° Can C were meager durin whole th ge period of study; less tha. 5% n s wa C However 0 5 e ratt th a ,e maintained higher value throughout the experiment. At the end 0 daye fil10 th osmf releasemetolachloe th f do % over95 r content. Fig. 2 shows the metolachlor release pattern from the film prepared with PE and EVA(PFF-II). The film released 33% to 46% e metolachloth of f wateo r y rda durine firse th on gt submersion additionae Th . l release rat metolachlof eo r during the next three days recorded 12% to 16%. The release profile from PFF-II durin whole th g e perio experimenf o d s slightlwa t y slower than that from PFF-I. The divergences in the metolachlor release rate from PFF-II submerge different a d t temperature revealed greater value compares a s f o dd witen he th PFF-I t A . the experiment after 100 days of release; however PFF-II retained a scanty amount of the metolachlor; 2% to 12% of the original content. In practic e herbicidth e beins i e g released froe filth m m condensee bth y d wate vaporizef ro d soil moisture. Stainless stee containet lva d distilled wate s coverewa r d wite filmhth s under greenhouse. Fig. 3 presents the released metolachlor by the condensed water during 50 days. The released rate of metolachlo % frowatee 22 th m s ry r b wa durins e firshr th 4 g2 t PFF-I and 18% from PFF-II. The rate became slow down there- after; the rate during the next four days was 24% from PFF-I

73 Y = 63. 4 e°-°058X Metolachlor G 0.8066*= r ) * Y = 86. 7 e°-0032X 54.Y= 0 0.8092*= r ( *

50.= Y 3 0.8071*= r ( ) *

A——A 10 °C ——• 22°C a—a 60°C

456 Elapsed time< Jdây )

Fig. I. Release profile of metolachlor from plastic film formulation (PFF- I ) in static submerged water at different temperatures.

100 Y = 67.2 e°-°°52X Metolachlor G r = 0.8440** ) 86.= Y 7 e°-°°32X = 57.Y 7 e°-0068X ( r = 0.8313** )

Y = 55.2 e°-0°72X ( r = 0.8274** )

A 10°C 22°C 050°C

456 Elapsed timeUday)

Fig. 2 Releas. e profil f metolachloo e r from plastic film formulation (PFF) I I - in static submerged water at different temperatures.

74 40

30

l •——• PFF-I (D o—o PFF-ÏÏ 20

10

0 l 10 20 30 40 50 Elapsed time (dayj Fig. 3. Release profile of metolachlor from the plastic films by condensed water under greenhouse.

% froan20 dm PFF-II. Total release dmetolachloe th rat f eo r from PFF-d PFF-I an d I89% an I % ,durin99 0 days 5 gwa s respectively. In field condition e fil,th m mulche e soi th s di alwayl s disclosed to the open air, therefore it is anticipated that rain gives metolachlo e th ris was o t ef hof r e filmfroth m. e releaseTh d amoun metolachlof o t r froe film s th measurem wa s d by using artificial rainfall device with different rain intensity. Tabl giveI resulte II ee metolachlo th sth f so r release rate from the films by rain. The release rate from PFF-I by 60mm per hour of rain intensity for one hour was 11.3% of the content, whereas that by same intensity for two hours was 16.6%. Meanwhile, the rate by 120mm per hour of rain intensity for one hour was 19.2%, while that by same intensity o fohour rtw s 20.1% swa . Even thoug totae th h l rainfall from 60mm per hour of the intensity for two hours was same as that fro mhoure e on releas120m r th hou,r pe fo mr e rate e froth m latte s higherwa r than that fro e formermanifestes i th m t I . d that the washing effect of the incorporated metolachlor by rain is more influenced by the intensity than the period. e releasTh e rate from PFF-I y b rais I evewa o ns littl r e affecte e intensit th periode y th b d r yo . In same manner e volatilizatioth , e incorporateth f o n d e majoth f ro e r on e frometolachlo ai fils th i me m th n i r dissipation way. Migrated amount of metolachlor from the films in the air was analysed under enforced air flow condition at

75 Table HI. Release of metolachlor from the plastic films by artificial rain with

different intensit periodd yan s

Percent releas f metolachloeo raiy b rn intensity Formulation 60 mm / hr 60 mm / hr 1 hr 2 hrs 1 hr 2hrs

PFF - 1 11. ± 0.63 * 4 16.0. 6± 19.2 ± 0.5 20.1 ± 0.7

PFF - II 3 0. 6 ± 10. 0. 1 79. ± 4 0. 11. ± 5 17.2 0. 9±

Standard error fro valuee mth f threo s e replications

177 Theoretical amount 1.0

o PFF- O)

. 50.4

;0.2

40 60 80 Temperature) C °

Fig. 4. Effet of temperature on migration of metolachlor from the plastic under ai e rfilm th enforce n i s flor dai w condition. different temperature. Fig.4 shows the migrated values for five days e amounTh . t from PFF-m c t 40°. a s I0.36usq Cwa r gpe of the film, while that from PFF-II was 0.25ug, which was equivalent to 2.06% and 1.40% of the metolachlor content. The increased temperature of air velocity accelerated the

76 migrated amount. Howeve e totath r l migrated amounf o t metolachlor in the air during the test period was negligible; contene lesth s f thao t % nregardles5 temperaturesf so .

3.3. Residue in soil and retained in the film Metolachlor soiresidue th l n i mulchee d wit e filhth r o m spraye dplanted witan pepped C hE re danalyses wa r comparo dt e e residue levelsurface Th th th n .i ee soil mulched wite hth films was kept in the range of 0.57ppm to l.lSppm with maximum level 8 days after mulching as illustrated in Fig. 5. The level in the soil mulched with PFF-I was all the time sustained higher than that with PFF-II. On the other hand, the residue in the soil sprayed wit C formulatioE h y aftes 4.83ppda wa n re mon application, thereafter the level was drastically diminished up to 30 days after application. From that day on the residue was maintained lower than that in the soil mulched with the films. The retained amount of metolachlor in the film was analysed day0 59 0 sdayd aftean s r mulchin croe th e soip th f glo field . Durin e firs th 0 daysg5 t , calculated release ratf o e metolachlor from the retained amount in the film was 85.5% from PFF- d 82.1an I % from PFF-II, which showed similae r th rate n i s lab experiment s presentea e retaines Th n Tabli . d IV e metolachlo e filth m n i after r crop harvest day0 9 ,s after mulching, was 0.28ug to 0.30ug per sq. cm, which was equivalent to 1.6 %o contente t 1.7th %f o .

5.0r

a——D Metolachlor EC A——t PFF- I PFF-E

0 1 8 10 15 20 30 40 Days after treatment

Fig . 5 .Persistenc f metolachloeo soin i r l mulched wit plastie hth c filmr o s applied metalachlor 40EC unde pepped re r r field.

77 3.4. Biological activity Weeding effect of the film was tested in red pepper field. Metolachlor s 40Ealswa C o involve referenca s a d e herbicide. e effecTh s calculatetwa d e fresfroth m h majoe weighth rf o t annual weeds 40 days after application. The film exhibited an excellent weed control effect irrespectiv f egraso broar so d leaf weed s showphytotoxicito sa N Tabln i n . eV s observeywa d during the whole growth period of the crop.

Tabl . RetaineIV e d amount f metolachloo s e plastith n i rc films under field condition

Retained amount Formulation Theoretical amount ( « /erf ) ( /«/erf ) 50 DAT* 90 DAT

PFI F- 17.5 2.5 ± 0.11*4 * 0.2 ± 0.08 3

PFF - II 17.9 3.20 ± 0.14 0.30 ± 0.04

* Days after treatment ** Standard error from the values of three replications

Table V. Weed control effect of the plastic films on annual weeds*

Weed control effect ( % ) Formulation Grass Broad leaf Digitaria Porttdaca Erigeron Qienopodium sangttinalis scopol olreracea L. canadenisis. L a&umL.

PFI F- 100 95 100 100

PFF - II 100 99 100 100

Metolachlor 40EC 100 93 100 75

Fresh wekjhtotae th f l o weedt s from untreated 869.3o/ns plowa t f

REFERENCES [1] GRAHAM-BRYCE, I.J., HARTLEY, G.S., "The scope for improving pesticidal efficiency through formulation", Advances in Pesticide Science (Th h Inte4t . Cong Pesticidf .o e Chem.) Zurich (1978) 781.

78 [2] KYDONIOUS, A.F., "Controlled release technology: Methods/ theory and application", Vol. l, CRC Press, (1980) 1 BAHADIR] [3 PFISTER, ,M. , G, LORENZ HERRMANN,R., ,R, , KORTE,F., Field trials with desmetryn containing cover and mulch sheet controo st l weed whitn i s e cabbage cultivationf o . ,J Plant Disease and Protection 94 (1984) 34 PFISTER] [4 BAHADIR,M., ,G. , LAY, J.P., Uptak f 14C-symetryneo e by Duck weed during release fro polymea m e r th matri d an x consequent herbicidal effects, J. of Controlled Release 7 (1988) 39 [5] STIKES, R.A., COPPEDGE, J.R., BULL, D.L., RIDGWAY, R.L., f o selecte e Us d plastic controllen i s d release granular formulations of aldicarb and dimethoate, J. Agric. Food Chem 1 (19732 . 3 10 ) [6] WHITE, M. D., SCHREIBER, M. M., Herbicidal activity of starch encapsulated trifluralin, Weed Sei 2 (1984.3 7 )38 HUSSAIN] B.Y., [7 OH , ,,M. Preparatio studd nan controlledf o y - release formulations of carbon-14 labelled butachlor, Toxicol. Environ. Chem 5 (19913 . 1 )10 [8] TRIMMELL, D., SHASHA, B. S., WING, R. E., OTHY, F. H., Pesticide encapsulation using a starch-borate complex as wall material, J. of Applied Polymer 27 (1982) 3919 CHOUDARY] w inter[9 Ne PRASAD , KANTAN, M. , P. - . L. ,. B , B . , M lamellar pesticide-metal-montmorillonite complex A nove: l technique for controlled release, J. Agric. Food Chem. 27 (1989) 1422

Next page(s) left blank 79 RADIOCHEMICAL PREPARATION OF POLYACRYLAMIDE HYDROGELD SAN AGRICULTURTHEIN I E RUS E

. YOUSEFZADEH*P , M.S. KHADJAVI**, M. SOHRABPOUR*

* Gamma Irradiation Centre, Atomic Energy Organizatio f Iranno , Tehran

* Chemistr* y Department, Science Faculty, Shahid Beheshti University, Tehran

Islamic Republic of Iran

Abstract

n thiI se resultworth , k f synthesio s f hydrogeio s s i s presented. These crossiinked polymers have interestinth e g property that they may release the retained additives to the surrounding medicontrollea n i a d manner. This property in turn may have application in agriculture (fertilizers, pesticides medicinn i r o ) e etc. In this work aqueous solutions of Polyacryiamide (PAAm) were irradiated to varying doses and concentrations of PAAm to find the optimum gelation dose . Specifically , the following objectives were studied in this work: - Radiation induced crosslinking of PAAm . e - concentratioEffecth f o t f additiveo n e degreth f n o o es crosslinking. - Measurment of the release rate of the additives (pesticides microelements) versus different soil type and the gel granule size. The results of this work show that the release rate of the additives is not a function of the soil type , but rather it does depend on the background moisture content, as well as the hydrogel particle size.

1. INTRODUCTION

Controlled release method f differeno s t type f chemicalo s s are postulate a mean f reducins o a sd addee th g d chemical e toxith burde f co ne materiaenvironmentth o t ln a n I . interesting techniqu e th toxicane e b f o choic ty ma e incroporated into a polymeric matrix which is then released into the environment which has a lower concentration for such compounds. According to the mechanism of the release of the active materials different types of controlied-reiease méthodes may be feasible. One which has recently received much attention , in various fields of application such as cancer therapy , plant growt pes, h t e hydrogecontroth s i , li technique.

81 Hydroge a wate s i i r insoluble homopolyme r o copolymer r which exhibit e abilitth s o o retait t yswel d lan ns i whe t i n mixed with water. The absorbed water is retained in the three dimensional matri f hydrogelo x s (1). Hydrogels or water - containing gels are polymeric materials that are characterized by their hydrophiiicity and their ability to reversibly sorb and desorb substantial amount f liquido s r o solutions s into their polymeric e networe th Th hydrophiiicit o . t k e du s i y presence of water solubilizing functional groups. The stability of shape in turn is due to the presence of the three dimensional network (2). The water retention value (WRV) of gel is the ratio e e retainee amounoriginath th oth f f o o t d l watege n i r gel (3). Synthetic hydrogels can be prepared by various methods including chemical and radiation - induced crossiinking . Some special polymers in aqueous solution readily undergo crossiinkin a radiatio y b g e producnth dos d s i alsan et o not contaminated with crossiinking agents . Therefore , the radiation technique seems to be superior to the chemical crossiinking method . e presenth In t e worirradiatioth k a solutio f o n f o n polyacrylamide (PAAm n watei )t a roor m temperaturd an e atmospheric pressure was carried out to yield the hydrogeis. Wate a r combinatio s evaporatio donwa y b el ge f o nf o n atmospheric or low pressure with elevated temperatures. r Ou studies focuse n o immobilizatiod f o typican l active agents such as Mg, Mo, B, Cu and some pesticides like borax and copper sulfate which , henceforth , shall be referre s activa o t ed agents.

2. EXPERIMENTAL WORK

Aqueous PAAm solutions were prepared with concentrations of 0.1 - 10 %(w/w) relative to water. The irradiation was performed in a research gamma irradiator GC-22 dosa t 3 kGy/hrea 0 9. rat d differenf o ean . t dose values. The resulting hydrogels were immersed in water r extractio- fo de e wate Th l e par so th r. t f r loadeo no l ge d watered gel were subdivided into small particle sizes and passed throug 5 microns4 d han sieve.0 8 f o s For preparation of this kind of hydrogel , two methods were used . The selected PAAm solution and the added active agents were irradiate e desireth d dan d hydroge s formewa l d (direct method ). In the indirect method the active agents were introduced into gel through a saturated solution. FTIR spectroscopy was used to determine the structure of e polymecrosslinketh d an r d PAAm. The release rate of the active agents from the gels was determined in the following way: e releasTh e rat n threi e e type f medio s a were studiede Th . media included distillated water, clay and sandy soils , with different moisture content . Concentration of released active agen s measurewa t y b atomid c absorption spectroscopy b r o y conventional titration methods. The gelation yield was determined gravimetricaliy from the e produceweigh th 6,7)( f l o t ge d.

82 3. RESULTS AND DISCUSSIONS 3.1. PAAm gels preparation

Gei point determination were done by the Charlesby- Pinner plot (o), which is shown in Fig. 1. Gelation dose for various concentration of PAAm solutions is shown in Fig. 2, e wateTh r retention valu f PAAo e m s geldeterminewa s d and the results of which is shown in Fig. 3.

8 1 6 1 4 1 2 1 0 1 8 6 4 2 0 l/D(KGy)

Fig. 1. Charlesby-pinner plot and gelation dose for PAAM aqueous solutio a dos t a 1 enkGy/h 9. rat f o e .

0.14

o Q

0.02 2345678 Concentration (w/w)

Fig . Dependenc2 . f gelatioo e n dose with PAAM concentration.

83 1460

8 14 20 Dose (kGy)

6%(w/w) 10%(w/w)

Fig. 3- Water retention value for different type of gel,

3.2. Active agent release characteristics

The hydrogels containing magnesium sulfate , ammonium molybdate, copper sulfate ,borax and ferric sulfate were immerse n i wate dr o mixer d wito typetw hf soilso s e Th . polymeric matrix upon absorptio f wateo n r swelle d latean d r release e activth d e agents e solutiointth o r soio n l matrix. s observeIwa t d tha direcy b t t preparatiot no n s methowa t i d possible to prepare gels which contain some special transition metal ions. This is because ions which contain empty d-orbitais capture the radicals which are essential e onsefoth rf crosslinkin o t g reactio r thiFo s . nreaso e th n indirect metho s r preparatiowa duse fo d e hydrogeith f o n s which contain these transition metal ions. The FTIR spectra of PAAm and cross!inked PAAm is shown in figures 4 and 5 respectively. During irradiation of PAAm the CO-NH-CO bond is forme . Therd thus i e a ssingle t pea t 344a k 7 (cm-1H N r )fo streching in crosslinked PAAm, whereas for uncrossiinked PAAm there is a doublet at this wave number . Carbonyl in PAAm n a absorptio s ha n bant a 1664{cm-1d r crosslinkefo t bu ) d PAAm it is shifted to 1674 (cm-1). e e effecsoiTh th lf o t typ s e studiewa eresultth d an ds of which is shown in Fig. 6 . The release of the active e diffusioth o t e ndu agent procese t seemb s i i o sd t s an s independent of the soil type. e relationshiTh p betwee e releasth n e rate e witth h granule e hydrogelsiz Th f gel o e. s show 8 i s d n Figsi nan .7 5 wit4 micronh s seemo releast s e faster tha 5 micro8 n n granule sizes .

84 flFfl-PAM]2R 99,9997^

co ZT 75,3766

69, 3000 200g 15ÖT3f 1000 500 WflVENUMßfR CM-J

e FTITh Fig R. A spectr. f PAAmo a .

RFfl=BR20R 50.0337

39. 0526.. 0 2000 1500 1000 500 WflVENUMBEl - CM R

e FTITh RFig - 5 spectr. f crosslinkeo a d PAAm.

85 ^ 120-1 .2 .o 100-

r<ü PH un T3 80- (D

60-

o Ö 40 H .2

S l Sand(ß5%) Oay(65%) 17 27 37 47 Time (day)

Fig . Effec6 . f soio t l typ n releaso e e rat f activo e e agent,

10 20 30 40 50 60 Time (day)

80 micro Meter micr5 4 * o Meter

Fig. ?• Relationship between granule size of gels and releasing rate of active agent.

86 100 a •g 3

d

3 I 0 Time (day)

80 micro Meter micr5 4 * o Meter

Fig. 8. Relationship between granule size of gels and releasing rate of active agents.

Fig.9 e showreleasth s e ratf o active e agent from hydrogeis versu e moisturth s e contene soilth f so t (65,80%). From this figur t i eappear s that higher moisture e th soiofl result n i highes r swellin n i d gelgan s consequently higher release rate of the active agent. The effect of the crosslinking dose on the release rate e activth of e agens alswa to studied s founwa t dI . thaa t y gavo 2kG t 5 e5 1 dosnearl f o e optimu th y m release ratr fo e the active agent. The release rate of gels containing element of Mg had a faster release rate whee polymeth n s crosslinkewa r d wita h dose of 15 kGy compared with that prepared at 20 kGy . This is because crosslinking is increased with higher dose values d swellinan g value lowered which consequently lowere th s release rate . The results of this phenamenon is shown in Fig. 10. The release rate of various active agents is compared in Fig.11. The solubility of these agents in water is usually a controlling e releasth facto r s showi fo e e rt rateth i n s A . ferric ions have higheth e r release ratd bora an e loweth x r release rate. To summerize , it can be said that 4 factors seem to be effectiv n diffusioi e n rat f theso e e active agent: s

e granul1Th - e hydrogeis th siz f o e . 2- The moisture of the background soil. e 3degreTh - f o crosslinkine n i PAAg m whic n turi s hi n controlled by the absorbed dose . e solubilit4-Th active th f eo y agent waten i s . r

87 Optimization of the release rate of the active agents under various conditions may be accomplished by modification of one or more of the above factors.

100

a .2 tS

45 50 Time (day)

Moistur% 65 e 80% Moisture

Fig. 9- Correlation between moisture content and release rate of active agent.

«f-oi o o

a I 6

90 100 Time (day)

(kGy25 01 (kGy" ) ""•"—• ) ' 10(kGy)

Fig. 10. Effect of gamma doses on the releasing rate of active agent.

88 Time (day)

-*- Borax% — *"' Cu% " Mo% -Q"" Mg % ">*" Fe %

Fig. 11. Effect of active agent type on release rate, (in soil with 65% moisture).

REFERENCES

[1] THOMAS.W.M."Acryiamide polymers".Encyclopedia of polymer scienc d Technologan e . y Interscience publishers Vol. 7(1964). [2] HUFFMAN.A.S . and RATNER.B.D, "Hydrogels for Medical and Related Application", Ed. by.ANDRADE,ACS, Washington(1976}. ] ZHAN[3 G ZICHENG, et.ai. "Preparatio f PAAo nm Hydrogeiy b s Radiation Technique".Radiât.Phys.Chem 0 (19873 . ) 307-308. [4] U.S. Patent, No. 3,336,129. "Plant Growth Medium" (1967). 15] YOSHITO IKAD d TOMOan A E MIT "Preparatio, A f o nHydrogei s by Radiation Technique". Radiât. Phys. Chem. 9(1977) 633-645. [6] CHARIESBY. A. Atomic Radiation and Polymers, Pergamon Press(I960 ) . [7] ROSIAK.J , BURCZAK.K , CZOLCZYNSKA , and PEKAIA.W. "Radiation Crossiinked Hydrogeis from AAm water solutions". Radiât.Phys.Chem. 22(1983) 917-928.

Next page(s) left blank 89 NOVEL CONTROLLED RELEASE (CR) AGROCHEMICAL FORMULATIONS: DEVELOPMENT AND EVALUATIONS

. RAJAGOPALANN . BHASKARC , , P.O. SHUKLA . AMARNATN , H Polymer Chemistry Division, National Chemical Laboratory, Pune, India

Abstract comprehensivA e programm e developmenth n o e f controlleo t d release (CR) agrochemical formulations is in progress at the National Chemical Laboratory (NCL). The significant accomplishments of this programme are as follows : (1) A crosslinked starch matrix has been develope applicatior dfo monolithis na c granule soile factore th Th .n si s governin kinetice gth f so agent release from this matrix have been investigated and the study of the structure of the crosslinked matri bees xha n undertake 13y nCCP/MAb spectroscopyR SNM . Procedur nover efo l single particle release measurement has been developed for understanding the mechanism of release. A granular soil-broadcast formulation of carbofuran has been developed and extensively evaluated in the cultivation of cotton, sorghum, safflower and mustard, giving protection against sucking foliar pests for over two months. (2) Proprietory processes based on the less expensive amino resins, specific to the individual agents and relase requirements, have been developed by microencapsulating a wide range of agrochemicals. Factors governing the release from these microcapsules have been investigated. ThumicrocapsulaA ) s(i r seed-coat formulatio f carbofurano bees ha nn developed and tested for the above crops with much reduced oral toxicity indicating it is over four times safer. Granulation of the above capsules in St-UF matrix has enabled its use in soil broadcast, especially under flooded conditions of paddy cultivation, (ii) An aqueous microcapsular dispersion of quinalpho bees sha n developed, whic uses hi s foliada r spray, doublin perioe gth f protectiodo n against aphid okrn si a cultivation, (iii) Microcapsule chlorpyrifof so s have been developed, applied both as seed-coat in the powder form and as a soil-broadcast after its granulation, being much superio othel al o rrt treatments including phorat controe th n e i whitf o lpolymeri R C e grubA ) (3 .c coating on commercial pesticide granules, applied as solvent-free self-curing spray in a coating pan has been developed which when applied to Furadan 3G is found to be as effective as St-UF-carbofuran granule fielde th n . so

1. INTRODUCTION Pesticide consumptio Indin ni a stand r hectarpe s compares arounea gm Ü d47 o dt about 17000 gms in Taiwan, 11800 gms in Japan and 3000 gms in USA. Despite such low average consumption, nearly 60% of it is consumed only in 40 districts out of a total of 467 districts, that too mainly in the cultivation of cotton and rice. The intrinsic toxicity of chemical pesticides, excessive application dosages and poor practices in handling and applications have led to increased concerns of both human healt environmentad han l hazards e researcTh . h programm TechnologR C n eo f yo agrochemicals at NCL is mainly focussed towards solving the twin problems besides improving the photo, thermal and soil stability of labile agents, thus improving their persistenc bioefficacyd ean .

2. CROSSLINKS») STARCH-UREA FORMALDEHYDE (St-UF) AS A MATRIX FOR ENCAPSULATION Starch, occuring abundentl naturen yi s beeha , n widely encapsulatin n usea n da g matrix for agrochemicals after derivatization and crosslinking[l-5]. The St-Up system developed by us has the unique advantage of crosslink!ng under mildly acidic conditions under which the pesticides are not degraded[6].

91 2.1 encapsulatioe Th n process Briefly, the encapsulation process consists ot'(i) gelatinising starch in water at >85", (ii) dispersing the agent in the starch paste under ambient conditions in a planetary mixer, (iii) reacting with UF methylals at acidic pH in the mixer, and (iv) finally t sievinwe rubbere gth y mas curind air-draughn san a n gi hours4 r fo t .° over50 t na The process as described incorporating the broad spectrum systemic insecticide, carbofuran studies wa , t differenda t urea/starch (U/ST) ratios whicmeasura s hwa e of the degree of crosslinking and at different loading levels[7].

2.2 Characterization 2.2.1 Matrix structure

e structurTh dynamicd an ee St-U th f o Fs matrices have been elucidateC 13 y db CP/MA studies[8,9]R SNM analyziny B . g 13chemicaC l shift f St-Uo s F samples having different degree f crosslinkinso g (Fig.l) establishes wa t i , d thaprimare th t y hydroxy group on the C carbon atom of the anhydroglucose units of the starch

molecul involves ei crosslinkine th n i d g reaction6 .

(a) Pure Starch (b) Gelatinised Starch St-UF samples with U/S-(c (d)0.2 0. ) 4 (e) 0.6 (f) 0.8, (x) Extracted UF resin (y) Extracted St-UF (U/S - 0.8) and (z) Extracted St-UF (u/ a- 0.2 )

13 Fig.l. C CP/MAS NMR SPECTRA

92 spin-spie Th n relaxation time (Tj) measurement St-Uf so F samples, which determine the long range motions of polymer chains, reveal that maximum degree of crosslinking takes place at U/St value of 0.6. Figure 2 shows that line width and T2 values, are well correlated with release and swelling rates.

0-6 0-6 s RATIu/ O—

700

0.2 0.4 0.6

.U/S RATIO'

Fig .2 Correlation between lin T emeasuremen widtd an - h C f o t

6 of starch, releas2 e rate constant K. and swelling rate constant k s

93 2.2.2 Release profiles

2.2.2.1 Releas waien ei r Using calrbofura e agentth e in-vitrs th a ,n o release profile e encapsulateth f o s d granules were studied under perfect sink conditions in excess water. The fractional release of carbofuran from the matrix at different U/St ratios confirms the earlier finding that ther decreass ei releasn ei e profilincreaso t e edu extenn ei crosslinkinf to g upto a U/St value of 0.6, beyond which there is obviously no further crosslinking, but only increased self-condensation of the methylals (Fig.3)[7]. 2.2.2.2 Releas aqueoun ei s methanol The release profiles of the St-UF carbofuran system were investigated in aqueous methanol of varying methanol content such as 100%, 70%, 50% and 0%, thereby varying the agent solubility to 100000 ppm, 27000 ppm, 10000 ppm and 1200 ppm respectively. The release profiles (Fig.4) show marked increase in release rate with increas agenn ei t solubility thereby indicating porous structur matrix[10]e th f eo . 2.2.2.3 Release mechanism watee Th r uptak agend ean t (carbofuran) releas St-Ue th f eFo system fit witn i se hth generalised equation

Mt/Moo = kt" ...... (1)

AI -Cu 3 • Q •/ K/U =1-5

o-o »000 2000 3000 4000 5000 Time (mins)

Fig 3 . Release profile f St-Uf-carbofurao s n granule t a differens t U/St ratios.

94 100

01 MeOH IN

MeOH- H20 MIXTURE

200 6OO 1000 1400 1800 2200

TIME (rnins)

Fig. 4 Release profiles of St-Uf-carbofuran granules in aqueous methanol of different methanol contents.

applicabl swellino et g controlled systems, where Mt/M< fractionae th s »i l releasf eo the agent at time t, k is a constant characteristic of the solute-polymer system and n is the diffusional characteristic of the release mechanism. A value of n equal to 0.5 indicates Fickian, diffusion controlled release mechanis nequad mindicate0 an 1. o lt s Case II or zero order transport for a device of slab geometry, and values in between are treate s non-Fickiada n (anomalous) transport. These limitinge valuear n f so somewhat lower for devices of different geometry like sphere or cylinder. In the case of a polydisperse system these values could be still lower[ll]. The St-UF system being polydisperse, made of irregularly shaped particles, n values computed' from the bulk release profiles do not indicate the exact release mechanism[l2-14|. Hence single particle release profiles were obtainey b d developin gnovea l procedur measuro et . eit Thu singlsa stainlesa en i particl t granule pu th s s f steeeo ewa l mesh boat which rested at the top of a 1 cm standard rectangular quartz cell provided with a magnetic needl bottoe th t ea m (Fig.5) magnetie th boa e d Th .tan c needle were outsid pate eth h of the optical beam. 3.5 ml of distilled water was put in each cell, which was then covered with a lid. Stirring and scanning were started simultaneously. Absorbance (Xman ru carbofuranr 8 xfo a 27 t obtaines wa ) regulaa s da rt specifi prina t ou t c time intervals using UV-V1S Spectrophotometer (Hitachi Model 220) provided with a cell programmer. Typical release profiles are given in Fig.6 and the n values computed

95 Lid

Stainless sfeel mesh boat

Gronul e

Water Optical cell

Magnetic neodle

Size of Optical cell'. !OmmxlOmmx45mm(hf.)

Fig.5 Single particle release set-up

38 fi o t : SAMPLE NO. Cl Q A ° V A O QU

* * * 4 ° n *

Q U Q ° | t *

QÙQ • ^ Q o * • SAMPLE NO. C2 LU Q »* LU n^ où LU K

0 100 200 "400400 500 TIME (min)

Fig.6 Single particle release profiles in triplicate of two particles - C d Can ,

96 from equatio thesr fo ) e nirregula(1 r particle sconfidenc % hav95 ea e level ranging from 0.46 to 0.74 with most of them averaging around 0.55. These values establish tharelease th t e mechanis mnon-Fickias i n (anomalous t Casno t e 11[10]bu ) . 2.2.2.4 Influenc solubilitf eo physicad yan l stat f encapsulaneo t gaio T n further insight int release oth e mechanism, selected model compounds were encapsulate e effecth f theid o t dan r physical stat solubilitd ean waten yi thein o r r release kinetics were studied[l5]. Thus carbofuran, acetanilide 4-nitrophenod an , i were used as solid encapsulants and 3-nitrotoluene, dimethyl phthalate and nitrobenzene as liquid encapsulants. Typical k and n values obtained are given in Table I. These studies show that solid encapsulants form a porous matrix having high release rates governed by Fickian or anomalous mechanism. Liquid encapsulants form a non-porous matrix (with limitation on extent of loading) governed by Case II or Super Case II transport. The release rates in the latter systems are much lower - see k values in Table 1 - by a factor ranging from 100 to 10000, and are not influenced by agent solubility. These systems could henc ideae e b increasinr fo l bioefficace gth persistencd yan f liquieo d agents for the control of soil pests. hydrophobicoif o e influenc e wels us a Th s la H p f eo l barrie aminn a r ro o resin barrier (double encapsulation) have also been investigated[16]. 3 2. Bioeflïcacy 2.3.1 Protection against foliar pest systemiy sb c action The St-UF carbofuran containing 3% of the active ingredient (a.i.), corresponding to the Furadan 3G granules marketed by Rallis India Limited, were evaluated extensivel e fielth dn yo bot experimentan hi llargn i plot d e san plo t trials (hn i , e cultivatio f sorghumno , cotton safflowe mustard Nimbkae an r th t da r Agricultural

Research Institute (N ARI), Phaltan. In these trials crops were planted in a randomised block design (plo 2replicatet) sizm 0 e3 timesd4 . Recommended agronomic practices were followe date th ad werdan e statistically analyzed. The bioefficacy was monitored as follows. Shootfly damage to the sorghum plant durin firs e day5 gth 4 t growtf so s measureh wa percentag e th y db e deade hearth n o t

tillersmaie th ncottonn n I plano . d infestatioe an t,th flyine th y ngb pests yphidd san jassids which attac foliai"e kth (lurin1 firse gth t 60-7 growts it 0 f co s monitorei s >h wa d

Table I k and n values of St-UF samples containing model compounds

Compound Physical Solubilitn yi K n state water(ppm) (min)'" Carbofuran Solid 1200 9.51xlO-2 0.31 Acetanilide Solid 5400 14.04x 10-2 0.29 4-nitrophenol Solid 16000 10.8IQ'x 6 2 0.40 3-.nitrotoluene Liquid 500 1.37 x ID'3 0.71 Nitrobenzene Liquid 1900 6.42 x IQ"3 0.54 Dimethyl phtha- Liquid 4300 I.17x 10'6 1.48 late

97 actuae byth intervalsy l insectda plancoune e 5 th th t f n a to t s o . Similarly coune th , t of safflower and mustard aphids on the respective crops was taken upto 70 days. A couple of illustrative examples are given below. Tabl give1 e1 percentage sth e dea de mai hearth nn o tplan tillerd an t f sorghuso m upt trial e day5 oth 4 t ssa conducte 1986-8n di NARy 7b t PhaltanIa date Th .a show improved control wit St-Ue hth F formulation ove longera r perio compares da o dt the commercial FURADAN 3G. There is significant difference especially in the 4Slh day tiller count. Table III gives the aphid count at 5 days intervals un cotton plant at the trials conducted by NARI in 1986. An improvement on aphid control as compared to Furadan 3G is discernible from the 30th day which become quite significan onwardsy da t5 4 fro e . mth Simila r data were obtaine cultivatioe th n di n of safflower and mustard as well, showing extended period of control as compared commerciae th o t l formulation l thesal n eI . case normae sth l protection periof do about 30 days obtained with Furadan 3G is doubled by the St-UF formulation.

TablI eI Bioefficacy of Sl-UF carbofnran formulation in sorghum cultivation conducted at NARI, Phaltan in 1986-87 (Values denote average dea shootflo t de hear . Soccatadu yA % t )

Treatment Dosage Main plant Tillers Main plant Tillers (21st day) (21st day) (30th day) (45th day) ST-UF-Carbofuran 15 kg/Ma 23.95b 16.22b 33.03b 20.18c FuradaG n3 15 kg/Ha 22.73b 15.84b 36.42b 27.64b Untreated control - 49.47a 27.82a 63.87a 34.80a

Same alphabet in a column denotes no significant difference.

Table III HioclTicac f Sl-UF-Carbolurayo cotton ni n cultivation conducte t NARIda , Phalta n 198ni 7 (The numbers denote aphid plant5 n so f cotto so nAphi- s gossypii) Days after ST-UF carhofuran Furadan 3G Untreated control treatment (15 kg/Ha) kg/Ma5 (1 ) 10 8b 8b 19a 15 8b 9b 24a 20 9b lOb 28a 25 lObc Itbc 3 la 30 I4c 19hc 42a 35 17c 21 he 47a 40 19c 24bc 52a 45 2 led 31h 64a 50 24cd 34b 68a 55 26cd 40b 72a 60 28ccl 44b 81a 65 3 Id 47b 9 la Same alphabet in a row denotes no significant difference

98 2.3.2 Protection against soil pests e whilTh e gru r roobo t grub larvae th , f 'chaffereo ' beetles consideres i , mose dth t serious soil pes groundnuf o t severad an t l other crop mann si y partl Indiaf sAl o n A . India Coordi nated Research Project (A1CRP) unde auspicee rth India e th f so n Council or' Agricultural Research has been functioning for developing effective means of controlling white grub with simultaneous evaluations at several centres spread out all over India. Their evaluation protocol includes monitoring percentage plant damage upt day0 ocro8 d san p yield differenr fo , t pesticide treatments, using Phorate granuleG standarde 10 th s St-Ue sa Th . F carboruran formulations were evaluated under this project for three consecutive years and their performance ratings compared with thos f FuradaPhorateo d an G ne3 10G. Tabl summarizeV eI findingse sth t I . may be noted that carbofuran is not generally very effective against white grub, but encapsulation o overals nit l performanc bees eha n considerably enhanced.

3. MICROENCAPSULATIO AMINN I O RESIN MATRICES The amino resin microcapsules have been widely reported in literature and an exhaustive survey hass been provided by Dietrich etal[18]. Amino resins used in microencapsulation are generally formed in situ in aqueous medium by the condensation of urea, melamine, guanidine and other amino compounds with formaldehyde. The patent literature[18] mainly reports on partially etherified products. 3.1 encapsulatioe Th n process

By suitably adjusting the reaction parameters stable non-leaky amino resin capsules have been prepared without partial etherification or other modifications, by a proprietory process (patent applie ddetaile A for). d investigatio agene th f nto matrix interactions has led to development of processes specific to the agents studied which include carboruran crystallina , e soli finn di e powder form, quinalphos hig,a h boiling liquid pesticide, and chlorpyrifos, a low melting solid. A judicious combination of amine th o compounds specifi agente eaco cth t f ho s use bees dha n arrive obtaio t t da n microcapsules with desired release properties and bioefficacy.

Table ÏV Performance Rating of St-UF carbofunm as compare Furadao dt againsG n3 t white grub under AICRP Centre Crop 1988-89 1988-90 1990-91

Durgapura Groundnut + + + + Kanpur Groundnut + + Kolhapur Groundnut N.S. Parbhani Sorghum + + Ranichauri Finger millet + + + + + + +

Rating = equivalen Furadao t G n3 + slightly better than FuradaG n3 ++ Superio Furadao t f secon- G n3 d onl Phorato yt G e10 + -f -f much superio Furadao rt equivalen- G n3 Phorato t t G e10 N.S, Data not significant

99 3.2 Carbofuran microcapsules

3.2.1 Release profiles Carbofura e a crystallinfor th f m o n i s i ne powder n microencapsulatinO . , git irregularly shaped capsules in the size range of 5-15 microns are obtained. The ami no resin shel permeabls i l wateo et wit d agenn ran ha t solubilit waten i m r pp f 120yo 0 capsulee at th 35°, s release measurable quantiries, lowesV t limiU y b t beinm pp g1 Spectrophotometry, in water. The release mechanism is apparently by permeation shele th f l o wal watey b l r followe dissolutioy db carbofuraf no elutios it s it d s nnan a saturated solution. The in-vitro release studies were conducted under perfect sink: conditions in excess wate interestinn A t 35°ra . g featur erelease observeth n i es profiledwa different sa t loadings variatioe Th . a.in n.i leve bees lha n studied from 66.3% o t 12.6% , keeping other matrix parameters constant. Figure 7 gives the plot of % release Vs time at different loadings. There is decrease in release rate with decrease in a.i. level which understandabls i e sinc shele eth l wall thickness increases correspondingly. However, at the lowest a.i. level of 12.6% there is a sudden jump in the release rate which could perhaps be explained in terms of the treatment of Thies etal[19]. With increased shell wall thickness, the core volume gets diminished and at very low a.i. levels, the amount of diffusing needed to cause B measurable change in a.i. level is also considerably reduced with the result that the overall fractional release rate is actually increased. 3.2.2 Bioefficacy 3.2.2.1 As seed coat Carbofuran microcapsules containing 50% of the a.i. (corresponding to the Furadan 50S f RalliPo s India Limited) were evaluate fiele seeth s da n do coat, once again cultivatioe inth sorghumf no , cotton, mustar safflowed dan t NARIa r , Phaltane Th . field data clearly establish that the period of protection is doubled in all these cases,

60 f

R e 1 e a s e

50 100 150 200 250 300 350 Time (Hours)

55.53% a.i. + 66.30% a.i. 33.04% a.i. c 12.60% a.i. n vitrI Fig.o 7 release profile f o carbofuras n microcapsulen i s water at different loadings.

100 otheo n r pesticide treatment being required durin firscropse e day0 gth th e 6 t f so Th . test protocol evaluatiod san n procedures were sam s describeea d earlie St-Ur fo r F carbofuran granules. Typical data obtained in two such trials are given below. Table V gives the bioefficacy data in the cultivation of safrlower in 1987. The aphid count data of the carbofuran microcapsules treatment clearly establishes its superiority over that of Furadan 50SP from the 35th day which becomes quite significant from 40th day onwards. Similarly, Table VI gives the bioefficacy data cultivatioe inth f Rapeseeno d mustar 1987n d i e aphiTh . d count data once again establish the significant superiority of carbofuran microcapsules over Furadan 50SP fro 35t e onwardsy mth hda . TableV Biocfficacy of carbofuran microcapsulcs as seed coa saillowen i t r cultivatio t NAKIna , Phalta 1986-8n ai 7 (The values denote average number of uphids, Dactynotus Carthami observed on five plants Days afier Curbot'uran microcapsules Furadan 50SP Untreated control treatment a.i % seedn .o (5 ) (5%a.i. on seed) 10 2.50 3.25 2.00 15. 4.00 3.00 2.75 20 7.75 8.00 10.25 25 7.00b 8.75b I5.25b 30 11.25b 10.25b 21.50a 35 I2.50b I6.25b 27.25a 40 12.()0c 22.50b 32.50a 45 13.25e 29.75b 41.25a 50 14.25e 40.25b 54.50a 55 16.50b 48.75a 60.75a 60 21.25e 60.50b 75.25a 65- 27.5üb 72.75a 86.50a 70 31.50b Sl.OOa 91.00a Same alphabet in a row indicates no significant difference

TablI eV liioelllcac carbofuraf yo n microcapsulcs sa seed-coat in mustard cultivation at NARI, Phaltan in 1986-87 (The values denote average numbe aphidsf o r , JL. erysimi observed on five plants) Days after treat- Carbofuran microcapsules Furadan 5ÛSP Untreated control ment (5% a.i. on seed) (5% a.i. on seed) 10 4.50 4.50 5.25 15 8.75 7.75 14.25 20 22.50b 28.25b 45.25a 25 26.75 b 32.25b 80.50a 30 31.00b 44.50b 108.25a 35- 40.25c 64.25b 127.50a 40 51.755 80.75 b 158.25a 45 I7.25c 36.25b 72.25 a 50 24.50e 60.50b 88.50a 55 34.00b 84.75a 105.25a 60 42.25b 117.50a 135.50a 65 60.50e 148.75b 207.75a 70 74.25e 220.25b 290.25a Same alphabet in a row indicates no significant difference

101 3.2.2.2 sois A l broadcast The carbofuran microcapsules were granulated in the St-UF martix to conform to 3% a.i. leve used sois an ld a l broadcas paddn ti y cultilvation. These 'double encapsulated' formulations were quite effective in the highly flooded paddy cultivation. The evaluations were carried out at the Tamil Nadu Agricultural University (TNAU), Coimbatore, in pots, introducing the actual insects enclosed in Mylar cages to evaluate the persistence (period of effective mortality to the insects) P and toxicity (extent of mortality) efficace Th evaluates . ,T ywa d against four insect species viz. brown plant hopper, white backed plant hopper, green leaf hoppe lead fran folder treatmente Th . s included double encapsulated microcapsules, St-UF carbofuran granule Furadad san n 3G. The PxT index was determined and the data analysed for statistical siguificance[20]. A typical evaluation against the leaf folder is given in Table VII. The granulated microcapsules are significantly superior, at higher dosage levels, to both St-UF and Furadan 3G. The St-UF formulation has much faster release under flooded conditions due to swelling and hence is only on par with Furadan 3G. At lower doses, the microcapsule maintainine sar levea f carbofuraw o l lo o watee gto th n ni r leadino gt lower efficacy. 3.2.3 Toxicity Furadan SUSP is not registered in India because of its very high toxicity. The acute oral toxicity of the microcapsules of carbofuran (containing 50% a.i.) were evaluated Indiay b n Institut Toxicologyf eo , Bomba Rallid yan s Agrochemical Research Station, Bangalore. These values given in Table VIII show a drastic reduction in toxicity microencapsulationn o . 3.3 Quinalphos microcapsules Quinalphos, a contact stomach insecticide and is a high boiling liquid containing ~70% of a.i., as commercially available. It is used as a foliar spray against the pest complex on rice, vegetables, groundnut, cotton, fruit trees etc., the emulsifiable concentrate (E.G.) containing aromatic diluents and emulsifiers being the conventional formulation. The microencapsulation of quinalphos gave spherical capsulels which were quite stable as dry powder 3.3.1 Release properties Quinalpho vera solubilit w s ylo s ha water(<25ppmn yi henced an ) e releasth , e rates were determined in 40% aqueous methanol in which it is soluble to the extent of 900ppm. The release profiles at 3 different loadings of 83,78 and 70.7/Msing the same

TablI eVI Biocfficnc granulatef yo d carbofuran microcapsules, St-UF carbofurad nan Furadan 3G against rice leaffolder, Cnaphalocrocis medinalis, in paddy cultivation at TNAU, Coimbatore in 1990 Dosage kg a.i. /Ha Formulation 1.0 0.75 0.5 0.25 FuradaG n3 1041c 72()b 687a 379a ST-UF carbofuran I382b W5a 483b 270a Granulated carbofu- I638a 960a 502b 311a ran microcapsules Values denot valueeT meanPx s3 calculatef so eacr dfo h replication separately. Same alphabe columa n i t n indicate significano sn t difference.

102 matrix composition is shown in Figure S.The release is quite fast at the highest loading, but drastically comes down when it is lowered just by 5%. This indicates agent matrix interaction leadin porougo t s shell walls beyon dthreshola d loading lowet .A r loadings the release profiles are quite linear after an initial burst effect. 3.3.2 Bioefficacy The bioefficacy of quinalphos microcapsular dispersion in water was evaluated as foliar spra Okrn yi a cultivatio NARIt na , Phaltan, using ECALUX E.Gn ,a . formulation of Sancloz India Li in i led, as standard. A typical evaluation of the aphid counts on S day intervals with pesticidal spray at intervals of 15 days is given in Table IX. These

Table VIII Acute oral toxicity ofcarbofurim microcapsules

Formulation Species LDSO Remarks Where determined (mgAg) Carbofuran micro- Rat 60.0 Using wates ra I.I. Toxicology, capsules vehicle Bombay (50% a.i.) -clo- Rat 170.0 m Usingu % g\ LI. Toxicology, acacia as Bombay vehicle -do- Rat 59.1 Rallis Agrochem. Res. Station -do- Mice 19.3 -do- Carbofuran 75DB Mice 4.7 -do- (75% a.i.) -do- Rat y.o -do-

120

E e 1 e a s e

0 0 15 50 0 10 200 250 Time (hours)

83.00% Loadin + 78.01g ^ Loading 70.69*; Loading n vitrI oFig releas.8 e profile f o quinalphos s microcapsulen i s 40% aqueous methano t differena l t loadings.

103 evaluations show that the E.G. is effective for 7-10 days whereas the microcapsules days0 2 lasr .fo t Apart fro increasee mth d protection period, this formulation being water based eliminates the dermal toxicity and fire hazards of the solvent based E.C,

Table IX Bioefficacy oi'Quinalphos microcnpsular dispersion on the aphid, Aphis gossypi okrn i i a cultivatio t NARIna , Phalta 198n i 7 (The values denote the average number of aphids on five random plants)

Days after M icroca pillar Quinalplios Ecalux Untreated treatment (250y a.i./Ma) (250ga.i./Ha) control P ret real in cut count -1 27.6 24.0 24.3 First spray 1 I23b 17.3b 28.3a 5 8.6b 8.3b 32.3a 1Ü 8.0b 12.6b 37.6a 15 8.0b 14.6b 40.3a Second spray 16 7. Ob 12.6b 44.3a 20 5.3c 14.3b 52.6a 25 O.Oc I7.3b 58.3a 30 O.Oc O. b19 58.0a Third spray 31 O.Oc 15.6b 60.6a 35 O.Oc 12.6b 64.6a 40 O.Oc 16.0b 66.3a 45 O.Oc 17.3b 67.6a Same alphabeindicatew ro a n significano i t n s t difference

3.4 Chlorpyrifos microcapsules Technical Chlorpyrifo meltinw lo a gs i ssoli d whic bees hha n microencapsulates da mels it aminn i t o resin matrix giving spherical capsules wela s i lt I .know n pesticide effective against subterranean pests like white gru termitesd ban .

3.4.1 Release profiles Chlorpyrifos has negligible solubility in water (~2ppm) and hence its in vitro release was studied in 60% aqueous methanol where its solubility is 860ppm. Its relelase profiles at the loadings of 90% and 82% (Fig.9) show that unlike quinalphos, stable capsule obtainee b n sca d eve higt na h loadings. 3.4.2 Bioefficacy Bioefficac microcapsulee th f yo s were evaluated against white grub bot sees ha d coat an sois da l broadcast (after granulatin capsulee gth St-Un si F matrix), under A1CRP during 1992-93 comparative .Th e dat evaluationo a obtainetw e th r dfo s (Table Nos. Durgapurt a ) I X d aan vis-a-viX s several other pesticidal treatments show spectacular performance by these formulations being even superior to standard Phorate 10G granules. Large scale trials are being conducted at present at Durgapura to confirm these results.

104 0 10 20 30 40 50 60 80 Time (hours)

• 90.0% Loading + 82.0% Loading

Fig. 9 In vitro release profiles of chlorpyrifos microcapsules aqueou% i60 n s methano o loadingstw t a l .

TableX liioefllcacy oftniaocapsular chlorpyrifo othed san r insccticidal seed treatments against while grub, Holotrichia consanguinea Blanch, in groundnut at Durgapura under AICRP in 1992-93 Sr.No. Treatments Average plant Average pod yield mortality (%) fi/Ha) 1. Chlorpyrifos EC (25 ml/kg 13.85 23.82a seed) (21.64)b 2. Chlorpyrifos microcapsulesI 4.17 23.92a (90% loading) 5 g ai/kg seed (11.63)a 3. Chlorpyrifos microcapsuleI sI 4.43 23.83a (82% loading ai/kg )5 g seed) (12.12)a 4. Quinalphos 25EC (25 ml/kg 15.63 23.75a seed) (23.09)bc 5. Lindane 20EC (25 ml/kg seed) 47.06 8.40cd (43.29)e 6. Diazinone 20EC (25 ml/kg 39.22 7.64cd seed) (38.67)e 7. Trizophos 4ÜE ml/k5 C(2 g 22.02 12.64bc seed) (27.91)bcd 8. Phorate IOG(25 kg/Ha pre- 21.26 14.72b sowing) (27.40)bcd 9. Imidacloprid 7UWS (5 g/kg 37.76 9.86bcd seed) (37.91 )e 10. Untreated connol 84.86 1.32e (165.00)f Figures in parentheses are angular transformed values Same alphabe columa n i t n indicate significano sn t difference

105 Table XI BioelTicacy u('granulated chlorpyrifos micro-capsules (MC) and other insecticidal presowiug soil treatments against white gru groundnun bi t Durgapura t a under AlCRl 1992-9n *i 3 Sr.No. Treatments Dosage Average plant Average pod kg/Ha morlatity(%) yield (q/Ha) l. PhoratG IO e 25 1.42 1 (27.b l) 5 IH.SOu 2. Granulated chlorpyrilbs micro- 20 14.55 (22.39)ab 18.62a capsules !OG 1* 3. -clo- 25 13.94(2l.«8)ab 19.00a 4. Granulaied chlorpyrilbs micro- 20 14.53 (22.3 l)ab 20.87a capsules IOG 11* 5. -do- 25 12.00 (19.92)a 21.75a 6. FuradaG n3 30 74.06 (59.47)e 4.37bc 7. -do- 35 59.83(5 1.05)d 5.83b 8. Iso[)rocarb IOG 30 39.13 (38.71)c 7.02b 9. Untreated control - 93.84 (77.30)f l.llc Treatmente Tablf Se o 3 ed Xfo an microcapsula e s2 th r r composition. II d an sI Figures in parentheses are angular transformed values Same alphabo columa n i t n indicate significano sn t difference

4. CR POLYMERIC COATIN PESTIDN GO EBg ANGLES Peslicide granules use soir dfo l broadcast application genrealle sar y mad f carriereo s suc clas ha san r yo d whic vere har y cheap. Replacing these wit polymeriha c matrix such as St-UF renders it comparatively expensive. Hence an approach was made to impart a polymeric coating on the commercial granules which will impart CR properties with only marginal add on cost. Thus a liquid spray coating formulation has been developed which has the special features of being solvent-free and self-curing at ambient conditions (patent applied for).

4.1 Release Profiles Release profilecoateR C f do s Furada G granule3 n n comparisoi s o St-Ut n F encapsulate uncoated dan d Furada granuleG n3 s were measured under perfect sink conditions in excess water (Fig.10). The coated granules appear to be even slower in release than the St-UF encapsulated.

4.2 Bioefficacy Bioefficacy of the coated granules studied in cotton cultivation at NARI, Phaltan, indicates that these are generally on par with the St-UF granules (Table XII). More trial e needesar confiro dt m this. Apart fro improvee mth d persistence, these coated granule alsy osma reduce dermal toxicity especially relevan highlo t y toxic pesticides like Phorate. Besides coatine ,th g techniqu alsn extendee eca ob incorporato dt e pesticide seedsn so impartinY B . ga stable, attrition resistant coating, the seed- dressing can be shifted to the factory premises fro normae mth sitn o le coatin fiele th thao ds n gto pesticide coated seeds can be directly marketed. Studies on these lines are in progress.

106 UJ

UJ

UJ K

40O 80O 1200 1600 Tim «min( )

R COATEC A D FURADAG 3 N B S1.UF. ENCAPSULATED CARBOFURAN C FURADAG 3 N

Fig. 10 Release profile of CR coated granules

TablI eXI Bioefficacy ot'CK coated Furadan 3G granules against the aphid, Aphis Cossypii cotton i . n cultivatio t NAKIna , Phalta 1992n ni (Values denote the average number of insects on five plants) Days after Furadan3G Furadan 3G St-UF carbofuran FuradaG n3 Untreate treatment wiih 2.5% with 5.0% kg/Ha5 (1 ) (15 kg/Ha) d control CR coaling coatinR C g 1 5 kg/I la 15 kg/Ha 15 I.Ob ().67b ().33b 2.67b 29.33a 20 ()l.67h 53.67b 1 14.67ab 176.67a 160-OOa 25 259.0ÜC 198.33c 423.675c 832.67a 741.33ab 30 205.3 3 b lU.33b 259.66ab 314.33a 399.0ua 35 176.33b 291.665 175.33b 199.00b 709.00a 40 117.33b 48.33b 25.33b 122.00b 376.66a 45 Ob 6.67b 0.67b 11.33b 34.67a 50 1 0.33 0 0 15.67 Same alphabet in a row indicates no significant difference

107 S. CONCLUSIONS In conclusion, a versatile St-UF matrix has been developed for soil application of pesticides which shows promise of maximum effectiveness with liquid agents. Proprietory ami no resin microencapsulating systems have been developed fur diverse applications such as foliar spray, seed coat, systemic action from soil and protection of plant roots from soil pests. A CR polymeric coating to commercial granules has been developed increasing persistence and with likelihood of redacting toxicity. The technique may be extrapolated to coating on seeds as well. In addition, other systems incorporatin n effectiva g e herbicid s r reduit fo eg cin phytotoxicity, an aquatic herbicide[21] and mosquito larvicide[22], a cellulose based matrix[23 a highl d an ]y toxic, reactive pesticide freely solubl waten i e r botfo r h reducing its toxicity and increasing its shelf life, have been investigated. The controlled release researc hpossesseL grouNC t pa a broas d skill-bas diversn i e e typef o s microencapsulation techniques such as aqueous, non-aqueous and inverse interfacial polymerizations, complex-coacervation and solvent evaporation. It is estimated that over two billion dollars worth food production, constituting nearly half of the total production in India is lost due to infestation by pests, plant pathogens, rodents, birds, weeds and nematodes. In such a scenario, the consumption of pesticides bouns i increaso dt e despite more stringent regulatiory measures being enforcey b d the Government on toxicily and pollution levels. Consequently Controlled Release Technology has a great role to play especially in reducing the toxicity of highly hazardous pesticides which are sprayed on the foliage and in improving the efficacy of newer agents applied on the soil (in place of the now' banned organochlorines) whic e degradear h d e rapidlsoith l y chemicalb y organismsd an s e alternateTh . , environment friendly, approaches using pheromoncs, pesticides of plant origin and biological control agents also- require CR delivery systems which can protect them from degradatio sunlighy nb head applicationn tan to , and/or from deactivation during storage.

REFERENCES

[1] SHASHA, B.S., DOANE, W.M. AND RUSSELL, C.R., Starch encapsulated pesticide slor sfo w release . PolymJ , . Sei., Polym., Lett., Ed.,14(1976) 417-420. ] [2 DOANE, W.M., SHASHA, d RUSSELLB.San . , C.R., Encapsulatio pesticidef no s withi starcna h matrix, Controlled Release Pestides SymS AC ,. Serie , (197753 s ) 74-83. ] [3 STOUT.E.L, SHASHA, d B.SDOANEan . , W.M., Pilot-plant proces starcr fo s h xanihide encapsulated pesticides . Appl.J . Polym. Sei (19794 .2 ) 153-159. [4] SHASHA, B.S., TRIMNELL, D. and OTEY, F.H., Encapsulation of pesticides in a starch-calcium adduct. J. Polym. Sci.,Polym. Chem. Ed.,19 (1981) 1891-1899. ] [5 SHASHA, B.S.,TRIMNELL,D OTEYd an . , F.H., Starch-borate complexe EPIr sfo C encapsulation.J.Appl.Polym.Sci.,29 (1984) 67-73.

] [6 BHASKAR , SHUKLA,C , P.O., RAJAGOPALAN MITRAd an . ,N , R.B. proces,A s e prepartaiofoth r f cotrolleo n d release agrochemical granules. Indian Paten. No t 167769 (1987).

[7] RAJAGOPALAN, N., SHUKLA, P.G., BHASKAR, C., BANKAR, V.S., DHARIA. J.R. and KHILAR,K-C., Starch urea formaldehyde matrix encapsulation of solid agrochemical Matricxsynthesi1 s characterizationd an s . J Appl, . Polym. Sei.5 4 , (1992)909-913.

108 [8] SHUKLA, P.G., SIVARAM, S. and MOHANTY, B., Structure of carbofuran in crosslinked starch NM C matri13 R Correlatioy : xb releasf no swellind ean g kinetics with dynamic behaviour of polymer chains. Polymer, 33(1992). 3611-3615.

[9] SHUKLA, P.C., SIVARAM, S., and MOHANTY, B., Structure and dynamics of starch crosslinked with urea formaldehyde polymer 13CP/MAC y R b s NM S spectroscopy, Macromolecules, 25(1992), 2746-2751. [10] SHUKLA, P.O., RAJAGOPALAN, N., BHASKAR d SIVARAMan , C. ,, S. , Crosslinked starch-urea formaldehyde (St-UF) as a hydrophilic matrix for encapsulation : Studies in swelling and release of carbofuran. J. Controlled Rel., 15 (1991), 153-166. [11] RITGER, P.L and PEPPAS,N.A., A simple equation for description of solute release FickiaI I anomaloud nan s release from swellable devices . Controlle.J d Rel., 5(1987), 37-42.

[12] GROSS, S.T., HUFFMAN, A., DONBROW, M. and BENITA, S., Fundamentals of rele-as'e mechanism interpretation in multiparticulate systems : the prediction of the commonly observed release equation from statistical population models for paniculate ensembles. Int . Pharm..J , 29(1986), 213-222. [13] HUFFMAN, A., DONBROW, M., GROSS, S.T., BENITA, S. and BAHAT, R.. Fundamentals of release mechanism interpretation in multiparticulate systems: determinatio f substratno e release from single microcapsule relatiod an s n between individual and ensemble release kinetics. Int. J. pharm., 29(1986), 195-211. [14] HUFFMAN, A, DONBROW, M., and BENITA, S., Direct measurements on individual milcrocapsuie dissolution as a tool for determination of release mechanism. J. Phar.,Pharmacol.,38(1986), 764-766. [15] SHUKLA, P.G., RAJAGOPALAN, N. and SIVARAM, S., Starch urea formaldehyde matrix encapsulation IV.Influence of solubility and physical state of encapsulant on rate and mechanism of release. J. Appl. Polym. Sei., 48(1993), 1209-1222. [16] RAJAGOPALAN, N., BHASKAR, C. BANKAR, V.S, SARAWADE, V.B, SHUKLA, P.OKHILARand . crosslinke,of K.C.Use , d starch-urea formaldehyde matrix for encapsulation of carbofuran -.Influence of pH and double encapsulation on release rate. Polymer Scienc eContemporar- y Themes" Sivaram. S . , volED . .2 Tata McGraw-Hill Delhw Ne ,i (1991) 1043-1047. [17] KULKARNI, N.V., RAJAGOPALAN , KALEN. , , R.PKHILARd an . , K.C.. Starch urea formaldehyde matrix encapsulation of solid agrochemicals: II Release mechanis releasd man e modelling . ApplJ . . Polym. Sei., 45(1992), 915-922. [18] DIETRICH , HERMAK. , , NASTICE.R.H. , , BONATZ TIEGEd an , Amin. E , W. , o resin microcapsules . Literatur1 . e surve patend yan t review. Acta Polymerica0 4 . (1989), 243-251. [19] THIES, C., ASIF, M, CHENG, P.S., DISTELRATH, D.L, SVOBODA, G.D., and ZHOU, J., An analysis of microcapsule mass transfer behaviour, Proc. Intern. Symp. Control. Rel. Bioact. Mater., 18(1991), 644-645. [20] MUTHULAKSHMI, M. Evaluation of encapsulated carbofuran formulations applied to rice with reference to bioefficacy and occupational exposure. M.Sc. (Agriculture) Thesis. Tamil Nadu Agricultural University, Coimbatore (1990). [21] GHATGE, N.DAMARNATHd an . , ChemicaN. , l contro watef o l r hyacintw ne a h: herbicide. Pesticides 19(6) (1985)73-75.

109 [22] MITRA, R.B., SHARMA, R.N., RAJAGOPALAN BHASKAR, ,N. , C., TUNGIKAR, V.u., RAG, J.V. and SUUKIA, P.C., Controlled release abate larvicidc and carbor'uran insecticide field studies in India. Proc.Intern. Symp. Control. Rel. Bioact. Mater. (19«7)4 1 , , 166-167.

[23] BOTE, A.N., NADKARNI, V.M. and RAJAGOPALAN, N., Cellusose xanthide as an encapsulating matrix: I comparison with starch xunthide on swelling and release properties. J.Controlled Release pressn I . .

110 PREPARATION OF CONTROLLED-RELEASE STARCH ENCAPSULATED PESTICIDES: ADVANTAGES AND OPPORTUNITIES OF EXTRUSION PROCESSING

M.E. CARR National Center for Agricultural Utilization Research, Agricultural Research Service, United States Department of Agriculture, Peoria, Illinois, United State f Americso a

Abstract

Recent advances have been mad technologn ei r encapsulatiofo y f no pesticides withi a starcn h matrix. This technology involvee sth entrapmen f pesticideo t e matrith n i xs without neer eithefo d r crosslinking chemicals, ancillary polymers modificatior ,o e th f no starch substrate e latesTh . t technology utilizes extrusion compounding technique achievo st entrapmente eth . Several typef so solid, crystalline, and liquid pesticides have been successfully entrappe unmodifien i d d cornstarch matrice y continuoussb , twin- screw extrusion processes develope thit a d s Center. This report reviews and summarizes various aspects of the encapsulation researc d developmenan h t carrie t ovee pasou d th r 3 yearst . Variables affecting the encapsulation efficiency and release rate propertie e productth f e discussedo sar s e preparationTh . , evaluation, and efficacy of starch encapsulated pesticides (herbicides) produce simulatea y b d d scaled-up extrusion process n anengineerina d g feasibility studr theifo y r large scale production will be reported by others.

1. INTRODUCTION Starch has been investigated considerably as an entrapment matrix for a variety of pesticides prior to starch encapsulation by extrusion [1-10]. Previous work has shown that starch controlled- release matrices can provide a number of benefits such as reduction in leaching, groundwater contamination, toxicity, odor, volatility, and decomposition problems e abundanTh . t availability costw lo , , and physical natur ordinarf eo y unmodified cornstarc Unitee th n hdi States is particularly attractive for such application. Other types of unmodified starches and flours are also suitable low-cost substrate r entrapmenfo s t applications. Various methods and procedures for incorporating pesticides into a starch matrix include the use of various types of natural and modified starches, crosslinking and/or coagulation chemicals, and combinations of starch with other polymers [1-4]. Methods that have involved, for example, ionic and covalent crosslinking of starch xanthate, coagulation of alkali-treated starches with calcium chlorid r borio e c acid d formatioan , f o ionin r o c covalently bonded interpolymers can be effectively used. However, the most simpl cost-effectivd an e e mean f entrapmene so th y b s i t natural process known as rétrogradation [11]. Starch rétrogradatio e phenomenoth s i n y whicb n h gelatinized starch

111 molecule n aqueoui s s media reassociat d reveran e o watert t - insoluble aggregates through hydrogen bonding. Remova watef o l r accelerate bondine sth g proces renderd san starce sth h essentially insoluble in water. The simplest procedure for incorporating a pesticide into a matrix of unmodified starch without other additives simply involves first aqueoue ,th s gelatinizatioe th f no starch at about 90-95°C; secondly, incorporation of the pesticide intexpandee oth d gelatinized starch matrix befor substantiay ean l exten f rétrogradatioo t n occurs; thirdly, remova f watero l d an ; finally, particulatin driee gth d matrix. Various mechanical means have been used to form the starch encapsulated pesticides with limited processing flexibility, control, and efficiency. However, the most versatile, simple, and cost-effective means is by continuous twin-screw processing [12- 17]. Twin-screw extruders are well recognized in the food, feed, plastics, and rubber industries as highly efficient mixer reactors, particularly for high-solids and difficult mixing requirements. We have found the corotating, fully-intermeshing, twin-screw extruder to be ideal for encapsulating bioactive agents in unmodified starch matrices. The purpose of the present report is to review and summarize starch extrusion encapsulation work carrie thit a st Centedou r over the pasyears4 t3- . Variables suc temperaturs ha e profile, screw design, screw speed, solids concentration, points of addition, additives, levels of addition, and other variables which can affect the matrix properties are discussed. The effect of these variables on encapsulation efficienc d rat an t whicya e h herbicidee ar s released from the matrix in aqueous media are presented.

2. EQUIPMENT The entrapmen f pesticideo t s withi a starcn h matris wa x carried out using a ZSK 30 twin-screw extruder manufactured by Werne Pfleiderer& r Corporation screwe Th .corotatinge sar , fully intermeshing, and composed of numerous slip-on elements of various designs for achieving desired feeding, conveying, and mixing capabilitie scred san w speed revolutions/mi0 50 s o frot 0 m5 n (rpm). The processing section include individua4 d1 l barrel sections (BS) having 7 zones and a die-head assembly (zone 8) to give a total L/D ratio of about 43:1 [12]. All barrel sections were heated by electric units and cooled by a closed-loop chiller as called for by preset conditions on a remote control panel. The die-head assembly was also equipped wit n electria h c heate witd an rh various dies. Work was also carried out by replacing the die-head assembly with a die-face cutter (pelletizer e )fitteb than dca t with various dies. Loss-in-weight feeders and high-pressure piston pumps were used for incorporating all materials into the extruder. A vacuum pump for removing water or volatiles has not been necessary in systems investigate dateo dt .

3. MATERIALS A variety of starches, flours, meals, pesticides, pigment carriers, and surfactants have been used to prepare the starch encapsulated pesticid twin-scre0 3 eK productZS e wth extrusioy sb n processes. Howeverf o e wor, s involvee th mosha k us f o te th d industrial-grade unmodified cornstarch because of its overall effectiveness, low cost, and ready availability. Som type5 1 e pesticidef so s have been starch encapsulatey db the ZSK 30 extrusion processes. However, most of the work has involved three widely-used herbicides. These include (1) granular

112 atrazine, 6-chloro-tf-ethyl-W- (methylethyl) -1,3, 5-triazine-- 4 2, diamine ) (2 liqui, d metolachlor, 2-chloro-N-(2-ethyl-6- methylphenyl)-N-(2-methoxy-l-methylethyl) acetanalide) (3 d an , crystalline alachlor, 2-chloro-W-(2,6-diethylphenyl)-N- (methoxymethyl) acetamide. Liquid "DUAL d EPTan "C (S-ethyl dipropylthiocarbamate) were also evaluated considerably. Dual is actuall metolachlo% y86 r active ingredient containin surfactantga . Alachlor was incorporated as a liquid melt at 50°C. Carriers that have been use o increast d e level f activo s e ingredient e finath s lf o (a.i.o abouproduc t % p 30 tu ) t weight include Kaolin clay, Fullers Earth, diatomaceous earth products and silicates e silicateTh . s were mosth et effectiv e mosth t t bu e expensive. Cost/performance aspects of the various carriers would require detailed consideratio studyd f o n an general n e I .us e ,th carriers (absorbents) or surfactants have not been required unless r moro abou e% 15 oily-typt e herbicid s incorporatei e d inte th o product A soli. d suc s atrazin a hwele b ln ca eretaine higt a d h level f incorporatioso n without additives.

4. PROCEDURES A variet f procedureo y s have been used o extensivto , o t e presen detain i t thin i l s report. However ,extrusioe th som f eo n procedure exemplifiee sar followss a d .

4.1. Exciinple 1 Starch (10% moisture) and atrazine were preblended and fed into barrel section numbe e (BS-1)on r d wate an ,s pumpe rwa d into BS-3 using an intense mixing screw and a high temperature gelatinization profile. Barrel temperatures (°C) were set at 70 in zone numbe (Z-l)e ron , 90(Z-2), 100(Z-3,4,5), 90(2-6,7)- (Z 0 9 d ,an 8 ). Feed rates wer eo giv t suc es a hvariou s total solids concentrations from 36-75 atrazind %an e incorporation level- 5 f so 20% (final product weight basis after drying to about 10% moisture). Total feed rates ranged from about 7-14 kg/h. The material was extruded through the die-head opening (about 320 mm2) using a screw speed of 400 rpm. A screw speed of 200 rpm was also used for the lower feed rates. In other work, the materials were extruded through dies with either four 2-mm-diameter o holetw r so 4-mm holes. Alternatively the die-head was replaced with a die- face cutter using dies with hole e Th diameter . mm 7 smals sa 0. s la materials were dried, milled (pin mill), and sieved (generally 1.4- 0.84 mm and 0.84-0.42 mm equal to 14-20 mesh and 20-40 mesh). 4.2 Example 2 Metolachlor or DUAL was pumped into BS-5 (starch at BS-1 and wate t BS-3a r ) using similar screw speeds, temperature profile, feed rates, and conditions as described above. In some instances, strands, at high solids concentrations, were chilled (BS-12,13,14) to reduce product expansio diee th preveno ;t t na frol toi m exuding around the die holes when high levels of herbicide were used (15- 25% active ingredient, final product basis); and to facilitate die- face cutting.

4.3 Example3 Various temperature profiles, screw configurations, screw speeds, and points of addition were used to feed either herbicide, absorbents, granular starch r o combination, f o materials s

113 downstream (generall y) followin BS-10 o 7t g gelatinizatioe th f no starch upstream. Additional information is found in the discussion sections. 4.4 Other procedures Procedures for determining encapsulation efficiency (EE) , swellability d releasan , e rate e mille(RRth f o d) products have been reported [12-17]. EE refers to the amount of herbicide retaine e granuleth y b d s after mild washing with chlorofors a m compared to the amount contained by the granules before washing (expresse %EE)s da . Unless specified amoune ,th t retained before washing compared to the amount introduced into the gelatinized starcextrudee th n hi essentialls rwa y quantitative extene Th . o tt which the granules swelled when soaked in water for 24 h at 25-27°C is referre swellabilitys a o dt ratee Th .whict s a herbicide e hth s were released e fromatriceth m n aqueoui s s medi t rooa a m temperatures while being swirled on an ORBIT SHAKER is referred to as release rate (RR). Product granules in early work were swirled in aqueous ethanol (90% water and 10% ethanol) to slightly facilitate the RR. Later in the procedure, only water was used.

5. RESULTS AND DISCUSSION Encapsulation efficiency (EE), swellability, release rate (RR) , and ultimately the efficacy of the starch encapsulated pesticides can be affected to various extents by the types of materials, extrusion conditions ex-sitd ,an u downstream processing conditions. These variables provide opportunit r preparinfo y g encapsulated products having a wide range release characteristics. For example, the use of either unmodified, acid-modified, high- amylose, waxy (amylopectin), derivatize other do r type starchef so s will each differently affect the physical properties of the matrix. Fillers, absorbents d surfactantan , s will further modife th y properties. Extrusion conditions alone can importantly affect release characteristic productse th f o f sexampler o e Fo .us e ,th a high screw spee d intensan d e mixing scre n resula ca w n i t relatively fast release product compare a lowe o t rd shear condition. Also, a partially gelatinized unmodified starch can effectively entrap level % resultinf herbicido s10 o a t n p i gu e relatively fast release product compared to fully gelatinized y alsstarch ma e affecte b oR R . y methodb d , rate d extenan , f o t drying the products. In general the RR will not be affected by dryin e starcth g h beyon de mos abouth t% moisturef 14 o t e On . importan tparticls i effect R R n e milleso th siz f deo products. Die-face pelletized particles woul e expecteb d o releast d e more slowly than milled product equaf so l surface area, althougt ye t hno studied. The preferred method of ex-situ processing is continuous die-fac) (1 e pelletizin desirea o gt d particle size suspende) (2 , d conveyin e pellet th n enclosea f o n gi s d strea ) f airo (3 m d ,an drying the pellets in a fluidized-bed dryer to about 12% starch moisture. Specific examples of variables that affect properties of the matrix are presented in the following sections. 5.1. Starch concentration Figure 1 shows that EE of the 2.03-0.84 mm (10-20 mesh) encapsulated atrazine productt appreciablno s wa s y affectey b d processin materiale gth extrudee th n si starct ra h concentrations of 35-65% (starch/starch -f water basis) and atrazine addition level f 5-20%o s . Percentag herbicidf eo e active agent refero t s

114 the amount contained by the product after drying to 10% moisture. Figure 2 shows that EE was very importantly affected at 65% starch concentratio smallee th r rnfo 0.84-0.42 (20-40 mesh) particles when atrazine addition level was increased to 20% (62% EE). TABLE I shows that increases in starch concentration (expresse % solidse extrude th s a % dvern 65 i )y ro t fro 5 3 m moderatel starce th yf ho reduceE encapsulateE e th d d metolachlor, DUAL, and alachlor products, except for alachlor at 65% starch concentration. Reductions in EE was very significant for the 20-40 mesh products at 65% starch concentration. DUAL, which contained a surfactant, underwent the least reduction in EE at 65% starch concentration. It should be emphasized that essentially all the herbicide that was incorporated into the products by the extrusion process, was retained, and that EE refers to the retention of herbicide after chloroform washing (Procedures 4.4).

1*100 n n ——— Os O s*' Hi 0 10 90 - A c 0 ———————————B —— O in o D

v> i iii 1 1 a. ca 20 35 45 55 65 70 u c LU Figur . 1 eEfficienc f encapsulatino y g atrazin n cornstarci e h matrice y extrusiob s n processing (2.03-0.8 m equam 4 l 10-20 mesh products). Atrazine additior fo % 20 n d levelan , s 10 wer , 5 e products witmoisture% 10 h .

atrazine% 110 . , 14-20m 2. 10% atrazine, 20-40 m . 3. 10% metolachlor, 14-20 m 4. 10% metolachlor, 20-40 m

Figure 2. Effect of clay on swelling of starch and starch/clay encapsulated metolachlo watern ri . Milled products were soaken di water 24 h at room temperatures.

115 TABLE 1

Effect of starch solids during extrusion upon the encapsulation of mctolachlor. Dual and alachlor and properties of the matrices*

Product Solids 10-20 Mesh 20-40 Mesh Swellability encapsulated Active agent (%) Active agent (%) in water11

recovered" encapsulated' recovered11 encapsulated1 35 97 92 98 89 150 Mctolachlor 50 96 90 96 82 170 65 96 86 95 69 ISO

35 98 92 98 90 160 Dual 50 97 90 96 88 170 65 96 86 95 83 190 30 98 90 97 85 200 Alachlor 50 97 87 97 85 180 65 97 72 96 68 360

•Starch feed WM constant and water rate was 32 to 132 ml/min. Total active agent recovered in dry product. 'Active agcnl remaining after extracting dry product.'Sample (0.2 g) in water (4ml).

In data not shown, the level at which liquid oily pesticides (or other oily materials such as corn oil, soybean oil, and mineral oil) coul e incorporateb d d intgelatinizee th o d starch substrate was importantly related to the starch concentration. Less oil coul e incorporateb d d retaine e e starcmatrian d th th s y ha b xd concentration in the extruder was increased. The extent to which encapsulated products (unwashed granules) swelled when soake waten i d r (Procedures 4.4 show)s i TABLn ni EI and Figur . 3 eDat a show that swelling increase s starca d h concentration in the extruder was increased. This is believed to be due ,o somt e extent o shear-initiatet , d breakdow f starcho n ,

350 65%

300

a? o>~ 250 Ï .£ 20° "05 W 150 "a o 100 a.

50

0 10 15 20 Figure 3. Effect of starch concentration (% solids) and atrazine addition level on swellability of starch-encapsulated products in water. Starch concentratio extrudee th n s i 35-65%n rwa . Milled products (0.84-0.4 m equa2m l 20-40 mesh) were soake waten i d t a r room temperatures.

116 particularl starc% 65 t hya concentration givea r nFo . l leveoi f lo addition, the breakdown of starch at high solids can be reduced or eliminate designy db , configuration e screw speed th ,an f o .d effece Th starcf to h concentratio starcf solids% o ( n R R hn )o encapsulated metolachlor, DUAL, and alachlor products (10% active ingredients, 10-20 mesh distillen )i d wate shows ri TABLn i n . EII e milledth f o , R unwasheR e Th d product s affectewa s d less than might be anticipated based on differences in their swellability. Releas nearls ewa metolachlor y fo complet h 1 2 DUALd n ei ran . DUAL releases wa d only slightly more quickly than metolachlor, whereas alachlo releases rwa d much more slowly. h) (41-49 1 2 n %i

TABLE 2

Effect of starch solids during extrusion upon rate of release of encapsulated metolachlor, Dual and alachlor into water1

Product Solids % Released (h) encapsulated (%) —_____——. 1 2 3 21 35 27 38 49 90 Metolachlor 50 27 39 51 94 65 26 43 53 100

35 27 40 52 95 Dual 50 35 46 57 95 65 39 54 65 97

35 11 20 23 49 Alachlor 50 9 16 21 41 65 il 18 23 46 •10-20 mesh unwashed samplewatel m 0 t r7 a n si (30) 0mg lOOrpm.

5.2. Herbicide addition level Figur showe1 s that atrazine addition smallevea d lha effect on EE of the chloroform-washed granules. Disregarding the small effec starcf to h concentratio thin ni s case average ,th valueE eE s % additio20 d an n , levels10 , 5 , r fo % 88 d an , 92 wer, 95 e respectively (2.03-0.84 mm equal 10-20 mesh products). In metolachloe contrastth f o E ,E r products decreased fro% (1067 m % addition) to 44% (20% addition). For products prepared at 65% starch concentration, swellability increased with increased level atrazinf so e (Figur) e3 but decreased for increased levels of metolachlor (TABLE III) . relativele partialle th b Thi o y t s ma e yydu lower shear stresn si the extruder that was encountered with use of metolachlor, resulting in minimal starch degradation. Addition leve atrazinf lo productn ei s prepare starc% 65 t hda concentration had little affect on RR of 0.84-0.42 mm (20-40 mesh) granules agitate aqueoun i d s ethanol (TABL . TABL) IV E V Eshow s that RR of the metolachlor products prepared at 65% starch concentration, was quite significantly greater for the 10% addition

117 TABLE 3

Effec metolachlof o t r concentration during extrusion upo propertiee nth matricee th f so s

Metolachlo) (% r 14-20 Mesh 20-30 Mesh Swellability

Metolachlo) (% r Metolachlo) (% r in water

recovered encapsulated recovered encapsulated 10 98 85 97 67 340 15 96 80 95 66 260 17.5 95 79 95 63 220 20 95 7! 96 44 220

level compared to the 15-20% levels. To some extent, this may be e reduceth duo t ed shear stress effec e highen starco tth t ra h additionlevell oi f s o shoult I . pointee db tesR R tha t te dou t th for the metolachlor products was carried out in water rather than aqueous ethano atrazinr lfo e product thao ss t direct comparisonf so products cannot be made. 5.3. Screw speed Screw speed may importantly affect the matrix properties depending upo extene nth sheaf to r stresstarce th n sho substrate. Variables such as starch concentration, type and addition level of herbicide, screw design/profil d othean e r factors plan a y interacting role exampln A . scref eo w speed matrie effecth n xto

TABU 4 Releas f Atrnineo e from Starcn-Encapsultfed Produc Aqueon ti m M edit' Atrazine Released Level of Addition Alrazine Encapsulated Percent (t/10 0productf jo ) (%) (t/100 {of product) <4hr) C4 hr) <72 hr) (4hr) (14 hr) f.2hr) 5 94 4.7 2! 29 32 1.0 1.4 1.5 10 87 8.7 20 30 41 1.7 1.6 3.6 20 61 12.2 18 29 35 2.2 3.5 4.3 'Chloroform-washed promue» (20-40 mesh) in »queouj 1095 ethanol. Extrusion procedure 1 (65% starch concentration).

TABLE5

Effec f melolachloo t r concentration during extrusion upon rate releasf th eo metolachlof eo r into water*

Metolachlor Percentage metolachlor release) h d(

21 10 36 46 55 87 15 18 24 29 60 17.5 15 22 27 56 20 23 32 38 61 •14-20 mesh unwashed samplewatel m 0 t ra 7 sn i (30 ) 0mg lOOrpm.

118 is show n Figuri n . 5 eStarch , clay (attapulgite) d atrazinan , e were processe e extrudeth n i d r usin n intensa g e mixing scred wan normal gelatinization temperatures (90-95°C) at 200 and 400 rpm. Starch/clay ratio was 4:1; total solids concentration was 65%; and atrazine addition level was 10%. RR of both 1.4-0.84 mm and 0.84- 0.42 mm (14-20 and 20-40) mesh products were considerably greater for a screw speed of 400 rpm than rates at 200 rpm. These differences were somewhat less pronounced as level of atrazine was depleted effechighee e Th th .f to r screw spee mucs dwa h lesr sfo metolachlogreates it o t r e lubricityrdu .

100 mesh 90 o c m £ so l c _g 75 70 D sW)- ë 60 LU . . 50 10 15 20

Figur . eEffec4 particlf to e sizatrazind ean e addition leven lo starch encapsulation efficiency.

100 80 60 40 30 . 20

14-20,400 10 i 8 20-40,200 6 10% atrazine 4 . 14-20,200 20% clay 4 14-20 mesh, 200 vs. 400 rpm 3 20-40 mesh, 200 vs. 400 rpm

l l l l l i i i i i i i I 0.51 2 5 10 30 50 70 909598

Figur . Effece5 clayf to , particle size extruded ,an r screw speed on release rate (RR) of atrazine from encapsulated starch and starch/clay products.

119 5.4. Tempera tore Processing temperatur affecn e ca matri e tth x propertiesn I . general, maximum level of loading can be achieved by fully gelatinizing the starch. Also completely gelatinized unmodified starch will underg a greateo r exten f hydrogeo t n bonding than partially gelatinized starch, resulting in a slower RR. Lower temperatures (e.g. 65-75°C usee b obtai o dt n )ca relativelna y fast RR. Various temperature profiles have been useful for special needs such as feeding materials downstream, cooling the material in orde decreaso rt e product expansio diee th pelletizinr ,o t na e gth material. 5.5. Feed rate Total feed rate can affect the properties of the matrix to various extents depending upon the net effect on the other variables such as mixing, shear, temperature, screw speed, and residence time. Thus, change feen si d rat requiry ema e adjustments in these and other variables such as screw design, die design, and L/o offset D t effec f feeo t d rate changes. Total feed rate (production rate) has been the least studied variable in using the ZSK 30 extruder. However, there has been indication that high level f oilo s y herbicide loadin mors i g e difficul o achievt t ea relatively high production rates.

5.6. Carrier additives The effect of various levels of clay (Fullers Earth) on EE of starch encapsulated atrazine and metolachlor products is shown in materiale FigurTh . e6 s were processed using normal gelatinization temperature profile and an intense mixing screw at 200 rpm. The materia s extrudewa l d throug a 2x4-mm-diameteh r diee Th . herbicides were quantitatively retaine extrudate th n di milled ean d after drying to 1.4-0.84 and 0.84-0.42 mm sizes (14-20 and 20-40 mesh) (10% active ingredients). The cloroform-washed atrazine

100

90

80 ^

4s- • 60 50 1. 10%alrazine, 14-20m 2. 10%alrazine,20-40m 4metolachlor0% 10 . 3 , 14-20m metolachlor% 410 . , 20-40m 30 0 20 40 60 80 100

Figure 6. Effect of clay and particle size on efficiency of encapsulating herbicides in starch and starch/clay matrices by extrusion processing. Cla Fullers i y s Eart 30/6f o h 0 mesh.

120 products retaine atrazine th f d o starch/clar ove% efo 90 r y blends containing up to 40% clay. EE then decreased significantly as clay content was increased. EE for the metolachlor products was much less unde same rth e conditions. Figure 7 shows a straight line relationship between swell of particles (0.84-0.4 percentage wateth n i d ) ran e 2 mm clath f e o n yi blends of up to 80% clay. As expected the granules with the most clay swelle e least th n ddatt I shownno . ae starch/clath , y granules with atrazine swelled slightly more than those with metolachlor. The effect of clay in the matrix containing metolachlor (10%) had a surprisingly small effect on RR as exemplified by 1.4-0.84 mm (14-20 mesh) granules in Figure 8. When clay without starch was processe extrudee th same n th di e t rconditiona starch/clae th s sa y blends, the release of metolachlor was complete in 3 h compared to blendr fo abouh s 8 witt4 h 20-50% clay. Clay only containin% 10 g metolachlor, but not processed in the extruder, released all of the herbicid aboun i e min5 1 t . In data not shown, RR of starch/clay extruded products containing 25% metolachlor was only slightly higher than those of the products containing 10% metolachlor. The RR was considerably greater for metolachlor than for atrazine at all levels of clay addition. Other type carrierf so s suc diatomaceous ha s earth, silicates, and ungelatinized granular starch have been used. Silicates were the most effective for obtaining high levels of oil loading (e.g., 20-25%). However cost/performanca n ,o e basi effectivenese sth f so these materials need considerably more evaluation. Granular starch has been incorporated downstream and found to cause a 2 to 3-fold increase in RR.

300

200

CD 5 C/3 T5 I 100 10%metoiachlor 20-40 mesh

20 40 60 80 100 Clay, %

Figur . e7 Effec claf to swellinn o y starchf go d starch/clay-an - encapsulated metolachlor product root sa soakemh 4 wate2 n di r rfo temperatures.

121 10%metolachlor 0-100% clay 14-20 mesh 200 rpm

0.2 0.5 1 2 5 10 20 30 40 50 60 70 80 Water Treatment Timeh ,

Figure 8. Release rates of metolachlor from starch and starch/clay encapsulated products. Particles were swirle. h 4 2 waten i dr rfo

100 Static 90 14-20 Mesh Air Dried „ 80 V) g 70 Œ 60

°r™ 50 uj 40 55 30 20 10 0 6 8 10 12 14 Time, Hrs

Figure 9. Effect of agitation on release rate of EPTC from starch encapsulated products.

5.7. Points or" addition Point additiof so whict na h starch, carriers, surfactantsd ,an herbicid incorporatee ear d intextrudee oth plan importanrn ca ya t role in both maximum level of herbicide and RR that is obtained. For example, addition of high levels of oil directly after water addition may inhibit gelatinization of the starch even using high temperatures (100°C) and intense mixing screws. Although addition of the oil further downstream may allow starch to be gelatinized, the oil may not be completely retained due to insufficient residence time, even with intense mixing problee Th . usuall n mca y

122 be resolved by using a proper balance of parameters. Normally, points of addition are much less important for oil addition level of up to about 10-12%. As another example, preblending of starch with a carrier vs. separate additions of the carrier at some point downstream can also affect loading and properties of the matrix. In general, the preferred procedures for optimization needs to be researched for each individual herbicide and extrusion system. 5.8. Other variables Other variables suc conditions ha dryinf so testind gan e gth products and of method of jet cooking compared to extrusion cooking of starc r encapsulatiofo h EPTf no C herbicide bees ,ha n reported previousl detain i y l [15]. Essentially, starch encapsulated EPTC products swelled slightly mor waten i e r after oven drying (50°C overnight released )an d more quickly than air-dried productsR R . was dramatically increase agitatiny b d g (swirling particlee )th s usin n orbitaa g l shaker (Procedures 4.4) compare o statit d c conditions in water [15]. Swirling at 300 gyrations/min released as much EPTC in 2 h as was released in 14 h at static conditions. The swirling method appeared to be a valid method for rapidly 5 2. o t e orde2 th f o rn o evaluatins wa R R e product. gth RR r sfo times greate r EPTfo rC encapsulate t cookinje y b dg thay b n extrusion. 5.9. Advantages The use of a corotating, fully intermeshing, twin-screw extruder for encapsulating pesticides in a starch matrix offers several advantages over previous procedures investigated. The extrusion process utilizes conventional equipment readily adaptable to continuous encapsulation and ex-situ downstream processing. Processing technique relativele sar y simple, versatile, efficient, and effectivet a 8 5 . K ZS Recen a d t an scale-u 7 5 K ZS p a wor n o k Werner & Pfleiderer has demonstrated that materials need not be preblended extrudee b n ;ca d through dies with hole 7 smals sa 0. s la mm; can be pelletized at the die face; continuously conveyed to a fluidize dryerd be d dried direc;r an fo d t packaging. e Costh f to downstream processing neglecting equipment cos bees tha n estimated to be less than one cent/lb of product. The use of unmodified starch (or flour), without need for chemicals, provides the most effective low-cost substrate available for the extrusion process r obtaininanfo d wida g e rang f releaso e e characteristicse Th . major disadvantag e initiath s i el cos f extrusioo t n equipment. However, high-production facilities should soon offset this initial investment in a successful operation.

REFERENCES [1] STOUT, E.I., et al.. Pilot plant process for starch xanthide encapsulated pesticides, J. Appl. Polym. Sei. 24 (1979) 153. [2] SHASHA, B.S., et al., Encapsulation of pesticides in starch- calcium adduct, J. Polym. Sei., Polym. Chem. Edition 19 (1981) 1891. [3] SHASHA, B.S., Starc d othean h r polyol s encapsulatina s g matrices for pesticides, Controlled Release Technologies: Methods, Theory, and Application (KYDONIEUS, A.F., Ed) CRC Press, Inc., Boca Raton, FL (1980) 207-224. ] TRIMNELL[4 al.t e , ,,D. Pesticide encapsulation usin starchga - borate complex as the material, J. Appl. Polym. Sei. 27 (1982) 3919.

123 [5] SCHREIBER, M.M., et al., Efficacy and rate of release of EPTC and butylate from starch encapsulated formulations under greenhouse conditions, Weed Sei. 26 (1978). [6] WHITE, M.D., SCHREIBER, M.M., Herbicidal activity of starch encapsulated trifluralin, Weed Sei2 (19843 . ) 387. [7] WING, R.E., et al., Effectiveness of jet-cooked pearl cornstarch as a controlled release matrix, Staren/Stärke 39 (1987) 422. [8] WING, R.E., et al., Amylose content of starch controls the releas bioactivf o e e agents . ControlleJ , d Releas 7 e(1988 ) 33. [9] SCHREIBER, M.M., et al., Bioactivity of controlled release formulation f starch-encapsulateo s d EPTC . J Controlle, d Release 7 (1988) 237. [10] SCHÖPFET, M.J., et al., Effect of starch encapsulation on atrazine mobility in small undisturbed soil columns, (Proc. Am. Soc. Agric. Engineering, New Orleans, LA, 1989), 1-13. [11] DOANE, W.M., et al., Encapsulation by entrapment within matrix of unmodified starch, U.S. Patent 4,911,952 (1990). [12] CARR, M.E., et al., Encapsulation of atrazine within a starch matri extrusioy xb n processing, Cereal Chem 8 (19916 . ) 262. [13] CARR, M.E., et al., Starch encapsulation of biologically active agent continuoua y sb s process, U.S. Patent 5,183,690 (1993) . [14] WING, R.E. t al.e , , Compariso f steao n m injection cooking versus twin-screw extrusio f o pearn l cornstarcr fo h encapsulatio chloroacetanilidf no e herbicides . Controlle,J d Releas 6 (19911 e ) 267. [15] TRIMNELL al.t e ,, . ,EncapsulatioD EPTf n o starc n Ci twiny hb - screw extrusion, Staren/Stärk 3 (1991e4 ) 146. [16] DOANE, W.M., Encapsulatio f pesticideo n n i starcs y b h extrusion, Industrial Crop Productd san 1 s(1993 . )83 [17] CARR, M.E., et al., Clay as a carrier in starch encapsulation herbicide prepare reactivy b d e extrusion, Starch/Stärken I , press.

124 PROCESS STUDIES, ENGINEERING FEASIBILITY AND COST ANALYSIS FOR STARCH ENCAPSULATIO HERBICIDEF NO E TH N SO CO-ROTATING TWIN SCREW EXTRUDER

. GRELLOJ , M.S. STARER Werner & Pfleiderer Corporation, Ramsey, New Jersey

M.E. CARR Agricultural Research Service, United States Departmen f Agricultureo t , Peoria, Illinois

. PAPAZOGLOE U FMC Corp., Princeton, New Jersey

United States of America

Abstract

The co-rotating, fully intermeshing twin screw extruder (ZSK) has been applied to the continuous compounding of starch encapsulated herbicide formulations. The ZSK is well recognized in the plastics, food and feed industries as a versatile mixer/reactor for the most difficult continuous processing requirements.Ove lase rth t five years continuoue th , s extrusion proces r compoundinfo s g bioactive agents into starch matrice s beeha sn developed. The continuous process of herbicide encapsulation includes the gelatinization of unmodified cornstarch in the extruder followed by the addition and incorporation of one or more active herbicides. The physical form of the herbicides ranges from oily liqui powdero dt y solid. This report will summarize the results of our process studies. Using our understanding varioue oth f s process interactions cosa , t analysis wil presentee b l scalea r p fo u dd commercial operation.

1. Introduction

Encapsulatio f herbicideno starcn i s preferentialls i h y accomplishe co-rotatine th n do g fully intermeshing twin screw extrude varieta r fo rf reasons yo . Foremos s thai t t continuous processing offers inherent economic advantage when compared to incumbent batch and fluidized bed processes. In addition, the twin screw continuous process offers distinct process advantages. These include versatilit n screi y d proceswan s design, efficienc n changi y w ene a ove o t r formulation, controlled mixing and product consistency. Conversio f cornstarcno incorporatiod an h f bioactivo n e ingredients characteristically include e operationth s f starco s h hydratio d gelatinizatioan n e additioth a f watevi o nn o t r cornstarch, incorporatio f additiveo n s inte starcth o h matrix, applicatio f mechanicao n d an l thermal energy, followed by heat transfer with the onset of starch retrodegration and finally die forming. Overall process enthalp typicalls i y y high dependin concentratioe th n go f wateno d ran presence th f otheeo r plasticizers. Quite conveniently twie th ,n scre s robustlwi y designer dfo the most difficult continuous processing requirements [1] and can provide sufficient power

125 without modificatio fundamentae th o nt l machine design therefors i t I . e well suite diversa r dfo e range of customized starch encapsulation products. Additionally twie th , n screw continuous proces s flexiblsi procesn ei s it s desigo t e ndu "building block" feature e procesTh . s sectio s composei n f segmenteo d d unitse whicb n ca h pieced together and tailored to suit a unique set of process and mixing requirements. The advantage e modulath f o s r design include high degre f freedoeo barren mi l arrangemend an t flexibilit alterinn yi sheare gth , mixin residencd gan e time distribution s therefori t I . e possible proceso t swida e arra f dissimilayo r formulation singla n o s e unit. Present work will first describ e experimentath e l studies. Next e resultth d , an s characterizations of the twin screw process will be presented. The various process boundary conditions which have been identified wil describede b l . Finally wile w , l presen modea t l based on the observed boundary conditions and apply a cost analysis of the twin screw process based on this model.

2. Theory

When encapsulating herbicide basie th , c functio extrudee th f accepo t no s i r t cornstarch, fully gelatiniz througt ei hcombinatioa f mechanicano thermad an l l energy, incorporate bioactive agent(s generatd an ) e sufficient pressur pas o et processee sth d matrix throug dieha . Figur1 e illustrates the twin screw process in terms of constituent operations. The initial task is to accept cornstarch inte feeoth d throat, simultaneously conve deaeratd yan bulw elo k density powder, and move it in a positive direction. Solid active ingredients (a.i.) may be introduced along with cornstarch in the feed barrel. In this region, the extruder barrels are cold and the screw is configured with conveying elements which progressively decrease in pitch. Water is introduced into the system using a pressurized injection apparatus. The two components are mixed using a combination of dispersive and distributive kneading devices. The objective is to promote association between wate starcd an r h while eliminatin starcy dr y h gan agglomeration s which could become impermeable. Liquid a.i. will usuall introducee yb d directly followin watee gth r onse(prioe th f o o tt r gelatinization) and be blended into the mixture using a distributive mechanism. In the case of chemically incompatible actives, incorporation is improved at high screw speed (high mixing intensity) while retention of starch structure and stable die forming are favored at low intensity conditions. In these cases, where opposing mixing criteria exist, optimums are sought and the application of experimental design is well suited. ingrediente Oncth l eal s have been introduced intsysteme oth reaco tast e e th s ,ki th t cornstarch with water to create a gel matrix structure which contains a dispersion of active agent. The physical nature of the dispersion, such as droplet size, influences functionality of the final product. The application of thermal and mechanical energy initiates the onset of structural chang starche th nature n ei Th . f theseo e change bees sha n subjec f numerouo t s s studieha d san been recorde literature th n di e [2,3,4]. Crystalline starch granule disruptee ar s givd y dan ewa tolargela y amorphous random polymeric melt watee Th . r plasticizea act s sa depreso t r e sth glass transition temperature. Therefore, the minimum water concentration must be such that the matrix will melt belo s decompositiowit n temperatur yield ean a finad l produc whicn i t e th h polysaccharide form continuoua s s polymeric phase. Starch transformation thin i s s regioe nar affected by extruder configuration, flow rate, temperatures, screw speed and moisture content of the starch. The matrix is then prepared for discharge and die forming. This will involve adjustment in melt viscosity via heat transfer. Simultaneously, the cooled matrix will be pressurized for die forming. Both affect e e soughsamar th s en i te extrudee constituenregioth th f d o n an r t operations are opposing. Therefore, optimization is required. The goal is to pressurize and discharge with minimal viscous dissipation such that the matrix is formable to the extent that hot

126 COPNSTARCH FEED

SOUD HERBICIDE FEED PUMP WATCH LIQUID HERBICIDE [dTiiluj]

NT/

SOLIDS_ LIQUID MIXING & GEIATION HEAT TRANSFER CONVEY INTRO. & DIE FORMING

MATERIAL TLDV

Figur - e1 Constituent Operation Herbicida f o s e Encapsulatio Starcn i n h Procese th n o s Twin Screw Extruder. s> face pelletizing at the die is possible via a high speed concentric type pelletizing system. In addition, the pellets should lack tackiness to the extent that they do not adhere together when contacted immediately after pelletizing. Related variables include flow rate, screw intensity - which is a combination of screw speed and element configuration, plasticizer concentration and designe di e desigDi . n considerations include numbe f holegivea o r r fo sn area, hole pattern, overall flow balance, land length (L/d) and material of construction. Hole diameter is usually determine functionay db l considerationdesiga t no producd ns i en variable d e th an t n si . Overall, process optimization results when screw intensity is sufficient to create an amorphous melt and distribute active agent within the melt, but gentle enough to minimize heat generation which results from numerous source f sheaso r stress. Shear stresses encounteren di the heat transfer, pressurization and die forming regions should be minimized in order to attain maximum flow rate additionn I . restrictioe di , pressurd nan e drop shoul e minimizedb a r fo d given numbe holee hold di f san eo r patter n- again minimizo t , e matrix heat generation.

3. Experimental

3.1 Process Configuration

Experiments were Werne a carrien o t Pfleidererd& ou r ZSK-57 twin screw compounder. This extrude intermediats ri usee b datbasis e dn a commerciascal r th o a t ca siz n sefo d i p eu ean l sized equipment. The process configuration is shown in Figure 2. The processing section consisted of 12 barrel segments with an L/d of 36:1. The cornstarch was typically metered into the first barrel while wate injectes wa r d intsecone oth d barrel liquie Th . d a.i.'s were metered directly downstream of the water while the solid a.i.'s were metered into the feed barrel along wit cornstarche hth . Twin screw gravimetric devices were use meteo dt solie th r d components whil liquide eth s were metered using assemblies which consiste triplea f do x piston pump, mass flow mete pressurd an r e nozzles. Figur platee hole di eshow3 Th e.m m plaa s 7 n0. viea f wo pattern show typicas ni thosf o l e use illustrated dan attempr sou optimizo t e open cross sectional area. The use of dies with hole diameters of 0.7, 1.0 and 1.5 mm are reported. The hole patterns were designed to yield the greatest number of holes for the available cross sectional area for minimum viscous e othedissipatioth rn o han t o allot ye dn r sufficienwfo t mechanical integrit sustaio yt n operationa pressuree di l additionn I . holee th , s were space t sucdou h thae th t pellets woul t sticdno k together upon cutting.

Discussion3.2

The objective of all experiments was to identify the set of input process variables which allow for simultaneous maximization of certain responses. Input variables include screw design, screw velocity, barrel temperature set points, order of addition and die design. The primary response variables included, but were not limited to, upper flow rate limit, highest concentration of smallesd a.i.an , t pellet size. Attainmen f maximuo t m total flow ratconcentratiod ean f no a.i. were considered most important since pellet size will var r eacfo yf ho t producse d an t application requirements l threAl . e condition s- flo w rate, a.i. concentratio pelled nan t e sizar e- directly opposing, and therefore, optimizations are again sought. We've identified the influencing variable d interactionan s s which affect flow ratd concentratioan e d wilan nl discuss each. Althoug e attemptew h o treat d t different herbicides "generically n additivea s a "r initia ou , l experiments proved that suc generalizatioha t possibleno s ni . Eac uniqun how systes eit s mha process requirements. This uniqueness is not unexpected, since chemical structure, rheology and mechanism of action are different for each of the tested herbicides.

128 PtSTON PUMP WATER PISTON PUMP UQUID HERBICIDE

SCREW FEEDER CORNSTARCH

CONCEKTWC DC FACED P6LLETTZEF

ZSK-57

CONVEYING

Figur - e2 Flow Proces Diagram Pilom e th 7 s5 tr mfo Figur - e3 Plan ViePlate f 0.7Di w eo m 0m

3.3 Results

TABLES I and II list the herbicides that were processed. In all cases, these chemicals formulationd an s were commercially available shele th f compoundsof , . None were formulated specifically for starch encapsulation or twin screw compounding. In addition to single active component formulations, numerous bioactive blends were encapsulated. These are listed in TABL . ScanninEII g electron microscop s use evaluato ywa t d e l formatioextenge f o td an n dispersion of active within the starch matrix. TABLE III summarizes the data and results for the various formulations date Th a. reported were attaine t stablda e steady state conditions with acceptable gelatinizatio dispersiond nan . Efforts were concentrate DUAe th n dLo system because it was one of the more difficult to incorporate at high levels due to its physical nature. Orde f additio o a criticara.ie s th .f wa no l variable whe active th n e ingredienn a s i t incompatible or semi-compatible (hydrophobic) oil. Both DUAL and corn oil were selected for this study uses placeb.a Corhighls s wa di a l d noi o yan incompatible wit cornstarce hth h system.

130 Three distinctly different process limiting phenomena were observed. First, the oily a.i. tends to compete with water for association with starch and therefore inhibits the formation of a gel matrix. This becomes increasingly prominen t increasina t g outpu a.id an .t concentration thin I . s case, the a.i. concentration limit can be offset by increasing water concentration. Next, chemical incompatibility limits maximum concentration. In this case, maximum incorporation required intensive mixin t higga h screw speed, which imparte dmatrixe heath o t , thereby creatin gheaa t transfer limitatioformine di e th g t na operation . Finally procese th , limites swa Theologicay db l dissimilarities between matri d oilxan . This phenomeno s studiewa n introduciny b d g a.it .a varying locations alon extrudee gth r barre observind an ldischargee th retentioe n i gth l oi d f no matrix extens A . f gelatioo t n increases l introductionpriooi o t r , miscibilit f a.iyo . into starch was reduced. At extreme conditions, two distinct phases of matrix and a.i. were noted at the discharge n generalI . e optimuth , s achievemwa y a.ib d . additio ne onse th priof o o tt r gelatinization, typically directly following water addition. It was also found that the use of

TABLE I: BIOACTIVE INGREDIENTS

Common Trade Chemical Name Name Name

alachlor LASSO 2-chloro-N-C2,6-diethylphenyt)-N- (methoxymethyl)acetamide

atrazine many 6-chloro-N-elhyl-N'-(l-methylelhyl)-l,3,5-tnazme- 2,4-diamme

butylate N US TA S-ethyl bis(2-tnelh\'lpropvl)carbamoihioate dicamba BANVEL ,6-dic,liloro-2-methoxybentoic3 acid

EPTC EPTAM S-ethyl dipropvt carbamothioale

ERADICAINE EPTC dichlormid+ (2,2-dichloro-iV-N-di-2- propenylacetamide)

metolachlor DUAL 2-Moro-N-(2-ethvl-6-methvlphenvl)-N-(2-methoxy-l- inet/iylelhyi)acetamide

metnbuzin SENCOR 4-amino-6-(l,ldimeth\'lelhyl)-3-(methylthio)- LEXONE l,2,4trtaztne-5(4H)-one

TABLE II: BIOACTIVE PREMIXES

Trade Name Common Name Manufacturer

SUTAZINE atrazine + butylate ICI Amencas

none dicamba + dual -

none dicamba + alachlor -

none dicamb alachlo+ a atrazin+ r e -

none dicamba 4- DUAL+ atrazine - BICEP METOLACHLOR + atrazine CIBA-GEIGY

BULLET alachlo atrazin+ r e Monsanto

none alachlo + metnbuzir n -

131 TABLE IH: EXPERIMENTAL RESULTS WHEN USING 1.0 & 0.7 mm DIE BORES

Active Total Bioactive Water Total Rate Ingredient Concentration Concentration (kg hr') (%) (%)

Atrazine 20 20 100

Metolachlor 10-22 15-20 110

Eradicaine 20 15 106

Sutazine 24.5 15 111

Dicamb DUA+ a L 18.5 15 103

Alachlor + Dicamba 41 15 103

Atrazin Alachlo+ e r 21.8 15 108 + Dicamba

Atrazine + DUAL n 15 108 4- Dicamba

Atrazine + DUAL 14 16 98

Alachlor 9 17 92

Alachlor + Atrazine 13 16 97

Metribuzin 9 17 186

Com Oil 9 20 106 (placebo) compatibilizers suc s surfactanta h d emulsifieran s s allowed increas n boti e h d ratan e concentration when adding the oily a.i. upstream. This clearly suggests that each new system requires specific formulation desig r compatibilitnfo y with cornstarch. In term f processo s variables optimue th , m rang f screeo w speed betwees d an wa s 0 15 n 300 min"1. Two screw designs were ultimately chosen which balanced mixing intensity on one hand d impartean , d minimum viscous dissipation. Figur 4 eshow s these designs e firsTh .t utilized conventional kneading blocks in the areas of starch gel formation and a.i. incorporation whil e seconth e d utilize E mixinTM d g element e regionth n i ss correspondin o a.it g . incorporation. Several die plates were evaluated to determine the size of granules that could be generated using starch matrices [4]. TABL presentV EI s data showin numbe e effece th g th f o t r of hole d holan se surfac e platedi e . th areOverall n i a , total open cross sectional ares i a diminished as hole diameter is reduced. In comparing processing behavior as a function of die diameter e observew , ddiametee di tha s s reducea l tothe al wa r d r an dproces s variables were maintained constant, die discharge temperature increased. This increased temperature results in the earlier onset of a flow rate limit. This behavior is expected since the smaller die diameter will yield greater pressure drop which results in increased discharge temperature. This increased temperature resultearliee th n i sr onse floa f wo t rate limit.

Scale4. up

f Scaltwio p neu screw extrusion utilize date th sa attained fro piloe mth t extruderr Fo . each herbicide system, the ultimate process rate limitation must be identified. The basic assumption is that the process will respond similarly on a larger extruder geometry. L/d is

132 I SCREW I

Figur - e4 Screw Configuration Extrudem s m Develope7 5 re th n do

TABLE IV: EFFECT OF EXTRUDER DIE PLATE CONFIGURATION FOR ENCAPSULATED HERBICIDES

Die Diameter No. of Total Open Die Granule Size (mm) Holes Cross Section mm (mesh) (mm2)

0.7 396 152 1(20)

10 308 242 2(10)

1.5 196 346 2.7(7) maintained constan d screan t w configuratio s scalei n geometricallyp u d assumptioe th f I . s i n correct procese e "finb th , en sca tuned " using extruder variables. screOftenw ne w a , design mus e evaluateb t production-scal e d testeth an dn do e extruder. Potential process limitations include extruder volume, available power or heat transfer surface area. We observed that extruder volume and heat transfer area were the limiting factors for the tested herbicide/corn starch matrices. Volumetric scale-u cubia s pi c functio f screo n w diamete s showa r equation ni n (1).

G -=a. .n(target) ,D (target) (target) ~moe 3 "(model (moD del)) where G is flow rate, n is screw speed and D is screw diameter. Maintaining L/d, screw geometr scred yan w speed constan volumetria r fo t c scal producep eu s equivalent degree-of-fill and average residence tim f withiscales i i p same u nth e famil extruderf yo s (targe moded an t l machine have equivalent geometrical clearances). Heat transfer limited processes are scaled up according to inner barrel surface area as show equation ni n (2).

n2 ' ° (* target} ^ (model) ~~ 2 ~~~~ D (model) All of the starch processes evaluated were not pure adiabatic systems. In most cases, direction of heat flow was from matrix to barrel surface. In this case, an overall heat transfer coefficient is calculated and then corrected for increased clearances between barrel wall and screw flight (larger screw diameter) and increased screw channel depth. pilo7 5 tK extrudeZS e th r r modelFo selecteds wa floa , whr , ratg basef k 102.eo 3 d extrudeo m l ninpumeaa m al 7 f n29.94o ts i r17 data scale e factoth p Th ,. eu o t 1 r7 5 fro e mth which yields a theoretical flow rate of 3046 kg hr on the 177 mm extruder. The model process is scaled up according to a volumetric limit, sinc1e deviation from empirical heat transfer data can be corrected by adjusting L/d. The commercial system is shown in Figure 5. The design of this process allows for two a.i. feedstreams - a solid and liquid. Multicomponent formulations greate possible ar r s ' that a.iwoulo . ebu ntw d requir epre-blendina gassumes stepwa t I . d that the starch would be supplied by p.d. railcar, solid a.i. in bulk bag and liquid a.i. in portable (250 galwaten ) to tanks0 r coole20 A . d chilling unicalculates wa t d base procesa n do s enthalpy of 0.175 kw-hr kg". Specification of the fluidized dryer is based on a reduction in matrix moisture concentratio reductioa d an % matrin !ni 10 n o xt fro temperatur0 m2 C ° 7 2 e o frot 3 m9 1 usin gspecifia kca6 0. l o t kg'c 4 hea0. °C"' f t o (c.p) 5. Preliminary Cost-to-Make Estimate

A preliminary cost estimat productioe th r efo n a starccos f o t h encapsulated herbicide indicates that for a hypothetical ZSK- 177 with an annual capacity of 18.2 million kg of product, the conversion 0.072cos$ s i t 6 kg'1. capacite Th evaluates ywa scalin y maximue db th p gu m production achieve ZSKe th - n do 57 as explained previously. This resulted in 3046 kg hr1. Assuming operational capacity of 24 h day'1, 250 day yr1 we arrived at the estimated annual capacity of 18. 2M kg of product. The rate of material loss was assumed 1%. The calculations also include 1.5 operators for 3 shifts per day, and capital of $5,000,000. Maintenance costs are estimated at 2.5% of the invested capital. Energy requirement extrude estimatem e 5 m ar scalee 17 d 7 sar . 5 fro0 an p r e dt u da mth kw-h cosa rt f $0.0 a o tkg' 6 (kw-h)" able ar estimato et e W . e energy requirements because eth 1 1

134 BAG DUMP STATION BAG DUMP STATION TRIAZINE HERBICIDE 152 TO 303 KG/HR

VATER METDLACHLOR 455 TD 305 TO l 761 KG/H KG/H9 60 R R CORNSTARCH 167O 5T 1980 KG/HR

AMBIEN SUPPLYR AI T -

Figur - e5 Commercial Twin Screw Based Facilit y- Proces s Flow Diagram (kw-hE SM r kg"1) remains constant when scalin frop gu m pilo familproductioo K t ZS ye th n ni f machineso e selecteW . highese dth valuE SM te attained fro l experimentmal arrivo t s a t ea conservative cost estimat productioe th f eo n costcose f $0.0Th o .t 6 (kw-hr0 also accountr fo s all process power requirements. This brings the Total Operating Cost to about $1.3M. TABLE summarize5 cose th s t estimation e conversioTh . n costotae bases i th t l n productiodo n output (doe t includsno materiae eth l lost during r shuo star p t u tdown) . The raw material cost depends on the type of product manufactured. Three scenarios were evaluated. Product A with a raw material cost of $0.66 kg"1, Product B with a cost of $1.32 kg'1 and Product C with a cost of $2.2 kg"1. The total production cost ranges therefore from $0.739 kg" r Produc1fo $2.3o t A t0 kg" r Producconversioe 1fo Th e . th C t f o n % cos3 s i t materiaw ra l cos r Producfo t (expensivC t w ra materials) w e era th f o increased % an , 10 o t s materials cost for Product C (the least expensive initial components).

TABL ENCAPSULATE: EV D HERBICIDE PRODUCTION COST

PRODUCTION RATE (kg hf') 3,046 ANNUAL CAPACITY (kg yr1) 18,200,000 LOST MATERIAL RAT) E(% 1

ANNUAL DEPRECIATION ($ yr') 500,000 ANNUAL OPERATOR COS yr')$ T( 154,440 ANNUAL MAINTENANCE COS yr$ T( 1 ) 136,230 ANNUAL ENERGY COST ($ yr1) 357,500 INDIRECT LABOR/OVERHEAD COST ($/yr) 150,000

TOTAL OPERATING COS kg')$ T( 2.855,974/yr

CONVERSION COST ($ kg') 0.073 PRODUCT A PRODUCT B PRODUCT C RAW MATERIAL COST ($ kg') 0.660 1.320 2.20 LOST MATERIAL ($ kg1) 0.0066 0.0132 0.022

MATERIAW RA L USAGE COS kg$ T( 1) 0.739 1.406 2.295

6. Conclusions

abilite Th versatilitd yan f encapsulatinyo gwida e rang f herbicideeo starcn i s h matrices using the twin screw extruder has been demonstrated. The process boundary conditions, including output rate, optimum position for feeding additives and screw speed, have been determined via experimental design techniques. Attempt generalizo st procese eth s wer t successfueno l because there is no representative average of the various herbicide chemicals that are commercially available.

136 In incompatibl r semi-compatibleo e system f starch-herbicideso , orde f additiomose o r th s t nwa critical variablee samth en I .systems , a.i. concentratio e s chemicai limitenth y b d l incompatibility with starch, the competition with water for starch association and the ability to die form the final product. Use of surfactants and emulsifiers greatly improved a.i. concentration and processability. However specifia , c surface modification system needdevelopee b o st r dfo every herbicide formulation, particularly when high level f incorporatioso e desirednar .

A Cost Analysis Estimat encapsulatioe th f eo n proces generates swa d base scalinn e do th p gu process from the 57 mm extruder. This evaluation suggests an encapsulation cost of $0.0726 per f producedo g k concentrate e totaTh .l production cost varies from $0.739 kg" $2.29o t 5 kg" 1 1 dependin utilizee costh e materialsf th w o t dra n go .

References

[1] DREIBLATT, A., Twin Screw Extruders, Process Plant Machinery, Butterworth, London . (1989pp 8 )58

[2] MIELCAREK, D., Twin Screw Compounding, Chem. Eng. Prog. 83 (1987) 41-49

[3] TRIMNELL, D., WING, R.E., CARR, M. E., DOANE, W. M., Encapsulation of EPTC in Starch by Twin Screw Extrusion, Starch 43 (1991) 146-151.

] CARR, WING [4 , E. DOANE . E. , Encapsulatio . M ,M. R , . W , f Atrazinno e withi Starcna h Matrix by Extrusion Processing, Cereal Chemistry 68 (1991) 262-266.

[5] WING, R. E., CARR, M. E., DOANE, W. M., Continuous Process for Starch Encapsulated Herbicides, Pesticide Formulations and Applications Systems 13 (1993).

Next page(s7 ) 13 left blank ENVIRONMENTAL FACTORS AFFECTING STARCH ENCAPSULATED HERBICIDE RATES OF RELEASE

T.J. GISH, B.J. WIENHOLD Hydrology Laboratory, Agricultural Research Service, United States Departmen f Agriculturo t e A. SHIRMOHAMADI Departmen f Agriculturao t l Engineering, University of Maryland, College Park, Maryland United State Americf so a

Abstract

Starch-encapsulation (SE) is an experimental control release technology designe exteno dt perioe dth timf o d e over whicha herbicide is released into the soil environment. The objectives of controlled release are to improve efficacy and reduce negative environmental impacts. Relatively little is known about how various environmental factors influence rates-of-release, or how controlled release influences herbicide environmental fate. Laboratory and field studies were conducted to evaluate how environmental factors influence the release rate of SE-atrazine (2-choro-4-ethylamino-6-isopropylamino-s-triazine) and SE- alachlor (2-chloro-2/,e'-diethyl-tf-fmethoxymethylj-acetanilide). Decreasing water availability, significantly reduced swelling and subsequent rates-of-release for both herbicides. As starch granules imbibe water they swell, allowin herbicide gth o et diffuse more readily out of the granule. At 0 MPa, complete releas atrazinf eo observes ewa d afte day1 afted 2 r san dayr7 s for alachlor. At -1.5 MPa, < 50% of the atrazine and < 80% of the alachlor was released from the starch granules after 28 days. Decreasing temperature also resulte decreasen i d d ratef so herbicide release. At 35°C nearly three times more atrazine and two times more alachlor was released than at 15°C. Soil microbial activity increased rates-of-release for both herbicides, likely the result of enzymatic breakdown of the starch matrix. Afte day1 r2 s twofola ther s ewa d increasn ei atrazine release relative to sterile soil. Effect of microbes on alachlor release was apparent only at early times. After 5 days increas% 20 thera alachlon s ei e wa r release compare sterilo dt e soilenhancee Th . d releas alachlof eo r relativ atrazino et e under varied environmental conditions was attributed to alachlor's greater solubility in water. Controlled release of atrazine reduced mobilit volatilitd yan y relativ commerciao et l formulation (CF). Although the SE-formulations are experimental, modification of herbicide behavior was observed that could reduce negative environmental impacts. Optimization studies based on herbicide characteristics should help maximize herbicide performance.

139 l. INTRODUCTION Over the past three decades, polymer chemists and soil scientists have been developing controlled release formulations of pesticides for agricultural use. The main rationale for controlled release pesticide formulation bees sha increaso n t e efficacy; however, with increase interest in environmental quality, controlled release pesticide formulations are also being evaluate their fo d r abilit reduco yt e environmental contamination. Determination of pesticide environmental impact is complex, and to a large extent is governed by how the pesticide is partitioned between the adsorbed, liquid and air phases. Adsorbed pesticides can be transported in runoff to surface transporter watero ) (1 s d preferentiall shalloo yt w groundwater through void root channel soir so l crack colloidaa n si l suspension (2). On the other hand, wet, warm soils have a propensity to favor vapor loss to the atmosphere (3). Additionally, the timing of precipitation events also effects environmental fat herbicidef eo s (2,4). Rainfall occurring shortly after application can result in significant preferential transpor pesticidf to shalloo et w groundwater. Starch-encapsulated (SE) pesticides maintai rate-limitena d release of the pesticide, limiting the amount of pesticide available for transported at any given time. Release of a SE compound is governed mainly by a diffusion process (5). When starch granules are applied to the soil they imbibe water, swell and the encapsulated compound diffuses out of the starch matrix. A strong correlation has been shown to exist between the degree of swelling exhibited by starch granules and the rate of release encapsulaten oa f d herbicide (6). Factors influencing ratf eo release froe starcmth h granule placee b n dsca into one-of-three categories; granule characteristics, characteristics of the encapsulated chemical environmentad an , l conditions. Characteristics of the granule include granule size (7), typ starcf eo h used (amylos amylopectio et n ratio) (8), agent used to crosslink the starch molecules, and amount of pesticide presen granule th n ti e (9). Granule size effect surface-toe sth - volume rati thereford oan distance eth chemicae eo th t s lha diffuse in order to leave the granule. Small granules exhibit faster rates of release than do larger granules (10). Granule composition (typ starcf eo crosslinkind han g agent) indirectly effects pesticide rat releasf eo modifyiny eb degree gth f eo granule swelling that occurpresence th n si water f eo . Wint ge ) reporteal(8 . dpercentage th tha s ta amylos f eo e starc th n i eh granules increased degree ,th swellinf eo g exhibitee th y b d granules declined and the rate of release of the encapsulated chemical also declined. Borate starch granules (granules prepared using borate to crosslink starch xanthate macromolecules) swell to a greater degree than do jet cooked starch granules. This results in faster release of the encapsulated chemical. Release of atrazine from borate starch granules was essentially complete after four days compared to eight daycompletr sfo e releas cooket eje froo d tw mformulation s (11). Finally amoune th chemicas f ,ta o l encapsulatee th n i d granule increases the rate of release (9). This paper review environmentae sth l factors influencing rate-of-releas atrazinf eo alachlod ean r from starch granules. The effect of this controlled release process on herbicide

140 behavior will be compared to commercial formulation (CF) herbicide behavior by assessing field-scale mobility and volatilization.

2. ENVIRONMENTAL FACTORS INFLUENCING RELEASE Environmental factors effect herbicide rates of release from starch granules by influencing either diffusion mechanics or by altering the integrity of the starch granule. Common environmental variables include water availability, temperature and microbial activity. Water availability may exert an influence on herbicide release by controlling water uptake by the granule which determines granule swelling and subsequent herbicide release. Soil temperature may effect granule swelling as well as diffusion mechanics. The starch granules are also biodegradable sucs subjece a har d an ,decompositioo t t d nan subsequent release of the herbicide.

2.1 WATER AVAILABILITY watee Th r potentia measuremene on s e availabiliti lth f o t y of soil water. As water is lost through evapotranspiration, a larger portion of the soil water becomes bound to colloids. The declin soin i e l water content leaves less unbound e wateth o s r energy statuwatere th soie f s,th o l water potential, declines. Polyethylene glycol (PEG) has been used to simulate a reduction in soil water potentia 12), hydrate(6 e l .Th moleculG PE d s i e very large and is unable to enter the pores of the starch granules, leaving only unbound water in the PEG solution to enter the starch matrix. When starch granules were place solutionn i d s of decreasing water potential, swelling and release of atrazine and alachlor decline wates a d r potential declined (Fig. 1) . Additionally, alachlor releas mors ewa e rapid than atrazine. After 5 days, 87% percent of the alachlor was released in a -0.5 MPa solution, while atrazine releas -0.a solutioa n i e5MP s wa n onlafte% 73 y days8 enhancee 2 r Th . d releas alachlof eo s rwa attribute alachlor'o dt s greater solubilit waten i y r (24 L'g 0m 1) than atrazine (33 mg L"1) . Rainfall events and subsequent evapotranspiration will generate several wetting-drying cycles over the growing season. e greatesTh t differentia waten i l r availabilite yth shoult a e b d soil-atmosphere interface, wher starce eth h granules resides A . a result, redistribution of water after a rain event may drawn the herbicide into small soil pores where rapid transport is less likely. Although there is no commercial instrumentation that can effectively monitor water potentials near the soil-atmospheric interface laboratore th , y studies indicate that soils with greater water retention wil more lb e effectiv maintaininn i e g high release rates tha sandna y soil alst I .o suggests that atrazine release will be less effective when subjected to drought conditions.

2.2 TEMPERATURE Since diffusio defines randoe ni th s mda thermal motiof no molecules increasn a , temperaturn i e e will also influence diffusion. Temperature has been shown to increase herbicide

141 rate releasf so e frostarce mth h granules (6) 35°t .A C nearly three times more atrazine and two times more alachlor was released from starch granules than at 15°C at all sampling times watea d ran potentiaC completa (Fi° MP g 35 0 t 2.)f A elo « releas atrazinf eo e occurred afte dayr8 s whil watea t ea r potentia encapsulatee th f -1.f o l o onl a % 0MP 30 y d atrazind eha been released frostarce mth h matrix afte daysr8 a 35°t d A .Can

100 - A

75 -

CO H 5 2 J Ü LJ o: o

0.0 MPa a -0.MP 5 a -1.MP 0 -1.5 MPa

0 0 0 2 10 30 TIME (days)

Figur Effec. e1 watef to r potential imposed using polyethylene glycol on percentage release of (A) atrazine and (B) alachlor from starch granule functioa s sa timef no . Error bars indicate ± 1 SE.

142 water potentia completa MP 0 f elo releas alachlof eo r occurred after 1 days while at a water potential of -1.0 MPa release of alachlocomplett no s rwa e afte daysr8 . springe Inth , when most crop plantede sar , soil surface temperatures vary depending on moisture status, plant cover and amoune th surfacf to e residue resulte Th . s discussed above

100 - A.

75 -

UJ 50 - CO < Lu 25 - LU KL Lu o O o

100 - B. UJ o 5 HL7 U ÛL 50 •]

25 -

0 0 3 6 TIME (days)

Figure 2. Percentage release of (A) atrazine and (B) alachlor from starch granules as a function of time, temperature 5 (•)° (•)[1 wate35 5 d (A), 2 ]d an an ,r potential [0.a (fille0MP d symbols -1.d )a an (ope 0MP n symbols)]. Error bars indicat. SE 1 e±

143 suggest that releas herbicideE S f eo s applie coolo t dsoi y ,dr l wil vere lb y slow compare releaso d t herbicide E S f eo s applieo t d warm, moist soil.

2.3 SOIL MICROBIAL ACTIVITY Soil microbes are able to produce the enzyme amylase which catalyze breakdowe sth starchf no . Through this procese sth microbe able sutilizar o e t e starcenergn a s ha y sourcet I . seems reasonabl assumo et e that microbial activity arouna d starch granule will result in granule decomposition and increased releas encapsulatee th f eo d herbicide (6). Schreibe . (13al )t re observed microbial activit vicinite th n i ysoif yo l applied starch granule Trimneld s. an (10 al ) t reportele d increased releas trifluralif eo n when starch granules were addeo dt solutions containing amylase. Soil microbial activity significantly increased the rate of release of atrazine and alachlor from starch granules applied to non-sterile soil compare granuleo dt s applie sterilo dt e soil (Fig ) (6).3 . After 28 days more than 90% of the encapsulated atrazine was released when incubate non-steriln i d e soil, compare <70o dt steriln %i e soils effece Th . microbia f to l activit less ywa s pronouncen do alachlor largely because alachlo releases ri quicklo s d y microbial activit shora d tyha time perio effeco dt t release.

3. EFFECT OF CONTROLLED RELEASE ON ENVIRONMENTAL FATE

1 3. MOBILIT PERSISTENCD YAN E Most herbicides hav higea h affinitsoie th l r matriyfo d xan are adsorbed quickly. However, a portion of the applied herbicide movn sca e preferentially, posin risga groundwateo kt r contamination. Since preferential transport is a convective process, increasing the role of diffusion may decrease mobility. The leaching potential for herbicides is greatest immediately after application, when herbicide concentrations are the greatest and plant evapotranspiration is negligible. Field- average atrazin havm 1 ~ e beet residua 1 nL" g e/i level 0 24 s> observe dayw fe s da afte r application consequenc,a f eo preferential transport (2, 4). To evaluate the effect of SE on herbicide transport ,laboratora y experimen conductes twa d using small soil cores extracted from a no-tillage field. Forty soil cores were subjected to water inputs that simulated preferential flow condition effluend san t collecte thao s d t transporf to technical grade atrazine and three experimental SE formulations could be compared (14). After 16 pore volumes > 35% of the technical atrazine had leached through the small cores compared to 1-10% for the three SE formulations. Although the duration of this latter study was < I month, it was an indication that SE atrazine woul lese db s susceptibl rapio et d transport durine gth greatese samth es timha t t i eleachin g potential. This study also served as justification for evaluating field-scale behavior of SE herbicides. In a two year field study, the mobility and persistence of atrazinF C d SEan e were compare adjacenn i d t 0.2 fielda 5h s (15). Persistenc herbicideE S f eo greates swa r tha F (FignC . .4)

144 Increased persistence was due not only to the gradual release of herbicide over time, but the manner in which it was released. Afte day1 day1 15 r1990n 1991n 16 si si d ,,an greater % tha36 n« of the atrazine applied as SE was in the top 5-cm increment while atrazine lesth s f o tha recoverees % 1 n wa applie e F C th s n a i d soif o l m differenco N (Fig1 to 1. p. .5) eF C betwee d an E nS alachlor mobilit observes ywa d (Fig. 6.).

100 - A.

75 -

Q - 0 5 U L GO 25 - LU

LU o 4 O 0 10 20

100 - B. LU

LJÜ CL 50 - * NONSTERILE • STERILE 25 -

0 0 0 2 10 30 TIME (days)

Figur Effec. e3 soif to l microbe percentagn so atrazin) (A f eo e alachlo) an(B d r released from starch granulea s sa function of time. Error bars indicate ± 1 SE.

145 LU Cd Lü >o o Lu 0 70 0 60 0 o:50 0 40 0 30 0 1020 0 Lu CT125 H o o B.

o 00 - <1_u Lu a. O o: 75 - Lu a. 50 -

25 -

0 0 100 200 300 400 500 600 700 TIME (days after Jun , 19901 e )

Figure 4. Field-scale recovery of herbicide as percentage of atrazin) appliealachlor) (A (B r d fo dean . Error bars indicate ± 1 SE.

Increase atrazinE S d e persistenc o likels t ei e ydu controlled release from the starch granule and its dependence on soil water availability. During a rain event, only a fraction of the applied SE atrazine is available for transport while all of the commercial applied broadcast spray may be transported in solutio colloidar no l suspension additionn I . , afte raie rth n even s ceasedtha atrazinF C portio, a e b e th y ema f n o

146 1990-91 1991-92

2 day1 s 18 days 1.25 - 1.25 -

1.00- 1.00 - 1 0.75 - 0.75 - \ 0.50 - \ 0.50 - \ 1 - , 0.25 - H 0.25 -

nn - «L. O 0 10 20 30 40 50 o 10 20 30 40 5

1.00 - 1.00 - 161 days I— r 157 days

0.75 - 0.75 -

• 0.50 - Lu 0,0- ;; O 0.25 - 0.25 - O O . V^ 0.00 0.00 -|— — — «=j-— *i —« —— | p- —• — . III! LJ o 10 20 30 40 50 •—r 0.25

0.05 -

0.00 20 30 40 50 SOIL DEPTH (cm)

Figur . Compariso5 e f field-scalo n e soil concentration (jig k g soil"1) profiles for atrazine applied as either SE or CF. Error bars indicate ± 1 SE.

147 00

o 1.25 -i 1.25 h- 1990 1991 12 days after application 8 day1 s after application h- 1.00 - 1.00 Z Lü 0 '^0.75- 0.75 z o 0 cr>0.50 - 0.50 et: l o ^0.25 - 0.25 I C5 n n n .——X—=^ a a. a — — a-^- a— e-»- e— - » n .n 0 0 10 20 30 40 50 0 10 20 30 40 50 SOIL DEPTH (cm)

Figure 6. Comparison of soil profile concentrations (ßg kg soil"1) quantifiablo alachlor N fo . CF eithe s a r o E erS amount alachlof so r were detecte e laso th ttw n i d sample period 1992d 199f so an 1. Error bars indicate ± 1. preferentially located along the walls of void root channels and soil cracks, susceptibl furtheo et r movement from subsequent rain events. Subjected to the same meteorological conditions, only a fraction of the SE-atrazine will be available for transport. Afte raie ceaseds rth nha soie ,th l water potentia soie th l f lo surface will be large, maximizing release from the starch granules concentratioe Th . n gradient aroun granule dth e should also be larger resulting in enhanced diffusive movement into the soil matrix (Fig. soie 7.)th l s .drieA s out, atrazine ratf eo release decreases. Since large soil pores will empty first, water will move into successively smaller pores, carrying pesticide into smaller pores that are less susceptible to preferential flow mechanics resulting from subsequent rain events e differencTh . mobiliteF C betwee d an largels i yE n S e ydu to the smaller flow pathways utilized with SE.

3.2 VOLATILIZATION Volatilizatio majoe th rf o lose on s s pathwayi n somr sfo e pesticides. Pesticide present in the vapor phase in the atmosphere may contaminate surface water by washout in precipitation, fallou particulatf to e materia whico lt e hth pesticide has become adsorbed, and by direct exchange between the atmospher wated ean r surface. Glotfelt . (16al )t ye estimated

HYPOTHETICAL PESTICIDE CONCENTRATION PROFILES CONVECTIO DIFFUSIO. VS N N DOMINATED TRANSPORT

COMMERICAL BROADCAST SPRAY STARCH-ENCAPSULATED (corrective transport) (diffusive transport)

2 m

Figur Hypothetica. 7 e mobilitF lC comparisod an soiln i yE S ,f no

149 that atrazine entering the Chesapeake Bay in washout and fallout tha f o abous runoffn twa % i enterint10 y ba . e Basegth n o d atmospheric and surface water concentrations of atrazine, and the distribution coefficient between air and water for atrazine they also determined that the Chesapeake Bay was undersaturated and t flu atrazinne f x o e th e shoul froe atmosphere db mth e th o et bay. However, they were unabl determino et magnitude eth f eo this flux. numbeA studief ro s utilizin herbicideE gS s have demonstrated increased efficacy when compared to commercial formulation s19)d an (17 ., ,The18 . increas efficacn ei E S f yo herbicide attributes swa reduceo dt d volatilization. However, these studies did not report any measured volatilization rates, but inferre frot i d m pesticide dissipation. Unfortunately, there are a number of loss pathways that effect pesticide dissipation (volatilization, chemical, biologica photo-degradationd lan , leaching pland ,an t uptake) thereby generating some speculation as to the impact of SE on volatility. Igreenhousna e study, volatilizatio atrazinf no d ean alachlo rcompares appliewa F C thret eithes a dr a d o eE rS temperature s35°Cd an (15 ), ,25 (20) . Cumulative volatilization of CF atrazine at 35°C was nearly two orders of magnitude greater than at 15°C (Fig. 8). Cumulative volatilization of CF atrazine was approximately four times that of SE atrazine at all temperatures. The effect of SE on volatilization of alachlor was opposite that of atrazine. At 25 and 35°C volatilization of SE alachlo folo tw d s greaterwa alachlorF C thar nfo r (Fig. 9) . These results suggest pesticidn o tha effece E tS th f to e volatilizatio dependene nar e chemicath n to l characteristicf so the encapsulated pesticide. When starch granules are applied to a moist soil surface they imbibe water and the encapsulated chemical goes into solution withi starce nth h granule where adsorption of the chemical by the matrix is low. For atrazine, the solution concentration withi granule nth e likely remainw slo due to its low solubility in water (33 mg L"1) , which combined with it's low Henry's Law constant (2.5 X 10"7), results in little volatilization. In contrast, the solution concentration of alachlor within the granule may be an order of magnitude greater (solubility 240 mg L"1) than that of atrazine resulting in a much steeper solution concentration-vapor density gradienr tfo alachlor thaatraziner nfo . This steep gradient, combined with alachlors higher Henry's Law constant (1.3 X 10"6) , results in much greater volatilization.

4. CONCLUSIONS Starch-encapsulation is a control release technology that shows potentia modifyinr lfo behavioe gth pesticidesf ro e Th . pesticidinfluencn o s ha E eS behavior appear variablee b o st , depending largelcharacteristice th n yo encapsulatee th f so d chemical. Starch encapsulation modifie behavioe sth a f ro chemical by controlling the rate the chemical is released into the environment. Rate of release is strongly influenced by environmental factors, especially soil water availabilitd yan temperature characteristicd ,an encapsulatee th f so d chemical, especially solubilit watern i y . Mobilit atrazinE S f yo d ean alachlor, both runoff and leaching, was reduced compared to that of CF atrazine and alachlor. Persistence of SE atrazine was

150 20 30 40 TIME (days)

Figure 8. Cumulative atrazine volatilized from soil at three soil temperatures after application as either commercial or as starch encapsulated. Error bars denot range eth f eo values observed. Note difference in Y-axis scale among graphs.

151 1200

10 20 30 40 TIME (days)

Figur Cumulativ. e9 e alachlor volatilized from soi thret la e soil temperatures after applicatio eithes na r commerciar lo starcs a h encapsulated. Error bars denot range eth f eo values observed. Note difference in Y-axis scale among graphs.

152 substantially greater tha natrazine F C tha f to . This increasn ei persistence was likely due to reduced losses of atrazine to leaching, and volatilization. Persistence of alachlor was not influenced by SE, likely because of the rapid rate of release for alachlor volatilization o e effecE Th S . f to surfacf no e applied SE herbicides is variable. Compared to CF, volatilization of SE- atrazine was reduced in a greenhouse experiment. These results suggest tha technologE tS usefua e b ly approacyma reducinr fo h g environmental contamination by agriculturally applied pesticides.

REFERENCES

1. GLOTFELTY TAYLOR, , ISENSEEE. W. , . ,D . R. , A . ,A JERSEY, R., and GLEN, S. Atrazine and simazine movement to Wye River Estuary. J. Environ. Qual. 13 (1984) 115-121.

2. GISH, T. J., ISENSEE, A. R., NASH, R. G., and HELLING, C. S. Impact of pesticides on shallow groundwater quality. Trans Soc. Eng. .Am 4 (1991. Ag 3 . ) 1745-1753.

3. CLAITHSPENCERd an FARMER, . M F. . J . , . M , . W , W Pesticide volatilization. Res. Rev 9 (1973.4 ) 1-47.

4. ISENSEE, A. R., NASH, R. G., and HELLING, C. S. Effect of conventional vs. no-tillage on pesticide leaching to shallow groundwater. J. Environ. Qual. 19 (1990) 434- 440. 5. SCHREIBER, M. M., and WHITE, M. D. Granule structure and rate of release with starch-encapsulated thiocarbamates. Weed Sei. 28 (1980) 685-690. 6. WIENHOLDGISH. d EffecJ an . , T . wate f J to . ,B r potential, temperature and soil microbial activity on release of starch encapsulated atrazine and alachlor. J. Environ. Qual 1 (1992.2 ) 382-386. . 7 TRIMNELL SHASHAd an . S Controlle . ,D . , B d release formulations of atrazine in starch for potential reduction of groundwater pollution. J. Controlled Release 12 (1990) 251-256. . 8 HAITIDOANE , d WING. AmylosE. an N . . . ,R W S e conten starcf to h control release sth encapsulatef eo d bioactive agents. J. Controlled Release 5 (1988) 79- 89.

DOANNEd an . M . , WING9 . , ,E. CARRW E. . ,R . ,M Encapsulatio atrazinf no e withi starcna h matriy xb extrusion processing. Cereal Chem. 68 (1991) 362-266.

. TRIMNELL10 , 1985OTEYd H. e an . SHASHA, Th .,F , , D. S. . ,B effect of a-amylases upon the release of trifluralin encapsulate starchn i d Controlle. .J d Releas e(19851 ) 183-190.

153 SHIRHMOHAMMADI d . WIENHOLD, GISHan 11 J. , . . ,T J . ,B . ,A Effect of starch-encapsulation on behavior of atrazine and alachlor. Am. Inst. Hydro. Proceedings of the Joint CIS/USA Hydrogeolog Hydrologd yan y8 , mt.6- y ,Ma 1993. Washington D.C. (1993) 109-124. 12. ZUR, B. Osmotic control of the matric soil-water potential Soil-wate. I : r system. Soil Sei 2 (1966.10 ) 394-398.

13. SCHREIBER TRIMNELL , WHITE, WING, E. M. D. . . ,R . ,M , M , D. and Shasha, B. S. Bioactivity of controlled release formulation starch-encapsulatef so d EPTC. J . Controlled Releas e7 (1988 ) 237-242.

14. GISH, T. J., SCHOPPERT, M. J., HELLING, C. S., SHIRMOHAMMADI, A., SCHREIBER, M. M., and WING, R. E. 1991. Transport comparison of technical grade and starch-encapsulated atrazine. Trans. ASA 4 (19913 E ) 1738-1744.

. , SHIRMOHAMMADIGISH15 J. . ,T WIENHOLDd an , . J ,A. . ,B Field-scale mobility and persistence of commercial and starch-encapsulated atrazine and alachlor. J. Environ. QualPRESS)N (I . .

. GLOTFELTY16 WILLIAMS, E. d FREEMAN, an . ,H. D . . P ,G . ,H LEECH, M. M. Regional atmospheric transport and deposition of pesticides in Maryland. In: D. A. Kurtz (Editor) Long Range Transport of Pesticides. Lewis Publ. Inc. Chelsea . (1990,MI ) 199-221.

. SCHREIBER17 ORWICKSHASHA, , ROSS, M. A. S. . . ,M . M , B EDGECOMBd an . L . ,P JR.. EfficacW ,. D ratd yan f e o release of EPTC and butylate from starch encapsulated formulations under greenhouse conditions. Weed Sei6 .2 (1978) 679-686.

. CENTNEd COFFMAN 18 an . Persistenc A . B . R W . ,C f eo several controlled release formulation trifluralif so n in greenhouse and field. Weed Sei. 28 (1980) 21-23. 19. COFFMAN, C. B., CENTNER, W. A. and SHASHA, B. S. Herbicidal activit controlled-releasf o y e formulations of trifluralin. Indian J. Agric. Sei. 54 (1984) 117- 122.

. WIENHOLD20 SADEGHI, J. GISH. d J Effec. an ,B . ,T . , f A t o starch encapsulation and temperature on volatilization of atrazin alachlord ean Environ. J . . Qual 2 (1993.2 ) 162-166.

154 MOBILITY, TRANSPORT AND ENVIRONMENTAL IMPACF TO STARCH ENCAPSULATED FORMULATIONS OF HERBICIDES

M.V. HICKMAN, M.M. SCHREIBER Insect and Weed Control Research, Agricultural Research Service, United States Departmen f Agricultureo t , West Lafayette, Indiana, United State f Americso a

Abstract The development and use of controlled release formulation of agrichemical products could alleviat ee currenmanth f o y t environmental concerns. Experimental starch-encapsulated (SE) formulations of herbicides have demonstrated their capabilit o reduct y e agrichemical o t losse e du s volatility, movemen surfacn i t e runoff water d leachinan , g throug e soith hl profile including movemen a macroporvi t e flow. Trials using packed soil columns and intact soil blocks have demonstrated the effectiveness of SE atrazine formulation reduco t s e leaching. These trials routinely show reductions in atrazine leaching of 65% to 85% of the leaching of commercially formulated (CF) atrazine. Field trials which monitor atrazin soin ei l cores ove course th r f eo the season have confirmed these findings at several locations. Atrazine concentrations in surface runoff from SE treated plots were reduced 80% compared with the CF atrazine in field trials under high rainfall. In season long fiel dtreatment E trialS e sth s reduced atrazine losses measure fiele th dt da edg y eb 40%. Extended residual activity in rotational crops from SE have been shown to problematie b t no labelen ci d crop rotations thu testedr sfa . Considering thal al t of the SE formulations tested are experimental, the results suggest that when the formulations are optimized for each active ingredient or cropping situation, the use of SE formulations could have a significant impact on the environmental contaminatio f soil nwateo d san r from non-point agricultural sources.

. 1 Introduction

Environmental concerns have come to the forefront in agriculture in the United States and Worldwide. These include potential contamination of groundwate surfacd an ] 2 e, water[1 r [3,4] resources. Modern agricultural production systems depend upon input pesticidef so s in order to meet the high demands for food and fiber. Herbicides are a significant portio e totath f l no pesticide s usedr exampleFo . e corth , n (Zead an mays ) L. soybean ) productio(GlycineL. x ma n regioe midwesterth f no n United States utilizes over 36,000 t of two herbicides, atrazine [6-chloro-N-ethyl-N'-(l- methylethyl)-l,3,5-triazine-2,4-diamine] and alachlor [2-chloro-N-(2,6- diethylphenyl)-A/-(methoxymethyl)acetamide] [5]. Atrazine has been frequently detected in groundwater [5] and surface water [6] and has been reported in rainfall [4]. Chemically, atrazine is a relatively stable molecule wit hwatea r solubilit , vapo mg/C 3 t 3 7 a f 2 ry a o L t pressura uP 9 3. f eo

155 averagn a d an e 2 5C persistenc t recommendea e year1 e rate< dus f . o s [7]. Atrazine herbicid bees eha n show leaco nt h through soimovd an l e with surface water under normal usage [8]. ] suggesteSchreibe[9 . al t e dr that starch encapsulated (SE) formulation f pesticideo s s could reduce leaching losses while maintaining required efficacy. Developmen f starco t h encapsulation technolog s progresseyha d froma batch chemical proces continuousa o st , mechanical extrusion process since it's , 15]14 . , Encapsulatio 13 beginning1970'd , mi 12 e s, th [10 11 n ,si starcn ni h involves the entrapment of pesticides within a solid starch matrix. The resulting granules exhibit controlled release propertie havd potentiae san e th controo t l l pesticide movement from site of application through leaching, water runoff, or volatility [8, 9,12,16,17] Current data suggest that experimental SE formulations significantly modif e mobilitth y f atrazino y n soili e n comparisoi , n with commercial formulations (CF) [18,19, 20]. This paper presents data from recent research to support the assertion that experimental SE formulations can reduce herbicide leaching through soil and movement with surface water flow. Additionally, data wil presentee b l o dt address concerns of increased residual activity from use of SE formulations.

. 2 Leaching

Schreiber et al. [9] suggested SE formulations as an approach to controlling the leaching of agricultural chemical. Many SE formulations, have been produce tested dan systemn di s ranging from packed soil column fielo st d scale trials. Several experimental SE formulations of atrazine were tested using packed soil columns at West Lafayette, IN to determine the initial mobility of atrazin n differeno e t soi formulationE lS typese Th . s were producey b d mechanical extrusion [15] using pearl cornstarc technicad han l grade atrazine. The extruded material was collected, ground and screened to size (TABLE I). Granules were sized into three categories with the larger granules able to pass a 1.4 mm screen but retained on a 0.85 mm screen, the midsize granules able to pass a 0.85 mm screen but retained on a 0.425 mm screen, and the fine granules passin g0.42a screenm formulation5m E S . s prepare thin di s manne havy rma e from 10 to 15% of the active ingredient (a.i) present at or near the surface resulting in a slightly faster initial release rate with some compounds. This material can be removed by briefly washing the granules hi an organic solvent. This was done in order to produce two of the formulations described in TABLE I (443 443D)d Can . Thidons swa orde n ei examino rt earl e effece th e th yf o treleas e leachine th n o atrazinef go .

2.1. Packed soil columns

Two soils Miama , i silt loam (fine-loamy, mixed, mesic Typic Hapludalfs, pH = 6.5, and 3.8% organic matter) and a Tracy sandy loam (course loamy, mixed mesic Ultic Hapludalfs, pH = 6.4, OM = 2.6%) were collected, air dried and screened through a 2 mm wire mesh. The <2 mm fraction was used to pack aluminum leaching columns columne Th . s consiste f aluminudo m tubes 47.2 cm long and 7.6 cm in diameter with an open slot 1.9 cm wide and 38 cm long cut int coveres slosidee e oth wa t Th sealed . dan d wit aluminun ha m strip heln di

156 TABLE I. CHARACTERISTICS OF ATRAZINE FORMULATIONS USED IN TRIALS WEST A T LAFAYETTE, IN.,

Formulation Granule Size Active Ingredient (a.i) Loading

passed-retained

mm %

149A 1.4 0.85 11.1

149B 0.85 0.425 11.1

149C 0.425 11.1

443A 4 1. 0.85 10.0

443 B 0.85 0.425 10.0

443C 1.4 0.85 9.0 a

443D 0.85 0.425 8.6a

4L commercial liquid 43.0

a Formulations 443C and 443D were surface washed with solvent to remove nonencapsulated herbicide following grinding and screening.

plac screy eb w clamps botto e colume Th th .s close f mo nwa d wit one-holha e rubber stopper and the columns were uniformly packed to a soil bulk density of 1.5g cm~3 with the air dried soils. The packed columns were secured in a vertical positio d saturatenan f distilleo l d e m witth d0 f hwate75 o p rto addee th o dt column. The columns were allowed to drain for 24 h before an additional 380 ml of distilled water was added and the leachate collected to verify column uniformit f packingyo . eacf Atrazino hp tubto applies rat a e t eewa ea th equivaleno dt 3.3o t 6g k of a.i ha'1 and a 1 cm layer of washed silica sand was added to limit the disruptio soie th l f surfacno e during additio waterf n o herbicide Th . e treatments consiste commerciaa f do l liquid formulatio formulationE S d nan s describea s d abov presented ean TABLn di . EI The columns were leached with 250 ml of distilled water (equivalent to 5 cm rainfall). Leaching was accomplished at an approximate water flowrate of 4 heam c watef m do maintaines 1 l min'a wa r colume d th an 1 n do n durine gth leaching. Columns were allowed to drain for 24 h and the leachate volume was recorde additionan a s da l chec f columko n uniformity seconA . d leaching event was performed in a like manner to the first using 150 ml of distilled water (3 cm rainfall). Followin e drainaggth e secone perioth e r th dfo f o leachin p to e gth columns were closed by insertion of a large rubber stopper and the columns were positioned horizontally in trays with the slotted side up. The slot covers were carefully remove columne th d dan s were seeded with approximatel seed0 y40 f so

157 bentgrass (Agrostis tenais Sibth. bioassaa s a ) y species e columnTh . s were maintained in the greenhouse for 14 days to allow for bentgrass growth and the exten leachinf o t determines gwa measuriny db e distance gth th f o p e to fro e mth colum poine th o tnt were bentgras unaffecteds swa triale Th s. were replicated thre edate statisticalls timesth a wa d an , y analyze analysiy db f variancso d ean means were separated by Fisher's protected least significant difference test [21]. SE formulations significantly reduced atrazine leaching on both soil types compared to the CF (Figures 1 and 2). Approximately 10 to 15% of the active ingredien r nea o surfac e formulation E th n rremovee S o b f s n i o et ca y db d an s

33.0

SANDY LOAM SOIL

LSD (0.05) = S.82

4.8

CF A3 44 443B C 3 44 443D FORMULATION

Figur . 1 eLeachabilit f pearo y l starch encapsulated atrazine wito tw h granule sizes and washed and unwashed granules on sandy loam soil. 44 443d 3 Aan C , retaine passemm screen 4 0.8m y d1. b 5m , 443 443d Ban D passed 0.85 mm, retained by 0.425 mm. 443C and 443D were solvent washe removo dt e unencapsulated atrazine. Tota. l rainfalmm 5 7 = l

25.5

SILT LOAM SOIL

(0.05D 2 4. LS = )

149A 149B FORMULATION

Figur . eLeachabilit2 f pearyo l starch encapsulated atrazin siln eo t loam soil. 149A passed 1.4 mm, retained by 0.85 mm, 149B passed 0.85 mm, retaine 0.42 y dscreenb m 5m . Tota. l rainfalmm 5 7 = l

158 washin witE S he solventgth t significantlno . d Washindi E S ye changgth e eth atrazine leaching in this trial although the washed granules tended to leach less tha nonwashee nth d granules. Granule size mad measurablo en e differencn i e leaching on either soil type tested (Figures 1 and 2). When considered over both soil types, both granule sizes and washed vs. nonwashed the SE formulations reduced atrazine leachin. compare% CF 75 e y th g b o dt Wauchope et al. [22] conducted similar trials on a Lakeland sand soil under unsaturated flow conditions. Treatments include atrazineE S d F C , atrazine applied witwithoud han polymeria t c adjuvant designe 'leachina s da g prevention plu E polymerie ' S agent sth e Th . c agent reduced atrazine movement by 90% compared to the CF alone. Bioassay data from these trials showed that atrazine th beed eha n -retaine uppee th d n di ran portionE S colum e e th th f so y nb SE plus polymeric adjuvant leading the authors to conclude that SE formulations may have a place in controlling atrazine movement. Boydston [19] packed columns with screened Quincy loamy sand dan treated the prewetted surface with a 90% CF of simazine (6-chloro-N,N'-diethyl- l/3,5-triazine-2,4-diamine), a herbicide similar to atrazine, and a SE formulation of simazine in pearl cornstarch. The treatments were allowed to equilibrate for 2 4beforh e leachingformulationE S e Th . s reduced simazine movemen ovey b t r 80% compared to the CF. The author concluded that SE formulations could significantly reduce the early season leaching potential of simazine. The data leae authoth d o exprest r s concern thaformulationE S t y resulma sn i t unacceptable level f residuao s l simazine availabl r earlefo y season leachinn gi followine th g year. Flemin . [23al t ]ge use d packed soil column bioassayd san determino st e the effect of granule size, clay amendments and processing parameters on the leaching potential and efficacy of SE atrazine formulations. The trials used a 7.5 cm diameter by 40 cm columns packed with Plainfield sand (Typic Udipsamment, mixed mesic) soi sand% l tha98 , s 0.7i t % organia cd matteha d ran pH = 6.2. Treatments included a 90% dry flowable CF and SE formulations in pearl cornstarch atrazinewit% h11 columne Th . s were prewetted, treated with herbicide thed san n subjecte leachino dt g wit htotaa watef f 15.o o l m 2rc applied at a rate of 1.3 cm in 20 min per day for 12 days. Leachate was collected and atrazine content determined. When leaching was complete, the soil columns were sectioned segment m intc o5 atrazind an s e conten s determinedwa t e Th . authors recovered 68% more atrazine from the top 5 cm of the SE treated columns tha treatmentsF n C resultefroF e C mth pulsa e n dTh i f atrazin e.o e that moved throug peae colume hth th kd concentrationan 25-3e th m t 0c a s nwa depth relateA . d efficacy trial also reported found that differencen i s bioavailabilit formulationyE S betwee d an F nC s coul overcome db decreasiny eb g the granule size.

2.2 Intact Soil Columns

A leaching study using intact soil blocks was carried out at West Lafayette, IN in the fall of 1990. Columns were removed from continuous soybean plots, wite on withe no-til on falha d l an chisel l plow tillage system chisee Th . l plow system included secondary tillage in the spring for seedbed preparation and one cultivation in early season. The columns were removed in the fall after soybean harves t beforbu t e fall tillage soie t thia lTh Treata .s s sitewa y silt loam witH hp

159 of 6.3 and 21, 50, and 29%, sand, silt and day, respectively. The columns were 76 werd an e m takec 0 n3 y fro b interron ma n o 6 7 w cy m spacb e that normalld ydi not receive tractor traffic. The columns were encased in cement in the field and supported with wire mesh beneat preveno ht t collapse during transpor raia o nt simulation facility. Herbicide treatments of atrazine at 3.36 kg/ha~l as liquid CF or as SE formulation 149B (TABLE I) were added 1 hour before rainfall began. The columns were situated to allow for the collection of all effluent water and to colume th f of t nallo n watesurfaceno y wru an o t r . Rainfal applies o wa ltw n di events using distilled water. The first rainfall was applied at a rate of 40 mm h'l . Effluenh secon2 a followe h"h r fores n d m I *2 fo r m an t a d 5 raiy 2 db f no water from the columns was collected for the first hour and then in 15 min increments until the end of the rainfall (300 min). The columns were allowed to drain overnigh thid an st drainag alss oewa collecte analyzedd dan . Atrazine content in the effluent waters was determined by solid phase extraction and gas chromatography. The treatments were replicated three times. The effluent volume from no-til chised an l l plow tillage block nearls swa y equa thin li lonse triath go t timl lace (Figure differentiaf eko Th du . e b e 3) y ma l that passed aftesecondare th r y tillagchisee th n eli plow tillage allowine gth syste reformo t m macropores havy thaema t been disrupte E tillagey S d b e Th . formulation reduced the total atrazine leaching by 80 and 60% over the CF, for chisel and no-till, respectively (Figure 4A and 4B). The SE formulations also eliminated the atrazine surge that is often present with early water flow on CF treatments. This is consistent with packed column data reported by Fleming et al. [23]. . [18Gisal t ]he measured atrazine movemen smalln i t , undisturbed soil columns taken from an established no-till management site. The columns (45 cin^ by 3 on) were treated with either technical grade atrazine or SE atrazine at the same rate. The columns were leached with a total of 16.1 pore volumes of water applied throug hdria p irrigation system watee Th .r applications were designed to favor preferential flow and all effluent waters were analyzed for

——— i:iiisËL~ - _ _ No/i ILL

0 10830 05 28 0 17 5 25 0 24 5 22 0 21 5 19 0 18 0 12 5 10 1 91 5 7 fid TIME (min)

Figure 3. Effluent water collected from intact soil blocks from no-till and chisel plow tillage systems under simulated rainfall .

160 Figure 4. Atrazine in leachate from intact soil blocks treated with commercial liquid and pearl starch encapsulated formulations and subjected to simulated rainfall. A, Chisel plow tillage; B, No-till. atrazine. After 16.1 pore volumee technicath f o f wate o s % l atrazin35 r d eha moved throug e columnth h s compare atrazinE S d e witth e hf o onl% 3 y formulation.

2.3 Field Studies

The ability of SE formulations to reduce leaching in packed and intact soil columns translate fielde th o .st Trial formulationE S sd usinan F gC f atrazin o s e were conducte Treata n do y silt loam soi t Wesa l t Lafayett 199n 1991i d 0an N e I . Atrazin applies ewa t ha"ratedg a k 199 2.2n f a i l 8 sha'o g n 14i 2. k 199 n 1i d 0an study plot tharandomizem a 5 n si t1 usey b d3 d complete block design with four replications. Soil cores were taken frolocationo mtw s within each plota o t ,

161 week8 2 , scm afte0 dept12 r f treatmenho week 0 1992 n i t d 0san after treatmenn ti 1991. The cores were divided into 15 cm sections except for the upper 15 cm whic sectiones hwa core sectionsth d m ef c int sectiono 5 l o7. Al . s were extracted and the atrazine concentration was determined. totae Th l quantit f atrazinyo e recovered corfros m entire c emwa th 0 e12 formulationo tw t differen e no th r fo 1990n ti s , however atrazine recovery from the entire core was three times greater for the SE that the CF in 1991. Below normal precipitation throughou 199e th t 1 growing seaso hely npma explaie nth increased recovery of the SE formulations. From a leaching perspective, the total s importanrecovera t e distributiono th s s yi a e therbicid th e soif no th l n i e profile. In 1990 over 95% of the atrazine recovered from the SE treatments was soif preseno luppee m (TABLc th 5 n i t1 r compare) EII . CF d e witth f ho onl% y63 Both formulations had over 90% of the recovered from the top 15 cm in 1991, a below normal precipitation year.

TABLE II. DISTRIBUTION OF ATRAZINE RESIDUES IN SOIL COLUMNS TAKEN FROM FIELD TRIALS AT WEST LAFAYETTE, INDIAN 1991991D N AI 0AN .

Treatment Formulation Rate ______Depth (cm)______JfcZ.5 7,5-15 15r3U 30-45 45-60 60-75 75-90 90-105. 105-120 Atrazine recovered

1990-28 WEEKS AFTER APPLICATION

kg ha- > /c Atrazine 4L 2.24 39.2 23.4 16.1 3.8 0.6 4.8 5.0 6.1 1.0 Atrazine SE, 14-20 2.24 73.4 21.9 4.0 0.0 0.0 0.0 0.0 0.0 0.71

1991-20 WEEKS AFTER APPLICATION

Atrazine 4L 2.8 82.2 10.5 2.9 0.0 0.8 1.9 1.7 0.0 0.0 Atrazine , 14-2SE 0 2.8 90.2 7.2 1.1 0.0 0.1 0.3 0.6 0.5 0.0

3 Surface Movement

An experimental s establishesitewa Junn di f 199eo determino 2t e eth effect f residuo s e cover, tillage practice [24 herbicidd ]an e formulatio soin o n l erosion and herbicide loss via surface water runoff. The study was located near Lexingtonfielo tw d n siteo L sI , with long term cropping historiesa .s Sitwa e1 conventionally tilled field manage corn-soybeaa n di n rotation previoue Th . s crop was corn. The soil was a Saybrook silt loam (fine-silty, mixed, mesic Typic Agriudoll). The trial was designed to utilize a rain simulator to apply controlled rainfall to 1 by 2 m, interill plots. Atrazine was applied as a spray for the CF s 149liquia granuleE d BS dan ha'g k s 18 (TABL 2. approximatel t a ) EI houy1 r before rainfall began. Treatment t sitsa weree1 : freshly , freshltilleCF d+ y tilled , freshlSE + y tilled+SE+residue removed d freshlan , y tilled+CF+residue removed treatmente Th . s were replicate timesd4 .s locate Sitlonwa a e2 n gdo term (15+ years) no-till fielcora n dni soybean rotation wit hprevioue corth s na s crop. Treatments were: no-till+Cf, no-till+SE, no-till+CF+residue removed and no-till+SE+residue.

162 Rainfal s appliewa l eaco dt n hh'lfo m intensitn mi plo a m 0 t 9 0 ra t 7 f yo usin gprogrammabla e rainfall simulator rainfale Th . l simulator using distilled water, was positioned 3 m above the plot surface. Runoff samples were collected intervaln mi a5 t s durin e rainfalgth runofd d infiltratioan l an f n rates were calculated. Sample herbicidr sfo e analysis were spiked wit internan ha l standard stored an d under refrigeration until processed watee Th . r samples were filtered, sediment content determined and the atrazine extracted using solid phase extraction techniques. Herbicide residues were quantifies ga y b d chromatography. Residue cove s estimatewa r d visuall d quantifiean y removay b d d an l weighing fro residue mth e removal plotsconventionae Th . l tillage plotd ha s approximatel cove% visuay y30 b r l estimat m~2eg k (0.2) 6 whil no-tile eth l plots had 100% residue cover (1.12 kg m~2). The infiltration rate was essentially 100% no-tile th n lo plots wit residue hth herbicido placn n ei o s e e lossurfaca vi s e movement occurred l comparisonAl . formulatiof so n were made usin plote gth s with residue removed. Runoff with residue removed occurred sooner froe mth no-till plots that fro e conventionallmth y tilled plots, howeve finae th r l runoff infiltratiod an n values were nearlr no-tilh"m fo ^m d y5 an l 2 identica . vs 3 (2 l conventional, respectively). Equivalen t meat no runof nd di equivalenf t sediment losses. No-till soils were structurally stronger tha conventionalle nth y tilled soils resulting in nearly a 3 fold increase in sediment loss from the conventional tillage treatments. Further surface th , e flo conventionan wi l tilled plots was more concentrated and channeling added to the sediment losses. When data were averaged over tillages, the CF lost 1.65% of applied atrazine in the runoff water compared with 0.35% loss for the SE formulations. This is a reduction of nearly 80% by the SE. A statistically significant interaction occurred between formulatio tillagd nan thin ei s trial (Figur . Runofe5) f losses from CF atrazine on conventional tillage were slightly less than losses from the no-tile th n SEl O plot. atrazine sth e los runofn i t f wate mors wa r e tha fol0 n2 d greater tha formulationsnE S fro e mth .

SE CF SE CF NO-TILL CONVENTIONAL

Figure 5. Atrazine loss in surface runoff water as influenced by herbicide formulation and tillage system under simulated rainfall at Lexington, Illinois, 1992. CF = commercial liquid atrazine formulation, SE = pearl starch encapsulated atrazine with granules werd thaan et m passem 4 d1. retained by 0.85 mm screen.

163 Mills et al. [20] conducted a study on corn plots near Topeka, Kansas

equipped to allow for collection of all surface runoff water. Herbicides were applied at 2 kg ha" an1 d incorporated into the top 5 cm of soil with a rotory tiller. Natural rainfall and sprinkler irrigation were used to add a total of approximatel plotwatef e o th sm o t r c ove 2 seasone y4 rth . SE formulations reduce e initiadth l flus f atrazino h e leavin e fielgth n i d runoff immediately after application by a factor of 20 compared to the CF. The authors concluded that the use of SE atrazine could reduce the mass of atrazine entering surface waters by 40%. Additionally, SE formulations would produce a more even flu f herbicideo x s through time resultin improven gi d surface water quality.

4 Residual Activity

Laborator d fielan yd studies have demonstrate e abilitieth dE S f o s formulations to slow herbicide availability, reduce herbicide leaching losses and reduce the movement of herbicides in surface water. It follows then, that SE formulations hav increasen ea d potentia residuar fo l l activit rotationan yi l crops. Atrazine has a residual life at recommended use rates of approximately 1 year [7]. Some crops, including oats (Avenu saliva L.) and other small grains are restricted from rotations which include atrazine becaus f thieo s long residual activity. Vail et al. [25] conducted studies designed to measure the dissipation of atrazine under field condition t Wessa t Lafayette , beginninIN , sprine th n f gi go 1991. Treatments include atrazinatrazinF E dC S d ean e formulations 149A, 149B, 149d an C (TABL applie) EI d preemergenc coro et ratet na 1.12,2.24f so 3.3d ,an 6g k ha'1 on June 10, 1991. Soil samples were taken before treatment and periodically throughou followine e seasoth th t d nan g spring soie lTh .core s were divided int5 4 oo t segment0 3 d an , s correspondin30 o t 5 1 , 15 o t depth7.5o g o t t 5 0 ,7. f so cm and were analyzed for atrazine by methanol extraction and gas chromatography. Oats were planted on the site on April 7, 1992 as a bioassay crop. The injury on oats was evaluated on May 25, 1992 and the oat crop was killed by treatment wit postemergencha e herbicide. Soybeans were no-till planted into the spray killed oats on May 27,1992 and were grown to maturity. Soybeans are a common rotational crop in the corn producing areas of the midwestern United sensitive Statear d san atrazino et e residual activity. Late corn plantin d treatmenan g n 199i t n combinatioi 1 n witn a h unusuall growiny ydr g season resulte poon di r weed control fro l treatmentmal s as well as reduced crop growth. The five months following planting had only 50 f norma%o l rainfall. Detectable level atrazinf so e were presen soin i t l samples fro l thremal e treatmen l formulationt al rate d an s s e (FigurmajoritTh ) f e6 o y E e soi S e th atrazin l e th f profilo th d m e uppec ean eth detecte 5 n 7. i r s wa d formulations 149 d 149A significantlan d Bha y greater level f residuaso l atrazine o differencetha N e smalle th . n CF n residua i rse 149th d Cl an levels were detected at depths below 7.5 cm except for the CF at the 30 to 45 cm level in the 2.24 kg ha~l rate. The pattern of greater residues in the upper portions of the profil expectes i e d give demonstratee nth d abilitformulationE S f yo reduco t s e leaching. Further, the effect of increasing granule size of atrazine release in SE formulation seee b thesn ni n ca se datalargee Th . r granules ten havo dt e eth highest residuals.

164 12(1»-i

HHW-

6(1(1

411(1-

2(1(1- NS «B-f*

12011

IUOII - •:-j P i* '5 i:!? a\ 10 — (•* :ji: b i| •»— (.110- r ^ B j-aS: i» s/ -Jlll) - j b_ i* ft* ' •I .4 NS ^ •? TjTn NS NS U 5î^ B:^f-t ^m_p5i — —, Ft 1S.- 6 2 7. 1S. 30.5- 26 7. . il 30. 45.- 5 7 SOIL DEPTH (cm)

Figure 6. Atrazine residues in soil approximately 10 months after applicatio commerciaf no starcd an l h encapsulated atrazine , 1.1A 2. kg/ha; B, 2.24 kg/ha; C, 3.36 kg/ha. 149A passed 1.4 mm, retained by 0.85 mm, 149B passe d, retaine 0.8mm 5 0.42, y 149db C5mm passed 0.42 'screenm 5m .

Tht bioassaoa e injure th yo yt cro p (Figur confirme) e7 soie dth l residual levels. All of the formulations resulted in carryover atrazine that injured the oats. The SE formulations had significantly more oat injury than the CF at the higher rates. At the 1.12 kg ha~l rate the larger granule SE caused more injury that the smallest sized SE or the CF. The substantial oat injury seen in this trial was not repeated in subsequent trials with more normal precipitation patterns. Oats are very sensitive to atrazine and residues that injure oats may be well below the level that would be problematic for labeled rotational crops. This is demonstrated by the soybean crop grown following the oats in this trial. No visible injury could be detected in the soybean crop and the yields (TABLE were not statistically different for any of the treatments.

165 Formulation

ce D -D

< o

336 2.24 1.12 ATRAZ1NE RAT ha-1g E(k )

Figur . eInjur7 oato yt s planted approximatel month0 y1 s after treatment with three rate f atrazinso s commerciaea l liqui r starcdo h encapsulated formulations. 149A passed 1.4 mm, retained by 0.85 mm, 149B passed 0.85 mm, retaine 0.42, 149y db 5Cmm passed 0.42 screenm 5m .

TABLE III. 1992 SOYBEAN YIELD ATRAZINN SI E RESIDUE TRIALS WEST A T LAFAYETTE . ATRAZININ , E TREATMENTS APPLIED JUNE 10,1991. Treatment Formulation Rate Yield

1 - k_ _ e _ ha' _ Atrazine 149A 3.36 3017 Atrazine 149B 3.36 3106 Atrazine 149C 3.36 3329 Atrazine 4L 3.36 3214

Atrazine 149A 2.24 3314 Atrazine 149B 2.24 3270 Atrazine 149C 2.24 3436 Atrazine 4L 2.24 3210

Atrazine 149A 1.12 3191 Atrazine 149B 1.12 3234 Atrazine 149C 1.12 3137 Atrazine 4L 1.12 3207

Untreated _r___ 0.0 3138 N.S.

5 Summary e trialth f so discusse l Al literature f thithosn di o th sl n ei al pape d e an rwer e performed using experimenta formulationsE S l . These formulations wert eno optimized or altered to improve their efficacy or their ability to control release or other factors. It is logical to assume that with a focused effort aimed at improving the SE formulation for a particular herbicide or cropping system the

166 advantages demonstrated with the experimental formulations could be even formulationgreaterE S e Th . s controlled leachin severan gi l different soild san under many conditions. They reduced the amount of atrazine moving off site in surface wate shorn i r t term, high rainfall situation wels sa oves a entire l th r e growing season. While SE formulations may result in elevated residue in the upper soil profil t doe i et appea no s preseno t r insurmountabln a t e problem. Something as simple as reducing the granule size resulted in substantially reduced residues. Normal rotational crops wer t adverseleno y affectee th y b d residues of high rates even following a dry growing season that would accentuate the residue problem.

References

[I] ROMANDER, L., Groundwater Logic, Agrichem. Age 32 (1988) 26-27, 19.

[2] LENARD, R.L., KINSEL, W.G., Evaluating groundwater contamination potential from herbicide use. Weed Techno (19882 l ) 207-216.

[3] Weed Science Societ f Americayo . Position Statemen Waten o t r Quality. Weed Techno (19915 l ) 921.

[4] BUSHER, H.R., Atrazin othed ean r S_-triazine herbicide rain i laken n si ni d an s Switzerland. Environ. Sei. Tech (19904 2 . ) 1049-1058.

[5] EPA, Another Look: National surve f pesticideyo drinkinn si g water wells. Phase II report (1992) EPA 579/09-91-020. U.S. Environ. Protection Agency, Washington, DC

[6] GOOLSBY, D.A., et al., Geographic and Temporal Distribution of Herbicides in the Upper Midwestern United States, 1989-1990. U.S. Geological Survey Water- Resources Investigations Report (1990) 91-4034, p. 183-188.

[7] Herbicide Handbook, Sixth Edition, Weed Science Societ f Americayo , Champaign, IL (1989).

[8] WIENHOLD, B.J., GISH, T.J., Effect f formulatioo s d tillagan n e practicn o e volatilizatio f atrazinno alachlord ean . EnvironJ . . Qual. (1993 Pressn )I .

[9] SCHREIBER, M.M., et al., Controlled release herbicides. Monogr. Ser. Weed Sei. (19874 . Soc)Am . 177-191.

[10] SHASHA, B.S. t al.,e , Starch-encapsulated pesticide slor sfo w release . PolymJ . . Sei., Polym. Letters Editio (19764 n1 ) 417-420.

[II] SHASHA, B.S., et al., Encapsulation of pesticides in a starch-calcium adduct. J. Polym. Sei., Polym. Chem. Ed. 19 (1981) 1891-1899.

[12] TRIMNELL, D., et al., Pesticide encapsulation using a starch-borate complex as wall material . ApplieJ . d Polym. Sei (19827 .2 ) 3919-3928.

[13] WING, R.E., et al., Factors affecting release of butylate from calcium in ion- modified starch-borate matrices. J. Controlled Release 5.(1987) 79-85.

167 [14] WING, R.E., et al., Effectiveness of jet-cooked pearl starch as a controlled release matrix. Starch/Starke 39 (1987) 4Z2-425. [15] CARR, M.E., et al., Encapsulation of atrazine within a starch matrix by extrusion processing. Cereal Chem. 68 (1991) 262-266.

[16] SCHREIBER, M.M., et al., Starch-encapsulated atrazine: Efficacy and transport. J. Environ. Qual. (1993) In Press.

[17] SCHREIBER, M.M., WHITE, M.D., Granule structure and rate of release with starch-encapsulated thiocarbamates. Weed Sei. 28 (1980) 685-690.

[18] GISH, T.J., et al., Transport comparison of technical grade and starch encapsulated atrazine. Trans. ASAE 34: (1991) 1738-1744.

[19] BOYDSTON, R.A., Controlled release starch granular formulations reduce herbicide leaching in soil columns. Weed Technol. 6 (1992) 317-321.

[20] MILLS, M.S., et al., Dissipation of starch-encapsulated herbicides: A field comparison with powdered herbicides. U.S. Geological Survey, Water Resources Investigation (1992) 91-403 203-209p 4p .

[21] STEELE, R.G.D., TORRIE, J.H., Principles and Procedures of Statistics: A Biometrical Approach. Second edition. McGraw-Hill Book Company Yorw Ne k, (1980).

[22] WAUCHOPE, R.D., et al., Mobility and efficacy of controlled-release formulations of atrazine. WSSA Abst. 30 (1990) 72.

[23] FLEMING, G.L., et al., Release and persistence comparison of starch-encapsulated atrazine. Weed Tech (19926 . ) 297-302.

[24] BRADFORD, J.M., HUANG , InterrilC , l soil erosio s affectena tillagy b d d an e residue cover. Soil Tillage Res. (1993 Pressn I ) .

[25] VAIL, G.D. t al.e , , Atrazine residues from commercia d starcan l h encapsulated formulations. Proc. North Central Weed Sei. Soc. 47 (1992) 19. Chicago.

168 EFFICAC STARCF YO H ENCAPSULATED FORMULATIONS OF HERBICIDES

M.M. SCHREIBER, M.V. HICKMAN Insect and Weed Control Research, Agricultural Research Service, United States Department of Agriculture, West Lafayette, Indiana, United States of America

Abstract

Regardles environmentallw sho y saf cosr eo t effectiv formulatiow ne ea y nma be, to be considered for adaptation, its efficacy must be equal to commercial formulation (CF) presentl markete th n yo . Herbicide efficac measures yi e th y db contro weedf o l s exhibite resultane th d dan t affect cron o s p yield followine Th . g herbicides have been formulated as starch-encapsulated (SE) granules and have had extensive laboratory, greenhouse field ,an d testin severar gfo l years: EPTC, butylate, trifluralin, alachlor, metolachlor, and atrazine. These herbicides represent a wide range of water solubility and volatility characteristics. They have been formulated as single component granules and two or three component granules, the latter acknowledging that most preemergence herbicide usee sar combinations n di E S . formulations of EPTC, butylate, and trifluralin have been shown to give excellent weed control under delayed incorporatio e cas th f triflurali o en i d o an nn n incorporation. Extensive field trials with SE formulations of atrazine with metolachlo alachlor ro r wit withoud han t dicamba gave excellent contro wida f lo e range of weed species equal to that obtained with CF. In most cases, crop yields from plots treated wit formulationE hS s were equa better o o lt r than those obtained with CF under conventional and conservation tillage systems on light and heavy soils.

1. Introduction

Regardless of how cost effective or how environmentally safe new formulation technologyw basee ne b n y do ma sfinae effectivth , w l tesho s i t e these formulation rean i e l worlar s d situations case th f herbicides eo n I . tese s th ,i t whether weed contro subsequend an l t crop yields obtaine goos a r bettee do d ar r than commercial formulations presently on the market. This concept is somewhat ironic since there neve bees rha ncleaa r consensu whaf so t constitutes commercial control. Le loos hypotheticaa u t t ka l exampl illustrato et e this assums pointu t Le .e that a herbicide formulation can consistently control weeds at a 90% level ("commercial higa control" s h ha potentia t )bu environmentar fo l l contamination, 80% chance. For sake of argument, a new starch-encapsulated (SE) formulation of same th e herbicid consistentln eca y control weed 80-85a t sa % reducn levelca t e,bu potentiae th l environmental contamination leve 20%o t l this I . s trade-off worth pursuing whatever Fo ? r reason agriculturae ,th l chemical industr insistes yha d that formulatioanw yne n must show efficacy equivalen thao t presenf o t t commercial

169 formulations (CF) without consideratio trade-offsf no regar n efficace I . th E S o d t f yo formulations have w , e accepte challenge dth standarde th f eo industrf o s y within sound statistical limits. In term herbicidaf so purpose th lr actio fo thif eo d sn an paper , efficacy means the capacity to produce weed control regardless of the loss factors that contribute to or detract from this capacity, such as by reducing losses due to leaching, volatility, and photodecomposition. The resultant effects of the weed control is the maximization of yields under the environmental conditions that prevail. Under field conditions, efficacy mus e demonstrateb t d ove ranga r f weeo e d species, soil textures, tillage systems seasond an , s wit withoud han t moistur temperaturd ean e stresses. The data presented herewith fall well within these criteria. TABLES I and II liscommone th t , trade chemicad an , l nam f herbicideeo d commoan s d an n scientific nam specief eo s mentione thin di s paper, respectively. Most of the efficacy studies with herbicides progressed from the laboratory to the greenhouse and finally to the field. Close collaboration was maintained with the research chemists during these periods so that formulations could be designed to fulfill the criteria desired. It cannot be stressed enough the importance of the teamwork between the research chemists and weed scientists in the ultimate design performancd an formulationE S f eo s that have been developed formulationE S l Al . s reported in this paper are considered experimental.

TABLE I. COMMON, TRADE, AND CHEMICAL NAMES OF HERBICIDES MENTIONED IN THIS PAPER

Common Trade Chemical name name name

alachlor LASSO 2-chloro-N-(2,6- thylphende i y 1)- JV (methoxymethyl) acetamide atrazine "many" 6-chloro-N-ethyl-A/'-

Premixes

BICEP atrazine+metolachlor QBA-GEIGY BULLET atrazine+alachlor Monsanto SUTA2INE atrazine+butylate ICI Ameriais

170 TABL . ECOMMOH SCIENTIFID NAN C NAME WEEF CROSO D DAN P SPECIES MENTIONED IN THIS PAPER

Common name Scientific name

dodder Cuscute campastris l~ giant foxtail Setaria faberi Hexnn. ivyleaf morninggloty Ipomoea hederacea (LJ Jacq. jimsonweed Datura stramonium. L common lambsquarter Chenopodium album. L kochia Kochia scoparia (L.) Schiad. vcnice mallow Hibiscus trionum L. Pennsylvania smartweed Polygonum pensylvanicum L. redroot pigweed Amaranthus retroflexus L. common ragweed Ambrosia artemisiifolia L. common coddebur Xanthium strumarium. L longspine sandbur Cenchrus longispimts (Hack.) Fern, velvetleaf Abutilon theophrasti medic. maysa . Ze L m co soybean Glycine (L.x ) ma Merr .

. 2 Efficacy f Volatileo Herbicides initiae Th l efficacy studies conducted wit formulationE hS s were concerned with herbicides that required immediate incorporation into the soil because of their rapi higo dt loshe volatilitsdu y and/or photochemical decompositiono tw e Th . main groups of herbicides were the thiocarbamates (now called carbamothioates) dinitroanilinee th d an s represente EPTy db butylatr Co trifluralind ean , respectively. researce Th h questions addressed with these groups were coul) a : formulation E dS s allow one to delay incorporation for some significant period of time and b) indeed, could the need for incorporation be eliminated, both without reducing the efficacy of the herbicides.

2.1 Chlorpropham

Although our initial work was with EPTC, one SE formulation of chlorpropham was prepared for use on dodder seedlings, a species susceptible to this herbicide. The problem was that chlorpropham has a short persistence in soil and dodder because of hard seeds germinates over time. Work by Dawson [1] indicated that the SE formulation of chlorpropham was superior to the CF and equal to a microencapsulated formulation for the control of dodder up to 5 weeks.

2.2 EPTC and Butylate

The development of SE formulations of EPTC and butylate followed the progression of SE technology, from the starch xanthates up to the jet cooking and extrusion processes. Fro originae mth l xanthate formulation were sw e abl shoo et w initial release was adequate for weed control and slow enough for residual activity [23,4]. Thoug t directlhno y measured, undoubtedls thiswa controo t e f ydu o l volatility losses.

171 Field studies using SE formulations of EPIC and butylate made with the borate process continue shoo dt increasee wth d efficac f thesyo e formulationsn I . 1989 formulationE S , EPTAMf so , ERADICANE, ERADICANE EXTRA, SUTAd Nan SUTAN PLUS gave excellent weed control and significant yield differences compared to their emulsifiable concentrate (EC) formulations (TABLES IÏÏ and IV). These treatments were applied when the air temperature was 32°C and the soil rainfalm c t fro surfac5 we mday3 a l s sewa befor e treatment. Even thouge hth

TABLE HL EFFECTS OF COMMERCIAL (EC) AND STARCH ENCAPSULATED (SE) FORMULATION RATED S EFTAN F SCORO N CO N YIELDS, 1989.

Herbicide (formulation) Rate Yield

-kg ha"-

EPTAM (EC) 1.68 568c 8ab EPTAM (EC) 336 4728 abc EPTAM (SE) 1.68 9237 def EPTAM (SE) 3.36 9664 ef

ERADICANE (EC) 1.68 4365 ab ERADICANE (EC) 336 5004 abc ERADICANE (SE) 1.68 9607 ef ERADICANE (SE) 336 10272 f

ERADICANE EXTRA (EC) 1.68 3694 a ERADICANE EXTRA (EC) 336 4189 ab ERADICANE EXTRA (SE) 1.68 8579 def ERADICANE EXTRA (SE) 336 921f 8de

Untreated check 3023 a Handweeded check — 9858 ef

a Means within yield column followed by the same letter do not differ significantly base Fisher'n do s Protecte (0.05D dLS )

TABL . EEFFECTIV COMMERCIAF SO L (ECSTARCD )AN H ENCAPSULATED (SE) FORMULATION RATED S BUTYLATAN F SO N EO CORN YIELDS, 1989.

Herbicide (formulation) Rate Yield a

a h ————— g —k ,— , -1

SUTAN (EC) 1.68 2916 a SUTAN (EC) 336 4490 ab SUTAN (SE) 1.68 6578 bed SUTAN (SE) 336 967f 6e

SUTAN + (EC) 1.68 543c 7ab SUTAN + (EC) 336 518c 6ab SUTAN •*• (SE) 1.68 7306 cde SUTAN + (SE) 336 866f 7de

Untreated check 3023 a Handweeded check — 9858 ef

Means within yield column followed by the same letter do not differ a significantly base Fisher'don s Protecte (0.05)dLSD .

172 formulations were incorporated within 6 hours after treatment, the EC formulations lost almost all of their activity regardless of rates used. The yields of corn from plots treated wit formulatioE hS n showe rato dn e differences from 3.3 1.6ha'lo 6g t 8 k . These data wer surprisint eno g base laboratora n do y study conducte 198dn i 7 to compare volatility losses of SE formulations of EPTC using different processes (FIGUR borate Th . e Eproces1) s produce formulatioda superior nfa thas twa r than t morBu e. importantlyEC e th jet-cookine th , g process (the fore-runnee th f o r extrusion process) exhibited significantly much better contro volatilitf o l y loss than borate th e formulatio hour4 n14 t soiseveo t sittin lwe p nsurfaca u n go e [5].

BLS D0.6= 4 EXPOSURE TIME = 100) • OHR R H ®4 2 D 48 HR El 96 HR 144 HR

CHECK EC BORATE JET COOK EPTC FORMULATION

t shooFigurOa t. ebioassay1 EPTf so C starch-encapsulated formulations exposed on a wet soil surface up to 144 hours.

A field test in 1992, originally designed to study efficacy of SE extruded formulations of SUTAZINE + and ERADICANE with four times of delayed incorporation s unsuccessfuwa , l becaus f laceo f rainfalko l following application. The herbicides were applie ver a incorporationsoi d y o dt ydr an l s were mad, 8 , e0 24, and 48 hours after application but the first rainfall of 1.1 cm did not occur until 10 days after application. Whe weee nth d control datpooles awa d over incorporation times significano n , t difference ween i s d control were foun r formulationfo d s (TABLE V). Based on all our previous studies and the preliminary data from similar studies in 1993, we can only speculate that the SE formulations would have produced significantly better weed control than the commercial granules and EC formulations when incorporation was delayed under moist soil conditions. formulationE S e Th EPTf so butylatd Can e have also been show overcomo nt e accelerated thiocarbamate degradatio soiln i proble,a m reporte severan di l arean si the United States. It has been reported that the slow release of the active thiocarbamates from the SE formulations reduces the rapid buildup of soil microbes capable of degrading the thiocarbamates [6]. The SE formulations offered significantly better control of this phenomenon than the EC or microencapsulated formulations tested. Many volatile herbicide amenable controb o t y E S sma lo et volatilit y losses which should increase their efficacy and safety. Recently a SE formulation of clomazone was shown to reduce volatility compared to its EC formulation [7J.

173 TABLE V. EmCACY OF SUTAZINE-i-AND ERADICANE FORMULATIONS INCORPORATED IN CONVENTIONAL TILLED CORN, 1992.

Control by speciesa

Herbicide Rate Giant Velvet- Jimson- (formulation) foxtail leaf weed

kgha-1 %,..,.

Sutazin (CPC )24 e+ 63 96 76 82 Sutazine + 24 SE 6.3 96 60 62 Sutazine + EC 6.3 96 76 92

Eradican (CFG 5 )e2 4.7 99 96 92 Eradicane 27 SE 4.7 82 93 87 EradicanE S 5 e1 4.7 99 97 90 Eradicane EC 4.7 99 97 86

signifcano N a t differences withi herbicidy nan weed ean d species

3 2. Trifluralin

Trifluralin, one of the most widely used dinitroaniline herbicides, is volatile, subject to rapid decomposition by ultraviolet irradiation, and very insoluble in water. Because of these physical characteristics it was a good candidate herbicide formulationE foS r r earlOu y. work thosd [8,9an f other]e o s [10] showed that application of trifluralin as a SE formulation had a lag phase before released activity was high enough to produce weed control equal to that of its EC. In both conventional tillage and no-till systems this meant early application long before soybean planting. This would make it very adaptable for custom operators and to the farmer by reducing the workload in the spring. Dat n TABLi a I indicatEV e that early applications were necessarn i y conventional tillage whether surface applie incorporatedr do applicatios A . n times became earlier, it became apparent that in the case of SE trifluralin incorporation

TABL L EFFECTEV TIMF S APPLICATIOO F EO PLACEMEND NAN F TO COMMERCIAL (CF) AND STARCH-ENCAPSULATED (SE) FORMULATIONS OF TRIFLURALIN ON SOYBEAN YIELDS (1982). (TRIFLURALIN RATE 1.7 KG HA'1)

Application Yields ab

Time Placement SE EC Untreated

-1

12-7-81 Incorporated 295A 0a 2460 bB 2010 aC 12-7-81 Surface 2990 aA 2160 bB 1850 aC 4-19-82 Surface 2850 aA 226B 0b 2320 aB 5-10-82 Surface 296A 0a 2160 bB 1900 aB 6-21-82 c Incorporated 2610 aB 3120 aA 1960 aC

a Values within column followed by the same lower case letter are not significantly different according to Duncan's Multiple Range Test (O.OS). b Values within a row followed by the same capital letter are not significantly different accordin Duncan'o gt s Multiple Range Test (0.05). c Date of planting

174 necessarye b woulo s t f i no dd y offe an ,excellen ma n a r t herbicid r soybeanefo s grown no-till. Thiconfirmes swa severan di l greenhous field ean d studies. Datn ai TABLE Vu summarizes the potential for trifluralin use in no-till when formulated as a SE formulation [9]. Although pendimethyli prepares formulationE nwa S a s da extensivo n , e field trials were conducted with this herbicide.

TABL L EFFECEVI COMMERCIAF TO STARCD AN D HL(C ENCAPSULATED (SE) FORMULATION TRIFLURALIF SO TIMD NEAN OF APPLICATION ON NO-TILL SOYBEAN YIELDS 1985. (TRIFLURALIN RATE 1.4 KG HA"1.)

Yiel> d* a Trifluralin ____Applicaffon time formulation 3-29-85 5-7-85

-kg ha"1-

CF 2165 bA 3107 aB S EB 326 A 2a 332A 9a

Untreated check 238 A 1b 234A 0b a Means withi yielna d column followe samy db e lower caset letteno e rar significantly different according to Duncan's Multiple Range Test (CL05) k Means with a row followed by same capital letter are not significantly different according to Duncan's Multiple Range Test (0X5).

3. Efficacy of Herbicides Subject to Leaching

3.1 Chloroacetamides

The two most widely used chloracetamides applied to both corn and soybeans for grass control are alachlor and metolachlor. Since both compounds are water solubl havd ean e been reporte groundwatern di , they appeare leno dt d themselves benefit o . t WheSE f no s either herbicide formulatioE s applieS wa sa s da n ni combination with a CF of atrazine, excellent grass control was obtained usually equa r betteo l r tha availableF nC almosn I . t every case, yield f coro s n were significantly higher from plots treated with the SE formulation than CF. This was true whether the borate or jet-cooking process was used for SE formulation. Results were simila conventionaln ri , chise no-tild lan l tillage systems.

3.2 Triazine

Although atrazine has low solubility in water, it is one of the most commonly detected herbicide groundwatern si . Sinc er laborator mosou f o t field yan d data indicate significant reduction in leaching when atrazine is formulated in starch compared to CF, it was important to evaluate efficacy of SE formulations of atrazine under field conditions. The critical question is, since atrazine has low solubility in releaswatere th e e,ar rate granulesE froS e mth s high enoug controho t l weeds.

175 Many studies were conducted from 1988 through 1992 culminating with demonstration studies in the midwest in 1990 to 1992. Ten sites in seven states were selecte evaluato dt efficace atrazineth E S f yo e wit formulationE hS alachlof so r ro metolachlor. The sites selected represented different soil types (sands to silty clay loams), a range of weed species, diverse weather conditions, and variations of tillage systems (conventional moldboard-plow, chisel plow, and no-till). Atrazine, alachlor, and metolachlor are herbicides that are rarely used alone in preemergence applications. Many tank mixe approvee sar thesr dfo e herbicides. markee Alsth aboun e oo t ar premixe 3 t1 s containing atrazine containin,4 g alachlor, containin5 d an g metolachlor t logicallI . y followed that mixture f theso s E eS formulated herbicides be evaluated together with their CF. SE using the extrusion process easily allows for more than one herbicide to be encapsulated in the same granule at the same time. Several two-way and three-way SE formulations containing atrazine or metribuzin were prepared and field tested.

. 4 Efficacy f Two-Wayo Mixtures of Separate FormulationsE S Two-way mixtures imply two herbicides applied but each formulated separately. The CF were tank mixed (unless otherwise noted) and each SE formulation was applied separately. TABLE VTfl shows the efficacy data obtained from the midwest demonstration studies in 1990 and 1991 from applications of atrazine applied with alachlor or metolachlor. The data show the control obtained on individual weed species listing the number of sites at which the weed species was presentmeae rangeth d nd an f controan o , l over sites. Contro f giano l t foxtaia ( l common weedy grass in the midwest) and the small seeded broadleaf weeds such as lambsquarte smartweed ran equas d wa formulations E l S betwee d an F nC . Control of velvetlea variabls wa l sitef al t s ea range baseth f controln eo do s recordedo T . some extent relatee depte b , y thith germinatiof o dht so ma n note velvetlear dfo d fan concentratioe th f atrazinno t thosea e depths. Contro f commoo l n ragweed dan cocklebur was usually better with the CF than with SE formulations. The control of four species only found at one site in 1991 is shown in TABLE IX. Statistically, control with the SE formulations was equal or better than the CF for all species. It is interesting to note that the broadleaf species (kochia, Venice mallow and jimsonweed) are all large seeded species. In the demonstration sites in 1990 and 1991, the granule size of all SE formulations was 1.4 to 0.85 mm. In spite of the variation of weed control of some weed species, the overall yields of corn obtained in 1990 and 1991 at sites containing sandy soils (TABLE X) or heavier soils (TABL wer) EXI e equa greater lo r wit formulatioE hS comparee nus o dt untreateyielde e th Th n s. o CF d e plotth s giv indication ea weef no d pressurd ean moisture stress during the growing season. A stud two-waf yo y mixture metribuzif so n with alachlo metolachlor ro r both as CF and SE formulations in no-till soybeans was conducted in 1991. The metolachlo formulationE rS ssize o wertw sf e (1.4-0.8o 0.85-0.4 d an m 5m 3 mm)e Th . data show TABLn ni indicatI EXI e tha mosn i t t case smallee sth granule th r e eth better the weed control obtained with little variation in soybean yield.

5. Efficacy of Two or Three Herbicides in Same SE Granule As stated earlier, the extrusion process easily allows for two or more herbicides to be entrapped in the same starch granule. The following combinations were prepare field dan d teste r efficiencydfo : two-way ) atrazine-alachlora , ) b ,

176 TABL . EFFICACEVm COMMERCIAF YO L STARC (CFD )AN H ENCAPSULATED (SE) FORMULATION ATRAZBSfF SO E (AT), ALACHLOR METOLACHLO(AL)D AN , R (ME 199199D N )I 10AN FIELD STUDIES.

1990 1991

Herbicide Sites Mean Range Sites Mean Range formulation weeds control of weed control of present control present control

No. -% ———— No. c GiajvLImtail

(CF(CFT L A A ))+ 8 93 64-100 9 94 59-100 AT (SE) + AL (SE) 8 89 78-99 9 90 61-100 AT (CF) + ME (CF) 8 94 90-100 9 92 46-100 (SE(SE)E T A M )+ 8 90 77-98 9 79 9-100

Velvetleaf

AT(CF) +AL(CF) 9 81 0-83 8 71 0-100 (SE)T A (SEL +A ) 9 59 1-88 8 56 0-95 AT(CF) + ME (CF) 9 82 0-100 8 78 0-100 AT(SE) + ME (SE) 9 65 0-89 8 50 0-96

Lambsquarter

AT(CF AL(CF)+ ) 5 99 96-100 4 99 99-100 AT (SE) -c AL (SE) 5 99 99-100 4 99 97-100 (CF(CFE T A M ) )+ 5 99 99-100 4 99 99-100 AT(SE (SEE M ) )+ 5 96 95-100 4 96 87-100

Smartweed

(CFT L A A )+ 3 97 91-100 2 96 93-100 AT (SE) + AL (SE) 3 93 84-100 2 95 91-100 AT (CF) + ME (CF) 3 95 86-100 2 97 95-100 (SET E A M )+ 3 92 85-100 2 96 92-100

Common Ragwppd

AT(CF) + AL(CF) 2 93 86-100 1 89 . (SE (SEL T A A ))+ 2 82 73-90 1 78 - (CF(CFE T A M )+ 2 94 83-99 1 50 - (SE(SEE T A M ) )+ 2 60 84-75 1 78 - Cocklebur

AT (CF) + AL (CF) 2 92 85-99 1 92 . AT (SE) + AL (SE) 2 72 69-74 1 58 - AT (CF) + ME (CF) 2 86 74-98 1 94 - (SE E (SET M A ))+ 2 73 56-89 1 60 -

TABLE DC. EFFICACY OF COMMERCIAL (CF) AND STARCH ENCAPSULATED (SE) FORMULATIONS OF ATRAZINE (AT), ALACHLOR METOLACHLO(ADD ,AN R (ME SINGLT )A E SITES IN 1991.

Control bvSnecies Herbicide Sandbur Kochia Venice Jimson- formulation mallow weed .

AT(CF) + AL(CF) 73 100 97 82 AT(SE) -f AL(SE) 64 100 88 65 AT(CF) + ME(CF) 82 0 98 82 AT(SE) + ME(SE) 91 100 88 82

177 TABL . EX EFFEC COMMERCIAF TO L (CFSTARCD )AN H ENCAPSULATED (SE) FORMULATIONS OF ATRAZINE

1990 Herbicide formulation Kilboume Wanatah Westport IL IN MN -1 ———— Kg na —

AT(CF) + AL(CF) 4580 9660 5080 AT(SE) + AL(SE) 2760 9660 5080 AT(CF> + ME(CF) 3830 10280 4890 AT(SE) + ME(SE) 3260 9280 5020 Untreated check 1570 9530 3890 LSD(O.OS) 815 1190 380

1991 Kilboume Wanatah Westport Madrid IL IN MN NE ————- — k,a gh ^ -1 AT(CF) + AL(CF) 6710 5520 7460 11100 AT(SE) •*- AL(SE) 4890 6150 6960 11040 AT(CF ME(CF)+ ) 5330 6080 7650 6150 AT(SE ME(SE)+ ) 7530 5580 7150 12100 Untreated check 3640 5080 0 10790 LSD(O.OS) 3260 1000 750 1130

TABL L EFFECEX COMMERCIAF TO L (CFSTARCD )AN H ENCAPSULATED (SE) FORMULATION ATRAZINF SO E (AT), ALACHLOR (AL)METOLACHLOD ,AN R (ME CORN )O N YIELD SILTN SO Y LOA CLAD MAN Y LOAM SOIL 199D N SI 0AN 1991.

Herbicide W. Laf. Urbana Ames Rosemount (formulation) IN n. IA MN

122Q ————— ha"—g k 1

AT(CF AL(CF)+ ) 9280 8340 6520 7900 AT(SE) -i- AL(SE) 10600 8460 6840 7530 AT(CF) t- ME(CF) 10850 8400 7460 7960 AT(SE)-(-ME{SE) 9780 6650 5330 7340 Untreated check 7210 6270 4330 250 (0.05D LS ) 1230 1630 NS 440

1991 AT(CF) + AL(CF) 7710 3760 4580 9970 AT(SE) t- AL(SE) 6400 3450 3890 9340 AT(CF) + ME(CF) 7020 3070 4200 10030 AT(SE) + ME(SE) 6900 3140 4890 9160 Untreated check 3760 2822 4140 5800 LSD (0.05) 940 NS NS 820 atrazine-metolachlor; three-way ) atrazine-alachlor-dicambaa , ) atrazineb , - metolachlor-dicamba. In 1990, a study in conventional tilled corn was conducted using atrazine and alachlor in the same granule. Two rates and two granule sizes were used in this study date Th a. show TABLn ni E XIII indicate that these herbicide same th en si granul more ear e effective whe granule nth smalls ei . However, there wero en significant difference cor n i formulationssE S n e yieldth d s an betwee .F C e nth

178 TABLE XII. EFFICACY OF COMMERCIAL AND STARCH ENCAPSULATED (SE) FORMULATIONS OF ALACHLOR(AL METOLACHLOD )AN R (ME) WITH METRIBUZIN NO-TIL (METN I ) L SOYBEANS, 1991

Contro Speciey lb sa

Treatments Granule Giant Velvet- Jimson- Soybean (formulation) sizes Rates foxtail leaf weed yield

1 mn kg ha' et kg ha"1

ME(EC)+MET(EC) 2.24+0.56 84 ab 53b 82 a 2489 ME(EC)+MET(EC> 2.80+0.56 94 ab 68 b 60 a 2871 ME(SE) +MET(SE> 1.4-0.85+0.85-0.43 2.24+0.56 80 ab 79 ab 95 a 2662 ME(SE)+MET(SE) 1.4-0.85+0.85-0.43 2.80+0.56 98 ab 80 ab 97a 2484 ME(SE)-t-MET(SE) 0.85-0.43+0.85-0.43 2.24+0.56 98 ab 76 ab 68 a 2507 ME(SE)+MET(SE> 0.85-0.43+0.85-0.43 2.80+0.56 100 a 85 ab 92 a 2613

AL(SE)+MET(SE> 1.4-0.85+0.85-0.43 2.24+0.56 90 ab 80 ab 55a 2441 AL(SE)+MET(SE) 1.4-0.85+0.85-0.43 2.80+0.56 57b 72b 94 a 2542 AL(SE)+MET(SE) 0.85-0.43+0.85-0.43 2.24+0.56 82 ab Tlb 69 a 2698 AL(SE)+MET(SE) 0.85-0.43+0.85-0.43 2.80+0.56 94 ab 95 a 68 a 2613 LASSO n+MET(SE) 0.85-0.43 2.80+0.56 100 a 83 ab 85 a 2642

Untreated check 1978 LSD (0.05) 290

Mean* s within colum weea f no d species followe same th e y diffet db lette no o r rd significantly base Fisher'n do s Protecte (0.05D dLS )

TABLE Xm. EFFECT OF GRANULE SIZE AND RATE OF APPLICATION ON WEED CONTROL AND CORN YIELD FROM STARCH ENCAPSULATED ATRAZINE (AT) AND ALACHLOR IN SAME GRANULE UNDER CONVENTIONAL TILLED CORN, 1990.

Granule Rate Giant Velvet- Redroot Lambs- Jimson- Yield b size AT-AL foxtail leaf pigweed quarter weed

11UI kg ha-1 „„or. kg ha-1

Commercial* 1.68+2.80 97 56 97 100 82 10000 c Commercial 2J4+3.70 99 84 100 100 86 9730c

1.4-0.85 1.68-2.80 75 12 95 100 27 9580 be 1.4-0.85 2.24-3.70 88 30 99 100 48 8800 ab

0.85-0.43 1.68-2.80 92 16 94 100 59 9570 be 0.85-0.43 2.24-3.70 97 45 100 100 80 9210 be

Untreated check 8150 a LSD (0.05) 17 23 9 NS 40

a Commercial formulations of atrazine and alachlor was premixed BULLET. b Means within yield column followed by same letter do not differ significantly based on Fisher's Protected LSD Test

Control of velvetleaf and jimsonweed, two large seeded weeds, was variable as noted in the extensive demonstration studies reported earlier. Atrazine is a weak herbicide for velvetleaf control even in the CF available. When used in a SE formulation, the rate of release appears more borderline. Could additioe th excellenn a f no t broadleaf control herbicide suc dicambs ha a brine gth

179 . DicambCF controe premithaa th o t s f ai tp o lu abou n xi commercia2 t1 l herbicides markete th n 1992n o I .investigatee w w , no efficace dth two-waf yo y combinations of atrazin alachlod ean formulationindividuas E ra S combinea d s an a F d C l an sd mixture in the same granule. These formulations were compared to three-way mixtures of atrazine, alachlor, and dicamba each as a CF tank mix and all in the sam granuleE eS . This stud conductes ywa conventionan di l tilled corn usine gth larger SE granules, 1.4 to 0.85 mm. The results of this study are shown in TABLE XTV and indicate that the addition of dicamba, whether in the granule or in the tank mix, increases the efficacy of contro velvetleaf lo ivylead an f f morningglory formulatioE efficacS e e Th .th f yo n was equal to that of the CF and so were the corn yields.

TABLE XIV. EFFECT COMMERCIAP SO L {CFSTARCD )AN H ENCAPSULATED (SE) FORMULATIONS OF ATRAZINE (AT), ALACHLOR

Control by species

Giant Velvet- Ivy leaf Lambs- Redroot Herbicide formulations foxtai leaf morning quarter pigweed Yield glory

% . .- kgha-1

AT

AT(CF)+AL(CF)+DC(CF) 98 96 97 98 98 11040 AT-AL-DC(SE) 92 93 93 94 94 11080

Untreated check ...... 9310 LSD (005) 7 14 11 14 5 920

* All commercial formulations were liquid and tank mixed. b Starch encapsulated mixtures wit sig+ ha n were separate grannie formulations, wit- ha sign indicate component herbicides in the same granule.

A similar stud 199n yi s conducte2wa no-tiln di l corn using metolachlor instead of alachlor and two granule sizes. The results shown in TABLE XV indicate similar weed contro notes a l conventionan di l tilled corn with lower corn yields. The extremely low rainfall received in May and June of 1992 affected the weed control and corn yields more in the no-till than in the conventional tilled corn.

6. Relationship of Granule Size and % A.L Encapsulated on Efficacy

formulationE S e Mosth f o t f atrazineo s , alachlor metolachlod an , r usen di these studies contained approximately 10% a.i. Since uniform distribution of granule soie th l n surfacso criticas ei effectivr fo l e control, ther significana s ei t relationship between size of granules and a.i. This relationship changes slightly with different physical characteristics of the herbicides. As an example, in the case of EPTC highla , y volatile herbicide vers i t i y, likely tha granula t e containin% g25 a.i. would give adequate distribution spher e witTh granulh a f . eo mm e4 siz1. f eo influenc eacf eo h granule should give enough overla goor pfo d weed controle th n I .

180 TABLE XV. EFFECTS OF COMMERCIAL (CF) AND STARCH ENCAPSULATED (SE) FORMULATION ATRAZINF SO E

Contro speciev b l s Herbicide Granule Giant- Velvet- Yield formulation size foxtail leaf

kg ha-1

AT(CF)+ME(CF) 96 68 6370 AT(SE)+ME(SE) 1.4-0.85 96 74 6980 AT(SE)+ME(SE) 0.85-0.43 97 70 7280 AT-ME(SE) 1.4-0.85 95 70 7040 AT-ME(SE) 0.85-0.43 96 79 6940

AT(CF)+ME(CF)+DC(CF)- — 98 97 6620 AT-ME-DC(SE) 1.4-0.85 95 90 6840 AT-ME-DC(SE) 0.85-0.43 97 80 5790

Untreated check __ •»• .. 6060 LSD (0.05) NS 143 NS

cas atrazinef eo , whichighlt no s hi y volatil water eo r soluble sphers it , influencf eo e from individual granules is small and a 25% a.i. formulation would require a very fine granule for adequate control. We believe that the a.i. percentages and granule size have sw e worked wit mos e close th h ar to e t desirabl e level eacr sfo h herbicide studied.

7. Summary

Our data indicate that SE formulations can be efficacious based on weed contro crod lan p yield. Volatile herbicide thosd san e subjec photodecompositioo t n more ar e efficien formulationE S markete n th i t n o , w eves no tha nF undenC r adverse weather conditions. Combination of herbicides in the same SE granule can broade spectrue nth weemf o d contro soir lfo l applied herbicide ligh n sheavi d tan y soils and under various tillage systems. Coupled with economics of twin-screw extrusion and the ability to reduce environmental impact, SE offers an excellent new technolog herbicidr yfo e formulations.

References

[1] DAWSON, J.H., U.S. Dept. of Agriculture, Agric. Res. Service, Prosser, Washington, personal communication, 1980.

[2[ SCHREIBER, M.M. t al.e , , Efficac ratd f releasan yeo f EPTeo butylatd Can e from starch encapsulated formulations under greenhouse conditions, Weed Sei. 26(1978) 679-686.

[3] SCHREIBER, M.M., WHITE, M.D., Granule structure and rate of release with starch-encapsulated thiocarbamates, Weed Sei. 28(1980) 685-690.

181 [4] SCHREIBER, M.M., WHITE, M.D., Effect of formulation technology on the release of starch encapsulated EPTC, (Proc. British Crop Protection Conf.- Weeds, 1980) Vol. 15,225-229.

[5] SCHREIBER, M.M., et al., Bioactivity of controlled-release formulations of starch-encapsulated EPTC, J. Controlled Release 7 (1988) 237-242.

[6] HALL, F.R., et al., Preemergence application of herbicides encapsulated in starch for overcoming accelerated thiocarbamate degradation (Proc. Intern. Symp. Control. Rel. Bioact. Mater. 1989) Vol. 16,424-425.

[7] MERVOSH, T.L. t al.e , , Granular formulations reduce clomazone volatility for surfaceu mso , WSSA Abstract (19935 s3 . )81

] WHITE[8 , M.D., SCHREIBER, M.M., Herbicidal activit starch-encapsulatef yo d trifluralin, Weed Sei (19842 .3 ) 387-394.

[9] SCHREIBER, M.M., et al., Efficacy of controlled-release formulations of triflurali no-tiln i l soybeans (Glycine max). Weed (1987Sei5 3 . ) 407-411.

[10] COFFMAN, C.B., et al., Herbicidal activity of controlled release formulations of trifluralin, India Agric. nJ . Sei. 54(1984) 117-122.

182 STARCH ENCAPSULATIO MICROBIAF NO L PESTICIDES FOR SUSTAINED ACTIVITY

M.R. McGUIRE, B.S. SHASHA Plant Polymer Research, United States Departmen Agriculturef o t , Peoria, Illinois, United States of America

Abstract

Many environmental and biological factors act to reduce the insecticidal potentia microbiaf lo l pesticides. Eac thesf ho e factor addressee b n sca d through formulation. Over the past five years, we have investigated formulatio microbiaf no l insecticides within starch matrices. While chemical pesticides have been encapsulate starcn di manr hfo y yearse ,th biological nature of microbial insecticides has prohibited the use of the harsh chemical extremr so H necessarep ensuro yt e gelatinizatioe th f no starch and subsequent entrapment of active ingredient. Pregelatinized starches and flours have facilitated formulation of microbial pesticides. Three distinct type starcf so h formulations have been developed:a sprayable and two granular baits. The sprayable formulation is composed of a pre-mixed combination of pregelatinized cornstarch or pregelatinized flou sucrosd ran etane b tha kn mixetca solidt da s rate 2-6%f so . Bioassay cottof so cabbagr no e leaf tissue treated wit sprayable hth e formulations demonstrated increased residual activity of Bacillus thuringiensis kurstaki (Btk) after simulated (greenhouse actuar )o l (field) rainfall. Similarly, experiments with small field plots of cabbage treated wit sprayable hth e formulations demonstrated efficacy simila thao r t conventiona f to l chemical insecticides typeo tw f se o Th . granular formulation conventionaa e sar l type granule which remains discret period y adherenn edr a througd d san an tt granulhwe e which will slightly dissolv remaid ean n stuc leao kt f tissue upon contact with water. Following drying granule ,th e remains tightly attache leae th f o dt surface. The granular formulations have been tested extensively against European corn bore whorn ri l stage corn. Sunlight screens incorporated within the granules significantly increased residual activity of Btk when granules were expose direco dt t sunlight. Under field conditions, feeding stimulants allowed a decrease in Btk concentration without significant loss of insect control. Work with these formulations is continuing with the addition of viruses and fungi to our research program. Clearly, improvements to formulations of microbial pesticides will enhance the acceptanc reliabilitd ean thesf yo e important pest control tools.

1. INTKODÖCTICN Chemical pesticides have long been used to control insects of agricultural and medical importance. In the U.S. alone, more than 300 million pound activf so e agen appliee tar d annually. Recent acknowledgment of environmental and health concerns associated with the wide scale use of chemical pesticides, however, will lead to a shift towards the use of more health and environment friendly active ingredients. Our challenge, now, is to integrate these new agents into a feasible and economical program of pest control. Micrcorganisms have long been know causo nt e diseas deatd ean n hi insects. Epizootics of these organisms in populations of insects are often very strikin virtualls ga y every insec naturalla n ti y occurring

183 population may be dead or dying late in the infection cycle. Unfortunately, however, these epizootics usually occure latth n ei population cycle of the insect after economic damage to the crop has occurred. Further, epizootics occur sporadically dependene ,ar n to environmental conditions and are extremely difficult to predict. Over the past 100 years, attempts have been made to use these disease agents in a proactive way; i.e. introduce the pathogens into field populations of insect causo st epizootin ea c and, thus, contro populatioe lth n [1]e .On method of introduction involves the infection of insects in the laboratory followed by mass release of the infected insect into field populations of the same species. Pathogens release thin di s inoculative manner then should slowly spread throug populatioe hth eventualld nan y brine gth insect under control othee Th ,r major strategy involve mase sth s productio pathogee th f direcnd o nan t applicatio damagina o nt g population of insects. This inundative approach requires formulatin pathogee gth n such that it survives in the environment until it is fed upon by the target species. The discovery of Bacillus thuringiensis Berliner subsp. kurstaki (Btk) , a spore forming soil bacterium lethal to insects, has greaa leo dt t easmans masi o yyt k sa productBt producs . ha ] d s[2 ean relatively resistant spore and protein crystal that are responsible for insect mortality. Individual strains of Btk have small host ranges and fit well into integrated pest management programs. However presentk ,Bt s a challenge to formulation scientists. It is a particulate and must be suspended uniformly through a spray mixture. Also, Btk is a living organis formulationd man s mus designee tb proteco dt frot ti m breaking down in the environment before it is fed upon by the target insect. Sunlight, especially that portion in the ultraviolet range, denatures spores [3] and crystals [4], and rainfall washes the particulates from the surfaces of treated plants. Further, Btk is not especially palatable to an insect insece ; th feey contaminatetn ma do d foliag onlr efo shorya t time without ingesting a lethal dose of Btk. If the insect then encounters unsprayed plant tissue, it may feed and recover from the sublethal infection. These factors have acte inhibio dt thitf o mas se sus valuable pest control tool. Ove pas e yearsw rth t fe have ,w e been workin adapo gt t starch encapsulation technology for formulation of Btk. However, due to the biological natur Btkf eo , past starch formulations involving high hear to harsh chemicals wer acceptablet eno availabilite Th . colf yo d water dispersible pregelatinized starches have, allowed formulation of Btk in protective, rainfast palatabld ,an e formulations. Three major typef so starch formulations have been develope typeo tw Btkgranuler f so :dfo s an sprayableda .

2. GRANULAR FCEWDLKTIGNS OF BXXLLDS TIVJPJNGTEKSIS 2.1 flbn-Aifoerent Pregelatinized cornstarches such as Miragel (A.E. Staley, Inc. Decatur, IL) , upon contact with water at room temperature will form a gelled mass. Usin rati ga part3 f oo s wate part2 o rt s Miragel, Dunkle and Shasha [5] developed a process to macro-encapsulate Btk in a cross-linked starch matrix. This matri desiregroune s xth wa o dt d particle size, usually in the 16-40 mesh range. The granules flow easily through standard on-farm granular applicators and are effective for contro Europeae th f lo n corn borer. These granules have been thoroughly tested ove pas e yearw rth t fe s against the European corn borer, Ostrinia nubilalis (Lepidoptera: Pyralidae) . To determine if the granules were palatable to corn borer

184 larvae bioassa,a developes ywa tes o effectivenese t t] th d[6 variouf so s possible feeding stimulants incorporated withi starce nth h granulese Th . assay involved placing two types of granules at opposite sides of a round plastic petri dish that had been lined on the bottom with a plaster of Paris and charcoal mixture. Larvae (less than 12 hours old) were then middlplacedishe e disth th e cappes ,n f th dei h o wa hel d 16-2r dan dfo 4 hours in the dark, and then quickly frozen. Dishes were opened and the number of larvae at each site was recorded. Although many additives were tested followine ,th g conclusions were drawn: First, starch itsely ,b s fi highlt no y preferre larvaey db , especially when tested agains tcora n leaf disk or other feeding stimulant; second, individual additives such as glucose, amino acids or individual corn borer diet components are only slightly phagostimulatory; third, and most important, combinations of ingredients were highly preferred, even when compared with fresh corn tissue. One of these multicomponent additives, the commercial feeding stimulant Coax (CCI Corporation, Litchf ield Park, AZ), was the most highly preferred additive tested. When additives were tested under greenhouse conditions (Figure 1), granules containing Coax were more effectivt ea killing corn borer larvae than were granules formulated with corn oil or without additives [6]. Under field conditions, the effect of feeding stimulants became more obvious [7]. Granules were prepared with Coax at 1 solidf o % osr10 weight sunligha ( % 1 , t conga protectantd ore , [8]r )o with no additive. Each of these granules were also formulated with no BtklU/m0 40 ,160r go 0 IU/m starcf go weighty hdr . Field tests were initiate infestiny db g corn plants with European corn borer e larvaon d ean week later granules were applie witw d ro hove e applicatorrth s calibrated to deliver 11.2 kg/ha. After 6 weeks, plants were split from base to tassel and the amount of vertical tunneling was recorded (Figure 2). It was concluded that if Coax was present in the formulation, the dose of Btk could be reduced by 3/4 without significant loss of activity. If no Coax presents wa commercialle ,th y standard dos 160f eo 0 IU/m juss gwa s t a effective as the commercial product Dipel 10 G (Abbott Laboratories, North Chicago, IL).

100 80

n 60 I 40 s Additive to Starch * 20 -B-COAX -H-CORL NOI NO STIMULANT 0 0 40 1600 3200 Bt DOSE (IU/MG) Figure 1. Response of European corn borer larvae to starch formulations of Bacillus thuringiensis treated with different feeding stimulants. Greenhouse-grown corn plants were treated with 75 mg granules and then egg masses were pinne whorle th do t . Plants were dissected 5 days latepercentagd ran e mortalit obtaineds ywa . Numbers represent mean plant5 f formulationr so spe ] 6 .

185 t-

III Z I2

ü

0 NONE CONGO COAX COAX DIPEL10G RED 1% 10%

Additive to Starch Figure 2. Efficacy of starch formulations of Bacillus thuringiensis applie fiel o s dinfestet dwa corn m d.Co with European corn borer neonate larva thed ean n treated with granules. Six weeks later, stalks were split and length of tunnels measured. Untreated control showt no t e nbu s ar stalkr averagepe . m Shadec 2 d1 d bars represent granules formulate Btk/mD I weight y 0 gdr 40 t da , unshaded bars represent formulations made with 1600 lU/mg. Note that in the presence of Coax, rate of Btk can be reduced without significant loss of efficacy. [7]

Besides reductio dosef n o granula e ,th r starch formulations alsn oca protect Btk from breakdown due to sunlight [9]. Starch granules prepared with potential sunlight protectants retained activity afte day2 r1 s exposur direco et t sunlight (Figur whil) e3 e those granules without protectants lost their activity within 4 days. To further test the hypothesis that encapsulation of Btk would prolong residual activity serie,a testf so s were conducted under field conditions. Granules were prepared wit additiveo hn , Congo red Coa,r o x and placed in the whorls of corn plants. At 1, 2, 4, 6, 8, 10, and 12 days, granules were recovered and assayed with a droplet technique [10]. For the assay, granules were digested with amylase to release the bacteria and neonate corn borer larvae were subsequently allowe imbibo dt e from droplets of the suspension. Larval mortality was assessed 2-3 days later. This experimen conductes twa threr dfo e years, each with different weather conditions threl al en I years. , temperatures wer unusuat eno r lfo the period of the test. However, rainfall did vary. In 1989, no rain occurre differenceo n d dan s were observed overall, amon granulese gth .

186 100

0 8 12 Day Exposurf so e

Figure 3. Protection of Bacillus thuringiensis from sunlight. Starch formulations were made with either 0.1% (shaded bars) or 1 % (unshaded bars) of sunlight protectants and expose direco dt t sunlight. Granules were then assayed for biological activity against European com borer neonate larvae. Granules without protectant lost all activity before the four day sample and are not shown, [after 9].

That is, the starch granules did not necessarily protect Btk better than a commercial formulation (Dipel 10G) which is prepared by coating corn grit with spore crystalsd san . However 1990n ,i , when rain occurree dth night after application of granules, Dipel 10G lost a significant amount of activity compared with the starch granules. Over the course of the 12 days, starch treatments were significantly more effective e th tha s nwa Dipel 10G. In 1991, rain fell 5 days after application. In this year, intermediats Dipewa G l10 activitn ei certais ya n starch formulations were better than Dipe (TablG l 10 . TheseI) e datsomewhae aar t contraro yt previously published reports describing los activitf s o afte k Bt rf yo application to foliar surfaces. Residual activity of granules in whorls of corn, however, has not been previously measured and different factors may be involved with loss of activity. Measurement of sunlight in the plante whorth f slo help examino st differencee eth s between foliar surface whorld san s (Figur . Clearlye4) , ther vers ei y little sunlight penetrating into the whorl of the plant and almost no ultraviolet light. The importance of providing UV protection for granules in the whorl of the plant may be less than previously believed. However, a definite benefit was observed when rain occurred. Starch granules encapsulating Btk were superior in retaining insecticidal activity compared with the commercial granule.

187 TABL . EORIGINAI L ACTIVITY REMAININ GRANULEF GO S FORMULATED WITH BACILLUS THURINGIENSIS AND EXPOSED FOR 12 DAYS IN THE WHORLS OF FIELD-GROWN OORN.a

Averacre Original Activity Remaining Treatment- 1989 1990 1991 No Additive 79.8 a 56.9 b 45.e 7b Congd oRe 63.5b 45.8 b 62.9 a Coa% x1 66.7 b 56.7 b 57.b a 5 Coax 10% 90.5a 77.2a 52.5 abc Dioel 10G 87.3 a 22.5 c 40.4c aGranules were assayed by allowing European corn borer neonate larvae to fee dropletn d o digeste e th f so d granules;. Four samples were take eacr nfo h granule type 1, 2, 4, 6, 8, 10, and 12 days after plant inoculation. nffeans withi ncoluma n followe same th e significantl t y dletteb no e rar y different (P<0.05, Least squares means procedure). GAdditives were mad Mirageo et l granules durin formulatioe gth n process. Dipel 10G is an non-encapsulated commercial product used as a control for this study.

2.0—1

0. 0

330 430 530 630 730 B30 S30 1030

Wavelength (nm)

Figur . eSunligh4 t penetratio foliagem co f no . Very little light, especially in the UV range penetrates into the whorl or behind the leaf axils of corn plants. Readings taken wit hLi-Coa r portable spectroracliometer fitted witha remote cosine .

188 In additio Btko nt , other insect pathogens have been successfully encapsulated wit Miragel/watee hth r formulation HeliothiA . s nuclear polyhedrosis virus [11], and a grasshopper entomopoxvirus [12] were both shown to survive this encapsulation process. Field tests with the grasshopper virus formulated with Miragel, molasses, corn oil, and charcoal demonstrated that grasshoppers would feed on the granules and become infected wit virue hth s [13]. 2 Adherent2. Granules: drawbacke th Miragel/wateOn e f eth o o st r formulation procese th s swa amount of water that was necessary to form the granules. Even at a 1:1 rati starcf oo waterd h an amoun e ,th grindinf to dryind gan g (without heat) that was necessary to finish the granules placed an enormous burden on scaling up the process. Therefore, we initiated a study to determine if the amount of water going into the formulation could be reduced. By simply reducing the amount of water, an unacceptable product consisting of a few large pieces and a lot of dust was created. However, by mixing the water with alcohol watee ,th r became disperse eved dan n mixins gwa possible. Propanol (15 ml), which will not gel the starch, was mixed with 35 ml water and added to 50 g Miragel. Upon mixing, individual particles formed that require grindino dn littld gan e drying. Test biologicaf so l and physical properties demonstrate loso d n activit f s o (Tablk Bt ) f yeo II aninterestinn da g propert adherencf yo surfaceo et s (Table III) [14]. Apparently, as granules are formed during mixing, they become coated with a small amount of ungelled starch. Upon contact with moist surfaces, this ungelled starch gels and acts to glue the granule onto the surface. During rétrogradation of the starch, the glued particle becomes insoluble and resists wash-off by simulated rainfall. In addition to propanol, other solvents such as ethanol, butanol, and acetone were effective in dispersing the water. Besides solvents, other water additives coul usede db . Salts, sugars, plant tissu insecd ean t tissue have all been found to be excellent water dispersers [15]. For example, CaCl2 (90 g) was dissolved in 60 ml water. Eight ml of this solution was added to 26 g Miragel and mixed to form discreet granules. Therefore totae ,th l amoun watef t o reduce s rwa o t d l from 6 m2 approximatel stilld an l ,m discree4 y4. t granules wit coulk hBt e db formed. In addition, we have determined that the Miragel, which sells for approximately US $1.55/kg could be replaced with pregelatinized corn flour which sell abour s fo $0.55/k S tU g without los activityf so factn I . , feeding preference tests demonstrated that granules made with flour (which contains abou proteistarch% % t10 90 d n)an were actually preferred over granules made with Miragel (Table IV) [16]. Granules formulated in this

TABLE II. EFFECT OF ENCAPSULATING BACILHJS ŒWRENGIWSIS IN 2-H»PANOIr<30NTAINING STARCH GRANULES ON PERCENT MORTALITY OF OffTRINIA NUBHALIS.

Percent mortality Trial______With propanol______Without propanol 1 52 48 2 42 48 3 40 43 4 23 52 5 55 43 P=0.53 (Paired t-test)

189 TABLE III. EFFEC ORGANIF TO C SOLVENT TYP ADHERENCN EO MIRAGEF EO L GRANULES.

Mean % loss (SD) Mean % loss (SD) of of granules granules from cotton Solventa from slides leaves (n=10) (n=5p> Days after application5^ 0 7 Water 100.0 (O.O)a 48.2 (26.7)a 91.9 ( 9.2)a 2-propanol 2.3 (1.4)f 8.1 (11.9)od 44.3 (13.7)d Methanol 82.9 (1.8)b 21.4 (15.1)b 76.3 (16.4)b Ethanol 13.3 (6.7)de 6.1 ( 3.6)d 52.9 (17.8)cd n-butanol 10.4 (2.1)e 4. 2.8)0( d 47.9 (21.2)cd Acetone 39.3 (1.5)c 13.7 (10.3)bcd 75.9 (21.6)b 1,4-Dioxane 18.3 (7.2)d 19. 6.4)b( 4 c 60.4 (17.1)c

Granules were prepared Icy mixing 50 g Miragel with 35 ml water and 15 ml solventa . bGlass microscope slides were wetted with distilled water and then granules were applied. After drying, slides were rewetted by allowing 40 ml water to flow over the slide in a 2 minute period. Slides were air dried and the washing procedur repeates ewa d three more times. Slides were then weigheo dt determine granule loss. °Cotton leaves were wette granuled dan s were then applied. After drying, 0-day leaves were harveste granuled dan s were remove scrapingy db . Granules were then drie weighedd dan simulateA . d rain treatmen applies e twa th o dt other leaves by spraying approximately 5 ml water onto the leaf three times ove sevee periodrth y nda . Leaves were then harveste granuled dan s weighes da above.

TABL . FEEDINEIV G PREFERENC EUROPEAF EO N CORN BORER LARVAE ALLOWEDA CHOICE BETWEE GRANULO NTW E FORMULATIONS31.

Percen larvaf to : eon Granule A______Granule B______A_____B_____

Mirage7 7 CaCll+ 3 2 CaClFlou+ 1 r96 2

Miragel + CaCl2 2 + Coax Flour 961 + CaCl2 + Coax 29 71

Miragel + CaCl + Coax Flour 961 + CaCl 25 75 2

Miraae CaCU______Floul+ 2 CaCl+ Coa+ 1 rx96 15_____85

Larvae were placed in petri2 dishes with a pile of granules at location . AfteB d hours6 r1 an A a , dishes were numbe e frozeth insectf d ro n an t sa each site were recorded. Percentages based on five dishes per comparison.

manner shoul same d th have l attributeeal granules sa s formulated with excess water, i.e. incorporation of sunscreens and feeding stimulants, and resistanc washouo et rainfally tb . Field test determino st e efficacy against European corn borer conducte 199n di 2 wit adherene hth t granules demonstrated tha granulee tth s were effective formulations (Figur. e5)

190 1 4 0. "05 «

1600 400 1600 400 Dipe G 10 l Untreated

Flour 961 Flour 980

Figur . Efficace5 adherenf yo t starch granule formulations for control of European corn borer larvae in the field. Granules were formulated wit flouhr o Flou r 1 980r96 , with 0 lU/m 40 wit160d r go an h0 Coax (shaded bars withour )o t Coax (unshaded bars).

SraÄYAELE POradÄEICK BXXLTDSF SO While the use of granular formulations of is widespread and growing vase ,th t majorit pesticidef yo appliee sar sprayabls da e formulations. Mos composee tar dispersinf do g agent helo st p suspene dth particulate Btk in the spray tank. Also, spreader-stickers are added to the tank to enhance application of the formulation. These formulations are particularly sensitiv degradatioo et sunlighy nb washofd tan y fb rainfall [17]. In 1990, McGuire and Shasha [18] reported on a novel spray formulation consistin mixtura f go Miraspersf eo pregelatinizea e( d starch simila Miragelo rt sucrosed )an . When mixe weighe 2-4t th da f f %to o wate sprayed ran d onto cotton plant Mirasperse-sucrose sth e combination effectively resisted washoff by simulated rainfall in the greenhouse. Unde simulatee rth d rainfall conditions, Dipe losX l2 t activity quickly. However, when the Mirasperse-sucrose formulation was added to the tank mix, more than 80% of the original insecticidal activity was retained after seven days [18]. Field tests conducted sinc manuscripe eth t appeared have supporte greenhouse dth e data. Test residuaf so l activity on cabbage leaves as well as full season insect control have demonstrated the utilit formulatione th f yo . Test efficacf so residuad yan l activit sprae th yf y o formulation s were conducte cooperation di n wit Illinoie hth s Natural History Survey, Champaign, IL in 1989, 1990, and 1991. In 1989 [19], three formulations

191 were tested; Mirasperse/Sucrose, Mirasperse/Sucros% 1 t (a Conge+ d ore spray solids) Mirasperse/sucrosd ,an spra% 10 eCoay+ t solids)(a x l Al . formulations contained Mirasperse/sucrose at 4% of the weight of the water; e.g liter5 .2 solidsg sk wate1 d .r an Cabbag sprayes ewa t da approximately weekly intervals, spray liters/h 0 volum27 s eleved wa aan l studye th f ,o cabbagd en e s billio 6 acree1 th r wa t pe s A .U wa n I k ofBt evaluate [206 scala 1- n ]df o e o wit representinh1 perfecga t6 head dan representing a thoroughly damaged head. Heads rated above 3 are considered as unmarketable in the fresh market. The results of this study (Table V) suggested that the starch formulations provided season long contro insecf lo t pests better than Dipecommerciaa , l2X l formulatiof no Btk and as well as Ambush, a pyrethroid insecticide. To examine residual activity, cabbage leaves were brought int laboratoro e t o th d fe d yan diamondback moth larvae (Plutella day5 d xylostella)s an afte 3 , r0 t a application. In these tests (Table VT), all formulations performed equally well after 1 and 3 days of exposure in the field. However, after 5 days, only the starch formulations retained significant levels of activity. Additive formulatioe th affeco t st no d tn di residua l activity significantly so, apparently, the starch and sucrose provided protection for the active agent.

TABLE V. EFFEKT OF MIRASPERSE/SUCROSE SPRAYART JF. FORMULATIONS OF BACHJMS THURINGIENSIS PROTECTION O CABBAGF NO E FROM INSECT PESTS.

Mean Injury Mean Percentage Treatment Rating^ Marketable Headsff Starch/Sucrose 1.98 ab 82.5 a Starch/ Sucrose + Congo Red 1.48 a 95.0 a Starch/ Sucrose + Coax 1.58 ab 92.5 a DipeX l2 2.23 b 82.5 a Ambush c 1.35 a 100.0 a Untreated 5.0c 3 0.0 b Based on a head rating scale of 1-6 [20]. Means followed by same letter are not significantly different (Fisher's least significant difference test, P<0.05). rate highet d a suitabl t no r e thafresr ar e fo n 3 h market, Means previoun i s a s column.

TABLE VT. RESIDUAL ACTIVITY OF BACILLUS THURTNGIENSIS FORMULATIONS ON CABBAGE LEAVES. 1989.

Mean Percentage Mortality3 Days After Application Treatment 0 3 5 Starch/Sucrose 100. 0 a 100. 0 a 88.0 a Starch/Sucrose Cong+ d oRe 100. 0 a 100. 0 a 84.0 a Starch/Sucrose Coa+ x 100. 0 a 100. 0 a 90.0 a DipeX l2 100. 0 a 87.0 b 22.0 b Untreated 3.3 b 15.3 c 0.0 c Diamondback moth larvae were enclose dishen di s containing cabbage foliage selected from treated plants. Means are based on five dishes with five larva dishr epe . Means followe samy db e significantl t letteno e rar y different (P<0.05 Fisher's least significant difference test).

192 Due to responses from industry, the next tests were conducted with lower level solidf so witd san h flour instea Miraspersef do 1990n .I , [21] similar tests were conducted wit pregelatinizeha d flour designated Film Forme Illinoiy b rA s Cereal Mills. Thi mixes swa d with 1 sucros1: a n ei ratio and added to the spray tank at 1% solids. Because no effect was observed due to additive in the previous year, nothing else was added to the spray tank. Results from this study revealed tha t solid% t1 no d sdi provide season long control or extend residual activity (Table VII) any better tha commerciae nth l formulation, Dipe . Heal2X d ratinge th r sfo cabbage treated with the flour-sucrose formulation averaged 2.63 whereas ratings for Dipel 2X treated plots averaged 2.33. Control plants averaged ratina 5.43f go . Therefore 1991n ,i , field tests were conducted with solids levels of 1, 2, and 4% with a different flour (designated 22191) with and without charcoal as a sunlight screening agent [22].

TABLE VII. RESIDUAL ACTIVITY OF BfCILUJS THUPINGIENSIS FORMULATIONS ON CABBAGE LEAVES. 1990.

Trial1 Mean Percentage Mortality3 Davs After Application Treatment 0 3ç1 5 Flour/ sucrose 98.1 a 47*7 a 5.0 a DipeX l2 97.0 a 52* 3 a 4.1 a Untreated 2.0 b 0 0 b 0.0 b Trial 2 Mean Percentage Mortality ______Davs After Application Treatment______Q______3______5______7-c Flour/ sucrose 100.0 a 90.8 a 32.1 a 12.9 a DipeX l2 99.0 a 58.6 b 14.9 ab 2.0 b Untreated 14.1 b 3.0 c 5.1 b 5.0 ab ^Diamondback moth larvae were enclose dishen di s containing cabbage foliage selected from treated plants. Means are based on five dishes with five larvae per dish. Means followed by same letter are not significantly different (P<0.05 Fisher's least significant difference test). b!9 mm rain fell on the test plants before the 3 day test. C20 mm rain fell on the test plants before the 7 day test.

Unfortunately, this tes confoundes twa periody db raif so n which prevented timely application pesticidesf so . Therefore, season long control results indicated that cabbage treated wit formulationl hal s averaged ratingf so more than 5 as heads were damaged beyond market requirements. However, results froresiduao mtw l activity bioassays revealed that residual activity was related to percent solids in the spray volume. Formulations with 4% solids protected Btk better than formulations with less solids and a commercial formulation (Table 8). Apparently, 4 percent solids is required to form a film thick enough to protect the active ingredients. As an additive, charcoal did not have any dramatic effect on retention of activity.

193 TABLE VIII. RESIDUAL ACTIVIT BSCTTffTSF EO THUPINGIWSIS FORMULATIOISW CABBAGE LEAVES. 1991

Trial1 Mean Percentage Mortality3 Davs After Application Treatment- 0 3 5 7 1% Solids 100.0 a 97.5 a 72.5 be 63.5 bc 2% Solids + Charcoal 100.0 a 90.0 a 88.b 8a 53.0 c 2% Solids 100.0 a 89.7 a 81.b 0a 75.0 ab 2% Solids + Charcoal 100.0 a 90.0 a 90.0 ab 60.0 bc 4% Solids 100.0 a 93.8 a 94.4 a 81.8 ab

4% Solid Charcoas+ l 100.0 a 97.5 a 93.8 a 89 •6 a Dipel 2X 100.0 a 70.0 b 57.8 c 51.3 c Untreated 35.3 b 2.5 c 1.3 d 8.5 d Trial 2 Mean Percentage Mortality Davs After Application Treatment 1 3 6e- 8*2 1% Solids 100.0 a 100.0 a 60.5 d 56.3~bc 2% Solids + Charcoal 100.0 a 100.0 a 76.7 be 43.8 c 2% Solids 100.0 a 95.9 a 82.3 ab 45.3 c 2% Solids + Charcoal 100.0 a 100.0 a 95.0 a 67.5 bc 4% Solids 100.0 a 100.0 a 86.3 ab 71.1 ab 4% Solid sCharcoa+ l 100.0 a 100. 0 a 89.6 ab 92.5 a DipeX l2 100.0 a 96.3 a 62.1 cd 61.3 bc Untreated 8.8 b 11.3 b 11.5 d 5.1 d

Diamondback moth larvae were enclosed in dishes containing cabbage foliage a selected from treated plants. Means are based on five dishes with five larvae per dish. Means followed by same letter are not significantly different (P<0.05 Fisher's least significant difference test). ^Percent solids refer amouno s lout sucros f d rf an t o e mixturee addeth o dt spray tank. Charcoal, if present, was cidded at 1% of the solids content. C51 mm rain fell before the 6 day test. ^2 mm rain fell before the 8 day test.

. SUMMAR4 Y In conclusion, starch formulations can be used to extend and enhance activite th Bacillusf yo thuringiensis . Granula r formulatione sar particularly effective when use periodn di raif so n when retentiof no active agent within a bait is necessary. Similarly, sprayable formulation effective sar relativelt ea y high concentration could san d probably be effectively used for aerial applications or where spray volume lowe sar . Wor continuins ki laboratorr ou n gi y with reducing solids content and extending the technology to formulate insect pathogens from other groups.

REFERENCES [1] STEINHAUS, E. A., Principles of Insect Pathology, McGraw-Hill, New York (1949) . [2] BEEGLE, C.C., YAMAMOTO, T., Invitation Paper (C.P. Alexander Fund) History of Bacillus thuringiensis Berliner research and development, Can. Ent. 124 (1992) 587-616.

194 [3] BEEGLE, C.C., DULMAGE, H.T., WOLFENBARGER, D.A., MARTINEZ, E., Persistence of Bacillus thuringiensis-var-kurstaki insecticidal activity on cotton foliage, Environ. Entomol. 10 (1981) 400-401. [4] POZSGAY, M., FAST, F., KAPIAN, H., CAKEY, P.R., The effect of sunligh proteie th n to n crystals from Bacillus thuringiensis var. kurstaki HD1 and NRD 12: a raman spectroscopic study, J. Invertebr. Pathol. 50 (1987) 246-253. ] DUNKLE[5 , R.L., SHASHA, B.S., Starch-encapsulated Bacillus thuringiensis: potentiaA methow lne increasinr dfo g environmental stability of entomopathogens, Environ. Entomol. 17 (1988) 120-126. ] BÄRTELT[6 MGGUIKE, J. . ,R , M.R., BLACK, D.A., Feeding stimulantr sfo the European corn borer (Lepidoptera: Pyralidae): Additivea o st starch - based formulation for Bacillus thuringiensis Berliner, Environ. Entomol. 19 (1990) 182-189. [7] MCGUIRE, M.R., SHASHA, B.S., LEWIS, L.C., BÄRTELT, R.J., KLNNEY, K., Field evaluation of granular starch formulations of Bacillus thuringiensis against Qstrinia nubilalis (Lepidoptera: Pyralidae), J. Econ. Entomol. 83 (1990) 2207-2210. [8] SHAPIRO, M., Congo red as an ultraviolet protectant for the gypsy moth (Lepidoptera: Lymantriidae) nuclear polyhedrosis virus, J. Econ. Entomol (19892 .8 ) 548-550. [9] DUNKLE, R.L., SHASHA, B.S., Respons starcf eo encapsulateh- d Bacillus thuringiensis containing ultraviolet (UV) screeno st sunlight, Environ. Entomol. 18 (1989) 1035-1041. [10] HUGHES, P.R., WOOD, H.A. synchronou,A s peroral technique th r efo bioassay of insect viruses, J. Invertebr. Pathol. 37 (1981) 154-159. [11] IGNOFFO, C., SHASHA, B., SHAPIRO, M., Sunlight-UV protection of the Heliothis nuclear polyhedrosis virus through starch-encapsulation technology Invertebr. ,J . Pathol (19917 .5 ) _ __ 134-136. [12] MOGUIRE, M.R., STREETT, D.A., SHASHA, B.S., Evaluation of starch encapsulation for formulation of grasshopper (Orthoptera: Acrididae) entomopoxviruses, J. Econ. Entomol. 84 (1991) 1652-1656. [13] ONSAGER, J.A., STREETT, D.A., WOODS, S.A., Grasshopper pathogen field evaluation: virus, USDA-APHIS-PPQ Grasshopper Integrated Pest Management Project Annual Report (1989) 220-228. [14] MOGUIRE, B.S., SHASHA, B.S., Adherent starch granuler sfo encapsulation of insect control agents, J. Econ. Entomol. 85 (1992) 1425-1433. [15] SHASHA, B.S., MOGUIRE, M.R., Adherent starch granules, U.S. Patent Appl. S/N 07/913,565, (1992) [16] GILLESPIE, R.L., MOGUIRE, M.R., SHASHA, B.S., Palatabilit flouf yo r granular formulation Europeao st n corn borer larvae (Ostrinia nubilalis (Hubner); Lepidoptera: Pyralidae), J. Econ. Entomol. (submitte publication)r dfo . [17] MORRIS, O.N., Protectio thuringiensis. B f no from inactivation by sunlight, Can. Ent. 115 (1983) 1215-1217. [18] MOGUIRE, M.R., SHASHA, B.S., Sprayable self-encapsulating starch formulation Bacillusr sfo thuringiensis, Econ. J . Entomol3 .8 (1990) 1813-1817. [19] KENNEY, K.K., OLDUMI-SADEGHI, H., STEFFEY, K.L., Illinois Insecticide Evaluations Forage, Field Vegetabld ,an e Crops, (1989) 61-67. [20] GREENE, G.L., GENUNG, R.B., WORKMAN, R.B., KELSHEIMER, E.G., Cabbage looper contro Floridan li cooperativ:a e program Econ. ,J . Entomol. 62 (1969) 798-800.

195 [21] OLOUMI-SADEGHI, H., GRAY, M.E., STEFFEY, K.L., Insect Management and Insecticide Evaluations, Illinois (1990) 90-96. [22] ODOUMI-SADEGHI, H., GRAY, M.E., STEFFEY, K.L., EASTMAN, C.E., Insect Management and Insecticide Evaluations, Illinois (1991) 121-131.

196 THE USE OF CONTROLLED RELEASE TECHNOLOGY TO IMPROVE THE PERFORMANCE AND REDUCE THE ENVIRONMENTAL IMPAC HERBICIDEF TO S

A. FLYNN . STORKP , . G1TTINJ , S . FINLAYM . WILLIAMK , S Daratech Pty Ltd, Victorian Institute for Dryland Agriculture, Horsham, Victoria, Australia

Abstract

Daratec s researcha h h projects f investigatino Controlle e us e dth g Release (CR) technolog a numbey areas n ke i y f .o r These include remedies for; solvent removal, groundwater pollution, volatile loss, off-target damage, crop phytotoxicityV U , degradatio inadequatd nan e foliar penetration.

Active ingredients researched include; triazines, sulfonyl ureas, dinitroanalines, phenoxys, aryloxyphenoxypropionates, thiocarbamates and pyridines.

In this paper example e giveb n volatile o nar s e loss reduction, reduced leaching, improved crop safety and solvent removal from foliar herbicide.

Daratech has a wide range of CR types. CR technology has enabled the reduction of volatile los f trifluralio s o 18%s nt enable ha fro % t e I los .m52 f th chlorsulfurodo s n from leaching to be decreased from 90% to 20%. It has reduced the phytotoxicity of trifluralin to wheat by 45 % and has been used in post emergence herbicides to recover the activity lost when solvents are removed from a herbicide.

1. INTRODUCTION e e Departmencommerciath th f s i o Daratecd m f Lt Agricultureo ar t l y Pt h , Victoria, Australia Controllea s i t I . d Release (CR researc0 )6 specialis s ha hd stafan t f workinn go CR technology for agricultural chemicals. Forty of these staff specialise on herbicides. It has current projects addressing the major environmental and performance problems of; solvent removal, groundwater pollution, volatile loss from soil applied herbicides, off- target damage, crop phytotoxicity, UV degradation and inadequate foliar penetration.

Active ingredients researched include; triazines, sulfonyl ureas, dinitroanalines, phenoxys, aryloxyphenoxypropionates, thiocarbamates and pyridines. The research group is resourced from government research grant organisations, chemical companies and private investors.

In this papee herbicide'prograth I wilre us l demonstratmo t technologieR C e eth s used. Example givee b volatiln e no ar s e loss reduction, reduced leaching, improved crop safety solvend an t removal from foliar herbicide.

197 2. BIOASSAY TECHNIQUES projece Th s placeha t d considerably more emphasi bioassan o s y expertisf o e e us tha e nth chemical analysis infeo t . t rThi no tha s i s t chemical r valianalysia t ou dn no i toos t i s bu l circumstances bioassays allow for a direct measure of herbicidal activity. Chemical analysis, generally destroy agenR detectd C e an t sth s total active present.

Daratech's bioassays have been extensively e mosvalidate th fielde tn th I advancn .i d e project, trifluralin, early products are registered and later, improved products, have successfully undergone extensive international field trials.

3. CR TECHNOLOGY USED

Unlike most other similar groups Daratech doet have specifino s on e technologyR C c . Experienc dat o eshowt s eha n that each active require specialisesa d system.

In general, however e technologth , y involve e loadine activth sth f eo g ingredient into porous powders without the use of volatile solvents. These powders are then coated to achiev e requireth e d release profile. Cost constraint d eas f an manufacturso e e ar e addressed throughout the project life. Formulation aims at additional costs of less than 10% over current formulations.

4. CR TECHNOLOGY TO REDUCE VOLATILE LOSS.

4.1. Introduction

Daratech work n fivo s e volatile, soil applied a herbicidescommercial s ha t I R .C , trifluralin formulation that offers farmers reduced volatile t losequabu s l herbicidal activity under high volatility conditions. Later formulations improve on this performance. l glasshousAl e experimental techniques have been extensively validate fielde th n .di objective Th technologR f thiC eo se projecus o removo t yt s i tcultivato neede t eth ) (1 ; e these herbicides into the soil, thereby reducing soil degradation of fragile soils and, (2) for volatile solvents, thereby reducing flammability.

4.2. Glasshouse method

A number of experimental protocols are used to test for volatile loss. The following protocol relateresule th o tst given below.

Pots were prepared with an alkaline loamy sand (pH=8.5), pre-moistened to 10% moisture. Herbicid s appliei e thio dt s soil throug ha calibrate d laboratory tracksprayer. Nine herbicide rate seved an s n replicates were used. Soil fro s thoroughlm wa eac t hpo y mixed either immediatel yafteh (Oh48 r r )o sprayin ceaso g(t e volatilisation returned an ) d potso t . Volatilisation occurs under controlle r movementdai . Pot bioassayee ar s d using Lolium rigidum as the test species. Pots are assessed for emergence after 10 days and data is tested for 'goodness of fit' and fitted with logistic curves.

198 4.3 Results

formulatioR C e Th n 129-0 s foun wa 1o havt d e equivalent herbicidal activite th o t y commercial EC when incorporated immediately after spraying (Fig.l). The LDSOs for both formulations were approximately 115 g ai/ha. When incorporation was delayed for 48h, 129-01 was significantly (P<0.01) more active than the EC. LDSOs were 140 g ai/ha for 129-01 and 240 g ai/ha for the EC. Extensive previous studies have shown that almost all of this loss is due to volatile loss under the test conditions used.

£100

0 40 0 30 0 20 0 0 10 500 600 Dos triflurlaig e( n ai/ha)

*Eh CO 129-0h 1O EC48h 129-0h 148

Figure 1. Activity of trifluralin formulations with 0 or 48h incorp.

4 4. Conclusion

CR technology can greatly reduce volatile loss. Caution must be taken not to reduce vapour pressure too far. Herbicides, such as trifluralin rely on vapour pressure for transport throug agente R soilC th h . s which excessively suppress vapour pressure also reduce herbicidal activity.

e formulatioTh n 129-0 s optimise1ha d releas r thiefo s formulation type. Approximately 18% of 129-01's trifluralin has volatilised in 48h compared with 52% of the EC standard. Such a formulation will offer farmers considerably more flexibility in cultivation than the commercial standard.

199 . 5 REDUCED LEACHING

5.1. Introduction

Daratech researches two herbicide groups that have problems with leaching, triazines and sulfonyl ureas. Leachin f activgo e ingredient belo rooe wth t zon f weedeo s presents three problems ) ris (i f ;groundwate ko r pollution, (ii) increased residue microbias a s l activits yi greatest neae soith rl surface, (iii) reduced herbicidal efficiency (any herbicidt no e retained in the weed root zone is wasted).

The objective of these projects is to develop CR technology to reduce herbicide leaching increasd an e herbicide retentio weee th dn ni root zone.

5.2 Glasshouse method

The following chlorsulfuron protocol relates to Ihe result given below. Special care was taken to avoid, soil subsidence, waterlogging at depth, excessive leach rates and preferential water flow paths. Failure in any of these areas will result in inaccurate result pood san r interpretation.

PVC columns of 63mm internal diameter and in 50mm segments were fixed together to a total lengt f 400mmo h . Columns were filled sievedm wit2m h , alkaline, loamy sand (pH 8.5 = preleache d )an d under constant suction (-15 kPa) wit equivalene hth por1 f eo t volume. This resulte botn di hunifora m water content (9.5% soil moisture soid an )l bulk density (1.5g/cm3) with depth.

Herbicid s applieewa pipetty surface db preleachee th th o f et o e d soil colum covered nan d with coars efiltea san ensurd o t rd an e even distributio f waterno . Three pore volumf eo water was applied to the surface by peristaltic pump at the equivalent of 50mm of rainfall/day. Suction was retained at -15 kPa. 24h after the completion of leaching the columns were dismantled and soil from each 5mm segment was bioassayed for the active ingredient. Leachate collected from each colum uncontaminatey s addedr nwa o t d d soil and bioassayed.

5.3. Results

Approximately 96% of the chlorsulfuron as Glean, was detected in the soil and leachate appliee f thith O f so d onl. % (Fi2) chlorsulfuroy6 g 0-50me th n i ms segmentnwa % 60 , was 51-400me foun leachatee th th n i dn i m% . dept30 d han chlorsulfuroe th f o % tota 48 A formulatio R nf o C lapplie e th s da n 3-67 detectes awa y db bioassay. It is expected that the remainder of added chlorsulfuron had remained in its CR matrie 0-5mth n appliee xi formulatiomR th C segmentf e o d th % chlorsulfurof O 16 n . n s detecte e 0-50mwa th n di m segment s detecte e 51-400m. wa th Onl% n i d 19 y m depth e leachateth n i % .1 1 d an

5.4. Conclusion

CR technolog greatln yca y reduc lose f herbicideeth so s activ leachingo et doinn I . g this the herbicide is retained in the topsoil where herbicidal efficiency is greatest. It is likely that this technology could also be used to reduce the application rate of herbicides.

200 Glean

3-67a

% Recovery Glea% 97 n % 3-6748 a

20 25 30 35 40 Leachate Depth (cm)

Figur . Effec e2 simulatef o t distributiode th rai n no f no chlorsulfuron formulations in an alkaline sandy soil.

6. REDUCED PHYTOTOXICITY

6.1. Introduction

Many herbicide e marginallar s y phytotoxi e cropth o st c thaR C t thee use. ar yon d formulations have been shown to be able to reduce the effect of some herbicides on the crop without loss of herbicidal activity. Daratech has extensive experience in this area with a CR trifluralin formulation which has undergone a test market across 50,000 ha. objective Th f thieo s projecreduco t s phytotoxicite ewa tth f selecteyo d herbicides without los f herbicidaso l activity.

6.2. Glasshouse method protocoe Th r phytotoxicitfo l f trifluraliyo modificatioa s nwa f thano t use r herbiciddfo e efficacy. The test species was Triticum aestivum and the duration of assay was 15d.

6.3. Results

Wheat treate formulationdR C witCR1 d e an hfouns th 65 wa havo dt CR s esignificantla y greater emergence than wheat treated wite commerciath h l standard emulsifiable

201 concentrate (EC) (Fig 3). LDSOs were 240 g ai/ha for the EC, 340 and 350g ai/ha for CR5 and CR16 respectively. activite t Th significantl weedno n yo s swa y (P<0.05) differen three th r e fo formulationst .

6.4. Conclusion

CR technology can greatly reduce the phytotoxicity of herbicides. In the above experiment the CR formulations were approximately 45% less phytotoxic potent on wheat than the EC.

0 200 400 600 800 1000 Dose (g triflurlain ai/ha)

EC CR5 CR16

Figure 3. Comparison of the phyto- toxicit trifluralif yo n formulations.

7. SOLVENT REMOVAL FROM FOLIAR HERBICIDES

7.1. Introduction

Many major foliar herbicide basee ar s solventsn do . These solvents emulsif active th y n ei water and often also aid in leaf penetration. Solvents are flammable and represent a risk in transport userd addition en I , .e storagth nd spillean f solveno s t based chemicale ar s notoriously difficul f I thes o cleat e. t up formulationn e reformulateb n ca s d into powdered, solvent free formulations packed in water soluble bags these problems can be overcome.

202 Approximatel millio0 y3 n pesticide container manufacturee ar s d each yea holo t r d solvent based formulations. This causes problems with container disposa stockpiled an l f theso s e containers litte worlde th r .

7.2. Glasshouse method

The protocol for the solvent based foliar applied aryloxyphenoxypropionates (Fops) is as follows. Pots were filled with potting mix, sown witseedo htw f oat o s s (Avena sativa) growd an n in a glasshouse at 24°C for 2Id (pots are thinned to Iplant/pot at 7d). Herbicide was applied throug hcalibratea d laboratory trackspraye potteo t r d soil. Nine herbicide rates and seven replicates were used. Pots were assesse r fresdfo h weigh datd an ta afted 14 r fitted with logistic curves.

7.3. Results

The solvent-free powder, 2-14a was 68% less active than the commercial standard Fop formulatioR C e Th . (Fi nt 4) significantlgno 1-7 s 2wa y differen n activite i tth o t y commercia formulatioC E l n (Fi. g5)

S3 -

W)

10 15 20 25 30 35 Fop (g ai/ha)

Commercia -*p - lFo 2-14 a

Figur . Effeceremovae 4 th f o t f o l solvent on the activity of Fops.

203 15 20 35 Fop (g ai/ha)

Commercial Fop *- CR 1-72

Figure 5. Effect of CR additives efficace onth Fopsf yo .

7.4. Conclusion

CR technolog e useb o improvdt n yca e recoveth e e efficacth r y lost from solvent based herbicide when the solvent has been removed. In the above experiment CR agents were selected for improved leaf uptake and through optimisation of this effect efficacy is improved. Care must be taken as CR techniques often result in foliar herbicides becoming less rainfast. e removaTh f solveno l t often allow r increasefo s d active content. This often offsety an s additional formulation costs.

8. CONCLUSION

CR technology has much to offer the world as a means of improving pesticides. CR technology, however simpla , e looke s b musa t en no do tpanace r pesticidfo a e problems but rather a formulation tool. The advances described have only been achieved after considerable research and development in the fields of chemistry and biology.

204 KINETIC STUDIE CONTROLLED-RELEASF SO E FORMULATION DIUROF SO N CONTAINING PAL MILL MOI L EFFLUENT

B.M. YAMIN . SAHALA , I MARDI Facult Physicaf yo Applied an l d Sciences, Universiti Kebangsaan Malaysia, Bangi, Selangor Darul Ehsan

R.B. MOHAMAD Faculty of Agriculture, Universiti Pertanian Malaysia, Sedang, Selangor Darul Ehsan

Malaysia

Abstract

Controlled-release formulations of diuron herbicide containing sodium alginate as binde kaolid an r r palno l milmoi l effluent (POME s fillera ) s were studied. Small ratios of alginate to kaolin or POME in the formulation produce less spherical granular products. The kinetic of release in static water was studied spectrophotometricall t 248nma y . Both products wito differentw h t fillers showed good first order plots with rate constants about axlO'l dayl. Preliminary screening on several species of weeds in one square meter boxes in glasshouse showed good effectiveness of the slow release products. Further studies are being carried out especially with the POME formulations which contain quite high major nutrients.

1. INTRODUCTION

Although there have been considerable progres slon so w release formulatiof no pesticides works internationally mor agriculturae datth n ao l field test mann si y parts of the world especially the tropical countries are still necessary. One of the formulations which received interes numbea y tb researcherf o r alginate th thas si f o t e based. Several herbicide d insecticidean s s have been formulate thiy b ds technique [1,2,3,4,5] mosn I . t formulations alginat s usei e s bindea d kaolid an r r clano s a y fillers. The effect of absorbants like charcoal, alumina, celite and silica gel in the formulations have also been reported [5]. Among pesticides that were formulatee dar 2,6- Dichlorobenzonitrile, carbofuran, Diquat, Dichlobenil, DDT, Dieldrin, endosulfan, metribuzi thiobencarbd nan .

Based on the availability of natural resources and waste of agricultural and chemical processes attemp mako t t e formulationth e s usin e availablth g e materials were recently carried out.

This paper presents some properties of the alginate based formulations of diuron f palo l milme oi us l e effluenth d s fillera an t . Diuro r 3-(3,4-dichlorophenyl)l-lno - dimethylure produces ai Malaysin di Ancoy ab m Sdn. Bhd.

205 2. MATERIALS AND METHOD

2.1. Chemicals

Sodium alginat s perchasewa e d from BDH. Kaoli d diuroan n n were kindly donate Kaoliy db n (Malaysia) Sdn. Bhd Ancod .an m Sdn. Bhd. respectively. Pall moi mill effluent (POME) cake was obtained from Golden Hope Plantations.

2.2. Productio Alginate th f no e Granules

Alginate gel beads containing dispersed diuron were prepared according to the method described by many previous workers [1-5]. A homogenised mixture of diuron, kaolin or POME in aqueous solution was added dropwise to calcium chloride via a peristaltic pump and through a glass tubing of a specific size. The granules were seperated, washe dried d an t roo da m temperatur abour e fo days 0 e shape 1 t th Th .f eo granules were foun varo t d y wit e compositiohth e e mixturdistancth th f d no an ee between the nozzle and the surface of CaCl solution.

2.3. Kinetic Measurements

Kinetic experiments were carriedeionisen i t dou d water under static conditions at room temperature (30°C). 100 mg (about 20 granules) of dried beads were transfered into a 250 ml stoppered erlenmeyer charged with 30 ml of deionised water. smalA l amoun f aqueouo t s solutio takes nwa n fro flase m th t certai ka n intervald san absorbance th measures ewa t 248nmda .

Aqueous solutio f diurono n (about lO'^M maximus ha ) m absorbancm n 8 24 t ea 1545= £ ( 8 cnrlmoHl). UV-Visible spectrophotometer Hitachi-200 uses r 0fo wa d recording the spectrum of the solutions and absorbance measurements. Perchloric acisodiud dan m hydroxide were use studo dt release yth t differenea . pH t

2.4. Preliminary Test on Efficacy

Efficgcy test on several species of weeds was carried out in a glasshouse. Four blocks consistin f fouo g r fiberglass boxex 45c m mI s witx wer hm eI siz f o e arrange randomnizea n di d complete block design (RCBD) filles . wa Eacd x wothbo h fine sandy loam soil (Bungo heightm c e arrangemen0 r 3 Th .Series o t e p th u ) f o t boxes is shown in Figure 1. Each block consists of four boxes treated with conventional diuron (WP80), diuron CR(5%) mixtura , f WP8eo CR(5%d 0an d an ) control (untreated) respectively amoune Th . f pesticideo t s give s 1.2nwa g a.i/mr ^pe wateres wa soie eac n i x lTh boxd hbo .wit litrh4 e wate r dayrpe .

3. RESULTS AND DISCUSSION

3.1. Palm Oil Mill Effluent (POME)

Pal l milmoi l effulèn s essentialli t a ynon-toxi c mil- producby l t comprising 70% clarification sludge, 19% strelizer condensate and 11% hydrocyclone washing

206 CContro= l Wettabl= P W e Powder (Conventional) C RControlle= dReleas- e formulation ( 5% diuron)

FIGURE 1

ARRANGEMENT OF BOXES FOR EFFICACY TEST

[6,7]. As the results of the extensive research on the elemental recycling of palm oil mill effluen n economicalla plantationsn i t w no s i t i ,y viabl d environmentallean y acceptable waste through proper procedure management e rol f POMTh o e.n a s Ea inorganic fertilizer substitute has also received much interest.

Chemical analysi s r eaccarriefo wa s ht batcou d f POMo h E cake user fo d preparation of the slow release formulations. Table 1 shows a typical nutrients content in the different types of effluent [6]. The chemical analysis of dried sludge collected from various sources is shown in Table 2.

207 TABLE 1 TYPICAL CONCENTRATION OF VARIOUS NUTRIENTS IN POME

Percentage of Nutrient Type of Pome N P K Mg

Raw Pome 0.10 0.02 0.20 0.04 (1.3) (4.7)** (4.0) (2.6)

Digested Lagoon Pome 0.03 0.01 0.15 0.03 (1.4) (0.6) (3.0) (1.9)

Digested Sludge Cake* 4.75 0.88 1.41 0.63 (226.2) (56.0) (28.2) (40.2)

Dewatered cake with 65% moisture Figur parenthesin ei s indicat respective eth e fertilizer equivalen n kg/tonni t e of pome, that is N as ammonium sulphate, P as rock phosphate, K as muriats kieserita g f potasM eo e d an h

TABLE 2 CHEMICAL COMPOSITIO DRIEF NO D SLUDGE FROM VARIOUS SOURCES

Item Composition Source

Dry matter, % 93.30 UPM Protein (Nx6.25),% 11.10 UPM Crude fat, % 13.90 UPM Crude fibre, % 11.70 UPM N-free extract% , 45.25 UPM Ash, % 26.35 UPM Ca, % 0.55 UPM P, % 0.44 UPM Energy, grass (kcal/g) 4.10 UPM Energy, metab (kcal/g) 3.10 UPM Salts% , 1.06 Gold Coin Silica% , 36.04 Gold Coin K,% 3.98 UM Mg, % 0.27 UM m . p . p , B 19.00 UM C, p.p. m 34.00 UM Zn, p.p.m 42.00 UM Fe,p.p.m 1317.00 UM

UP MUniversit= i Pertanian Malaysia

U M= Universit y Malaya

208 3.2. Formulation Products

It is known that swollen beads thav formed after the droplets were in contact with Calcium Chloride solution undergone shrinkage as much as 30% in size after drying. Although a good spherical granules can be easily obtained, the shape depends very ratie muc th f alginat oo n ho kaolio et r POMEno . Small quantit f alginatyo e will give flake like granule. Under the present conditions the diameters of the dried granules were in the range of 1.40-1.94 mm when prepared with different composition of alginate, fillers and diuron. Under the same conditions and compositions granules from both kaolin and POME formulations showed slightly different diameter and weight.

3.3. Kinetic Studies

Figure 2 shows the typical plot of release concentration or absorbance against time. On most experiments the release takes about 4 weeks to reach equilibrium. Although the quantity of diuron released into the solution or at equilibrium is less in the order of (Diuron-Alginate) < (Diuron-POME-Alginate) < (Diuron-Kaolin- Alginate), the kinetic does not necessarily follow the same order. Plot of log(Aao-At) versus time, where e Aabsorbanced Afar ^-an t infinita s y timd timan , et e respectively, gave good straight linr morefo e tha half-lifen3 s indicatin googa d first order kineti f releasco e reproducibilite (FigurTh . e3) f ratyo e constan repeatiny b t g the same experiment 5 times gave standard deviation of +. 0.015. The rate constants diuroe % releasfoth 10 r nf eo formulation Alginate% 90 n i s , (20% alginate% 70 , POME) and (20% alginate, 70% Kaolin) are 0.119, 0.132 and 0.112 day"1 respectively.

Generall release yth f diuroeo deionisen ni r distilledo d wate s significantli r y faste rpresence thath n ni f acie o r base do .simpl o Thern s eewa relation betweee nth e solutionth f o .H releas p Howeve d an e r certaifo r n formulations thera smal s i e l decreas e ratth e n i froe m acidi o neutrat c d increasan l a elittl n basii e c solution indicating some interactions between acid or base with the alginic carboxylate gvoup. Perchloric acichoses dwa n because perchlorat knows i n ehavio o nt e less interaction or coordination with many metal species in solution. Although the kinetic measurement is limited by the solubility of diuron in aqueous solution, a linear increas f rateo e constants with increasing percentag f diuroeo observes nwa somn di e formulations.

3.4. Efficacy Test

The RCBD arrangement of pesticide treatments is shown in Figure 1. A few but small germinated weed were observe boxe0 1 f contron do i s slod an lw release diuron formulations afte a weekr . Afte 2 weekr s only control boxe d growinha s g germinated weeds. No weeds was observed in other boxes, indicating that high enough concentratio e activth f eo n agen s releasedwa t o product , e effecth en o t germinating wee do supres t seed d an ss thei r e controgrowthth l Al l .boxe s wa s completely filled with herbs afte weeks0 1 r . Other boxes remained clean until five and a half months of the observation period. Some seeds must have been dislocated into some treated boxes, especially those close o controt d l boxe s growta s s wa h

209 10 15 20 25 30 35 40 45 TIME/DAY

(1) Formulatio Diurori,90% 10 : n % Alginate. ) Formulation:(2 Diurori,20% 10 : % Alginate,70% POME. (3) Formulation : 10% Diuron, 20% Alginate, 70% Kaolin.

FIGUREZ

RELEASE OF DIURON FROM DRIED CALCIUM ALGINATE GRANULE IN STATIC WATER

210 ) Log(t A - A

0.4

0.2

-0.2

-0.4

-0.6

-0.8

-1.0

-1.2 0

-1.4

-1.6

10 15 20 25 30 35 40 TIME/DAY

FIGURES

PLOT OF LOG (Av- At) AGAINST TIME RELEASFOE RTH DIUROF EO STATIN I C WATER Slow Release Formulation : 10% CRDP(4-14-2)

211 observed afte halra fivd f emonthsan . Tabl showe3 countine sth numbee th f go f o r weed t specifia s c weeks. Weed species were identified after four weekd an s presented in Table 4.

TABLE 3

EFFICACY TEST

Week Treatment 1 2 4 8 10 20 22 1 Control 10 55 119 100 % 100 % 100 % 100%

R C + P W 2 6 - - - - - 33 3 WP 15 - - - - - 12

4 CR 13 ------

5 CR 0 - - - - - 30

6 WP+ CR 1 ------

7 Control 1 10 14 30% 100 % 100 % 100 %

8 WP 2 ------

9 WP 0 - - - - - 10

10 Control 0 4 10 25% 100 % 100 % 100% 11 CR 0 - - - - - 30

R C + P 1W 2 4 ------

13 WP + CR 0 - - - - - 15 P 1W 4 0 - - - - - 9

15 CR 3 - - - - - 25

16 Control 23 30 63 40% 100 % 100 % 100 %

212 TABLE4 WEEDCONTROE TH N SI L BOXES AFTER FOUR WEEK

BOX WEED NUMBER NUMBER 1 Cyperus distans 87 Borreria latifolia 21 Cleome rutidospenna 4 Mimosa pudica 3 Mikania micrantha 3 Amaranthus spinosus 1

7 Cyperus distans 6 Mimosa pudica 4 Borreria latifolia 4

10 Cyperus distans 4 Borreria latifolia 4 Mimosa pudica 2

16 Mimosa pudica 27 Borreria latifolia 24 Cyperus distans 12

. 4 CONCLUSION

Palm oil mill effluent can be used as filler for the alginate based slow release pesticides formulations. Further studie beine s ar variou e gth carrien o st aspectdou f so application of products in the soil system and field applications.

ACKNOWLEDGEMENT

We thank The Malaysian Government and Universiti Kebangsaan Malaysia for their support to initiate the project (Grant No: IRPA 2-07-03-08). The authors would also lik thano et . BarnkDr Ancof eo m Sdn. Bhd providinr .fo diuroe g th Kaoli d nan n (Malaysia) Sdn. Bhd. for free sample of kaolin.

213 REFERENCES

[1] CONNICK, W.J., BRADOW, J.M., WELLS, W., STEWARD, K.K, VAN, T.K., Preparatio d Evaluatioan n f Controlled-Releaso n e Formulations 2,6- Dichlorobenzonitrile, J. Agric. Food & Chem. 32 (1984) 1199.

] [2 SCHACHT , VANDICHELE. , , J.C., Improved alginate Based Slow Release Perticides: Food and Environmental Implications, IAEA,(1988) 267.

[3] VOLLNER, L., KUTSCHER, R., DOMBOVARI, J., ONCSIK, M., Distribution and Metabolism of Carbofuran in Paddy Rice from Controlled Release Formulation, Proc. Int. Symposiu n Changinmo g Perspectiven i s Agrochemicals: Isotopic Techniques for the Study of Food and Environmental Implications, IAEA- SM-297/33(1987).

] [4 PEPPERMAN, A.B., KUAN, J.C.W., McCombs , AlginatC. , e Controlled Release Formulation Metribuzinf so . ControlleJ , d Release, 12(1991)105.

] [5 HUSSAIN , GANM. , , RATHOR,J. , M.N., Preparatio f Controlledo n - Release Formulations of 14c_ Labelled Thiobencarb Herbicide and Study of their Environmental Behaviour, Pestic. Sei. 34(1992)341.

] [6 , HUATCHINC. . L , P.T., WENG, C.K., Nutrient Recycling Through Utilization of Palm Oil Mill Effluent, PORIM Int. Palm Oil Conf. Agriculture (1991)261.

[7] HASHIM, M.T., Treated POME as Nutrient Source for Oil Palm, PORIM Int. Pal l ConfmOi . Agriculture (1991)244.

214 EFFICACY OF THE ALGINATE CONTROLLED-RELEASE FORMULATION (CRF) OF THIOBENCARB HERBICIDE WEER FO D CONTRO TRANSPLANTEN LI D RICE

H.R. SOLTAN Pesticide Chemistry Department, Facult f Agricultureyo , University of Alexandria, Alexandria, Egypt

Abstract Field trial was conducted during the summer season of 1992 at the Agriculture Research Station, Alexandria Universit determino yt e effec th ef thiobencar o t b formulations (alginate controlled-release formulation and granular formulation), their combination, application times and rates of application on weed control and on economics of transplanted rice. Thiobencarb at 0.75 or 1.5 kg/ha was applied twice, 1 day before transplanting and 6 days after transplanting. The differences in response among weeds and rice to thiobencarb formulations, their combinations, application times and rates were variable. All of the herbicidal treatments as e hanwelth ds a l weede gave don e superior yields over un-weeded control. Controlled release thiobencarb formulation applie dday6 s after transplantin mose th kg/h5 s t1. effectiv t awa ga e treatment up to 40 days after transplanting, followed by the same formulation applied 1 day before transplanting, base ween do d density, weed heigh weed weigh y an t provide i d dr d tan e dth highest grain yield.

INTRODUCTION loss by evaporation and degradation. Ishikawa et al Weeds still cause about 15% reduction in (1977) have shown that thiobencarb was rapidly yield and quality of rice each year and reduced lost from aqueous solution expose sunligho dt d an t efficiency of production inputs (Smith, 1977). only 12.4 e originath % f o l concentratioe th f o n Among troublesome annual weeds in rice (Oryza herbicid s recordewa e d after 12h n alternat.A e saüva L.), in Egypt are barnyardgrass (Echinochloa metho f o reducind e environmentath g l loss, crus-galli (L.) Beauv.), Jungle rice (Echinochloa extending the effective period of control and d umbrellcolonuan ) L. ma sedge (Cyperus reducin e potentia th gn targe no f to l injuro t s i y difformis L.). The use of herbicides is a major regulate the bioavailability of the active ingredient factor contributing to high yields of rice in Egypt. through the use of controlled-release formulations Herbicides such as thiobencarb (S-4-chlorobenzyl (Schreibe , 1988)al t e r . Alginate gels have been diethyl thiocarbamate) are used to control use s matricea d r preparatiofo s f controlledo n - graminaceous weeds, particularly barnyardgrass. release formulation f herbicideo s s (Soltan, 1991, Barnyardgras n averaga t a s e densit f o 10.y 7 Hussain et al, 1992). The present investigation was plant/m a ric n ei 2 reducy fielma d e yields a y b s undertaken to evaluate the efficacy of alginate (Hill% muc25 , s h1984)a . controlled-release formulation of thiobencarb for Generally, the effective period of control weeds weed control in transplanted rice. Emphasis was on in rice can depend on how readily a herbicide is the influence of alginate formulation on the depleted through volatilization, leaching, sorption economic f o transplantes d rice production by organic matter, absorptio y plantsb n r o , comparable to conventional granular formulation chemical, photochemica d an microbial l when it were applied at different times and degradation (Kearny, 1977). Herbicide different rates. formulations containing thiobencarb, mostln i y granular form which quickly release the herbicide Material Methodd san s into water and the released herbicide is subject to e experimenTh e Agricultura s donth wa n ti e l Research Station, University of Alexandria at Abis area, Alexandria, Egypt durin summee gth r season Research carried out with the support of the IAEA under of 1992soie Th l . under this claa stud s y ywa loa m research contract No'6958/RB with 1.42% organic matter, 39% sand, 26% silt

215 and 35% clay in top of 15 cm horizon. The soil PH when applied at a rate of 1.5 kg/ha 6 days after s 1.9 . mohs/cmwa 5 m C E. e .th Gizd an a wa7 7. s transplanting (DAT). This closely followee th y db 17 1higa h yielding shord an , t duration variets ywa combination of CR formulation plus granule used. Thre fou o eday-olt 1 2 r d rice seedlings were formulation (0.75 kg/ha + 0.75 kg/ha) when it was transplante aparthillr m e dc pe ,0 Th .space 2 x 0 d2 applie6 DAT t a e d controlle.Th d release experimen a randomize n i s t laiou dwa t d block formulation of thiobencarb at 0.75 kg/ha was the design with 12 treatments replicated thrice. The least effective l formulational (50.04% e th f o ) s treatments included were: (a) no weeding, (b) hand during the early stages of crop growth when the weed [20 and 40 days after transplanting PAT)], formulatio y beforda s 1 applieewa n t a d (c) thiobencarb herbicide [formulate s alginata d e transplanting (DBT). Durin e lateth g r stages controlled-release formulation (CRF granulad )an r thiobencarb (CRF) at higher doses (1.5 kg/ha) formulation (G)j was applied at 0.75 or 1.5 kg/ha maintained their effectiveness up to 40 DAT and beforony eda e transplantin DBT1 dayg6 ( d s an ) gave the highest total weed control efficiency after transplanting (6 DAT), and (d) conventional (79.91%-82.81%) more than both the conventional granule formulation plus CRF of thiobencarb at a granule formulation (69.08-72.7%) and the hand rat 0.7f eo 5 kg/h 0.7a+ 5 kg/hd aan applieT DB d1 weeding twice (Tabl. e2) 6 DAT. This is consistent with the expected delayed Observation of Echinochloa crus-galli, K herbicidal effect of the controlled-release colonum and Cyperus difformis were recorded formulation e higheth , whico t r n e turi h du n (from 1.0/m2 area whic permanentls hwa y marked concentratio f thiobencaro n b availabl n watei e r DAT0 4 d .an plott 0 2 ne t )a e inth than the corresponding levels for the conventional Durin growine gth g season, weed samples were granule formulation. randomly taken from each plot by using 50x50 cm e datTh a indicated thae combinatioth t f o n quadrates after 55 days from transplanting. The dry granule formulation Plu formulatioR sC n coult dno weight was determined after drying in an oven at provided more weed control comparable to the 48hr 100fo .C ° Wee d contro evaluates wa l e th n di granule formulation alon t morebu e effective that form of (I) average weed height, (II) weed density, e hanth d weeding twice, base n totao d l Weed (III) weed control efficiency (%) comparable to the control efficiency (Table 2). un-weeded ckec d (IVan k ) percentage reduction The result obtained in this study are in weighy dr f e eaco t(%Rth hn i )individua l species agreement with that obtaine Krishnamurthy db t ye of weeds as well as total weeds. al (1983) who reported that thiobencarb at 2.0 Some of the agronomic character of rice, at kg/ha applied within 3 to 5 DAT gave good control harvest, were recorded, e.g. grai strad nan w yields of weeds including Echinochloa Spp (70 to 90%). (tons/ha), numbe f panicleo r r unispe t area, grain numbe r paniclpe rd 1000-graian e n weight. Chemical analysi f Proteio s n conten s carriewa t d Weed height e effec Th f o :thiobencar t b out accordin e methoth g f A.O.A.Co d . (1975). treatment weee th dn so plan t heigh showe ar t n ni Data were statistically analyzed accordino t g (Tabl . Thes3) e e results show that weed plant Snedecor and Cochran (1967). heigh s moswa t t significantly l effectedal y b thiobencarb treatments comparabl e controth o t e l RESULTS AND DISCUSSION during the early stage of crop growth (20 DAT). Weed density: Data given in table (1) In this case, however, granule formulatiof o n indicated that the Echinochloa crus-galli and & thiobencarb applied 1 DBT, CR formulation Colonum were dominanth e t weeds, their applied 1 DBT and 6 DAT and the combination percentage f o infestatios n were 50.71d an % formulations seemT e appliemorb DA o et s 6 d 32.56% respectively f 83.27%o m , su wit, a h effective in reducing Echinochloa crus-galli. E whereas the Cyperus difformis was 16.73 colonu d Cyperuman s difformis height thae th n percentag infestationf eo . other treatments correspondine Th . reductiog% n As for the doses of thiobencarb used, higher . crus-gallE e inth i heigh thesr tfo e treatments were dose f eacso h formulation provided more effective 39.17%, 36.63% , 37.6 38.09d 3an % whereae sth control thadosew nlo s bot termn hi totaf so l weed respective value r Efo ,s colonum were 36.46%, density and the effect on individual species (Table 35.29%, 35.89% and 35.6 respectively and for weee Th d . contro1) l efficienc population yo n basis Cyperus difformis showed 100% reduction of it. over un-weed contro maximus wa l case th f eo mn i Durin e lategth r stage f croso p growt DAT0 h(4 ) thiobencar 5 kg/h1. t aa b formulate s granulaa d r thiobencarb formulated as CR formulation at the T DA 6 d an formulatioT n DB boti 1 ) h (G n higher rate (1.5 kg/ha) reduced Echimochloa Spp. treatments durin earle gth y stage f croso p growth height significantly more e thaotheth n r 0 DAT)(2 . Thes r ewit pa wer hn o ethiobencar b thiobencarb treatment t bot a srat w helo higd han formulated as controlled-release formulation (CRF) levels.

216 Table 1: Effect of Time of Application, Rate and Thiobencarb Formulations on Weed Population (No nr2) m Transplaned Rice

Time Rate Echinochloa crus-galli Echinochloa colonum cypems difformis Total weeds Treatment applied (kga.i/ha) 20 DAT 40 DAT Mean 20 DAT 40 DAT Mean T 20DA 40 DAT Mean 20 DAT 40 DAT Mean

CRF1 1DBT3 075 466 1366 9160 0 16 366 6 6 6 8 200 700 4500 1032 2932 19820 G2 1 DBT 075 400 1533 9660 333 933 6330 1 66 800 4830 0899 3266 20820 CRF 1DBT 1 50 233 0733 4830 0 83 232 3 3 3 3 000 233 1 165 0466 1299 08825 G 1DBT 150 1 66 1066 6160 0 2083 3 0 6 56 000 300 1500 0366 1999 11825 CRF+G 1DBT 0 75+0 75 266 1033 6495 233 6 00 4 160 000 333 1665 0499 1966 12325 CRF 6 DAT4 075 400 1233 8165 5 49 234 3 6 6 6 166 566 3660 0799 2489 16440 G 6 DAT 075 433 1566 9995 266 8 00 5 330 166 666 4 160 0865 3032 19485 CRF 6 DAT 150 166 0633 3995 0 33 2 166 0 0 3 033 200 1 330 0365 1166 76550 G 6 DAT 1 50 200 1066 6330 0 83 12 33 3 43 000 266 1330 0333 1765 10490 CRF+G 6 DAT 075+075 233 0933 5830 166 4 66 3 160 000 300 1 500 0399 1699 10490 Hand weeding 20 & 40 DAT — — 866 1000 9330 666 4 00 4 830 300 266 2830 1832 1666 17490 Un-weeded control — — 1000 3400 2200 733 1766 12495 3 33 1300 8165 2066 6466 42660

% of weed infestation 4867 5276 5071 0 56 2 3753 7 6 5 7 2 1354 1991 1673 10000 10000 10000

..... L. S. D0 05 1 86 0639 1313 0515 1 50 0530 ......

1 CRF = 7 56% Controlled-release formulation of thiobencarb granul% 5 = eG 2 formulatio thiobencarf no b Day= T s beforDB 3 e transplanting 4 DAT = Dnys after transplanting to to l—* 00

Table 2: Effect of Time of Application, Rate and Thiobencarb Formulations on Weed Control Efficiency

Time Rate Echinochloa crus-galli Echinochloa colonum Cypcrus difformis Total weeds Treatment control efficiency (%) control efficienc) (% y control efficiency (%) control efficienc) y(%

applied (kga.i/ha) T 20DA 40 DAT Mean T 20DA 40 DAT MeanT DA 0 2 40 DAT Mean T 20DA T 40DA Mean

CRF1 1DBT3 075 5340 5982 5836 5006 5096 5070 03993 46 15 4488 5004 5465 5339 G2 1DBT 075 6000 5491 5609 5447 47 16 4933 050 15 3846 4084 5648 4948 51 19 CRF 1DBT 1 50 7670 7844 7804 6821 81 14 7735 10000 8207 8573 7744 7991 7931 G 1DBT 150 8340 6864 7200 7271 6795 6934 10000 7692 8162 8228 6908 7228 CRF+G 1DBT 0 75+0 75 7340 6667 7003 6821 6602 6670 10000 7438 7960 7584 6959 71 10 CRF 6 DAT4 075 6000 6373 6288 6821 6228 6402 05015 5646 5517 6132 6150 6146 G 6 DAT 075 5670 5394 5456 6371 5469 5734 050 15 4876 4905 58 13 53 10 5432 CRF 6 DAT 1 50 8340 81 38 8184 7735 8301 8135 09009 8207 8371 8233 8281 8205 G 6 DAT 1 50 8000 71 58 7579 8185 7548 7735 10000 8461 9230 8388 7270 7541 CRF+G 6 DAT 0 75+0 75 7670 7255 7350 73 13 7361 7470 10000 7692 8162 8068 7372 7541 Hand weeding 20 & 40 DAT — — 0000 7058 5759 0000 7734 6134 00000 7953 6533 0000 7423 5900 Un-w ceded control — •— " 0000 0000 0000 0000 0000 0000 00000 0000 0000 0000 0000 0000

1 CRF = 7 56% Controlled-release formulation of thiobencarb granul% 5 = eG formulatio 2 thiobencarf no b Day= T s beforDB 3 e transplanting Day= T s afteDA 4 r transplanting Table 3: Effect of Time of Application, Rate and Thiobencarb Formulations on Weed Plant Height (cm ) in Tranplanted Rice

Time Rate Echinochloa cnis-galli Echinochloa colonum Cyperus difformis

TVi1 * Aff*I HIHet t mI l P n f applied (kga.i/ha) 20 DAT 40 DAT Mean T 20DA 40 DAT Mean T 20DA T DA 0 4 Mean

CRF1 1DBT3 075 3800 7333 5566 4152 80313 60916 1500 4825 31620 G2 1DBT 075 3775 7733 5754 41 16 870905 12 4 6 1500 5358 34290 CRF 1DBT 150 33.83 61 18 4750 3751 73070 55290 0000 3950 19750 G 1 DBT 150 3250 6984 51 17 3683 82920 59870 0000 47.83 23910 CRF+G 1DBT 0 75+0 75 35 11 71 29 5320 3933 83 119 61 220 0000 4766 26305 CRF 6 DAT4 075 3762 7070 54 16 4063 80480 60550 2200 4403 33010 G 6 DAT 075 3755 7440 5597 4025 88250 64250 1866 5033 34500 CRF 6 DAT 150 3333 5920 4626 37 16 71 440 54300 1500 3666 25305 G 6 DAT 1 50 3305 6560 4932 3733 78020 57670 0000 4466 22500 CRF+G 6 DAT 0 75+0 75 3553 6683 51 18 3770 80910 59300 0000 4633 23 166 Hand needing 20 & 40 DAT — — 51 10 61 33 56 14 5460 60660 57 130 5600 3666 45910 Un-w ceded control .... — - 5339 91 52 7245 5797 10482 81380 5533 7049 62910

JU»O. JL/ n ne — — 04 79 0233 05 770 03370 16 56 06760

1 CRF = 7.56% Controlled-release formulation of thiobencarb 2 G = 5% granule formulation of thiobencarb 3 DBT - Days before transplanting 4 DAT = Days after transplanting

to •3 There wer o significann e t differences among formulatio e loweth t ra n rate produce e leasth d t the thiobencarb treatments and also between grain yield Other treatments gave yields between thiobencarb treatments and hand weeding twice on these results reducin heighe gth Cyperuf to s difformis Yield component particularly panicle number/m2, grain number/panicle, weigh f 100o t 0 Wee weighty ddr date TablaTh :m 4 shoe w gram and straw yield were substantially influenced a clea rweighy e weereductiodr th f e do tth n i n some what by the various thiobencarb treatments tested under various thiobencarb treatments m All of the thiobencarb treatments produced panicle comparison to the un-weeded check The dry number/m2 and straw yield significantly more than weigh t excee no f wee o td ddi d zero undee th r e un-weedeth d control Amongs e thiobencarth t b effective treatment treatments formulatioR C , n applie highee th t da r Concernin e percentagegth f eradicatioo s f no rate produced the highest panicle number/m2, gram weed in nee field, it was clear that the effectiveness number/panicle, weight of 1000 gram and straw of suppression was altered by changing the time of yield Also e lateth , r treatment improvee th d application, application rate and formulation type protein content of grains This might be due to of thiobencarb herbicide (Tabl ) However4 e e th , more weed control leadin increasn a o gt yieln ei d conventional granule formulatio s generallwa n y components There was no significant difference less effectiv reducinn i e e wee weighgth y ddr s a t between the treatment with CRF alone at 1 5 kg/ha compared with the controlled-release formulation and its combination with the granule formulation e mosTh t dramatic effec f thiobencaro t n o b in yield components (Tabl) e5 reducin weighy dr E crus-gal f e o tgth 16%9 h(8 ) In general higher doses of herbicides favoured was caused at the higher rate by CR formulation more yielth e de th components r fo t bu , when it was applied at 6 DAT This was closely conventional granule formulation applieT DB 1 d followed by the same formulation (15 kg/ha) no significant differences betweew nlo higd an h applied at 1 DBT, which in turn was followed by dos s a wel es a higl e hgranul th dos f o e conventional granule formulation (1 5 kg/ha) and kg/h5 7 formulatio0 aF CR d nan applieT DB 1 d granule formulation Plus CR formulation (0 75 applied either before or after transplanting were kg/ha + 0 75 kg/ha) when both were applied at 6 observed (Table 5) e minimuTh T mDA effec n reducini y t dr e th g Once again, it is worth noting that nee gram weight of the total weeds (51 37%), as compared yield could be improved by using CR formulation with the control was shown by the granule of thiobencar kg/ha5 1 t ba , particularlT DA y6 formulation at a rat of 0 75 kg/ha when it was T applieDB 1 t da These results contradict with those obtained by Bnsht et al (1987) who reported that thiobencarb at 1 5 kg/ha controlled weeds effectively and the biomas reduces % swa 42 y db It is worth mentioning that application of CR T formulatioDA kg/h5 6 7 r o 0 at T na eithe DB 1 r gave weigh y reductiodr E crus-gall f e o tth n ni i which was statistically insignificant with the granule formulatio f thiobencaro n e higheth t ba r rate when it was applied at 1 DBT

Grain yield yielan d d components: Gram yields obtained from various thiobencarb treatments as well as those of the hand \\eeded ones, were significantly higher than e thosth f o e un-weeded plots (Tabl ) Maximue5 m grain yield 5 t/ha (R s 579 recorde C e caswa ) f th o e n i d formulation of thiobencarb at 1 5 kg/ha when it was applied at 6 DAT It was significantly different than the other treatments, but the performance of the CR formulation applied 1 DBT was not significantly different than that of CRF applied 6 DAT Such results are quite logic by referring to the data of Table (4), it seems that there were significant negative correlation between gram yield d weey an weighdr d t Hence e granulth , e

220 Table 4: Effect of Time of Application, Rate and Thiobencarb Formulations on Dry Weight (gm) of Weeds/Square Meter, 55 Days after Transplanting Rjce

Time Rate Echinochloa crus-galli Echinochloa colonum cyperus difformis Total needs Treatment applied (kga,i/ha) M. %R5 wt. %R wt. %R tvt. %R

CRF1 1DBT3 075 38735 5557 17089 5548 2073 6483 57897 5596 G2 1DBT 075 41952 51 88 197 18 4863 2261 6251 63931 5137 CRF 1DBT 1 50 13097 8497 8799 7708 1448 7543 23344 8224 G I DBT 1 50 21559 7527 12770 6673 1594 7296 35923 7267 CRF+G IDBT 75+0 5 07 26023 7015 132 10 6559 0837 8580 40070 6952 CRF 6 DAT4 075 32747 6243 15949 5895 1351 7708 50047 6193 G 6 DAT 075 37345 57 16 7 1 2 18 5259 1290 78 11 56852 5675 CRF 6 DAT 1 50 9445 8916 88 14 7704 527 9106 18257 8611 G 6 DAT 1 50 16735 8080 11026 7127 499 9135 2826 7850 CRF+G 6 DAT 075+075 173 16 8013 12403 6769 974 8393 30693 7665 Hand weeding 20 & 40 DAT — — 16996 8051 112 19 709 1023 8264 29238 7775 Un-wccdcd control .... — 87183 0000 38391 0000 5895 0000 131469 0000

L.S.D00, — — 241 80 .... 13089 .... 1488 .— — - ....

1 CRF = 7 56% controlled-release formulation of thiobencarb granul% 5 = eG 2 formulatio thiobencarf no b 3 DBT = Days before transplanting Day= T s afteDA 4 r transplanting 5 %R = Percentage of reduction

to K) s

Tabl : Graie5 n Yiel Yield dan d Component Trasplantef so d Ric Affectes ea Timy db Applicatonf eo , Rat Formulatiod ean n Type f Thiobencarso b Herbicide

Time Rate Grain yield Panicles Grains 1000 Grain Straw Yield Grain Protein Treatment applied (kga. i/ha) (t/ha) (No. Im z) (No. /panicle) weight (g) (t/ha) Content (%)

CRF1 1DBT3 075 4810 24500 11868 2412 09361 9280 G2 1DBT 075 4600 23500 11344 2403 09 187 9 150 CRF 1DBT 150 5460 27700 13850 2483 10306 9610 G 1DBT 150 5003 26100 12720 2434 10047 9520 CRF+G 1DBT 0 75+0 75 4993 25600 12450 2411 09973 9480 CRF 6 DAT4 075 4930 24300 11735 2422 09691 9360 G 6 DAT 075 4727 24000 11434 2410 09540 9230 CRF 6 DAT 150 5795 28500 14058 2523 10649 9720 G 6 DAT 150 5 173 26800 12926 2463 10296 9630 CRF+G 6 DAT 0 75+0 75 5 106 26000 13126 2419 10211 9560 Hand weeding T DA 0 4 20& —— —— 4965 25400 13350 2393 09865 9400 Un-w ceded control ------...... 3803 20900 11070 2380 08372 8990 ion Iv.S.U QQ5 0558 23201 11251 0996 01 059 0096

ControIIed-releas% 56 7 = F CR 1 e formulatio (hiobencarf no b 2 G = 5% granule formulation of thiobencarb 3 DBT = Days before transplanting 4 DAT = Days after transplanting ACKNOWLEDGMENTS 5. Kearny, P. C., 1977. A Challenge for The Author wish to thank the International controlled-release Pesticide technology. Atomic Energy Agency authoritie r supportinsfo g Page . SchersB 30-3 . . ControlledH ed , n 6i - the work and Faculty of Agriculture, Alexandria release Pesticides. ACS. Symb. . Am Ser . 53 . University, EGYP r providinfo T e researcth g h Chem. Soc. Washington. ,DC facilities. Thanks are also due to Dr. Manzoor 6. Krishnamurthy, K.; Prased, T.V.R and Hussain Head, Agrochemical unit, Joint Kenchaih . 1983K , . Integrated Weed control IAEA/FAO program, Agency's Laboratories for in transplanted rice. Proc. Asian-Pacific providing the controlled-release formulation of Weed Sei. Soc. Conf. 9:349-356. thiobencarb herbicide. 7. Official Method f o e Analysisth f o s Associatio f officiao n l Analytical Chemists (A.O.A.C.), 1975. Edited by William Horwitz 12th Edition, Washington, USA. REFERENCES 8. Schreiber, M. M.; Wing, R. E.; Scasha, B. Brish, P s.; Pandey, P. C. and Lal, P., 1987. , d White1988D. San .. M ., Bioactivit f yo Agronomi d economian c c evaluatiof o n controlled-release Formulations of herbicides in transplanted rice. International herbicide starcn i s h encapsulated granules. Rice Research News letter (Philippines)2 1 , Proc. Int. Symp. Control. Rel. Bioact. (2), 36-37. Matter, 15,223 PP. , 1984HillM. . J , . Weed riced . 63-67san P . . 9. . SmithJ Jr. . R ,, 1977 . Comparisonf o s Proc. California Weed, th Conf. 6 3 , herbicide treatment r weefo s d contron i l Sacramento A 1980C , . California Weed rice. Agricultural Research Service United conference, El-Macero. States Departmen f Agricultureo t , Report Hussain ; d RathorGanan M. , . J ,, M.N., series 233. 1992. Preparatio f o Controlled-releasn e 10 . Cochrand Snedecor , an 1967 . G. . W . W ., G , Formulations of 14C-labelled thiobencarb Statistical Methods. John Wile and Sons, herbicid studd an ef thei yo r environmental Inc. N.Y. behavior. Pestic. Sei. 34: 341-347. 11. Soltan, H. R. 1991. Environmental behavior Ishikawa ; d K. Nakamuraan , . Y , of the Slow-release Formulations of Kuwatsuka, S., 1977. Volatilizatiof o n butachlor herbicide . ThesisD . Ph ., Faculty thiobencarb herbicide from aqueous solution of Agriculture, Alexandria University, soild Pestic. an J . . SCI 49-57: 1 . . Alexandria, Egypt.

Next page(s) left blank 223 EFFICACY EVALUATION OF CONTROLLED RELEASE FORMULATION THIOBENCARF SO RICN BI E

C.M. BAJET, L.E. FABRO National Crop Protection Center, University of the Philippines at Los Banos, College, Laguna, Philippines

Abstract

Field experiments were conducted from 1991-1993 to evaluate the efficacy of controlled release thiobencar t farmera be U.P.L.Bth st a fiel .d an d Experiment Station. These were evaluated under irrigate d rainfean d d conditions durind gan bott we h dry seasons. Alginate (TAL 2490) and corncob (TA1 and TA2) formulations of the thiobencarb were evaluated at the ai/hg K 5 arat 1. unde f eo r rainfed paddyy durindr e th g season d significantl,ha 19911 TA L 249.d TA an 0 y greater number of tillers/hill compared to the commercial formulation. However e seeth ,d weighr pe t n paniclete s wert significantlno e y different between e controlleth d release formulationse (CRFsth d an ) commercial formulation. Weed density and fresh weight of the CRFs against Echinoloa sp. at the 40 days after application (DAA) were significantly lower than the commercial formulation. Efficacy evaluatio f corncoo n b (TA d TA2d an 1an ) alginate (TAL 26391) thiobencar d 2,4an bD kraft lignin formulations at 3 rates of application were done during the wet season, 1991. Visual assesment of weed control A showeDA o significan5 n d4 d an 5 3 t t a difference between commercial and CRF, fast and slow, corncob and s mixturalginateit r o e F witCR , h 2,4D. Kraft lignin e least 0.7th a ts 2 2,45 g wa K TA effective D d an 1 TA . ai/ha had significantly greater yield over the untreated control. o alginatTw e thiobencarb formulations (TAL 070192 an a localld y produced formulation L 0892TA , ) were evaluated at 2 rates and 2 times of application during the wet season, 1993. Total number of weeds at 40 days after transplanting (DAT) showe a trend d advantagf o e the CRFs when applied at 1 DAT at the rate of 0.75 Kg ai/ha. Trend also indicated that efficacy decreased when the CRFs were applied at 4 DAT but the delay coul offsee b d usiny tb g higher application5 1. rat f eo Kg ai/ha. These indications were not reflected on yield and yield component ricef so .

225 INTRODUCTION

The total land area devoted to rice production in the Philippine e tota th s aboui smillio2 f l o 3. t % . whic23 nha s i h land devoted to agriculture (PhilRice,1992).This area is distributed to about 52% irrigated, 44% rainfed and 4% upland rice. This could therefor dividee eb d into 1.65, 1.4 d 0.10an 4 r irrigatemilliofo . ha n d lowland, rainfed lowlan d uplanan d d rice, respectively y season.dr Durin e , th gonl y abou1 t million hectares (0.8 irrigated arid 0.2 rainfed) are planted to a second crop season of rice (Mabbayad et al.,1983). Thus, the actual area planted for the two cropping seasons total to 4.2 millio. nha In 1980 average ,th e rough rice yiel 2.85s dwa , 1.7d 4an 1.0 tons per hectare for irrigated lowland, rainfed lowland and upland rice, respectively with an average of 2.16 tons/ha (Mabbaya t al.,1983)e d n experimentaI . l plot f researco s h institutions like the International Rice Research Institute where condition e ideal ar se averag th , e8 tons/ha yiel6- s i d . Recently e yieldth , s have been steadily increasin n farmeri g s fields with an average of about 3»4 tons/ha yield for irrigated lowland paddy (PhilRice, 1992). However ,deceivine b thiy sma g since the area devoted for rice is steadily shrinking as farmers are shifting to more profitable crops or are converted to residential or industrial lands. With this situation there ia compelins g nee o increast d e rice productivit r unipe yt area. The presence of weeds could decrease yield up to 40% dependin e timd duratioth an en o g n whee wees th i nd present. Zimdahl (1980) reported that 40 Echinocloa. crusgalli/sq t a m7-4 0 days after emergence could reduce yield by 40% whereas Mercado (1979) reported that competition against transplanted ric s greatesi e t. E. whee densit0 th n2 s i y crusgalli/sqm wit criticaha l periodays0 4 f .do Usually there is only one application of the herbicide during the crop seaso t thi bu s noftei s n supplemented with handweedinr o g rotavation. Handweeding has been estimated to be 144 man days per hectare and costs being higher in direct seeded and upland rice (Mabbayad et al.,1983). Mechanized weeding using a rotary weeder is possible when rice is transplanted in rows by incorporatin e growinth g g weeds inte soilth o . However, weeds growing parallee directio th e weede th o d t thos lf an ro n e growing nearice rth e plan sparee tar thiy b d s method. Herbicide usage has been steadily increasing as farm labo s becomini r g more expensive r transplanteFo . d irrigated crop where submergenc controllede b n eca , preemergence granular herbicide is convenient and where soil moisture is 80-100% field capacity, liquid formulation e morar se effective pr s a e emergence treatment (Mabbayad et al., 1983). In the Philippines ,8 recommende 1 ther e ar e d herbicide formulations for transplanted rice with 9 different active ingredients.There are no controlled release formulation of any pesticide availabl e marketth n i e. Sprayable e ar mors e popular over granuleo addet e d du a sresul cos s a f tproductioo t d an n transportation.

226 Thiobencarb (S-4 chlorobenzyl diethyl thiocarbamatea s i ) commonly used recommendean d d pre d earlan - y post-emergence herbicid r fo controe f weedo l n direci s t seeded an d transplanted rice. However, thiobencar easils i b y lost through photodegradation and co-volatilization with water (Ishikawa et al.,1977; Hussei f controllet al.,1992)o e n e us e dTh . release formulation (CRF) is an excellent alternative approach in reducing environmental losd contaminationan s t alsI . o takes care of mid season weed growth as a result of irregular emergence of weeds. e eas Th f incorporatioo e f bioactivo n e materials makes e alginatth l procesge e r granulafo s r productio f CRFo n s useful.Connick et al.(1984) investigated the herbicide dichlobeni alginatn i l e formulation y varyinb s g kaolin content and relatin releaso t t gi e rate granuld san e yield, Pfistet e r . (1986al ) also usee alginatth d e matri o investigatt x e th e herbicides monolinuron, desmetryn, chlordazon, atrazine, MCPB, simazine and chloroxuron. The factors affecting the permeability from calcium alginate films were studiey b d varying immersion time d calciuan s m acetate concentrations (Julian et al.,1988). It was suggested that short immersion times could result in permeable calcium alginate film. The applicability of calcium alginate formulations to butachlor (Hussein and Oh, 1991; Soltan,1991) and thiobencarb (Hussein et al.,1992) were likewise studied.

METHODOLOGY I. Small plot evaluatio F thiobencarCR f o n b against Echinocloa unde. sp r rainfed paddy conditions. January-April, 1991 Field plots measurinm werl farmer'n x ei m lai t gl ou d s fiel Maahast a d s BanosLo , , Laguna. Ten-da d ricol ye seedlings cv. IR-66 were transplanted and herbicide application at 1.5 Kg a.i./ha s madwa . e when seedling . glabrescenE f o s d an s E. crusgalli were at 1 to two leaf stage. Three controlled- release formulation f thiobencaro s b were evaluated namel: y TA1 (corn cob based, 4% ai, fast release formulation) , TA2 (corn cob based, 4% ai, slow release formulation) , TAL 2490 (alginate based a ,commercia d 6.23an ) %lai granular formulation (Saturn G). The study was conducted during the dry season (Januar - Aprily , 1991) following randomized complete block design (RCBD) with three replications. Parameters assessed were crop injury at 7, 14 and 21 days after application (DAA) through morphological symptoms sucs a h leaf chlorosis, leaf tip burning and death of the plants expresse n percentagi d e compared wite controlth h . Percentage grass control was evaluated at 7, 14, and 21 DAA. Weed counts and fresh weigh grasf to s weeds were determineusinA DA g0 4 t da a sample quadrat of 0.5 x 0.5 m. Height of rice plants was measurenumbee th tillerf d ro d an s were counted fro hill4 m t sa the center of each plot at 40 DAA. Before crop harvest, the numbe productivf ro e tiller hilr weighd spe lan grainf to s from panicle0 1 s were determined froe samplth m e quadrat. Data were subjected to statistical analysis and means were compared using leve% 5 DMR significancef t lo Ta .

227 II. Efficacy evaluation of CRF thiobencarb at different rates of application under irrigated, rainfed paddy conditions, August-November, 1991. Field plot a m wers 2 n measurinei x lait m ou d 2 g farmer's field at Sto. Domingo, Bay, Laguna following the randomized complete block design (RCBD) with three replications e experimenTh . s donwa te durin e 199th t g 1we season (August - November, 1991) and the following treatments were evaluated.

Treatments Rate (Kg ai/ha) 1. TA1 3.7% thiobencarb 0.75 2. TA1 fast release, corn cob 1.00 3. TA1 1.25 4. TA2 4.1% thiobencarb, 0.75 2 sloTA w release. 5 , corb nco 1.00 2 TA . 6 1.25 7. Saturn G 9% thiobencarb, 0.75 . 8 Satur nG commercia l granular 1.00 9. Saturn G formulation 1.25 10. TAL 26391 7.4% thiobencarb 0.75 2639L TA 1 . alginat11 e based 1.00 L 2639TA 1. 12 1.25 13. 2,4 D Kraft Lignin based 0.75 F CR , D 4 2, % 50 D 1.04 2, . 14 0 15. 2,4 D 1.25 . Saturthiobencar% 16 4 nD % b+2 0.75 17. Saturn D 2,4 D, local EC formulation 1.00 18. Saturn D 1.25 19. TA1 + 2,4 D 0.75+0.4 20. TA2 +2,4D 0.75+0.4 21. Saturn G+2,4D 0.75+0.4 22. Untreated control 23. Handweeded contro Handweedinl- g twice withiT DA 0 n4

The plots were constructed in pairs and were surrounded by dikes to minimize cross contamination and facilitate management of each plot. Fertilizatio N/hg k s 0 aspli5 wa f no t into basal applie t lasa d t harrowinp dressinto d gan g applie t bootina d g stage. Fourteen d daydapool s g seedling f o IR7s 2 were transplanted on August 24, 1991. To ensure weed infestation of at groun f o leas % d25 t cove t 15-4a r 0 days after transplanting (DAT), seed f Monochorio s a vaginalis. E_ . d Cyperuan . sp s crusgalli wer ey befor sowda 1 ne transplanting. Applicatiof o n weed seeds was done weekly from 2 DAT up to 40 DAT. The herbicide sd croan wer pT e DA injurapplie 2 yt a ddat % ( a necrosis % , survival % stunting, ) were evaluate t 7,15,3a d 0 d weeDAAan d , A countweeDA 5 d d4 san contro , 35 , l 20 dat t a weights at 45 DAA. Yiel calculates dwa hectarr pe n do e basis.

228 III. Field efficacF thiobencarCR f o y t differena b t rates and time f applicatioo s n under irrigated rainfed conditions, April-August, 1993 plotm 2 sX werm 2 e e constructeTh blocksn i d , surrounded by dike d enclosean s n plastii d c sheet o minimizt s e cross contamination. Fertilizatio f 90-15-1o ns wa g . N-P-k 5ha r Kpe split into basal applie t lasa d t harrowing t maximua , m tillering .an p dressindto g applie t panicla d e initiation. Fertilizer equivalent to 30-15-15 Kg NPK/; was applied basally usin mixtura g completf eo e fertilize uread ran . Fourteen d dapodayol sg seedling . UPLRi-1cv f o s0 were transplanted on April 20, 1993. Natural weed infestation was allowed to occur and the following treatments were evaluated: Treatments Rate Time of Kg a.i./ha application (DAT) 07019L 1.TA 2 0.75 l 2. TAL 0892 0.75 l 3. Saturn G 0.75 l 07019L 4.TA 2 1.50 l 5. TAL 0892 1.50 l 6. Saturn G 1.50 l 07019L 7.TA Satur2+ nG 1.50 l 07019L 8.TA 2 0.75 4 089L 9.TA 2 0.75 4 lO.SaturnG 6.75 4 11.TAL 070192 1.50 4 12.TAL 0892 1.50 4 13.Saturn G 1.50 4 14.TAL 070192 1.50 4 15.Untreated control lö.Handweeded control 2X 0 4 d an 0 2

TAL 070192 and TAL 0892 are both alginate formulations with 7.5 5.1d 6an 3 percent active ingredient, respectivelye .Th forme s provide e wa Internationar th y b d l Atomic Energy Agency e anlate th s dproduce wa r d localle Nationath t a y l Crop Protection Center using locally available materials. e followinTh g data were obtained: densit 0 f weed2 o y t a s an0 DAT4 d , heigh d tillean t r numbe f riceo r , numbef o r panicles/sqm, number of grains/ panicle, 1000 grain weight, grain yiel d totaan d l biomass e harvesTh . t indices were calculated by dividing the yield with the biomass multiplied by one hundred.

RESULTS AND DISCUSSION I. Small plot evaluation of CRF thiobencarb against Echinocloa sp. under rainfed paddy conditions. Jan-April, 1991 Ten day old rice seedlings were not injured by thiobencarb applied at 1.5 kg ai/ha in any of the formulations used. Thiobencar s recommendei b r prefo d d earl-an y post-emergence control of grasses in transplanted and direct seeded rice.

229 Seedling . glabrescenE f so . crusgall E d san i treate1 t a d lea2 o ft stage showed yellowin f leavesgo , pale appearancd ean burning of leaf tips at 7 DAA. Intensity of leaf chlorosis and yellowing increased with time leadin o dryint g f leaveo g d an s death of seedlings. At 21 DAA, the effect of the herbicide was contro% 95 notes o t lwa eviden 5 d8 d (Tabltan . 1) e Plots treated with CRF thiobencarb had significantly lower weed count than the commercial formulation (Table 1) . However, plots treated with TAL 2490 had smaller weeds than the corn cob based CRF's. These weeds emerged later which indicated adequate herbicide presen soile th n .ti Fresh weights of Echinocloa sp. weeds in TA1 and TA2 formulatio 1 time1 nd swer an lowe 9 e r than thane thosth f o e untreated e othecheckth rn . O hand , fresh 9 timeweigh1 s stwa e alginatloweth n i r e formulation whil 4 fole d reductios wa n commerciae founth n i d l formulation (Tabl Result. 1) e s indicate thaL 249TA t0 extende e presencth d f adequato e e herbicide levele soith ln si tha t suppressed seed germination. Plant height measured at 40 DAA showed no significant differences among treatments (Table 2). However, plants in the untreated plots were slightly o competitiot talle e du r r fo n light with towering growtglabrescen. E f o h. crusgalli E d an s . A significantly greater number of tillers were found in the herbicide treated plots than the untreated check. Total productive tillers in comparison to the total number TA2, 1 ,TA r fo % 53 d an 6 7 , 79 , o 73 fwerT , tillerDA e 69 0 4 t sa L 2490TA , commercial formulatio d untreatean n d check, respectively (Table 2). Close to half of the tillers produced in the untreated check did not bear panicle or were dead at crop harvest. More seeds were apparently producee controlleth n i d d release treatment thae commerciath n l t formulationo s wa t bu n significantly different(Tabl e untreateTh . ) 2 e d controd ha l lowest seed weight due to uncontrolled weed growth.

Tabl % wee. I e d control, densit d weigh. crusgalliE an y f o t treated witF thiobencarbCR h , Jan .- Apr. , 1991 Treatment % weeds control Weed Density* Fresh weight; (DAA) (no./0.25 sqm) {gm./0.25 sqm) 0 4 1 2 4 1 7

TA1 15 55 85 90 3.0c 9.9c TA 2 18 60 90 95 3.0c 8.3c TAL2490 20 80 95 95 3.0c 4.8c Saturn 15 75 85 85 13.0b 21.3b Untreated 0 0 0 0 59.0a 91.4a

cv 24.8% 14.7%

* DatA a collecteDA 0 4 t a d

230 Table II. Effect of CRF thiobencarb applied at 1.5 Kg ai/ ha on agronomic characteristics of rice, Jan.- Apr., 1991.

Treatment Heigh* t Tiller * Productive Seed (cm) (no./hill) tillers«« Weight** per hill (gm/10 panicles)

TA1 72.5a 29.5a 20.5a 18.3a TA2 71.8a 26.7ab 19.7a 18.1a TAL 2490 72.2a 27.1a 21.3a 19.6a Saturn 72.0a 23.7b 18.2a 13.1a Untreated 75.0a 12.2e 6.5b 5.1b

cv 25% 8.9% 7.4% 6.2%

: data collecte 40DAt a d A !* data collected just before harvest

II. Efficacy evaluation of CRF thiobencarb at different rates of application under irrigated, rainfed paddy conditions, August-November, 1991. Neither necrosis, stuntin r herbicidno g e damags wa e observed on all treatments from 7 to 30 DAA. At 20 DAA all treatments provided good weed control except 2,4 D treated plot 5 DAA 3 sl treatment t al (Tabl,A . ) 3 e s were significantly better tha untreatee nth controe d vaginalisth . plo M r f lto fo , Cyperus sp., Fimbristylis. milliacea. Sphenoclea. zeylanica and Ludwigia. octavalvis. Weed control improved from 20 to 45 DAA which indicates sustainabilit f controo y l (Tabl . Visua) 3 e l assessment of weed control at 35 and 45 DAA showed no significant difference between commercia d CRFan l , fasd an t slow, corn cob and alginate based,. mixture with 2,4 D and pure thiobencarb. Considering total weed coun d weean td weighf o t all weed species, TA1 + 2,4 D, TA2 + 2,4 D, TA2 at 1.0 kg ai/ha, and TAL 26391 at 0.75 and 1.25 Kg ai/ha plots had zero weed growth (Table 4). The kraft lignin formulation of 2,4 D showed similar control with the untreated plot for E crusgalli.. majoe Th r wee dmilliacea. F foun s dwa . t significantlYielno ricf o s d ewa y different betweee th n rate applicatiof so n cose (TablTh t . effective5) e rate appears e 0.7tb o 5g a.i./haK . There wer o significann e t difference between the formulations mainly due to the variability brought d biran d t damagaboura y b te which were sever n somi e e plots. Variabilit o grait s alse nwa y du omaturit n effeca s f a o yt repeated transplanting e goldee becausdamagth th f f no o ee snail which was severe for some plots. The difference in weed number and weights were not therfore significantly reflected in yiele th d data. III. Field efficacF thiobencarCR f o y t differena b t rates and time applicationf so , April-August, 1993 Application of TAL 070192 at 1 DAT at the rate of 0.75 Kg ai/ha resulte n reducei d d densit f Echinocloo y . comparesp a d

231 Table III. Percentage weed controF treteCR f lo d plots, August-November.199a < 1

Treatments Rate Days After Application 5 4 (Kg ai/ha 5 )3 0 2

TA1 0.75 93 100 100 1.00 84 86 81 1.25 88 96 92 TA 2 0.75 85 92 92 1.00 100 100 100 1.25 100 100 98 SaturG n 0.75 98 99 96 1.00 8(5 94 91 1.25 100 100 99 1 TA39 6 L2 0.75 98 100 100 1.00 79 87 93 1.25 99 100 100 2.4D 0.75 62 82 85 1.00 72 82 87 1.25 54 78 80 SaturD n 0.75 100 100 99 1.00 100 100 100 1.25 100 100 98 + 2,4 1 DTA 0.75 + 0.4 10O 100 100 + 2,40T A2 0.75 + 0.4 1OO 100 100 Saturn G + 2.4D 0.75 + 0.4 92 95 95 Untreated - 0 0 0 Handweeded 2x 100 100 100

with applicatio T (Tabl DA Densit, . zeylanic) 4 S 6 e t f a o ny a am. milliaceF d e othe th t n rshono o a hand w di dtren d advantag earlief eo r application over tha f lateto r treatment. Nevertheless weeT showeDA d 0 densitd2 t significantla y y lower numbe l herbicid f weedal o r n i s e treatments compareo t d the untreated check. e heighTh f rico t et varno planty d wideldi s y among treatments and the tallest plants were found in plots that received a mixture of TAL 070192 and commercial formulation applied at 4 DAT (Table 7) . In terms of tiller production, highest significant tiller productio s notewa nd from plots treate ai/hg K d 5 a wit 1. e ratL 07019applie hth TA f eo t a 2t a d 4 DAT o significan.N t differenc s founwa e d however from other herbicide treatments compared wite besth ht treatments.The untreated check produced the least number of tillers. At 40 DAT total weed density in all treatments were significantly lower than the untreated check. Similar weed counts were obtained from plots treated witL 07019TA hd an 2 commercial formulatio t 0.7 a ng ai/h1 DAT K 5 t ,a a commercial

232 Table IV. Weed weight and count at 45 DAA

Treatment Rate Weed Count Weed Weight Weight g ai/ha(K ) (no./sqm.) (gm/sqm.) (gm/weed)

TA1 0.75 0.1 0.05 0.4 1.00 24.9 28.0 1.1 1.25 7.2 15.2 2.1 TA 2 0.75 3.1 4.2 1.4 1.00 0.0 0.0 - 1.25 0.8 1.4 1.9 SaturnG 0.75 5.1 11.2 2.2 1.00 5.7 14.2 2.5 1.25 0.2 0.8 3.4 TAL 26391 0.75 0.0 0.0 - 1.00 4.5 7.4 1.6 1.25 0.0 0.0 - 2,4D 0.75 19.9 48.6 2.4 1.00 4.8 4.4 0.9 1.25 68.6 80.8 1.1 SaturD n 0.75 0.7 1.0 1.5 1.00 0.2 1.0 5.4 1.25 0.8 1.5 2.0 + 2,4 1 DTA 4 0.70. + 5 O.O O.O - TA2 + 2.4D 4 0.70. + 5 0.0 0.0 - Saturn G + 2.4D 4 0.70. + 5 3.0 5.8 1.9 Untreated - 255.4 175.8 0.6 Handweeded 2x 0.0 0.0 —

233 Table V. Projected yield of rice treated with CRF thiobencarb, Aug. - Nov., 1991.

Treatments Rate Yield (Tons/ha} Grouping

TA1 0.75 3.6 a 1.00 2.1 abc 1.25 3.1 abc TA 2 0.75 3.7 a 1.00 2.9 abc 1.25 2.2 abc SaturG n 0.75 1.9 be 1.00 2..1 abc 1.25 2.2 abc TAL 26391 0.75 2,8 abc 1.00 2.9 abc 1.25 2.1 abc 2,40 0.75 2.0 abc 1.00 1.8 c 1.25 2.5 abc SaturD n 0.75 2.1 abc 1.00 2.1 abc 1.25 1.6 c + 2.4 1 D TA 0.75 + 0.4 2.3 abc TA 2 + 2.4D 0.75 + 0.4 2.3 abc Satur+ G n 2.4D 0.75 + 0.4 1.8 c Untreated - 1.5 c Handweeded 2x 3.2 abc

Projected yiel f rico d e treated with CRF thiobencarb , Aug. - Nov., 1992.

Treatments Rate Yield (Tons/ha) Grouping

TA1 0.75 3.6 ab 1.00 2.1 abc 1.25 3.1 abc TA 2 0.75 3.7 a 1.00 2.9 abc 1.25 2.2 abc SaturG n 0.75 1.9 bc 1.00 2.1 abc 1.25 2.2 abc TAL 26391 0.75 2.8 abc 1.00 2.9 abc 1.25 2.1 abc 2,4D 0.75 2.0 abc 1.00 1.8 c 1.25 2.5 abc Saturn D 0.75 2.1 abc 1.00 2.1 abc 1.25 1.6 c

+ 2T 41 D TA 3 2. 0.7 54 0. + abc TA + 2.42 D 0.75 + 0.4 2.3 abc Satur+ G n 2.4D 0.75 + 0.4 1.8 c Untreated - 1.5 c Handweeded 2x 3.2 abc

234 Tabl . Wee ricn ei VI T ed treatedensitDA 0 4 t dya thiobencarbwitF hCR , April-August 199a 3<

TREATMENTS RATE TIME OF WEED DENSITY (KgAPPLICATIOai/ N (No.70.25 sqm) ) ha (DAT) Total Weeds Echinocloa

L 070191TA . 2 0.75 1 1.75 j 0.5 k L 089TA 2 . 2 0.75 1 7.50 d 4.5 d 3. CF

Table VII Height and tiller number of rice treated thiobencarb,April-August,1993,weF withCR t season

TREATMENTS RATE TIMF O E PLANT HEIGHT TILLER NO. (Kg ai/ha) APPLICATION (cm) (DAT) 1. CRF 0.75 1 54.6 b 20.6 be B CM . 2 0.75 1 54.0 b 23.3abc 3. CF 0.75 1 57.3ab 21.7abc 4. CRF 1.50 1 55.5ab 22.6abc 5. CMB 1.50 1 54.3 b 25.9abc 6. CF 1.50 1 57.4ab 27. 7a 7. CRF+CF 0.75+0.75 1 59.4ab 26.4abc 8. CRF 0.75 4 59.8ab 26.9ab B CM . 9 0.75 4 59.1ab 25.4abc 10.CF 0.75 4 56.4ab 26.2abc 11.CRF 1.50 4 61.6a 28. 2a 12.CMB . 1.50 4 59.6ab 23.8abc 13.CF 1.50 4 57.6ab 21.9abc 14.CRF+CF 0.75+-0.75 4 61.8a 23.6abc 15.UNTREATED - - 56.9ab 19.5 c 16.HANDWEEDED 2X 0 4 d an 0 2 58.5ab 23,8abc

Value e meanar s f fouo s r 4 replication hillr d plope san t s Means followe a commo y b d t nsignificantl no lette e ar r y differen DMR% 5 t Ta t

235 Table VIII. Weed density at 40 DAT in rice treated with CRF thiobencarb, April-August 1993

TREATMENTS RATE TIMF O E WEED DENSITY (K/ APPLICATIOai g N (No.70.25 sqm) ) ha (DAT) Total Weeds Echinocloa 1. TAL 070192 0.75 1 1.75 j 0.5 k 089L 2.TA 2 0.75 1 7.50 d 4.5 d b < F C . 3 0.75 1 1.75 j 1.0 j 07019L 4.TA 2 1.50 1 3.0 h 2.5 g 5. TAL 0892 1.50 1 5.25 g 3.0 f 6. CF 1.50 1 1.75 j 0.5 k L 070192+C7.TA F 1.50 1 2.25 i 1.5 i 8. TAL 070192 0.75 4 8.5 c 7.0 c 9. TAL 0892 0.75 4 6.75 e 3.5 e 10.CF 0.75 4 6.0 f 3.5 e 11. TAL 070192 1.50 4 1.75 j 0.5 k 12.TAL 0892 1.50 4 21.5 b 9.75b 13.CF 1.50 4 0.5 k 0.25kl 14.TAL 070192+CF 1.50 4 3.25 h 2.0 h 15.UNTREATED - - 54.25a 29.75a 16.HANDWEEDED 2x 0.0 I 0.0 I

Table IX Effect of CRFs on yield components of rice, April-August,1993

no panicles no grains 1000grain Treatment Ratm sq e r pe Time per panicle weight(g) 1 TAL070192 0.75 1 249 a 66 a 31.7 a 2 TAL0892 0.75 1 236 a 59 a 31.3 a 3 CF 0.75 1 208 ab 79 a 32.0 a 4 TAL070192 1.50 1 b a 3 20 58 a 31.9 a 5 TAL0892 1.50 1 b a 8 20 71 a 31.5 a 6 CF 1.50 1 193 ab 76 a 31.9 a 7 TAL070192+CF 1.50 1 b a 7 19 64 a 30.5 a 8 TAL070192 0.75 4 235 a 70 a 30.5 a g TAL0892 0.75 4 233 a 72 a 32.4 a 10 CF 0.75 4b a 3 20 63 a 31.9 a 11 TAL070192 1.50 4b a 5 20 54 a 31.0 a 12 TAL0892 1.50 4 215 ab 72 a 30.7 a 13 CF 1.50 4 b a 0 19 76 a 31.7 a 14 TAL070192+CF 1.50 4 186 b 62 a 30.8 a 15 UNTREATED 162 ab 73 a 31.6 a 16 HANDWEEDED b a 4 19 74 a 30.6 a

236 Table X. Effect of CRFs on yield and yield components of rice, April-August,1993

Yield Biomass Harvest Treatment Rate/Time (Tons/ha) Indea < x

1. TAL070192 0.75/1 5.0 a 30.1 ab 16.8 a 2 TAL0892 0.75/1 4.4 a 29.2 ab 14.7 a 3 CF 0.75/1 5.2 a 26.7 abc 19.3 a 4 TAL070192 1.50/1 4.0 a 24.4 abc 16.4 a 5 TAL0892 1.50/1 4.6 a 25.6 abc 18.4 a 6 CF 1.50/1 4.5 a 29.4 ab 15.3 a 7 TAL+CF 1.50/1 3.9 a 25.8 abc 16.4 a 8 TAL070192 0.75/4 5.0 a 32.9 a 15.1 a 9 TAL0892 0.75/4 5.4 a 32. 1 ab 17.1 a 10 CF 0.75/4 4.1 a 20.6 abc 21.8 a 11 TAL070192 1.50/4 3.6 a 19.9 bc 18.0 a 12 TA0892 1.50/4 4.7 a 26. 1 abc 17.7 a 13 CF 1.50/4 4.7 a 24.4 abc 20.1 a 14 TAL+CF 1.50/4 3.4 a 16.3 c 23.4 a 15 UNWEEDED 3.4 a 26.3 abc 15.4 a 16 HANDWEEDED 4.7 a 26.0 abc 17.3 a

LITERATURE CITED

Connick, W.J., Bradow,J.M., Wells,W. , Steward, K.K. and Van, T.K. Preparation and evaluation of controlled release formulations of 2,4 dichlorobenzonitrile. J. Agric. Food Chem. 32 119-1205 (1984). Hussein, M., Can, J. and Räthor, M. N. Preparation of controlled release formulations of 14C labeled thiobencarb and study of their environmental behaviour . Pestic. Sei. (1992). . PreparatioY . B , d studOh an nf Hussein o d y an . M , controlled release formulation labeleC 14 f so d butachlor. Toxi. Environ. Chem. 35 101-110 (1991). Integrated pest managemen n ricei t . Philipp. German Biol. Plant Prot. Project, Bur. Pi. Ind., Dept. Agric., Philipp. 2nd ed. (1990) 136 pp. Julian, Radenbaugh d WisniewskiN. an . T ,. W . J . G ,. S , Permeability characteristics of calcium alginate films. J. Contr. Rel. 7 165-169 (1988). d CalendacionMabbayadan , Obias. O B. . T . . R ,B ,. R , Rice Culture Systems in the Philippines. NFAC/UPLB Rice Prod. Manual.Los Banos, Laguna, Philipp. 121-136 (1983) . Mercado,B.L. Introductio o weet n d science.Southeast Asian Reg. Cent. Grad. Study Res. Agric. (SEARCA) College, Laguna, Philipp. (1979).

237 Pfister , Bahadir d M. Körte an . G ,. . Releas F M ,, e characteristics of herbicides from calcium alginate filins. J. Contr. Rel. 3 229-233 (1986). PhilRice. Rational e establishmenth r fo e f PhilRiceo t . Philipp. Rice Res. Inst. Info. Bull. Series 3 (1991) 6 pp. PhilRice. The Philippine Rice Situation in Capsule. Philipp. Rice Res.Inst.Info. Bull. Series 1 (1992) 8 pp. . EnvironmentaH Soltan . M . R , l behaviou e sloth w f o r release formulation of butachlor herbicide. PhD thesis, Alexandria University, Egypt (1991). Zimdahl . L Wee . R ,d Crop Competition A :Review , Int. Plant Prot. Center, Corvalis, Oregon (1980).

238 CONTROLLED-RELEASE HERBICIDEN SI RICE-FISH CULTURE: EVALUATION OF THIOBENCAR BUTACHLOD BAN R FORMULATIONS

M. SOERJANI, S. WARNOUTOMO Centr Researcr efo Human ho n Resources anEnvironmente dth , Universit f Indonesiayo , Jakarta, Indonesia

Abstract

Innovation in weed control as an effort to increase food production has been implemented among otherdevelopmene th y sb controlled-releasf o t herbicidef eo overcomo st e weed problems. The impact of controlled-release formulations of thiobencarb and butachlor on rice-fish cultur s studieewa d using irradiated latex mixed with these herbicides. Comparative stud s donwa y e with commercia herbicideo l producttw e th d f an so s pretilachlor, also compared with handweeding. The results showed that although hand- weeding provided the highest rice production (5.39 tons/ha) butachlor controlled-release formulation kg/h6 0. t a s gave als reasonabloa y high rice production (4.65 tons/ha), while both treatments also gave high productions of fish. This is due to the high survival rat f fis etheso n i h e treated plots e lowesTh . t fish shortese weighth d an t t lengt f fisho h were in plots treated with 1.5 kg/ha butachlor commercial product, indicating that the slow-release formulation f butachloo s r provid betteea r condition cultivato st e fish thae nth commercial produc f butachloro t e concentrationTh . f herbicideo s e wateth s n i i rs relatively highehighee th n i rr dos f herbicideeo e slow-releasth n i s e formulations. This i s correlated wit lowee th h r residue e higheth f o sr dos f herbicideeo e slow-releasth n i s e formulations. In plots treated with commercial products of herbicides, there was no herbicide concentrations that can be detected in the water.

1. INTRODUCTION

Weeds are biotic components in a system that creates nuisance, risks and additional lost or cost to man, man's interests or properties, e.g. crops, animals, water, land and others. e anxiouar n extraco t sMe mose th t t from their lanfrod dan m technological inputo st assure a better way of life. However, this trend will probably result in numerous side effects, such as the use of pesticides in general and herbicides in particular. Therefore, in the long run it is unlikely that any production increase will always fulfill the objective of development efforts f thiI . s continues developinn i , g countries, suc s Indonesiaha e th , unpriviledged community will not only have unequal share of benefits, but they also likely to suffer most from the side effects. The situation may become even more serious if, in turn productioa , n increase result morn i s e serious pest problems (Soerjani 1993: 271)A . dramatic exampl histore browe th th s ei f yo n planthopper that damaging rice crops, which s developewa d fro mpesa f minoo t r importanc mose th to e t seriou s rice pes Southeasn i t t Asia, resulting from change agriculturan i s l practices, e.g ricw .ene varietied an s indiscriminate pesticide usage.

This create situationsa whicn i , more hth e inventio made nw agriculturan ei l technology, e morth e dynami e conditiocth e agriculturath f no l syste . Thereforemis , thera s ei continued search for noval innovations in agricultural practices for their sustainability.

239 On f theseo e searche controllede th s si slowr o delivery of pesticides. This report consisting of results from experience with slow-release formulation and commercial product f thiobencarbso butachlord an herbicide ricen si field witwithoud han t fish.

. OBJECTIVE2 STUDE TH YF SO

Controlled or slow delivery herbicides implies formulations that protect and release biologically active materials combined with delivery pathways appropriate to the target organism and its environment (Wilkins 1990). This approach is designed to increase the efficienc f delivero y bioactive th f yo e materials alss i t oI . aime t havinda g improved safety to operators, crops as well as expected persistence. specifie Th c objective stude th f yso are:

(a) Comparative information concernin efficace slow-release gth th f o y e formulation commerciae th d an l formulatio f thiobencarno butachlod ban r weefo r d contron i l rice-fish culture; (b) The impact of the slow-release formulation of thiobencarb and butachlor on certain components of the system, e.g. rice products, weeds, fish, and water.

3. METHODOLOGY

The study was conducted during and 1992-93 rice seasons, with the following procedures:

1 Fiel3. d Experiment Preparation

Experiments were conducted at the experimental field of CIBA-GEIGY at Cikampek, West Java. The rice field was prepared with 5 m x 5 m plots to make a total of 48 plots. The experimental plots were treated with N, P & K 14 kg/ha each, befory da e e on transplanting N/hag k witdayd 5 2 3 an ,,h 2 s after transplanting.

3.2 Slow-Release Formulation

The formulations were prepared at the chemical laboratory of the National Atomic Energy Agency, Jakarta with the following procedure. Thiobencarb of 0.89 (93% technical grade) adde detergentdg m wit 0 h7 , were mixed witirradiateg h2 3 ( d MRad) latex, wit additionan a hnon-irradiateg 0 2 f o l g detergent 0 d7 late d xan . e mixturr drieai Th roon s di emwa temperature (25°Chours4 2 e finar fo )Th . l products are films of 2 mm. The same procedure was applied to butachlor of 0.89 (93 % technical grade).

3.3 Materials

1) Herbicides: thiobencarb and butachlor both commercial product and slow- release formulations, and pretilachlor commercial product. 2) Rice variety used was Way Seputih seedlings of 20 days old to be transplanted three seedlings as a cluster in one hole, which has a distance of 25 x 25 cm. 3) Fish variety was Cyprinus carpio (goldcarp) of 5-7 cm long or 10-12 g, and was released 12 fishes per plot, or equivalent to 4,800 fishes/ha.

240 3.4 Experimental Designs

e experimenTh randomizea s wa t d block design with four replication eacn i s h block. The treatments were as follow:

Thiobencar butachlod ban r slow-release formulation kg/h 4 wit2. d ha 0.6an 2 ,1. were applied evenly at the transplanting date. Pretilachlo sprayes wa r dayd4 s after transplantin t 1,2ga 5 kg/ha. Commercial thiobencarb (1 kg thiobencarb + 0.5 kg 2.40 and commercial buta- kg/ha5 chlo1. f o r, were spraye dayd4 s after transplanting. Handweeding, twice: 7 and 21 days after transplanting. Control plots with fish. Control plots without fish.

3.5 Analysis of the Data

e datTh a analyzed were: numbe f productivo r e tillers, height y weighdr , f o t weeds, number and length of panicles, seed production harvested 90 days after transplanting, fish productio herbicidd nan e residue e remaininth n i waten i s d gan r latex of the slow-release formulations.

The chemical analysis was done at the Chemical and Biological Laboratory of the National Atomic Energy Agency, Jakarta.

. RESULT4 DISCUSSIOND SAN S

4.1 Number of Productive Tiller and Height of Plants

e tilleTh r numbe s reacheha r stabilizeda d growth afte day0 3 rf transplanting o s . Thi shows si Figurn ni . e1

Number of productive tillers Plant Height 70 (cm)

30 40 DAT

Figure 1. The average number of productive tillers of all treatments.

241 The maximum average productive shoots after 30 days were 23 - 24 shoots per blocksignificano n therd s an , ewa t differen e numbeth f o f productivt o r e tillers between treatments. Ther s als significano ewa on t differen heighe th n f rici to t e plants betweee nth various treatment, although there was an indication that the highest plant height plotaften i day0 s 4 sr treateswa d with slow-release formulatio thiobencarf no t ba any rate, while the lowest was of rice plants in plots treated with pretilachlor at 1.2 notee 5 b o kg/hat d s thai optimae t I th t. l tiller numbe s reachewa r d afte0 3 r days whil e riceth e plants were still increasin plann gi t heigh d levelinan t f aftegof r 40 days.

4.2 Dry Weight of Weeds

The weeds observed in the experimental plots were mainly the following species: a) Hydrilla verticilata Roy le. b) Monochoria vaginalis Burn. f. c) Marsilea crenata Presl. d) Limnocharis flava Berch. e) Scirpus juncoides Linn. f. f) Echinochloa crusgalli . BeauvL . g) Sagitaria guayanensis H.B.K.

The result of the various treatments on the dry weight of weeds after harvest showed cleaa r different between treatments (Figur. e2)

Dry Weigh f Weedo t s (kg/ha) 100

77.93 80 M 68 04 tabl 61 37 Ib) 57.46 56.61 56 03 55 10 60 (bel (bel (bl 50.05 Ibcl 47.95 47 78 led) 45.78 led! (cd) (cd)

36 71 40 (el

20

CM X ro d

Figure 2. The dry weight of weeds (kg/ha) of various treatments after harvest. Number followed with different letters are significantly different. kg/ha6 0. = ; TSF6 0. : thiobencarb slow-release formulation= F BS ; buthachlor slow-release formulation thiobencar= P TC ; b commercial product; BCP = butachlor commercial product; PCP = pretilachlor commercial product; H.W. = handweeding; C + F = control with fish; C - F = control without fish.

242 weighy dr e f weedo t Th s show followine sth g trends:

a) Butachlor (slow-release formulations) depressed weeds more than thiobencarb (compare 1 - 3 and 4-6) especially at 2.4 kg/ha; b) Thiobencarb slow-release formulation t controno d di sl weeds better than thiobencarb commercial product (compare 1-3 and 7) while butachlor controlled release formulation at 1.2 and 2.4 kg/ha controlled weeds better than butachlor commercial product (compare 5 - 6 and 8). c) Handweeding (two times) was the best to suppress weed growth (10). d) Fish gave certain suppressing impact on the growth of weeds (compare 11 and 12).

3 4. Rice Production

The impacts of the treatments on the number and length of stalk did not show significant different between treatment highese th s t numbe f stalko r s were control plots with fish (11.92), the shortest were control plots without fish (9.94) and plot treated with 0.6 kg/ha slow-release formulation of butachlor (9.55). The average numbe f stalko r s 11.12 longesse wa Th . t stalks were plots treated witkg/h2 1. h a slow-release formulations of thiobencarb (21.19 cm) and butachlor (21.12 cm), whil average eth e lengt f stalkh o 20.3s . swa 9cm

The treatments show different rice production (Figure 3).

Rice Productions (tons/ha)

6.0 5.39 (a 4.85 5.0 4.65 4.70abl 4.38 labcl 4.39 (abcl f-2* (bed) pa (bed) 4-28 T 4-24 3 93 (bed) __ ;' (bed) (bed) { I 3.92 d) 4.0 CVJ in o O 6 CM o csi

n (O

Figure 3. The rice production of different weed control treatments. Treatments 1 to 12 are explained the same as in Figure 2. Numbers followed with different letters show significant different.

It is shown from the results of the various treatments that the highest production of rice kernels was obtained from two times weedings (10. HW), followed by the following treatments: 1.25 kg/ha pretilachlor kg/h5 1. , a butachlo6 0. d an r

243 slow-release formulation of butachlor. The lowest products was control plots without fish. This indicates that fish cultur givy ricn ei eema certain impac ricn o t e production thad an ,t weed contro s urgentli l y needed, since weesignificantly dma y reduce rice production.

4.4 Fish Production

Fish production harvested 65 days after transplanting was affected by different treatments. Thiweighn i s slengti d an t shows ha Figurn ni . e4

fish weights | | fish lenght

Fish Weight Fish Length Increase Increase (g/fish) « -^ «n i (cm/fish) "v 01 60 2-« 9.0 in — ^ s 1 _1 50 - ^ S 2 in ^ 8.0 «r ^ — ^ Tt — — o- m — & y * 'T S o Ul o (0 ^ CO (O 40 - (0 | 7.0 n ' — y n -^ X •o — •D o M % l'*l SO'S \

I 6.0 30 S ni

5.0 20 1 4. 85 (st)

- 1 15.8 0 (e j

10 4.0 I 3.5 r - N o dat a (n fish ) 7/////////////////A 26.1 0 //////7?/////////A 24.60( 0 0 • | 3.6 3 fr i '0 u. u_ u. CL co v> . U -Q(/) .. U U- l- 1- 1- s a U. & % S £ CÛ U. to + N « n ti N *< ° u CNJ 5 0 r

Figure 4. Fish increase in weight and increase in length meanwed 65 days after transplanting (see text).

Fish growt s indicateha increase th y db weighn ei lengtd affectes an t e hwa th y b d treatments. The fish growth was affected by rather severely by the applications of slow-release formulation f thiobencaro s l dose al witd t commerciae ba san hth l product. The growth of fish in plots treated with slow-release formulation of butachlor (4-6) was faster than when plots were treated with commercial product of butachlor at 1.5 kg/ha (8). It is shown from number 10 that handweeding cause certain interference in the fish growth, almost the same as the control plots numbe . Ther11 r e mus certaie b t n impacplanktoe th n o t n food chai thesn ni e plots. Pretilachlor treatment numbe seem9 r s les affectinn i s g fish growth compareo dt butachlor commercial product numbe. 8 r

4.5 Fish Survival

As mentioned earlier s cultivatefiswa h fish/plot2 1 d fis8 r treatment4 pe hr o s e Th . surviva e fis day0 th h2 f o sl after being cultivate treatee th n di d plot s showi s n ni Figure 5.

244 Fish Survival (%) 100 100 93 3 93 .3 93 .3 9 3. 9 3 3 8 3. 3 89.E 85.4 80

60 -

40 3S 4 20 1 1 1 I 0 u. U. CL a. O u_ to V) to CO CO Ü CL 1- P CO CO CO £ CO in + to CD in in g o C3 O CM r~ «— T— O jj ID r~- 06 oi T-

Figure 5. The survival of fish 20 days after cultivation

The observation of fish survival shows clearly that butachlor commercial product is most harmful to fish survival (8) while thiobencarb (7) and pretilachlor (9) are much safer for the fish survival. However, the slow-release formulation of kg/h4 butachlo2. o l at rate al clearl6 t s0. a r y caus betteea r surviva fise th h f o l (4-6). Handweeding and control plots did not show any impact on the -fish survival (10-11).

6 4. Herbicide Residu Waten ei r

Sixty five days after treatments some water s samplelefwa t analysed dan r dfo possible pesticide residu watere th resulte n ei Th .show e sar Tabln ni . e1

Tabl . Wate1 e r analyse r pesticidsfo e residue day5 6 s s after application

No. Treatments Active ingredient a /h g k of pesticides (ppm)

1 Thiobencarb s.f. 0.6 0.017 2 Thiobencarb s.f. 1.2 0.029 3 Thiobencarb s.f. 2.4 0.040 4 Butachlor s.f. 0.6 0.049 5 Butachlor s.f2 .1. 0.113 6 Butachlor s.f. 2.4 0.154 7 Thiobencarb c.p. 1.5 ud 8 Butachlor c.p. 1.5 ud 9 Pretilachlor c.p. 1.25 ud

s.f. = slow-release formulation c.p commercia= . l product u.d undetecte= . d detection limit for thiobencarb 0.010 ppm detection limit for butachlor and pretilachlor 0.040 ppm

245 This indicates that the slow-release formulations still release herbicides that can be detected as residues in water. The higher the contents of the herbicides in the formulation more sherbicideth e eth s were released int watere oth . Thiobencarb treated plots showe loweda r herbicides residue sinc knows i t ei havo nt ehighea r ratf eo degradation compare butachloo t d r (see Tomizawa 1977: 30). Therefore th e residues thiobencarb formulation at all rates are lower than the residues of butachlor formulations.

7 4. Herbicide Residue Slow-releasn si e Formulation

The residues of herbicides in the slow-release formulations indicates the less effects of these formulations on fish. After harvest (90 days after transplanting) the residue f herbicideso thesn si e formulation showe ar s Tabln ni . e2

e residuTh Tabl f herbicid. o e2 e active ingredien e slow-releasth n i t e formulations 90 days after transplanting.

No. Herbicides Doses Active ingredient (kg/ha) (W/W)

1 Thiobencarb 0.6 10.70% 2 Thiobencarb 1.2 10.42% 3 Thiobencarb 2.4 10.19% 4 Butachlor 0.6 9.47% 5 Butachlor 1.2 9.07% 6 Butachlor 2.4 8.62%

In comparison with the data of herbicide concentration in water (Table 1) it is obvious thae lesth ts residu f herbicideeo e formulationsth n i s e higheth , e th r concentration herbicide th f so e active ingredien watere th n i t.

5. CONCLUSIONS

1) The highest rice production is when weeds are controlled manually by handweeding. This is due to the fact that weeds were reduced severely under this treatment (Figure 2 and 3). e highesTh t ) ric2 e production with herbicide treatment whes swa n treated wit6 h0. kg/ha of butachlor slow-release formulations and 1.25 kg/ha of pretilachlor, althoug t clearlno h s thiy wa scorrelate d wit weee hth d control result case th f eo n i s butachlor (Figures 2 and 3). e fisTh h productio ) 3 termn ni f fiso s h weigh fisd han t lengt s obtainehwa plotn i d s treated with 0.6 kg/ha butachlor slow-release formulations and with 1.25 kg/ha of pretilachlor (Figure 4). 4) Fish surviva higs i l h (100% plotn i ) s treated wit kg/h6 h0. a thiobencarb slow- released formulations, also in plots treated with slow-release formulation of butachlor and in plots treated with pretilachlor at 1.25 kg/ha.

246 5) The concentration of the active ingredient of herbicides in the water was higher was the higher concentration of the herbicides in the slow-release formulation (Table 1), which was due to the fact that the residue of the herbicide active ingredients in the resslow-release f lateth o t f xo e formulations afte day0 9 r s were highee loweth n i rr concentratio f herbicideno formulationse th n si .

ACKNOWLEDGEMENT

e authorTh s would . likHussain thanM o et . kDr , Research Coordinatoe th f o r "Development of controlled-release formulations of pesticides using nuclear techniques" Joine th f t o FAO/IAEA Divisio IAEAe th f no , Viennsuppors hi r guidancd afo an t thin ei s study. The study was conducted under the grant partially provided by the Research Coordination Program of the Joint FAO/IAEA Division. The assistance of Mr. Made Sumatr e Nationath f ao l Atomic Energ ypreparatioe Agencs stafth hi n i fe d th yan f no slow-released formulation chemicae th n i d l an sanalysi herewits si h acknowledgede Th . field experiments were conducte e CIBA-GEIGY'th t a d s experimental statiot na Cikampek.

REFERENCES

Soerjani, M. 1993. Integrated pest management and the environment. BIOTROP Special Publ. No. 50: 265-285.

Wilkins, B. M. 1990. Controlled delivery of crop-production agents. Taylor & Francis, London: xi - xv.

Tomizawa . 1977C , . Pespresend an t t statu f residueo s f pesticideo s s Japan Pesticides Information3. no.

Next page(s7 ) lef24 t blank CONTROLLED-RELEASE HERBICIDE FORMULATION RICN SD I EAN RICE-FISH CULTURE

M. SOERJANI, S. WARNOUTOMO Centr Researcr efo Human ho n Resources anEnvironmente dth , Universit f Indonesiayo , Jakarta, Indonesia

Abstract

There is an increasing trends of the need for technological innovations in weed control in food crops, mainly rice crops, to meet the demand of the population growth in developing countries novae Th . l techniqu ween ei d contro particulan i l pesd an rt managemenn i t general must also comply with the effort to maintain the sustainable support of all production components. The development of slow-release formulation technique of herbicide efforn a improvn o i t st e crop productivity, throug herbicide hth e efficacy, while also concerned wit possiblha e external impacts, e.g. residu soiln ei , grainsd fisan h . This paper is to conclude the result of a study of the use of thiobencarb and butachlor in rice- fish culture, with a result that there were some residues of these herbicide in the soil and e ricth e n i grain t no .t fis Thiobencarbu h b showe lesda s ris contaminatinn ki soie r gth o l leaving residue fisn i s h compare butachloro dt . Rice production (5.35 tons/ha s stile i ) th l highest in handweeding treatments, 7 and 21 days after transplanting, in which weeds tnay reduc yielde th e combinatiof A almoso . % 60 t f 0.7no 5 kg/h f thiobencarao b slow-release formulations with 0.75 kg/ha commercial product gave a relatively high rice yield of 4.93 tons/ha.

1. INTRODUCTION

Wit e ever-increasinhth g trende nee r footh pressure dfo n i th sd o cropt e f eo du s population growth, especially rice in most developing countries in the tropics, efforts to increase the efficacy and efficiency of the use of improved technology is unavoidable. This include bettea sf herbicides o safed e an r us r e concere increase t onlTh th . no n yi s ni , of products t als potentiae bu ,oth l externalitie f impactsso environmente th , n suco s ha , e.g. soil, non-target organisms, and contamination of the products.

Among effort o increast s e efficaceth efficiencd an y s als i e developmeny oth f o t controlled-releas r slow-delivereo herbicidesf yo . Controlled-release formulationf so herbicides imply formulations that release biological active ingredient, combined with delivery pathways appropriate to the target organisms and the environment (Wilkins 1990). This report consisting of the results of a study to compare the side impacts of the use of thiobencarb and butachlor on the soil, fish and rice grains. The second study is an experiment with slow-releas commerciad an e l product f thiobencaro s b treated separately or in combinations at different times of applications.

Thiobencar s furthebwa r studied reasoe baseth n do n that thiobencar rapidls bi y degraded (Tomizawa 1977) and has a low residual impact in plants (Ishii 1976), as well as low toxicit fisn yi h (Ishii 1977; Tomizawa 1977).

249 . OBJECTIVE2 STUDE TH YF SO

The study is designed to increase the efficacy of delivery of the biological active ingredient f herbicidesso alss i t oI . aime improvo dt e safet operatorso yt , cropse th , environment as well as the non-target components in the system. specifice Th objectives stude th f yo are:

(a) The residue and impact of the applications of thiobencarb and butachlor on certain components of the system, e.g. on soil, rice grains, and fish;

) (b Comparative informatio e efficac e slow-releasth th f o f n o y commerciad an e l formulation f thiobencaro s r weebfo d contro ricn i l e culture.

3. METHODOLOGY

The study was conducted during the 1990-91 and 1991-92 rice seasons, with the following procedures:

1 Fiel3. d Experiment Preparation

Experiments were conducted at the experimental fields of the Rice Research Station Sukamandi (for the first experiment) and the CIBA-GEIGY at Cikampek, West Java (for the second experiment). The rice field was prepared with 5 m x 5 m plots to make a total of 36 plots for the first experiment and 48 plots for the second. The experimental plots were treated with N, P & K 14 kg/ha each, one day before transplanting, and with 23 kg N/ha, 25 days after transplanting.

2 Slow-Releas3. e Formulation

The formulations were prepared at the chemical laboratory of the National Atomic Energy Agency, Jakarta with the following procedure. Thiobencarb of 0.89 (93% technical grade) adde detergentdg m wit 0 h7 , were mixed witirradiateg 2 h d 3 MRad( ) latex, wit additionan ha non-irradiateg 0 2 f o l g detergent d0 7 late d xan . e mixturr drieai Th roon s di ewa m temperature (25°Chours4 2 e finar fo )Th . l product. mm filme 2 sar f o s

3 Material3. s

) 1 Herbicides: thiobencarb slow-release formulatio commerciad an n l producd an t butachlor commercial product (the latter only for the first experiment). 2) Rice variety used is Way Seputih seedlings of 20 days old to be transplanted three seedlings in one hole which has a distance of 25 x 25 cm. 3) Fish variety of Cyprinus carpio (goldcarp) of 5 - 7 cm long was released 12 aftey fishe da r plot r e pe rics on , e transplantatio nfirse (onlth tr experiment)yfo .

4 Experimenta3. l Designs

e firsTh 1t )experimen randomizea s wa t d block design with four replicationn i s each block treatmente Th . s wer follows ea : Thiobencarb and butachlor commercial products were applied both at 0.4, 0.8, 1.6, and 3.2 kg/ha one day before transplanting in rice-fish culture.

250 Herbicide residues were analysed from soil, graifisd han n samples2 5 , 40 , day0 7 d s aftean r herbicide applications respectively.

desige experimene th Th f no shows i t Tabln ni . e1

Tabl . Desig e1 firse th tf nexperimeno t

No. Herbicides Doses kg/ha

1 Thiobencarb 0.4 0.8 1.6 3.2

2 Butachlor 0.4 0.8 1.6 3.2

3 Control -

2) The second experiment was a randomized block design with the following treatments (all without fish) with four replication blocksn si . Slow-release formulations of thiobencarb of 0.75 and 1.50 kg/ha applied evenly one day before and six days after transplanting of rice (without fish). Commercial formulation thiobencarf so 0.7f b o 1.5 d 5an 0 kg/ha, sprayee don day before and six days after transplanting. Combination of commercial and slow-release formulations of thiobencarb 0.7 50.7+ 5 kg/ha day x befor,y si treate sda d aftee ean don r transplanting. Handweeding, twice: 7 and 21 days after transplanting. Control plots.

The design of the experiments is shown in Table 2.

Table 2. Design of the second experiment

No. Treatment of thiobencarb Applications days from (kg/ha) transplanting

1 0.75 BF 1 day before 2 0.75 CP 1 day before 3 1.5F S 0 1 day before 4 1.50 CP 1 day before 5 0.75 SF + 0.75 CP y beforda 1 e 6 0.75 SF 6 days after 7 0.7P C 5 6 days after 8 1.50 SF day6 s after 9 1.50 CP 6 days after 10 0.7+ 0.7 F S 5P C 6 days after 11 Handweeding day1 2 sd an afte 7 r 12 Control —

251 5 3. Analysi Date th af so

The data analyzed were:

1) Residues of herbicides were analysed in soil samples 42 days after applications, in fish sample day0 s5 s after being released ricn i ed grainan , day0 s7 s after herbicide applications. 2) The results of the treatments were analysed as follow: the number of productive tiller ricd an se grain production weed weighy an , ddr t harvesa t t (92 days after transplanting).

. RESULT4 DISCUSSIOND SAN S

Residuee Th f Herbicide1 so 4. s

herbicide Th e residue soin si l (sample days)2 d4 , fish (sample days0 d5 ricd ean ) grain (sample day0 d7 s after transplanted showe ar ) Tabln ni . e3

Table 3. Average thiobencarb and butachlor residues in soil, fish and rice grains

Doses Residues (ppm) in No. Treatments kg/ha Soil Fish Rice grain

, * 1 Thiobencarb 0.4 u.d. u.d. u.d. 0.8 0.007 u.d. u.d. 1.6 0.067 u.d. u.d. 3.2 0.336 0.286 u.d.

2 Butachlor 0.4 0.008 0.012 u.d. 0.8 0.072 0.018 u.d. 1.6 0.164 0.033 u.d. 3.2 1.184 0.235 u.d.

3 Control - - - - u.d* undetected.= .

The figures show that:

a) Both thiobencarb and butachlor at all levels didnot give any residual effect on rice grains. b) Thiobencarb had less residual impact in soil and fish compared to butachlor. e highe Th herbicide dosee cth rth ) f so e treatment highee sth residuee th r soin si l and fish. e residuTh f thiobencareo ) d kg/h4 s 0. undetecte t ba wa soiln di - , whil4 0. t ea kg/h6 1. s alsaowa undetecte fishn di . e) In general it can be concluded that the residue or impact of thiobencarb on the environment is better than that of butachlor.

252 4.2 Comparison Between Slow-Release Formulations and Commercial Product

4.2.1 Impacts on weed growth

The result of thiobencarb slow-release formulations and commercial product combinationd day6 an d s an afte y rda f bottransplantin1 so t ha shows gi n ni Figure 1.

Dry Weight of Weeds kg/ha 100 89.02 (f)

80 - 71 39 (el 63 37 (el 60 -

3807 39.37 36S2 Id) 35.54 40 - led) 238•£.y oU0 r^ Udl (bed) 27.06 27.18 (be) (be) 20.76 Ib) 20 - 5 29 (a) l l 1 1 I I I \ I I m i

o0.

o + CL cc Q. O O a. CO O tc cc O ui t- O O

weighy FigurDr f weed o t. e1 t harvessa day2 (9 t s after transplanting) 0.70.7= 5 5 kg/haslow-releas= R S ; e thiobencarb= P C ; commercial product of thiobencarb; H.W. = handweeding. Numbers followed by the same letter did not differ significantly. beforey treate* da treate* e * ;d on day x dsi s after transplanting.

s showIi t n that handweeding suppres o t (11 stils bese i y ) th lwee e wa t th s d existence in rice crop. If herbicide is to be used, the best control action on weeds are 0.75 kg/ha thiobencarb slow-release formulation combined with commerciae th l produc t befory treateda dayx e edsi on trans s d (10an )- planting (5) and 1.5 kg/ha thiobencarb slow-release formulation treated one day before transplanting (3). This indicates tha generaln i t slow-releasee th , formulation controls weed growth better tha commerciae nth l product. Further on treatments one day before transplanting is in general also better than treatments six days after transplanting.

4.2.2 Impacts productiven o tillers

e impact varioue Th th f so s treatment productive th n so e tillers shoe wth following results (Figur. e2)

253 Numbe f Productivo r e Tillers/Cluster

IO 14 50 Ibl 14 -

11 50 12 lab) 10 75 10 75 10 50 10 50 10 25 (ab) 10 00 (ab) 10 00 lab) 9 75 lab) lab) (abl 69 0l.l (ab) lu 8 75 la) " (a)

8 -

6 -

4 -

2 -

n l 1 1 l t l l l l l l l

10 P-; d

cc a. O o cc o CO i- in r-- r-. o ö d O o d

Figur . e2 Number f productivso e tillerday0 4 i ssa after transplanting after different treatments. Number f treatmentso letted an s r notee ar s the same as in Figure 1.

The number of productive tillers was highest in the handweeded plots with averagn a f 14.eo 5 tillers/cluster, while onl tillers/cluste0 1 y controe th n i r l plots. This means a reduction of over 30% of the productive tillers due to presene th f weedso t treatmene Th . t wit combinatioha f thiobencarno b 0.75 kg/ha commercial produc 0.7d an 5t kg/ha slow-release formulatiot a n both times of applications, resulting in only 10.50 tillers/cluster, but when treate dayx dsi s after transplantin producet gi d rather high yiel f ricdo e (4.97 tons/ha) compared to the handweeded plots with 5.35 tons/ha (see paragraph 4.2.3).

4.2.3 Rice productions

Rice production e resultth s f varioua so s s treatment e presentear s n i d Figur. e3

s obvioui t I s that weed existenc ricn i e e field reduced rice production significantly e compariso e see.th b Thi n ni n ca s f handweedinno g treatment which produced 5.35 ton f rico s e grain/h controe th a d (11an l ) plots which produced only 2.15 tons/ha (12) accordancn I . resule th suppressinn o i et g e weedsth , treatment wit combinationa h f 0.7slow-releasg o sk 5 e formulation and 0.75 kg commercial product of thiobencarb applied six days after transplanting gav relativelea highese yth t rice productio f 4.9no 7 tons/ha.

254 Rice Productions (tons/ha) 6.0 5.35 4 97 lc> (bel 5.0

3.84 4.0 3 63 3 5s 3 66 (abc| l""" (abc) ' ' ' 3 OB 307 298 lab) 2 77 3.0 """ la) ,.70 la) (al 2 15 la) 2.0

1.0 -

1 1 1 I 1 1 1 1 l l 1 I

a. o O cc O i- in | o O o

O

Figur . e3 Rice production tons/h resulte th s f varioua o s s treatment. Number f treatmento s letted an s same rn i noteth s a e sar Figure 1.

CONCLUSIONS

1) Thiobencarb gave less residu soin ei s compare a l butachloro dt . Onl t higa y h ratf eo kg/h2 3. a thiobencar contaminaty bma e fish wit hresidua f 0.28eo 6 ppm. 2) Weeds severely reduce rice production from 5.35 tons/h handweedinn ai 2.1o gt 5 tons/h controe th n ai l plots r almosyielo ,% 60 d a tlost . 3) A combination of thiobencarb 0.75 kg/ha slow-release formulation + 0.75 kg/ha commercial products treated six days after transplanting gave a relatively high yield of 4.93 tons/ha.

ACKNOWLEDGEMENT

The authors would like to thank Dr. M. Hussain, Research Coordinator of the "Development of controlled-release formulations of pesticides using nuclear techniques" of the Joint FAO/IAEA Division of the IAEA, Vienna for his support and guidance in this study e stud s conducteTh . ywa d unde grane th r t partially provide Researce th y db h Coordination Progra Joine th f mt o FAO/IAE A Division assistance Th .. Mad Mr f eeo Sumatra and his staff of the National Atomic Energy Agency, Jakarta in the preparation slow-release th f o e formulation chemicae th n i d l an sanalysi herewits si h acknowledged. The field experiments were conducted at the Field Experimental Station of the Rice Research Centr t Sukamandea CIBA-GEIGe th d an i Y Field Statio t Cikampekna .

255 REFERENCES

Ishii, Y. 1976. Some evaluation trials and practical uses of Saturn in various rice cultivating countries. Japan Pesticides Information No. 26.20.

Soerjani, M. 1987. An introduction to the weeds of rice in Indonesian. In: M. Soerjani, A.J.G.H. Kostermans & G. Tjitrosoepomo (eds.) Weeds of Rice in Indonesia. Balai Pustaka & BIOTROP, Jakarta, Bogor: 1-4.

Soerjani, M. 1988. Current trends in pesticide usage in some Asian Countries. Environmental implications and research needs, In Pesticides: Food and Environmental Implications. I.A.E.A. Viena, Austria: 219-233.

Soerjani, M. 1993. Integrated pest management and the environment. BIOTROP Special Publ. No. 50: 265-285.

Wilkins . 1990M .. B , Controlled delivery f crop-productiono agents. Taylo Francis& r , London. xv - i x :

Tomizawa, C. 1977. Pest and present status of residues of pesticides Japan Pesticides Information3. no.

256 FIELD PLOT TEST OF EFFICACY OF THIOBENCARB FORMULATIONS FOR WEED CONTRO DIRECLN I T SEEDED RICE

D. OMAR, R.B. MOHAMAD Faculty of Agriculture, Universiti Pertanian Malaysia, Sedang, Selangor Darul Ehsan

Malaysia

Abstract Experiments were conducte fieln di d plot Tanjunn si g Karang rice growing area to study the effect of the timing of application and type of formulation of thiobencarb herbicide on the control of weeds, especially Echinochloa crusgalli. The area selected has a history of heavy weed infestatio farmern ni s fields thesn I . e test performance sth commonla f eo y used commercial thiobencarb formulation, SaturnR was compared with that of a controlled-release (CR) formulation. These d containe7.56an %0 5. d thiobencarb, respectively formulatioR C herbicid e e th Th d . ha n e uniformly disperse a mixtur n di f calciuo e m alginat kaolind e herbicidan e Th . s ewa befory da 1 e d direcappliean 6 tt dsowina grice (DBSth e f seedkg/ho 0 )6 t sa a herbicidratee Th . e formulations were applied 50/5 a singl s a 0r y o mixtur o et deliver 0.7 5r 1.5o a.i./ha0g k treatment e effece th f Th numbee .o t th n d so ran growt f weeho d plants growte th , h parameter rice th e rice f plantso th e d an s grain yield was determined. SaturnR at 1.5 kg/ha, CR formulation at 0.75 kg/ha and hand weeding were most effective in controlling the number of E. crusgalli plants. There was no significant difference between the formulations or the rates used regarding their effect on the growth of the rice plants. The highest rice grain yield was recorded from the plots treated with the CR formulatio formulatioR C kg/ha5 e 1. t alsTh na s . onwa very effectiv t 0.7a e 5 kg/ha rate and the rice grain yield from the plots receiving this treatment was much greater than those treated wit a similah e commercialr th rat f o e formulation.

Introduction

The field plot test was conducted according to the protocol given by IAEA. The objective of the experimen studo t effece s y th f timin wa to t f applicatiogo formulationd nan f thiobencaro s b against weed direcn si t seeded rice.

Material Method san d

Location e fiel ricn Th i d es plo growinwa t g are t Tanjuno a g Karang, Selangor, Malaysia e studs Th .ywa conducte farmer'n di s plot durin maie gth n growing season (wet season f rico ) e i.e. from Auguso t t December, 1992. The area selected has a history of heavy infestation of weeds especially E. crus- galli.

257 Land Preparation

e soiTh l typ s Bakaei u series (alluvial soil)e propertieTh .soie gives th i l f Appendiso n ni e Th . 1 x land was cultivated twice, once when the soil was dry and later when it was wet.

Chemical

The alginate controlled-release formulation of thiobencarb, TAL 070192 (7.56% a.i.), was obtained from IAEA conventionawhere th s ea l granular formulation, Saturn a.i.)s supplie% (5 wa , G ®y 5 db Agriculture Chemical Malaysia.

Treatment split-plo8 x 2 A e sizt . f experimentadesigTh eo in s use5 2 n wa d x witm 8 hl2 applicatioares awa n maie timth n s m . ea plo l m treatmend Ther 2 s an t subplot e x e wa th ploe m s a t2 Th t .siz s ewa spacing between the plots and 2 m spacing between the replicates. There were 2 timing of applications and 8 treatments in this study. All treatments were replicated 4 times. The summary of the treatment gives si Tabln ni . e1

The herbicide was applied at 6 days before sowing of rice seeds (DBS) and 1 DBS. Rice seeds was sown using direct seeded metho kg/ha0 6 fertilizee t da Th .kg/h s applie0 0 6 rat9 e wa : rf th eao N t da day0 2 t skg/h0 a kg/h6 afte . : aK aP r sowin f ricgo e seeds (DAS) e applicatioTh . (ureaN f no s i ) divide dDAS0 6 int applicatioo d oN .an thre applied S DAS0 ean 2 DA f insecticid t no da 0 4 , d ean fungicide were mad infestatioo n thers ea s ewa n observed.

Data were collecte quadratm t rando5 da ploe 0. th x . t m n i usinm Numbe 5 . cnts-gctllig0. E f o r s wa weighy dr DAS0 f d padd4 o t an weedd d t harvestt yA an an .recorde we S se DAS0 1 th ,DA t d0 a , 2 were recorded. Other data collected were number of penicles, number of grains per penicles and weight of 1000 grains. Data were subjected to the analysis of variance (ANOVA) and means were compared using DMRT.

Table 1. Timing of application and treatment.

Tim Applicatiof eo n Formulation Application rate (Main plot) (Sub plot) (kg a.i./ha) 1DBS Saturn 1.50 1DBS Saturn 0.75 1DBS TAL 70 192 1.50 1DBS TAL 70 192 0.75 1DBS Saturn + TAL70 192 0.75 + 0.75 1DBS Saturn + TAL70 192 0.375 + 0.375 1DBS Hand weeding - 1DBS Control - 6 DBS Saturn 1.50 6 DBS Saturn 0.75 6 DBS TAL 70 192 1.50 6 DBS TAL 70 192 0.75 6 DBS Saturn + TAL70l92 0.7 0.7+ 5 5 6 DBS Saturn + TAL70 192 0,37 0.375+ 5 6 DBS Hand weeding - 6 DBS Control -

258 Result and Discussion

numbee . crus-galliTh E f ro showe significano dn t difference betwee timinn2 f applicationgo s at 10 DAS (Table 2). However, at 20 DAS the number of E. crus-galli was significantly higher when chemicae th compareS applies DB wa l DBSI 6 t do da oppositt e .Th e resul observes wa t DAS0 4 t da .

Tabl . eEffec2 timinf o t applicatiof go thiobcncarf no numbee th n bo r ofE. crus-galli

Applicatiof no N Numbe . crus-galliE f ro thiobencarb 10 DAS S 20DA S 40DA 1 DBS 32 5.9 a 5.4 b 4.5 a 6 DBS 32 3.3 a 9.5 a 2.8 b Values in the same column having the same letter are not significantly different at p=0.05

Table 3. Effect of thiobencarb formulations on the number of E. crus-galli

Treatment Rate N Number of E. crus-galli* (kg a.i./ha) 10 DAS S DA 0 2 40 DAS Saturn 1.5 8 2.1 a 3.4 b l.lb Saturn 0.75 8 6.2 a 8.1 ab 4.7 a 2 19 0 7 L TA 1.5 8 3.7 a 7.5 ab 4.0 a TAL 70 192 0.75 8 2.6 a 7.4 ab 2.9 ab Saturn+TAL70192 1.5 8 4.2 a 7.b 1a 4.2 a Saturn+TAL70192 0.75 8 5.4 a 12.6 a 4.1 a Hand weeding - 8 8.1 a 5.9 b 3. lab Control - 8 4.4 a 7.6 ab 4.7 a Value same th en si column havin same gth et significantl letteno e rar y differen t p=0.0a t 5

The treatment did not show significant difference in the number of E. crus-galli at 10 DAS (Table 3). mixture formulatione Th th f eo Saturd an s t 0.7na a.i./hg 5k a (HD) showed relatively higher number of the weed which could be due to insufficient concentration of a.i. released to cause the effect.

significano n Ther s ewa t difference between treatment weighy t weighdr we f d so e tbase an tth n do paddy at harvest (Table 4). However, treatment with Saturn at 0.75 kg a.i./ha gave significantly greater weight of E. crus-galli indicating insufficient a.i. released to control the weeds.

Tabl . eEffec4 thiobencarf o t b formulation weigh y t weighdr we d f weede o t an tpaddth d n san o s t ya harvest.

Treatment Rate N Weeds Paddy (kg a.i./ha) Fresh weight* weighty Dr * Fresh weight* Dry weight* Saturn 1.5 8 124.9 ab 43.5 ab 642.5 a 245.5 a Saturn 0.75 8 286.2 a 102.2 a 438.6 a 173.4 a 2 19 0 7 L TA 1.5 8 134.b 6a 50.b 5a 574.6 a 226.9 a TAL 70 192 0.75 8 120.5 ab 45. 6 ab 577.9 a 237.1 a Saturn+TAL70192 1.5 8 139.b 0a 57.b 2a 535.6 a 218.7 a Saturn+TAL70192 0.75 8 176.b 6a 71.4 a 538.9 a 216.7 a Hand weeding - 8 4.2 1.9 580.2 a 253.0 a Control - 8 139.b 3a 45.b 7a 557.7 a 217.3 a /aluesame th en si column havin same gth t significantle letteno e rar y differen t p=0.0a t 5 *g/0-25

259 Table 5. Effect of thiobencarb formulation on number of panicles and seeds per panicle, weighf o t seed yield san d

Treatment Rate N No. of No of seeds Weight of Yield (kg a.i./ha) pcnicles per penicle 1000 seeds (g) (ton/ha) Saturn 1.5 8 114.b 6a 57.7 abc 18.6 a 4.92 Saturn 0.75 8 86.8 b 49.e 6b 17.1 a 2.95 TAL 70 192 1.5 8 103.9 ab 63. 2 a 19.0 a 4.99 TAL 70 192 0.75 8 102.b a 3 55.5 abc 18.4 a 4.18 Saturn+TAL70192 1.5 8 129.5 a 47.3 c 18.0 a 4.41 Satum+TAL70192 0.75 8 126b la . 44.0 c 16.6 a 3.68 Hand weeding - 8 93.b 2a 62.8 ab 19.5 a 4.57 Control - 8 92.0 ab 48.9 c 17.0 a 3.06 Values in the same column having the same letter are not significantly different at p=0.05 significano N t differenc s observe ewa l treatment al r weighde fo th n f 100so o t 0 seed f ricso e (Tabl. e5) Treatment with Saturn and TAL 070192 at 1.5 kg a.i./ha gave the highest yield i.e. 4.92 ton/ha and 4.99 ton/ha respectively. Thi followes si mixture th 07019y L db f Satur eo TA hand d 2nan an d weeding producing 4.41 ton/h 4.5d aan 7 ton/lia respectively yielde Th . s produced correspone th o dt greater weight of seeds, number of seeds per panicle and number of productive tillers.

260 EVALUATIO CONTROLLED-RELEASF NO E FORMULATION THIOBENCARF SO B TN RICE-FISH ECOSYSTEMS USING RADIOCHEMICAL TECHNIQUES

C.M. BAJET, L.C. ARAEZ, E.D. MAGALLONA National Crop Protection Center, Universit Philippinee th f yBanoss o Lo t sa , College, Laguna, Philippines

Abstract The rate of release of carbon-14 labelled thiobencarb from calcium alginat d corb controlled-releasan eco n e (CR) formulation distillen i s d an d paddy wate studies rwa d under laboratory conditions. Herbicidal efficace th f yo SaturnCd Ran R 10G commercia,a l formulation, against Echinochloa crusgalli and thei rgrowte effecth n f tranplanteo ho t d ricgraid ean n yiel s studiedwa d in potted plants. The formulations were applied at two rates of application. Tilapia ediblfisd han e kangkong plants were expose separatn di e experiments to water treated with slow and fast release corn cob formulations and the toxicit herbicidee th f distributiofisyd e o th han o st f residueno plantn si s were evaluated e releasherbicide Th th f .eo e fro alginate mth e formulations wa s slowed when kaolin content was increased. In corn cob formulations the release herbicideth f eo s controlle wa sproportioe th e polymery b dth f o n s coating the granules. Maximum release from the corn cob formulations was attained on the 7th day; whereas, the release from the alginate formulations kept steadily increasing during the 56 days of the experiment. Some herbicide was lost from the water, apparently due to volatilization. Herbicide concentration in the paddy water was lower than that in distilled water, indicating adsorption by the suspended matter. The efficacy tests showed that all formulations at both rate f applicatioso n were equally effectiv controllinn i e weedse gth e Th . commercial formulation caused significant stuntin phytotoxicitd gan rice th e o yt plants; whereasR formulationC e th , s were safer. Applicatiol al f o n formulations resulted in increased yield of the rice grain, and there was no

significant difference between the formulations or the rates of application. LC 0

of the technical thiobencarb to the fish was 0.23 ppm. However, the fis5 h formulationR C expose e th o t d s contained 0.2 60.2o t residuem 9pp d an s survived e death dt 0.3d fisbu , ha 0.3ho residues1t m 3 pp e fishTh . head contained higher residues than the fillet. Herbicide residues in kangkong plants wer leaved eol stalk d highelowed e san th an syounn n i ri r g leaves.

INTRODUCTION When pesticides are applied by conventional spraying-, about 60-90% of the chemicals fail to reach the target. The available

261 concentratio s i nreduce y y environmentab b ddrif d an t l degradation s i therefor t I . e necessar o d repeate o t y d applications or apply higher dosages to guarantee the effectiveness of these pesticides. However, this becomes a e environmenburdeth o t n d alsan to result o mort s e n food i se controlleTh . d release technologw ne a s i y approach to solving these problems by releasing only the necesary amoun f pesticido t t a eappropriat a tim et ea ratd intervaan e l whe pese presennth s t i mosr to t active. Thiobencarb, S-(4-chlorobenzyl diethylthiocarbamateN )N, s ,i commonla y used herbicid direcn ei t seede transplanted dan d rice {Ishii,Y.1974) s effectivi t I . e agains e barnyarth t d grass (Echinocloa crus-galli) which is a predominant weed problem in the rice paddy. Zimdahl (1980)reported that 40 E. crus-galli per sqt 7-4a m 0 days after emergence reduce d% wherea yiel40 y b ds Mercado (1979) reported that competition against transplanted rice is greatest when the density is 20 E_._ crus-galli per sqm wit criticaha l periodays0 4 f .o d f calciuo e us m e alginatTh e matrix materia r controllefo l d release formulations (CRF) was investigated by Pfister et al. e herbicideth (1986 r fo ) s monoliriuron, desmetryn, chloridazon, atrazine, MCPB, simazin d chloroxuronan e . Thiobencarb, (Hussein et al, 1992) butachlor (Hussein and Oh,1991) and dichlobenil (Connic. 1984al t e )k were likewise studief o e dus wite th h alginate based formulation. Biodegradable basic polymeric materials like natural organic macromolecules suc s barka h , sawdust, lignin, waste paper and plant derived fibers like corn cob or rice straw could be useful as carriers. This study therefore aim o t sevaluat e laboratory prepared alginat d corncoan e bthiobencarf o base F CR d y determininb s it g release rate n distillei s d paddan dy water using radiotracer

262 techniques. Since it is targeted to be used in the rice paddy and that irrigated rice field constitut n importana e t inland fishery resource, uptak d bioconcentratioan e n studie n Nilo s e Tilapia is very important. Data generated could provide an insight of the usefulness of CRF thiobencarb in rice fish culture. Uptak kangkonf eo g (Ipomoea aquatica usuall, ) y founn i d irrigation, drainage canals and waterways and which constitutes farmer's vegetable diet could assess environmental contamination and risk due to outflow of paddy water from the fields. The importanc f evaluatino e e effectivitth g f o ythes e CRF n rici s e compared to the commercially available form is necessary so this research wil farmere usefue b lth t la s level.

MATERIALS AND METHODS

Release rates in water

e followinTh g controlled release formulation (CRF)f o thiobencarb were tested usin e liteon g r distille d paddan d y water. These formulations were provided by the International Atomic Energy Agency.

C thiobencar14 % THln 5 alginati b6. - d kaolinan e , specific activity = 1.291 uCi/g. fast release CRF

C thiobencar14 % n alginat i TH21 b8. - d kaolinan e , specific activity = 8.071 uCi/g, slow release CRF

TA1- 4% 14C thiobencarb, 4% PVA, 8% POEG and 84% corncob, specific activit 5.40y= 5 uCi/g, fast releasF eCR

TA2- 4% 14C thiobencarb, 8% PVA, 4% POEG and 84% corncob, specific activity= 5.405 uCi/g, slow releasF CR e The formulations were added to water to make a concentration of 16 mg/L. The flasks were wrapped in aluminum foil to minimize photodegradation and water was sampled at 2, 4, 8, 24, 48 and 96

263 hrs then l, 2, 3 and 8 weeks for released radioactivity. Analysis was done by adding 15 ml of universal cocktail Ins tage l XF (Packard) before counting the radioactivity by Liquid Scintillation Counter (LSC) f Tricarb 1000, Packard Instrument Downer. Co s Grove USA], ,II eact .A h sampling day watee ,th s rwa stirred gentld fou watean l m rs replaces sample ywa wa r d an dd wit e samhth e amount using distilled water releasr Fo . e raten si paddy water,the sampl s centrifugeewa d before taking sampler fo s wit4 7. h f padde o 0.059 Th H yp LSC . %a wate totad rha l suspended particles. The average water and ambient temperature were 28 and 30°C, respectively. e eightth f ho weed en k e periodth t A , pellets were filtered and the flasks were rinsed with 100 ml each of water then acetone e filteTh . r paper, wate d acetonan r e rinsates were analyzed and total radioactivity was accounted for. The filtered formulation was analyzed by combustion using a Biological Material Oxidizer (BMO) [Model OX-400, Harvey Instruments Corp., Hillsdale , USA]NJ ,e releaseTh , d 14C s absorbeOwa a n i d 2 mixture of 10 ml absorbent and 5 ml prepared liquid scintillation cocktail e absorbenl Th . m a solutio5 s 12 wa t f o n ethanolamine with 875 ml methanol to make one liter. The cocktail s preparewa dissolviny b d POPOg m 0 P5 g (1,4 bis-5-phenyloxazole- O (2,2-ylPP 5g benzene0 diphenyloxazole5. d an ) n tolueni ) o t e make literon e . After e preparecombustionth f o dl m cocktai5 , l s furthewa r adde radioactivitd an d measures ywa LSCy b d . e wate Th s extractewa r y partitioninb d l gm wit0 10 h méthylène chlorid d drie ean passiny b d g through anhydrous sodium sulfate. The extraction was done three times with 50 ml solvent e pooleanth d d dried extracts were concentrate o almost d t dryness, taken in 4 ml methanol and was analyzed by Gas Liquid Chromatography (GLC). Extraction recover 89-90%s ywa .

264 The Hewlett Packard Gas Chromatograph Model 5840 was operated wit followine hth g conditions. Column: 6 ft X 1.5 mm ,glass, 3% SE-30 on chromosorb WHP 80/100 Temperature: column - 200 (°C) detector - 300 injector - 220 Gas: hydrogen - 3 (ml/min) 0 nitroge3 - n air - 50 Confirmatory analysi f wateo s r extrac s don wa y tthib e n layer chromatography (TLC) using benzene:ethyl acetate (10:1) mixture as developing solvents (Ishikawa et al., 1976) and the spots were viewed under ultraviolet light.

Toxicitv and uptake of thiobencarb to Oreochromis niloticus fingerlincrs

Two week old Nile tilapia, Oreochromis niloticus fingerlings with body length of approximately one inch were acclimatized for n aquariua e wee on n i kd wer an me given rice bran rationse Th . fingerlings d unti weran s e prio hr th elo t r4 starve2 r fo d completion of the test. The fish were exposed to a series of concentration e mortalitth d an s y count were done afte 8 hrs4 r .

LCe valuTh calculates ewa probiy db t analysis. 0 5 Nile tilapia used for rice fish culture with average weight 0 gram3 f so were expose o radiolabelet d F thiobencarCR d b (TA1=2.83% a.i., activity = 11,752 dpm/mg and TA2 = 3.05 % a.i., activit y12,22= 1 dpm/mg commerciad )an l formulatio t amaximua n m release of 0.2 ug/ml. The water level was maintained daily and sampled for 15 days then analyzed by LSC. Dead fish were 14 collecte d concentratioan d C thiobencar f o n s analyzewa b y b d BMOe amounTh . t bioaccumulated afte day7 rs als wa s o determined by combustion of the tissues of the surviving fish. A parallel test using commercial formulation (Saturn 10G) and untreated

265 contro s don o wa determinlt e e mortalitth e y differenco t e du e thiobencarb released from CRF.

uptakthiobencarC 14 f eo y b b edible kangkong (Ipomoea aquatica) Radiolabeled corn cob based TA1 and TA2 were applied at the ug/m1 rat f maximut eo la m release pare .On t kangkon part0 1 o sgt water was maintained. Sampling of water was done daily up to 15 dayd kangkonan s s samplewa g t 1,3,5,7,1a d 5 day1 sd an afte0 r application. A parallel test was done using commercial formulation to determine if there were any visible signs of phytotoxicit o thiobencart e du y b uptak y kangkongb e e planTh . t o youngestw , wite up th h t t shootcu s wa s classifie s youna d g leaves,all others as old leaves cind the stem. A 0.2 g sample was analyzed by combustion using a BMO.

Herbicidal activity agains crus-gall. tE i

Maahas clay loam sois collecte wa le Universit th t a d f o y e Philippineth s BanoLo st sa Experimental coveres Statiowa d dnan with a fine mesh cloth to prevent contamination of other weed species. The soil was left undisturbed for 2 weeks to ensure the germination of other weed seeds present in the soil. Equal amount of soil was weighed and put into plastic pots with 14 cm diameter. Two week old rice seedlings (variety IR66) were transplanted at three seedlings per hill per pot. Each pot was seeded with eleven Echinocloa crus-galli seeds to compensate for the 10% ungerminated seeds as predetermined by a germination test. Four days after transplanting (DAT) e testh ,t formulations [TA1 (corncob, fast release) 2 (corncobTA , , slow release) L 249TA ,0 (alginate formulation, medium release d commerciaan ) l

266 formulation] were applied at the rate of 1.5 and 0.75 kg a.i./ha wit 4 replicationsh . Control pots wit herbicido n h e application was done to determine the difference in yield due to weed control.

At 21 and 49 DAT, the number of germinated weeds and plant heigh f boto t h weericd an de plant were measured. Fertilizes wa r applied at the rate of 60 kg N + 14 kg P + 14 Kg K per hectare. Nitrogen fertilization was split into three applications: basal, 25 DAT and at panicle initiation whereas the P and K fertilizers were applied basall t lasa y t harrowing. Yiel s projectewa d n o d per hectare basis with the assumption of 16,000 hills/ha.

RESULTS AND DISCUSSIONS Release rat distillen i e d water

The release of thiobencarb from CRF alignate in distilled wate s relativelrwa y slower tha e corb baseth nco n d formulation Maximu. (Fig) 1 . m concentratio h wee8t k s releasee wa nth n o d equivalen o 68.5e totat d 61.56th an 0lf o amoun% t applier fo d

percentage released concentration

10 20 30 40 50 60 days after application

algmate.fast algmate.slow corncob,fast corncob,alow Fig 1 Release rate of thiobencarb formulation n distillei s d water

267 TH1 and TH2, respectively. In contrast, TA1 and TA2 released 50- 58% and 45-50%, respectively starting 7 to 56 DAA. TLC analysis e méthylènoth f e chloride extracts produce o distinctw d t spots

for the alginate formulation with average R value of 0.15 and f

0.68. The reported R values of the metabolite 4-chlorobenzyl f methyl sulfone and thiobencarb are 0.18 and 0.74, respectively (Ishikawa et al., 1976). However, further confirmatory tests should be done. Low total recovered radioactivity after the eighth week may be due to the high room temperature (28-34°C) and may effect volatilization with water (Table 1). Higher recovered radioactivity was found for fast than slow release formulations r botfo h alginat cord baseb ean nco d CRF. The corn cob based CRF of thiobencarb may be more efficient since maximum concentratio s attainenwa d faste d releasan r s wa e sustained ove a rlonge r perio f time o dmaximue Th . m releas7 t ea

Table 1

Total accounte thiobencarC 14 d b after eight weeks, aa percentags e totath f l o eamoun t applied Formulation alginate corncob fast slow fast slow

Water(%)* 88.50 61.56 53.47 49.68 Beads(%)** 22.28 15.46 17.66 16.48 Degraded(%)*** 7.82 4.61 Total recovered 90.78 77.02 78.95 70.67 (%)

' includes correctio o samplingt e du n , wated an r acetone rinsates and filter paper counts % remaining in granules determined by combustion difference betwee e maximuth n m concentration released and the concentration at eight weeks

268 days after application is an advantage because keeping the rice weed-free at 7-40 DAT is the critical period for competition (Mercado, 1979; Zimdahl, 1980). Weeds emerging 40 days after transplantin t affec no r late o gd tdi r yield (Zimdahl, 1980n I ). comparison e alginatth , s ejus wa baset F abouCR de releasth t e concentration of the corn cob CRF at approximately 30-40 DAA (Fig. 1). Release rate in paddy water

The release rate in paddy water indicates that degradation and/or soil adsorptio r absoptioo n s fastewa n r thae releasth n e of thiobencarb e maximu(FigTh . m2) . amount release s 35.7wa d 8 and 23.66 percent of the applied thiobencarb for TA1 and TA2, respectively. The amount released at 14 days after application was 13.30 and 11.08% in paddy water as compared to 58.23 and 49.0 n distillei 5 d TA2an ,1 d TA wate respectivelyr fo r . This indicate e higth s h adsorption capacite suspendeth f o y d soil

percentage of the original concentration

2 1 0 1 8 6 days after application

TA1(fast)-DW - TA2(slow)-D-* W TA1(fast)-PW -*- TA2(slow)-PW

DW-dlstllled water; PW-paddy water

Fig2 Release of fast and slow corncob formulatio distillen ni paddd dan y water

269 particles and the effect of the slightly basic pH. The concentration of thiobencarb in water and in the upper soil layer is important for efficacy.

Toxicit f thiobencaro v o fist b h

The LC of technical thiobencarb to 2 weeks old Nile tilapia 0 fingerling e 0.23s b founwa so 4t d mg/L. Thiobencarb coule b d lethal to fish present in the paddy based on the maximum concentration foun y Rosb d Sav an s a (1986 n paddi ) y water fo r commercial formulation which is 0.676 mg/L. At maximum concentration of 0.2 mg/L, released thiobencarb in water remains relatively constan d TA2an r bot1 .fo tTA h However, the concentration in fish was variable and no relation f lethao l concentration with tim f exposuro e s notewa e d (Fi3 g and 4) . This might be due to other factors like sensitivity as affecte y b sizd e differences» Average concentratiof o n thiobencar dean i b d fish fillet were 0.31 0.33d 2an 2 respectively

concentration (ppm) 1.2

1

0.8

0.6

0.4

0.2

0 4 6 8 10 days after application

water fillet * head

Fig 3 14C thiobencarb in Tilapia exposed to fast release corn cob formulation

270 concentration (ppm)

1.5

* 0.5 - 4-

4 6 s 10 days after application

water fillet head

Fig 4 14C thiobencarb in Tilapia expose sloo dt w release formulation d TA2an .1 fo TherTA r e wero significann e t difference between e concentratioth n accumulate y survivinb d g fish afte 7 dayr s exposur d concentratioan e f thiobencaro n n deai b d fishe Th . concentrations of thiobencarb in survivors were 0.256 and 0.285 ug/g respectiveld TA2an .1 ThiTA sr coulfo y d indicate that with controlled release formulation, the tilapia could survive even if the concentration in water is within the limits of LC50 value. More residues were found in the head compared to fish fillet. Tejad Bajed an a t (1993) reporte LC5n a d0 valu f 1.9 eo d 1.1 7an 4 ppm, respectively at 24 and 48 H, respectively for the commercial formulation. Zhang et al.(1991) reported a Tim of 3.2, 2.8 and 2.5 ug/ml respectively at 24,48 and 96H.

Uptak thiobencarf eo edibly bb e kangkong (Ipomoea aquatica)

Low level uptake of thiobencarb from CRF TA1 and TA2 was found for both young and old leaves (Fig 5 and 6) . However,the

271 concentration (ppm)

2 1 0 1 8 6 4 2 14 16 days after application

—— Water —— Young leaves —*— Old leaves Stalk | Fig. 5 14C thiobencarb in Kangkong exposed to fast release formulation

concentration (ppm)

o 6 a 10 12 14 16 days after application Water Young leaves —*— Old leaves Stalk Fig 6 14C thiobencarb in Kangkong exposed to slow release formulation

stalk absorbethiobencarC 14 r e dbotfo th hmos bf o t formulations. No visible signs of injury of kangkong like necrosi r phytotoxicito s s observed.Thiwa y s indicates high

toleranc f kangkono e o thiobencart g o maximut p u b m releas1 f o e ug/ml.The concentration in water remained relatively constant up 5 day1 s o t after application.

272 Herbicidal activity to Echinocloa crus-galli

t Thexperimenpo e t showed thal treatmental t s have significantly lower number of weeds than the control at 21 and 49 DAT Ther. (Tabl ) e2 e wer o significann e t differences between treatments and rates based on weed number data. It appears that the rate of 0.75 kg a.i./ha could be sufficient to control E. crus-galli. However e commerciath , l formulation applie t 0.7a d 5 kg a.i./ha resulte n tali d l weeds comparabl e untreateth o t e d at 49DAT. No phytotoxicity in rice was observed by visual assessment but ther s significanwa e t stuntin e ricth ef o gplant s treated with the commercial formulation compared to the control at 21DAT (Table 3). No significant difference in plant height and tiller numbe s observewa r d w tille9 DATlate4 lo t e .a rTh numbed an r o keepint e plante shade du gth th s planen wa si T tDA heigh1 2 t ta o weekfotw r s after transplanting .Thi s don wa o sprevent e t dilution due to heavy rainfall and loss of weed seeds due to splashing.

Table 2. Effect of CRF thiobencarb on pant height and numbe f Echinocloo r a crus galli seedlings

Treatment Rate Helght(cm) Number (kg al/ha) T DA 1 49DA2 T 21 DAT 49DAT

TA1 0.75 O.Od O.Od O.Ob O.Ob 1.50 O.Od a.od O.Ob 0.5b TA 2 0.75 O.Od 9.0d O.Ob 0.3b 1.50 1.1cd 18.3cd 0.5b 0.8b TAL 2490 0.75 3.0bo 27.0bc 1.8b 1.3b 1.50 O.Od 12.0cd O.Ob 0.8b SaturG n 0.75 4.4b 39.5ab 1.0b 1.5b 1.50 1.2cd O.Od O.Sfa O.Ob Control - 10.8a 49.3a 6.5a 6.0a

•means followe e samth ey b dlette r (a,b,c,d)art no e significantly different at the 5% level

273 Table 3. Effect of CRF thiobencarb on yield, height and tiller number of rice

Rate Height Tiller No. Yield Treatment Kg ai/ha 21 DAT 49DAT T DA 1 2 49DAT Tons/ha

TA1 0.75 18.4abc 54.4a 4.0ab 28.5a 4.04a 1.50 20.5abc 56.0a 3.0b 25.8a 3.43a TA 2 0.75 19.8abc 62.8a 3.0b 30.5a 3.75a 1.50 18.5abc 61. 5a 4.8a 26.0a 4.08a TAL2490 0.75 22.2ab 55.5a 3.0b 31.5a 3.56a 1.50 23.2a 55.8a 3.0b 26.8a 4.02a SaturG n 0.75 16.9bc 57.7a 3.5b 28.3a 3.48a 1.50 16.0a 62.0a 5.0a 26.0a 3.77a Control - 22.6a 57.1a 3.0b 21.8a 1.45b

* means followed by the same letter are not significantly different at 5% level

The control pots had an average of 6.5 weeds/hill. The low germinatio e weed th s enhance e f wa environmentaso nth y b d l conditions wherein the soil was partially or fully submerged most of the time. These conditions were not conducive to the growth of E. crus-qalli. Projected yiel f IR6do 6 treated with CRF f thiobencaro s s bwa not significantly higher than the commercial formulations. However l treatmental , highed ha s r yield thae untreateth n d control (Table 3). The application of 0.75 kg a.i./ha was the most cost effective rate having no significant yield difference over the recommended rate of 1.5 Kg a.i./ha..

Table 4. Projected yield of rice variety IR 66 treated thiobencarF witCR h b

Formulation Rate Yield (kg ai/ha) tons/ha

TA1 0.75 4.04 1.50 3.43 TA 2 0.75 3.75 1.50 4.08 TAL 2490 0.75 3.56 1.50 4.02 Commercial formulation 0.75 3.48 1.50 3.77 Control - 1.45

274 CONCLUSION

e corb formulatioTh co n e mors founb wa neo t dusefu l than alginate th e based formulation becaus maximue eth m release th f eo active ingredien n watei ts sustainewa r d from 7-56 DAA.This coincides wit criticae hth l perio weer fo dd competitio ricen ni . Yiel ricf o d e applied with fast (TA1 slod an )w (TA2) release corn cob formulation was not significantly different from the commercial formulation when applied 0.7an t 5g bot5 a K d 1. h a.i./ha. CRFs were shown to be effective against Echinocloa crus- e halrecommendee gallon th f t a i d rate. w leveLo f uptako l d leave y ol youneb d kangkonf so gan s wa g observed witC thiobencarb14 h e residueth , s being concentrated e stalkmostlth n i .y Fille f Tilapio t a accumulated 0.25d an 6 0.285 ug/g when expose a maximu o t d m concentratio 2 ug/m0. lf o n fo day7 r s using fast (TA1 d slo)an w (TA2) release formulation, respectively. More residues were accumulate heae th d n thai d n i n the fillet. LC50 of Tilapia was 0.234ug/ml using thiobencarb.

ACKNOWLEDGEMENT e authorTh s gratefully acknowledg e assisstancth e d an e suggestions of Dr. Manzoor Hussein of the International Atomic Energ ye suggestion th Agenc d an y f Lorenzo s . FabroE o , weed scientist of the National Crop Protection Center for improving e efficaccth y evaluation. This research projec e s fundei tth y b d Internatonal Atomic Energy Agency.

LITERATURE CITED Bahadir ,M.and G.Pfister. 1990. Controlled release formulations of pesticides. Chem Plant Prot.6:l-64. Connick,W.J ., J.M.Bradow . WellsW , , K.K.Stewar d T.Kan d . Van. 1984. Preparatio d evaluatioan n f controlleo n d release formulation 4 dichlorobenzonitrile2, f o s . J . Agric Food Chem. 32:1199-1205.

275 Hussein, M., J. Can and M.N. Rathor. 1992. Preparation of Controlled release formulations of 14C Labelled thiobencarb herbicide and study of their environmental behaviour. Pestic Sei: .34 Hussein,M. and B.Y. Oh. 1991. Preparation and study of controlled release formulation of 14C labelled butachlor. Toxi. Environ. Chem 101-110: .35 . Ishii,Y. 1974. Saturn: New selective herbicide. Japan Pestic.Info.19:21-25. Mercado,B.L..1979. Inroductio o Weet n d Science. Southeast Asian Regional Cente r Graduatfo r e Stud d Researcan y h in Agriculture. College, Laguna Philippines Pfister ,G.M. , M. Bahadir and F. Körte. 1986. Release characteristic herbicidef so s from calcium alginatl ege formulation. J. Controlled Release 3(4):229-233. Ross,L.J. and R.J. Sava. 1986. Fate of thiobencarb and molinate in rice fields. J. Environ. Qual. 15(3):220- 225. Tejada,A.W., C.M.Bajet 1993.Toxicit somf yo e pesticides used in rice. Paper submitted for publication to the Bull. Environ. Contain. Toxi. Zhang, Q.H., J.H. Sun, X.M D.F.Chang, .Li . 1991. Researcn ho the herbicidal efficacc d residuean y f controlleo s d release formulations of thiobencarb in rice fish ecosystem. Proceed FAO/IAEd .2n A Research Coordination Meeting on Controlled Release Formulation. Beijing Agric. Univ t 14-18,1991.Oc . Zimdahl,R.L. 1980. Weed-Crop Competition: A Review. International Plant Protection Center, Corvalis, Oregon

276 POTTED PLANTS GREENHOUSE STUDY ON THE EFFICACY OF THIOBENCARB FORMULATIONS AGAINST EcMnocbloa crusgaUi IN DIRECT SEEDED RICE (Abstract)

D. OMAR, R.B. MOHAMAD Faculty of Agriculture, Universilti Pertanian Malaysia, Sedang, Selangor, Darul Ehsan, Malaysia

An experiment was conducted under greenhouse conditions to study the effect of the timing of applicatio typd thiobencarnf an e o b formulatio controe th n nEchinochlof o lo a crusgalli grown together with ric potten ei d plants formulatione Th . s include controlled-releaso dtw e (CR) formulationd an s a conventional formulation. The CR formulations were i) 7.56% thiobencarb uniformly dispersed in a mixture of calcium alginate and kaolin and ii) thiobencarb adsorbed on corn cob garnules, which were then coated with a mixture of poly vinyl acetate (PVA) and polyoxyethylene glycol (POEG). The cor formulatiob nco thiobencarb% 4 d nha conventionae Th . l formulatio Saturns nwa R which contained 5% thiobencarb formulatione Th . s were applie a.i./hg 1.5d k 0.7t 0dan a day 51 a d rates beforan s6 e sowin pregerminatee g th (DBS f o ) d rice seedlings. Seed Echinochlof so a crusgalli were e sowth n i same pot t 3DBS sa herbicid e effece th f Th .o t e formulation germinatioe rated th san n so weee th f dno seed determines swa day7 t da s after sowing (DAS) growte weeeffece e th ric d th n Th f d.o ean ht o plants was determined periodically untill 70 DAS. The timing of application of the herbicide did not shogerminatioe effecy th wan n o tweee th f dno seeds. However, higher rat f applicatioeo e th f no herbicid mors ewa e effective tha lowee nth r ratel threAl . e formulation t botsa h rates were effective, anherbicide dth e treatment resulte increasen di d numbe tillersf ro , heigh weigh d rice tan th e f plantto s and reduced number of tillers, height and weight of the weed plants. These in turn had effect on the yiel ricf do e grain whic substantialls hwa y greate pote th s n rreceivini g herbicide treatment. However, highee th r cor e formulatiodosb th nco f ecommerciao e th d nan l formulation were more effective than other treatments.

Next page(s) left blank 277 STUDIES ON THE EFFICACY OF HERBICIDES IN RICE CULTURES (Abstract)

J.DOMBOVARI, M.ONCSIK, L.SZELVASSY Irrigation Research Institute, Szarvas, Hungary

Echinochloa species are commonly found in the rice growing areas of Hungary. Preemrgent herbicides, including thiobencarb, are used more frequently for the control of these weeds. Laboratory experiments were conducted to study the release of thiobencarb from two controlled- release (CR) formulations field an d, experiments were conducte comparo dt efficace eth f severayo l herbicide controe growte th th n n si f weedf ric o o hlo d e plantsan graid san n yiel direcn di t seeded as wel transplantes a l formulationR dC ricee Th . s include cor o formulationb dtw n co alginatn a d san e formulation cor n formulationI b . nco s thiobencar absorbes bwa corn granuledb o nco s which were then coated wit hmixtura f poleo y vinyl acetate (PVA polyoxyethylend an ) e glygol (POEG). These were TA1 ,proportio 2 proportion 1 containin o POEd t o t 1 an 2 TA2d e G n A i n ni an Th ,gPV . alginate formulation (CRA) contained 6.5% thiobencarb uniformly disperse mixtura n di calciuf eo m alginate and kaolin. Field experiments were carried out to test the efficacy of a number of herbicides in direct seeded rice. Thes commerciaa e d includean A dCR l formulation (Saturn thiobencarbf )o , Stomp 330E formulation of pendimetalin and Molinate 10GR . Similarly, the efficacy of CRA and Molinate at 1.5 and 3.0 kg.ha was compared in transplanted rice. In the laboratory tests the release of thiobencar fastes bwa r froformulatio1 mTA n than fro mfiele th TA2 dn I experiment. s with direct seeded rice Stomp 330 t E1.6a 5 kg/h mors awa e effectiv controe th n i ef Echinochloa o l species than the other herbicides; whereas commerciae th , l formulatio thiobencarf no somewhas bwa t better than the CRA formulation. However, yield of rice grain and total rice biomass was the highest with Molinate used at 5 kg/ha. Grain yield with the CRA formulation was better than with the commercial thiobencarb formulation. Total rice biomas thiobencaro s wittw samee e hth th s . bwa Whe effece nth t of thiobencarb CRA formulation was compared with that of Molinate in transplanted rice, the rice grain yield for the two herbicides was the same at 1.5 kg/ha, and both treatments resulted in an increase of about 50% rice grain as compared with the control plots. However, when the herbicide increaserats A ewa kg/hCR 0 e 3. th o da t r rate fo yiele Molinatr th ,% fo d 40 increas % d 45 e an s ewa formulation. Similar results were obtained wit totae hth l rice biomass.

Next page(s) left blank 279 STUDY ON THE DYNAMICS OF RELEASE CARBON-1F O 4 LABELLED HERBICIDES FROM CONTROLLED-RELEASE FORMULATION WATEN SI R (Abstract)

Fujun WANG, Mengwen QI, Huaguo WANG Laborator r Applicatioyfo f Nucleano r Techniques, Beijing Agricultural University, Beijing, China

Experiments were conducte studo dt processee yth s governine gth release of carbon-14 labelled butachlor and thiobencarb herbicides from two type of controlled-release (CR) formulations into water. The formulations were i) herbicides unformly dispersed in a mixture of calcium alginate and kaolin an herbicide) dii s adsorbe cor n granuleb do nco s which were then coated with a mixture of poly vinyl acetate (PVA) and polyoxyethylene glycol (POEG). Theoretical models and corresponding equations were developed and the experimental data for the release rate of the two herbicides was compared with the models. The results indicated that the release of the herbicides from the alginate formulations was governed by diffusion process; whereas, the release from the corn cob formulations was thermodynamic process. The diffusion of the herbicides from the alginate granules was inversely related to the thickness of the granule, with slower release from the larger granules and vice versa. The release of the herbicides from the corn cob granules was related to the type compositiod an polymee th f no r mixtur surfac e granulese th th n f eo e eo Th . study showed tharate releasf th teherbicide o e th f eo fastes swa r fro core mth n cob formulations than the alginate formulations.

281 STUDIES ON THE EFFICACY OF THE CONTROLLED-RELEASE FORMULATIONS OF HERBICIDES AGAINST WEEDN SI TRANSPLANTED RICE (Abstract)

Fujun WANG, Mengwen QI, Genhai YANG, Huaguo WANG, Quinhua GUO Laboratory for Application of Nuclear Techniques, Beijing Agricultural University, Beijing, China

Weed control efficacy of controlled-release (CR) formulations of thiobencarb and butachlor were compared with the commercial formulations of these herbicide transplanten si d rice field experiments during 1990, 199d 1an e controlled-releas1992Th . e formulations comprise ) herbicidedi s uniformly disperse a mixtur n i d f calciuo e m alginat ) herbicideii d kaoli an d e an n s absorbed on corn cob granules which were then coated with a mixture of polyvinyl acetate (PVA) and polyoxyethylene glycol (POEG). Some of the alginate formulations also contained sandherbicidee Th . s were applie thret da e different rate f 1.50so , 1.0 0.7d 5 an a.i/h5g k day4 a3- s after transplanting. The weed control efficacy was directly related to the rate of application of the herbicid thid turn esi an n resulted into relevant increas graie th nn ei yielde Th . contro f weedo l hany b s dapplicatioe weedinth e y herbicideb th r f go o n s resulted in significantly higher yield of the rice grain than the control plots. On the other hand, there was no significant difference in the yield of the rice grain from hand weeding or the use of herbicides. There was also no significant difference between the increase in the grain yield resulting from the use of the commercia controlled-release th r lo e formulation botd san h typ formulationf eo s provided equally good weed control.

Next page(s) left blank 283 STUDY ON THE FATE OF CONTROLLED-RELEASE FORMULATIO CARBON-1F NO 4 THIOBENCARB IN A MODEL PADDY ECOSYSTEM (Abstract)

J.H. SUN, X.M , Q.ZLI . . ZHANG, Z.Y. CHEN Institute of Nuclear Agricultural Sciences, Xhejiang Agricultural University, Hangzhou, China

calciua e fat f o eTh m alginate controlled-release formulatiof no carbon-14 labelled thiobencar studies bmodewa a n di l paddy ecosysteme Th . concentration of the released thiobencarb first increased with time and then declined e herbicidTh . s tranformewa e d int2 radiolabelle1 o d degradation product watern i se majo Th . r metabolite remained unidentified e soith l n I . most of the herbicide was found in the upper layer and there were at least 6 metabolites, som whicf eo h were conjugate fore th soil-bounf mo n d i d residues. e timA th f tharves o e t 45.6 e totath %lf o thiobencarb-relate d radioactivity present in the soil was extractable with the solvents, and the rest remained bound herbicide Th . e translocated ricinte partl th oeal f plantso s growe th n ni treated soilresidued an , s were foun strawn i d , roots, grai huskd nan . More than 92% of these residues were in the form of tissue-bound residues. In the plant the herbicide was degraded into 9 metabolites. The relative proportion of the metabolites changed with time. The residues which were extracted from the plant containe d11.2d 55.9an %1 6. , thiobencar strawe th n bi e , th grai d nan husk, respectively. The prédominent metabolite in the grain remained unidentified.

Next page(s) left blank 285 STUDIECONTROLLED-RELEASE TH N SO E PESTICIDE FORMULATIONS FOR PEST CONTRO COTTON LI D NAN MAIZE USING ISOTOPE TECHNIQUES*

F.F. JAMIL . TAMIM , L QURESHI . HAQA , , S.H. MUJTABA NAQVI Biological Chemistry Division, Nuclear Institut r Agriculturefo Biologd ean y (NIAB), Faisalabad, Pakistan

Abstract

Studies were conducted on controlled release 14C- carbofuran formulations with EVA for pest control in cotton and maize to investigate the fate and degradation of the release carbofuran from the formulation. Cotton plants both in the field and pots were subjected to three different treatments at the time of sowing: application of 14C-carbofuran formulation with EVA; cold carbofuran formulation with EVA granulad an , r carbofuran pesticide. It has been found that insect attack was much more on control than on treated plants for three months after germinatio t afte5 monthbu n4- r s granules treated plants offered less resistanc o insectt e s comparea s o t d formulation treated plant both hot and cold. The granules treated plants produced much more cotton lint in the first picking which reduced gradually and consequently hot and cold formulation treated plants produced maximum yield. Radiometrie analysis indicated that recovere t formulatiodho n pieces retaine t leasradioactivitda % 24 t y aftex si r months. More activit s recovereywa plann i d t leave s comparea s o root t d d ste d similarlan sm an y from soil samples, highest radioactivit observes y wa sample trene a di Th resultf do dept a . m t froc s a uptf cm 0 sh mo 1 3 o1 from pot and field experiments were almost similar. The usedC-carbofuran formulation with EVA recovered

from the field after cotton harvest was reused for maize crop at th H e time of sowing. Radiometrie analysis revealed that the reused formulation pieces after three months still retained 56% radioactivity of the original dose applied. This confirms thaformulatioe th t s indeenwa d slow releas eved lowee ean nth r amount usee th dn i sformulatio n pieces were available to the next crop. The corn ear-worm attacked the crop at the time of maturity but this proble s minomwa r with formulation treated plants resultin betten gi r yield over control plants.

1. INTRODUCTION Cotton crop wits multifariouit h s advantagee b n ca s rightly calle e lifth d e lin f Pakistao e n economys i t I . cultivate n 2,83o d 6 thousand hectares providing about 12.867 thousand bales. It accounts for sixty percent of the export earning d fiftan s y five percene domestith f o t c edibll oi e production in the country. It also provides raw material to 1035 ginning factories 2 Textil26 , e mill d 13,00an sl expellinoi 0 g units, besides providin opportunitieb jo g millionso st peoplf 'o e e fiel icottod nth an d n based Industries [1]. The crop season is suited to insect development and it is heavily attacke severay b d l insect pests majoe ,th r ones being

* Research Carried out with the support of the FAO/IAEA/GSF under Research Contract No. 3694/GS.

287 Empoasca devastans (Jassid), Bemisia tabaci (White fly), Earias insulana and Earias fabia (Spotted bollworms,) Pactinophora gossypi&lla (Pink Bollworm) and Hiliothis armigera (American Bollworm) e cottoTh . n crop needs heavy insecticidal protection and consumes well over 80% of the total insecticides used in the country e maizTh .e cro cultivates i p d bot summen i h winted an r r e majoanth d r pest e aphidsar s , leaf hopper d maizan s e borers [2], The insecticides used are systemic as well as others. The existing method f usino s g pesticide r controfo s f cottoo l d an n maize pests results in wastage by way of degradation, accumulation in other organisms, etc. The Controlled release pesticide technology can be effective in improving the efficiency of some existing pesticide n reducini d e environmentaan sth g l problems associated with others [3,4,5]. Isotope techniques offer precise quantitative method r measurinfo s g release th f o e pesticide from the formulation, the stability of the pesticide withi e formulationth evaluatind an n effecte gth factorf so s such as soil moisture, irrigation, rains and temperature in releasing the pesticide froe formulatioth m n [6,7,8] e worTh k. plan

envisaged the study of the fate arid degradation of slow release -C14 carbofuran formulation n cottoi d maizan n e using radiometeric techniques under conventional agricultural practices and environmental condition f Pakistao s o obtait n n information about its efficacy and for obtaining the conditions for its use.

. 2 MATERIAL METHODD SAN S 2.1. The chemical: C-Carbofuran formulations with EVA (Polyethylene- polyvinyl acetate co-polymer) with a specific activity 0.016 mCi/ d granulaan g r carbofuran pesticide were supplie y GSFb d . Munich, Germany.

288 2.2. studies with 14C-Carbofuran on cotton: One hundre d twentan d y five a plastidi m c c0 3 pot f o s lines with polyethylene Sheet were filled with 22 kg of good homogenised soil. To forty eight of these prepared pots 1 gm of formulation pieces (1.5 - 2 cm) containing 19.2 mg carbofuran and 16.i radioactivituC 9 d y an wer a edi appliem c a circl 5 n i df o e . Foucm a a deptt5 r f cottoo h n seed f varieto s y NIAB-78 were placed in the centre of each of these pots. Inorder to get a comparison between the conventional application of carbofuran and

s formulatioit tha f o t n carbofuraC) n solutio+ (C n i n

14 2 1 acetone (2 ml containing 19.2 mg and 16.9 uCi) was applied to another similar set of forty eight pots with four seeds of NIAB- pote th e ress Th f wero t . e 78 planted with cottont seedge o t s control plants. A single plant was maintained in each pot. The first irrigatio s appliewa n o daytw ds after germination. Normally the pots were irrigated twice a week for the first five week d thean sn after every e weekfielTh .d experimens wa t conducte a fiel n i dd plo f 30/3o t m 0siz e divide o fout rn i d small plots for the three treatments and control, with plant to plant distance of 35 cm and row to row 80 cm. All normal agricultural practics like fertilization, irrigation etc. were followed throughou croe tth p period. Sampling from pot dons swa e every week in duplicate for cotton plants, soil and also for residual formulation piece n cas i sf formulatio o e n treatments. Soil samples were collected wite hel f mini-soith ho p l samplers for 10 cm dia (which were again divided into upper 12.50 cm and lower 12.50 cm portions, 10 cm - 15 cm and 15 - 30 cm to get information about the migration rate of carbofuran from the polymeric matrix to soil and its further movement in the soil. e fielIth n d experiment, only cotton plants were sampled. Samplin f soio d gformulatio an l n pieces froe fiel th s mdon wa d e

289

experimente th f o d en C-Carbofura. e atth n formulation treated 14 cotton plants, their residual formulation piece soid san l samples were combusted using sample oxidizer (Packard 306-TR1-CARB) and finally counted by Liquid Scintillation counter. Data for insect attac alss kwa o recorded throughou croe tth p season.

2.3. Studies with c-Carbofuran on maize: The useC-Carbofurad14 n formulation with EVA recovered from the field after cotton harvest (six months from the time of application e glass storeth wa ) sn i dvial r eighfo s t months. This was reused for maize variety Akbar at the time of sowing in August, Observin l normaal g l agricultural practicese Th . radioactivity of maize plants, soil and residual formulation pieces was measured by combustion in a packard sample oxidiser and liquid scintillation counte s describea r d earlie r cottofo r n samples. Data for insect attack was recorded throughout the crop.

3. RESULTS AND DISCUSSIONS The insect attack was much more on control than on treated plant threr sfo e months after germinatio aftet 5 nbu r4- months granules treated plants offered less resistanc insecto et s s comparea o formulatiot d n treated plant sd col an d bott ho h (Fig. 1). In line with these result, granules treated plants produced more cotto e firsn th lin tn i tpickin g which reduced gradually and consequently hot and cold formulation treated plants produced maximum yield (Fig. 2) . Cotton plants in pots treated with carbofuran formulation looked healthier than those treated with carbofuran solutio controe th r o nl plants e corTh .n ear worm attacked the maize crop at the time of maturity but this problem was minor with formulation treated plants resulting in better yield over control plants (Table 1).

290 Fig: Pos1 . t incidcncc(/6 n cottoo ) n plants trcd :be with different form f insecticido s e aflcr 5 monbh d , n an ;5 ii i'icl-G d potc»an d .

291 Fig. 2: Effect of different treatments of carbofunn on cotton plant height, weight and yield grown in field and pots.

292 Tabl i eEFFEC4 1 C-CARBOFURA F TO N FORMULATIO PERCENE TH N O NT ATTACK OF CORN EAR WORM ON THREE MONTHS OLD MAIZE PLANTS AND THEIR FINAL YIELD.

Parameter t formulatioHo s studien d Control

Corn ear worm (% infected 4 40 plants) 5 9 Yield (g/plant 7 14 , averag 5 2 f o e plants).

Radiometrie analysis indicated that recovered hot formulation pieces retaine t leas a d% radioactivit 24 t y aftex si r months (Table 2). The plant tissue in the formulation treatment exhibited activity around three percen 5 week2 td o froan st 4 m then decrease weeks2 3 hale t on a f , o t dwhil casn i ef solutio eo n treatmen e activitth t plantn i y s increase % uptil3 o t d9 weekl s and then decreased graduall o 0.08 t yweeks2 3 e activit %t Th a . y e sois ihigheth nwa l r initially (upt e solutioweeks5 o th n i ) n treatmen d thean tn gradually decrease de formulatiowhilth n i e n treatmen w initiall lo e activit e soith ts th wa ln yi y (upt5 o weeks d the 2 week)an 1 nd the t an increasesa n % graduall52 o t d y decreased. The results related to movement studies of pesticides in soil and in different plant parts showed that the highest activity in soil of the formulation treatment was observed from 10 cm dia samples uptill 12.50cm depth as compared to other soil zones. Roots, sted leaan mf samples coule analyseb d d separately only when the plants were 14 weeks old. More activity s recoverewa d froe leaveth m s comparea s o roott dd stem an s . Similarl n cas i yf applicatio o e f solutiono n , radioactivits wa y

293 TABL E2 Percent recovery f 14C-carbofurano t a various time intervals fromvholethe cotton plant, total soilpot and recovered formulation pieces.

Pots Field TimeSoilPlantRecovered Total %PlantSoil Recovered inter- tissue formula- recovery tissue formula- val tion tion (month) pieces pieces

1 33. 5 5.7 a 40. 1 79.3 4.6 a * * (4 .8+. 9 ) (3.5+1.1) 2 42. 0 11.6 a 31. 8 85. 3 8.2 - - 0 ( 10+ 0 . 82 .70) 3 48. 8 15.0 a 28. 9 92. 7 12.5 - - (14.6+4) 4 51. 3 17.5 a 26. 5 95. 3 17.7 a - (15 .7+1. 8) . (159+ . 9) 5 54. 1 17.4 a 22. 9 94. 4 15.3 - - (16 . 1+1.3) . (144+ . 9) 6 49. 9 17.6 a 26. 1 93. 6 11. 4 a 51.5 24.2 (15 .1+2. 5) (10.2+1.2) * = Sampling was done after 6 months only. = Percen a t recover f 14C-carbofurao y n from extra cotton plants removed during thining afte week3 r f germinatioo s e th r fo n purpose of maintaining single plant in each pot.

recovered 10 cm dia soil sample uptill 12.50 cm depth and leaves had more activity than stem and roots. The studies are in confirmatory to our previous study (10,11) and lead to encouragement since leaves, which constitut majoe eth r foliagf eo plants and their exposure to insect attack is maximum and had contained more activit a longe r fo yr spa f timee o ntren Th f .o d results from potd fielan s d experiment s almoswa s t similar (Table 3, 4) . e radiometriTh c analysi maizf so e crop samples revealed that the reused formulation pieces after 3 months still retained 1.7 uCi/g radioactivity which was 56% of the original dose applied i.e. 3.1 uCi/g (Table 4). This confirms that the formulatio s indeenwa d slow releas eved ean n lower amounte th n si used formulation pieces were availabl e nexth t o t crope . Like

294 14 Table 3 PERCENT RECOVERY C-CARBOFURAN FROM DIFFERENT SOIL * ZONE D PLANSAN T PARTS AT VARIOUS TIME INTERVALS.

POTS FIELDS

Time 0-10cm 0-10cm 0-15 15-30 Root Stem Leaves Root Stem Leaves inter- uppem c r lowem c r val 13cm 13cm (month)

1 12.7 5. 6 4.7 10.5 _* whole , , ^ whole «• plant plant 5.7 4.6 2 15.0 6.3 14.0 6.7 - whole - - whole - plant plant 11.6 8.2 3 18.5 9.3 10.2 10.8 2.0 1.1 11.5 1.4 2.5 8. 6 4 19. 6 9.0 11.8 10.9 3.8 4.4 7.5 2.3 4.2 9.4 5 20.6 11.7 13.8 8.0 1.8 5.6 8.7 2.1 4.8 7.5 6 17.3 10.3 8.4 13.9 3.0 3.6 8.5 0.65 2.7 6.2

= Plan * t parts coul e analyseb d d separately only whee th n plants were 3 months old.

Table 4 PERCENT RECOVERY OF 14 C- CARBONFURAN AT VARIOUS TIME INTERVALS FROM THE WHOLE MAIZE PLANTS, SOIL, RECOVERED FORMULATION PIECE SEEDSD AN S .

Time Soil Plant Recovered Seeds Total (months ) tissues formulation recovery flakes (%)

1 19.8 * 0.24 66.8 86.84 (1350894) (16374) (4557562) (5924830) 2 26.7 3.58 60.6 90.88 (1821660) (244252) (4134554) (6200468) 3 28.1 6.6 56.1 0.0390.83 (1917178) (450298) (3827533) (2046) (6197056)

Applied dos r 682269o e i =3.uc 81 dpm/g/plant (100%) * = Radioactivity in dpm.

295 previous findings (9,10) soil samples contained highest a sampledi a radioactivitdeptm c m t f c a upts0 o h 1 3 1 o e th n i y and more activity was recorded initially for (2 months) from plant leave s comparea s o roodt ste d t afteman tbu months3 r e th , amount of radioactivity was almost similar in the roots and e facth t o t tha leavee tdu maiza se shollow b whics y i e ma h - rooted cro s comparepa e higheso cottot dth d nan t activite th f yo pesticide was confined to upper 13 cm soil zone arround the main diam c )ste 0 resultinm(1 n highegi r uptake root th thed y an sb ne through ste o mt leave d ultimatelan s y some activit s alswa oy e seedth y takesb providinp nu g them protection agains e peststh t

(Table 5).

TABLE 5 PERCENT RECOVERY OF 14 C - CARBOFURAN FROM DIFFERENT SOIL ZONES AND PLANT PARTS.

Soil Zones Plant Parts

Time 0-10cm 0-10cm 10-15 15-30 Root Stem Leaves Seed (month) upper lower cm cm 13 cm 13 cm

9 2. Whol4 1 .2 3 e 3. plant11. 3 2. 3 (770965)* (156922) (225149) (197858) (16374) 2 12.9 3.2 5.4 5.2 0.86 0.98 1.74 (880128) (218326) (368426) (354780) (58675) (66862) (118714) 3 13.3 2.7 2.3 9.8 2.4 1.9 2.3 0.03 (907418) (184213) (156922) (688624) (163744) (129631) (156922) (2046

Applied dose =3.1 uci or 6822698 dpm/g/plant (100%) Radioactivit= * n dpmi y .

4. CONCLUSION These finding o indicatd s e that controlled release formulations increase e efficac e dth pesticid th f o yeved e ean th n formulation pieces user cottofo d n proved quite effective when reuse n maizi d e crop a cro, f shoro p t duration. These studies

296 also indicated that controlled release formulation carbofaraf so n increase s efficacit d d protectean y e croa th d p r bettefo d an r longer period of time and reduced environmental pollution as well.

ACKNOWLEDGEMENTS

e senioTh r author wishe o thant s k Drs. BahadiM . d an r G. Pfister, G.S.F. Munich, German r providinfo y C- 14 e th g Carbofuran formulations with EVA. International Atomic Energy Agenc d Pakistaan y n Atomic Energy Commissio e alsar n o acknowledge e financiath r fo d l suppor researcd tan h facilities.

REFERENCES

1. Annonymous., "Economic surve f Pakistao y n 1992-93", Annual Report. Govt. of Pakistan, Finance Division Economic Advisors Wing, Islamabad. (1991-92), 89 - 97. 2. Annonymous., Agricultural Statistics of Pakistan (1990-91), Govt f Pakistano . , Ministr f Foodo y , Agricultur- Co d an e operatives; FooAgriculturd an d e Advisory Division, Economic Wing, Islamabad (1990-91). 34 - 0 ,3 3. . PfisterBahadirG d an . ,M Controlled release formulations of Pesticides; Chem-Plant-Prot. Perlin, W. Ger: Springer- Verlag (6) (1990), l - 64. 4. El-Nagar, M; Pfister, G., Bahadir, M. and F. Körte. "Preparation and release characteristics of controlled release formulations". Toxicol-Environ-Chem. London: Gorden and Breach Science Publishers 1991. 31/32 p. 383 - 393. 5. Wilkins, R.M., Batterby, S., Heinrichs, E.A., Aquino, G.B., Valencia, S.L. "Management of the rice tungro virus vector Nephotehix Virescens with controlled release formulations of Carbofuran". J. Econ-Entomol. College Park, Md. Entomological Societ Americaf yo ) (1984 499- (2 . 5 2 ),49 7 6. Hussain, M. , Oh, B.Y. "Preparation and study of controlled release formulation f carbon-1o s 4 labelled butachlor". Toxicol-Environ-Chem. Reading: Gordon and Breach Publishers; 33 (1/2) (1991) 101 - 110. 7. Jamil, F.F., Qureshi, M. J., Bashir, N. , Naqvi, S. H.M. "Studies on the controlled-release carbofuran formulations for pest contro n cottoi l n using isotope techniques". (Proc. 13th Int. Symp. on controlled release of Bioactive materials Norfolk. VA (1986) 238 - 239.

297 8. Oh, B.Y.,"Release profile of C-butachlor from controlled release formulation prefered with alignate-kaoline materix". Korean-Journal of Weed-Science (Korea Republic) 10 (2) (1990) 122-132.

9. Jamil, F.F., M. J. Qureshi, A. Haq, N. Bashir, S.H.M. Naqvi. "Efficacy of the controlled release of C-Carbofuran formulatio pesr fo nt contro cotton"n i l . (Proc. Int. Symp. on changing prospective in chemicals. Isotopes techniques e fostudth r f food o environmentay an d l implications". Neuherberg. Germany (1987) IAEA-SM 297/31 (1988) 169 - 175. 10. Jamil, . F.F.J Qureshi . M , , A.Haq, S.H.M. Naqvi. "Effectivenes e controlled-releasth f o s e carbofuran formulatio r pesfo nt contro n maizi l e using isotopic technique fieln i s d conditions". (Proc. 18th Int. Sympn o . Controlled Release of Bioactive Materials. July 8-11 Amsterdam Netherlande Th . s (1991).

298 CONTROLLEF O E US D RELEASE FORMULATION INSECTICIDEF SO S FOR Tip: CONTROL OF TERMITES AS PEST CROPF S O FORESTR D SAN Y

J.W.M. LOGAN Natural Resources Institute, Chatham Maritime, Chatham, Kent, United Kingdom

Abstract

Termite majoe sar r pest cropf so forestrd san tropican i y d lan sub-tropical regions of Africa and Asia. Until recently, they were controlled by organochlorine (cyclodiene) insecticides whose persistence protected the crops till harvest and exotic trees through the susceptible seedling stage. These insecticides have been banne withdrawr o d agriculturn i froe mus mosn i e t countrie d existinsan g alternative insecticides lace th k persistence to provide protection against termites. Controlled release formulations of some of these short-lived insecticides have been show provido t n e protectio r tree fo d ncrop san goo s sa d s thaa t provide e cyclodienth y b d e insecticides withoue th t environmental problems. Current formulation muce sar h more expensive than conventional formulations using the same active ingredien limites i d thei an te higo rus t d h value cropd san forestry.

. INTRODUCTIO1 N

r morFo e tha year0 4 n s unti 1980se th l , many soil insect pestf so crops and forestry were controlled by the use of persistent organochlorine (OC) insecticides such as aldrin, dieldrin and chlordane. The insecticide was used as a soil treatment or seed dressing which create barriea d r aroun plane th d t roots which, becauslone th g f tereo m persistence, meant tha singla t e application at planting protected the crops through to harvest. Howeve lone rth g persistenc insecticidesC O e th f eo , coupled with their tendency to accumulate in animal fat and build up in food cyclodiene th o chaint d ele s insecticides being withdrawr no mann i ye bannecountriesus r fo d . Thi resultes sha searca n i d h for new insecticides capable of controlling these soil dwelling insect pests. A wide variety of soil pests were controlled by the persistent OC insecticides. In some cases the persistence was not necessary. Larva f varioueo s beetle lepidopterd an s a which hatch earln i y controllee seasob n nca shory b d t lived insecticides appliet a d plantin wil d occut an glno r again unti nexe th lt season. However, other pests, including some adult beetles, millipedes and termites, are either present throughout the season or appear seasone th lat n i e . vere b Theyn migraty mobilca yma d ean from untreated to treated areas once the insecticide has lost its toxicity. croe Alsoth pf i , remain e fiel r morth fo n di es than

299 one year, the soil insecticide may need to remain active for most of the life of the crop to protect against an annual recurrence of soil pests such as whitegrubs or termites. Many soil insecticide mose sar t effectiv mixef i e r witn o soie i d t th hla before planting. If they are applied to the soil surface later they may not penetrate sufficiently to control attack on plant roots, either because of low water solubility or a tendency to surfacee bin th soio t dt la . Consequently, with some pests, ther neea r soi s fo di el insecticides which remain activr fo e long periods. Although conventional formulation f somso e currently available insecticides persist in the soil for several weeks or months, there are several situations where they do not have the required persistence to give adequate control. A major soil pest proble Australin i m attacs awa n sugako r cane by white grubs (Scarabaeidae larvae) cans A .e plants remain ni the ground for one plant crop plus several ratoon crops and the whitegrub yeao tw r r larvao s e havon l ea period n insecticid,a e s requirewa d that would provid least ea yearo ttw s protection. An Australian company, Incitée Ltd, develope controllea d d release formulation of chlorpyrifos capable of releasing the insecticide into the soil over a long period [1]. The Incitée CR formulation technology involves the incorporation of insecticides, with inert release rate modifying agents, into thermoplastic matrix granules wit microporha e structure which allow controllee sth d releasinsecticide th f eo e over predetermined intervals ,froe b whic month3 mn years3 hca o st . The insecticide is protected from degradation in the granules releases untii t i l d intenvironmente oth A stead. y releasf eo insecticide ensures that as the insecticide is broken down in the replaces i soi t i l y furtheb d r insecticid maintaio et constanna t level aroun granulese th d . This action provide same sth e long- term protectio persistene th croe s th pa f no C insecticide tO t sbu withou environmentae tth l hazards associated with their persistence [1]. In addition, the controlled release formulation safee sar handlo t r e than conventional formulations. e insecticidTh incorporates i e d withi plastie th n c granulo es duso thern verd ts an yi e little possibilit f skio y n contact. Oral (rat) and dermal (rabbit) LÜ5Q for the carbosulfan formulation are >1000 and >2000 mg/kg body weight respectively [2] and dermal studies on phorate controlled release granules (CRGs) showed a one hundred fold reduction in toxicity compared wite technicath h l active ingredient [1]. The controlled release formulation of chlorpyrifos successfully controlle ranga d whitegruf eo b pest f sugaso r can Australian i e . At application rates of 3-4 kg a.i./ha they provided 2-3 year control resulting in yield increases of between 10 and 30 tonnes per hectare [1, 3]. Trials in other countries (Indonesia, Reunion, Philippines, Tanzani Chinad aan ) have confirmee th d efficacy of chlorpyrifos CRGs against a range of whitegrub pests of sugar cane and also against the root bug Stibaropus spp and wireworms (Elateridae) [3]. From this formulation, a range of controlled release insecticides and were developed, including som replaco et persistene eth insecticideC tO s user fo d termite control. This pape controllef o r e reviewus e dsth release insecticides for the control of termites.

300 Naturae Th l Resources Institute (NRI) became interestee th n i d controlled release insecticides as a method of controlling termites in tropical forestry and agriculture. Termites are a major pes cropsf to , tree buildingd s an tropic e th n d i ssan subtropics, particularl semi-arin i y d areas widA . e rangf eo tropical crops are killed or weakened by termites cutting through the ste grount a m d leve r tunnellinlo througp gu roote th h s [4], Similar damage occurs to exotic trees such as Eucalyptus in Afric firse yearo Asith d tw t an an i agrowthf so . Mature trees can also be attacked by termites such as Coptotermes spp. which tunnel out the trunk [5], Some harvester termites (Hodotermes, some Macrotermes Nasutitermitinaee th d som an f eo ) fee n deao d d and live grasses and can remove, grass cover, particularly in areas which suffer from overgrazin livestocy gb k [6] morr .Fo e than 40 years, organochlorine (cyclodiene) insecticides such as aldrin, dieldrin and chlordane were used to control termites in all these situations [7]. Replacement insecticides have been foun controllinr fo d g termite damage to buildings. Chlorpyrifos has been shown to prevent termite damag buildingo t e 0 year 2 betweer d d sfo san an 0 1 n pyrethroid insecticides (cypermethrin and permethrin) for between 5 and 10 years [8]. However, generally they are not effective for prolonged contro f termitelo agriculturn i s r forestreo s a y they break down more rapidl thesn i y e situations [9].

2. TERMITE CONTRO AGRICULTURN I L E USING GRGs Cereals, suc maizes ha , barley, whea sorghumd tan , legumes, including groundnuts, pigeon peavarioud san s beans, sugar cane, cotton, yams, tobacco and vegetables are attacked by termites which feed on the roots or base of the stem causing the plants to wil died tan . Controlled release formulations were tester fo d the control of Microtermes spp. damaging groundnuts in India and western Sudan [10]. Termites, including Microtermes spp. are a major pest of groundnuts in Africa and parts of Asia. They attack the roots causing wilting and death and damage the pods, either borin d feedinkernele gan th througd n r gpo o so e th h feedinsofe th d shelt po n tissuego e l th (scarifying) f so n I . additio croe th p o lost n s which resultsd ,po e damagth o et encourages fungal attack and infection by Aspergillus flavus leading to the production of aflatoxins, which are carcinogenic and toxic [10, 11, 12]. Controlled release trials carrieIndin i t a ou dcompare d chlorpyrifos CRG (4 kg a.i/ha) with chlorpyrifos granules (4kg a.i./ha) and seed dressing 5g a.i./kg), isofenphos granules 5 kg a.i./ha), and aldrin dust (1 kg a.i./ha) and seed dressing (5 g a.i/kg). In Sudan chlorpyrifos, carbosulfan and phorate CRGs (4kgs ai/ha) were compared with aldrin dust (1kg ai/ha) [10]. e trialTh Indin i s a were complicate presence th y b dothef eo r soil dwelling pests including white grubs (Scarabaeidae), false wireworms (Tenebrionidae groundnue )th antd san t root borer (Spfenoptera perroteti Guerin-Meneville) which also attackee th d resultroote th podd san o s s reflec e effece th th t f to insecticides on all these insects. In Sudan, pod damage was due solel Microtermeso t y spp n IndiaI . mose th ,t effective

301 treatments were aldrin dust, chlorpyrifos CRG, isofenphos granule d chlorpyrifosan s granulesformulatioG CR e e th Th . s nwa most effective formulatio chlorpyrifof no s with significant increase yieldn i s s equivalen aldrie thoso th t n ni e treatment and significant reductiond rooan t d damagpo n i s e (Tabl. eI) Yield assessments were not made in the Sudan trial but there was significantly less root damage in the aldrin, chlorpyrifos CRG and carbosulfa G treatmentnCR scontrole th tha n i n s (p<0.01). Significantly less pods were damaged in the aldrin, chlorpyrifos CRG, carbosulfan CRG (p<0.01) and phorate (p<0.05) than in the control pland san t mortalit s significantlywa yaldrie th les n ni s and chlorpyrifos CRG treatments (Table II) [10].

Table I The effect of insecticides on yield and pod damage to groundnuts at Hyderabad, India (after Logan et al. ,1992).

Treatment yield/plo damagd po et pod% s kernels a O 1 4 aldrin dust 290a 3.9b Aldrin seed1 dressin25 g 166 11 .0 chlorpyrifos6 23 granule 170 9.7 chlorpyrifos CRG 387a 285a 3.1b chlorpyrifos seed dressing 273 185 10.8 isofenpho7 s30 granule 222 5.3 7 19 control 139 20.9 CV%* 28.28.34 34.1 SED 83.59.41 0.7

•f Square root (X+0.5) transformation a significantly higher (P<0.05) than the control b significantly lower (P<0.05) thacontroe th n l

effece Th Tablcontrollef to . eII d release insecticiden so termite attack to groundnuts in western Sudan (after Logan et al. ,1992).

Roots Pods Pods Plants attack bored scarified killed

Control 7.34 42.19 7 7. 36.96 Aldrin dust 0.04a 0.48a 1.56a 0 . 4a Chlorpyrifos CRG 0.96a 13.48a 10.54a 0 . 1 6a Carbosulfan CRG 2.34a 13.64a 22.76b 3.68 Phorate CRG 6.54 24.64b 25. 2b 3.02 cv l 27.6 24.2 29.3 41 .6 SE 0.3 0.06 0.74 0.41 Square root (X+0.5) transformation a significantly lower (P<0.01) than the control b significantly lower (P<0.05) than the control

302 3. TERMITE CONTRO FORESTRN LI Y USING CRGs Termites are a major problem in tropical forestry. In Africa and India, exotic tree species, particularly Eucalyptus, are heavily attacked in the first 2-3 years and termite attack can result in total los younf so g trees [5]. Termites suc Microtermess ha spp., small Odontotermes strotermesd An sppd .an spp. tunnep lu inside the root to the stem of the tree. Larger Odontotermes spp., Pseudacanthotermes sppd Macrotermes.an spp. remove th e bar d cambiuroote kan th r lowe so f o m treee r botn th steI .f h o m cases the trees wilt and die. Attack can occur either in the nurser r afteo y r plantin t intfielde ou ge oth latte Th . mor s i r e commo , 13][5 n . Controlled release granule formulation fouf so r insecticides (chlorpyrifos, carbosulfan, carbofuran and phorate) have been teste r termitfo d e control. Extensive trial forestry sb y research institutes in South Africa and Zimbabwe compared the insecticides and assessed different application methods, concentrations, granule size d releassan e rates [14, 15]. Additional trials were carried out against termites in Malawi, Kenya, Tanzania Ethiopi d Braziaan l, 19] 18 [16 , 17 , Phorate, carbosulfan and carbofuran CRGs were equally effective in controlling termite damag werd ean e frequentl s effectiva y s ea aldrin or chlordane. However, phorate was not as effective under very dry conditions [13, 14]. A similar effect was noted in the groundnut trial western i s n Sudan [10]chlorpyrifoe Th . G sCR performed less well than carbosulfa G (FignCR [13) 1 . , 14]. Chlorpyrifos is a contact insecticide with little or no systemic r fumigano t action whereas carbofura phoratd nan contacte ear , systemic and fumigant. Carbosulfan, although not systemic or fumigant, is broken down rapidly in soil to carbofuran which is. Atkinson [14] considered that the termites might be able to pass between the chlorpyrifos granules whereas the others, which are systemic, would be taken up by the tree and protect it but Mitchell [13] discounte e systemith d c effec Macrotermess ta d fe bare treef th o k n o s whose roots wer carbosulfan i e n treated soil seemt I . s more likely thafumigane th t t effec phoratf to e and carbofuran extends the insecticidal action well beyond the granules creatin mora g e effective barrier, wherea effece sth f to chlorpyrifos is much more local, particularly in dry soils and may allow termites to pass between the granules. In addition, both chlorpyriphos and carbofuran were phytotoxic at the concentrations needed to control termites [13, 14]. Atkinson [14] compare R formulatioC e th d carbosulfaf no n with three OC insecticides (chlordane ec, aldrin dust, gamma BHC ec and dust) and 8 conventional formulations of non OC insecticides (aldicarb granules, alphamethri microencapsulate, ec n d diazinon and fenitrothion, flufenoxuron ec, Isazofos ec, phorate granules and alphamethrin suspension concentrate). Chlordane and aldrin were consistently effective in controlling termite damage and, in some trials, were significantly (P>0.05) better thaothey an n r treatment controllee Th . d release formulations performed better thaconventionae th n l formulation l excepC O al e f sth to insecticides and the highest concentration of the alphamethrin suspension concentrate. Consequently, further trials evaluated carbosulfan CRGs (10% a.i.) as the most promising of the insecticides.

303 control Chlorpyrifos Carbosulfan Aldrin 14%CR 10% CR 2.5 dust

insecticido N e {conlfot)

0,02 a.l2g .

0.05L a. 9g

0.161 ga.i,

Figure 1. Contro f termito l e damag Eucalyptuseo t Zimbabwn I applicatioe th y eb n of clilorpyrifos or carbostilfan in Ihe nursery, 142 weeks after planting (after MltclteM, 1989),

Carbosulfan CRGs were effectiv protectinn ei treee th g s whether they were mixed with the potting soil and used in the nursery or applied to the planting hole when the trees were transplanted to the field. Mitchell (1989) found that dose betweef d so an 2 n0. 0.5 g a.i. Carbosulfan CRG (2 and 5 g product) per tree applied nursere ith n y provided protection against termites equivaleno t that of aldrin without any phytotoxicity (Fig 2.). Higher application rates (1.0-1.5 g a.i. (10-15 g product) per tree) were needed if trees were treated at planting out because of the greater volume of soil to be treated (Fig. 3) [14, 15, 16]. The size of granule had no effect on the efficacy of the treatment when the application rate of Carbosulfan active ingredient was the same.smallee Th . r granules provide greatea d r numbef ro possible contact points betwee termitee granulee nth th d t san sbu there were no differences in efficacy possibly due to the fumigant principas effecit f to l breakdown product, carbofuran [14, 15]. In practice the choice of treatment method will depend on the local circumstance botd san h methods have advantaged san disadvantages. For instance, if there is no termite damage in the nursery e insecticid,th e release pottine th n i dg sois li wasted and, in some African countries, seedlings are kept in the nurser morr fo ye than three months. However, field application is more expensiv requiret i s ea s greater quantitieR C e th f so granules and labour to mix them with the soil in the field.

304 control Aldrin Marshal Marshal 2.5% Dust suSCon suSCon

Trial 1: 58 weeks after planting

Trial 2 : 94 weeks after planting

Figure 2. Control of termite damage to Eucalyptus in Zimbabwe by the application of Marshal/suSCon in the nursery (after Mitchell, 1989).

IUU 1 ml/tree 1g/tre0 e g/tre5 1 e -

80 5 g/tree

\ J 60 '> 3 t ! 40 \ s 20 I \ o ——————— control Chlordane Marshal Marshal Marshal 60EC suSCon suSCon suSCon

Trial 1 : 57 weeks after planting

Trial 2 : 52 weeks after planting

Trial 3 : 51 weeks after planting

Figure 3. Control of termite damage to Eucalyptus in South Africa by application of Marshal/suSCon at field planting (after Atkinson, 1989).

305 There is less control of the application and careless mixing can resul largn i t e gaps betwee granulee th n s whic n allohca e wth termites access to the tree roots. Also, there can be considerable variation in planting hole size and hence the volume of soil mixed with the insecticide. If the hole is greater than 1-2 litre volumen si granulee th , s becom o diluteto e d with soil effectivete ob . Wher treee nurserth e e raise th sar d n i dan y then give r sol no smallholdero t d r farsfo m woodlots, nursery application ensures proper application rate confined san e sth handling of the granules to trained personnel. 4. DISCUSSION Trials of controlled release granules have shown that termites cacontrollee nb d successfull botn i y h agricultur d forestryean . There have bee coso n t benefit analyse r termitfo s e contron li crops. In small scale farms in Africa and Asia where termites majoa e rar proble maizen i m , groundnut othed san r crops i t si unlikely that control release insecticides will ever be cost- effective. However largen i , r farm witd an sh high value crops such as sugar cane, tobacco and cotton their use may be cost effective in areas with consistent high levels of termite attack. In commercial forestry, studies in Malawi and Zimbabwe on cost effectiveness have shown positive returns with a cost benefit of 1:7 in the latter [20]. In addition, there are the socioeconomic benefits which are more difficult to cost. There is an urgent neereforestatior fo d provido nt e firewood, construction timber and for reducing soil erosion and desertification. In many area Africf so impossibls i Asid aan t i a groo et w suitable trees without protection against termites. In these cases, the social and environmental benefits will justify the use of carbosulfan CRGs. There is a need for an appraisal of the negative environmental effect controllef so d releas f insecticideseo concentratioA . n of insecticide in the soil capable of controlling termites will also affect other soil biota including those which are harmless or beneficial and this has not been assessed. However, it seems likely thaapplicatioe th t n R formulatiomethoC d an d n minimise the effect n forestryI . , only small quantitie a.ig 1 ( r so . less) of the insecticide are used and restricted to a small volume of soil around the roots. The amount of treated soil is insignificant compared with tha untreatef to d soil betweee nth o shouls tree d dsan have little effec othen to r organismsr Fo . example, for an average planting of 1300 trees/ha only about 1% e lanoth fd wil e treatelb plantinf i d g hole applicatio useds i n ; about 0.02% with nursery application. Withi treatee th n d soil, the majorit e insecticidth f o y withis ei granule th n d onlean a y y time an e soi th t . la littl n i releaseds Afte i s i t ei r , carbosulfan has a half life of 2-3 days and is degraded to carbofuran. Carbofuran persist longer sfo r (half life 30-60 days) [21]. There appear little b o st e movemen f eitheo t r insecticide within the soil as Mitchell [13] found that roots which grew outside the treated soil plug could be attacked by termites. Thi supportes si studiey db Francn si e which showed that although concentrations of up to Smg/kg of carbofuran were found in trees protected with carbosulfan CRGs, levels in surrounding vegetation were either very low or not detectable [20].

306 A greater percentage of the soil will be treated to protect crops becausgreatee th f eo r densit plantingf yo . Howevere ,th controlled release formulation likele sar provo t y e less environmentally damaging than the soil application of any conventional formulations. With conventional formulations, it is necessar applo t y y very high concentration soie th t la o st plantin ensuro t g e that sufficient remain proteco st croe tth p throughout the season and these high concentrations are likely to be more damagin soie th l o biott g a thae levelth n s releasey db thR formulationsC e . The forestry trials have resulted in the registration in a number of countrie commerciaa f so l product Marshal/suSCon' containing 10% carbosulfan, which releases the,insecticide oveyearso rtw . To date, only two methods of application are recommended for the contro termitef lo pests sa Eucalyptus;f so nursery application of 10g/L of potting soil or field application of 10g/tree mixed with 1-2 1 of soil in the planting hole. Research into the use of chlorpyrifos for the control of white grubs in sugar cane has shown that application rate d formulatiosan nvariee b nee o o t dt d control different whitegrub speciedifferenn i d san t soil types [3]. The same may apply to the use of controlled release formulations for the control of different species of termite and in different soil types. In Afric d Asiaaan , attac Eucalyptusn ko mainls i y e causeth y db fungus growing termites (family Termitidae; sub-family Macrotermitinae) Soutn I . h Americ damage th a causes i e y db termites belongin Rhinotermitidaee th o t g . Even withie th n Macrotermitinae, different genera and species attack the trees in different ways and have differences in behaviour patterns. In some areas, most termite attack occurs in the first six months and could probabl controllee b ywitG CR shortea h a y db r release time othersn i ; threo continuet t ,i p eu yearr fo swould san d require granules wit lona h g release time. Combined wite th h rang f soieo l type d climat an probabls i t i e e thae th t formulation, application rate or application procedure will need to be modified to deal with different conditions. There are already some indication f thiss o Sout n I . h Africae ,th carbosulfa applieG soie nCR th l o t surfacd r controefo . M f lo natalensis s effectivwa arean i e s whert bee groune no neth d ha d ploughed but elsewhere it had to be mixed thoroughly with the soil froplantine th m g hole [14]. Applicatio f carbosulfano o nt the surface 5 cm was not effective in protecting trees against Microtermes spp. or Microcerotermes spp. in Tanzania but controlled thes othed ean r termites when mixed thoroughly with the soil [19]. However Braziln ,i , application e eitheth o t r base of the planting hole or lightly incorporated into the surface soil was successful in controlling 3 species of Rhinotermitidae [15] e choicTh . f applicatioeo n methos ha d implications for labour costs. It requires less labour to apply the granules to the surface or base of the hole than to incorporat soile th .n i t i e

Marsha trademara s CorporatioC li FM f ko suSCod a nan s i n registered trademar f Incitéo k e Ltd.

307 Controlled release formulation non-persistenf so t insecticiden si plastic granules have been shown to control termites attacking groundnut Eucalyptusd san plantations mann I .y cases, control goos a couls da s providee wa db persistene th y db t organochlorine insecticide without sbu hazarde th t environmene humano st th r so t associated with these insecticides.

ACKNOWLEDGEMENTS The author would like to thank Mr. Peter May and Dr. Mario Valvasor providinr f Incitéio fo d Lt e g unpublished information and for commenting on this paper. He should also like to thank Dr. Tom Wood of NRI for his criticism of the paper.

REFERENCES [1] MAY, P.D., BOEHM, N. Controlled release soil insecticides for control of sugar cane pests. Sugar-y-Azucar 81 (1986) 127-129, 132. [2] Marshal suSCon Technical Manual. Incitée Ltd. Morningside, Australia. ] DEGROOT[3 VALVASORI,M., ,R. f suSCoo e nUs , Blus ea alternative to persistent organochlorine insecticides for soil pests of sugar cane. Int. Sugar J. 91 (1989) 210-212. [4] SANDS, W.A. The role of termites in tropical agriculture. Outloo Agricn ko .9 (1977 ) 13(5-143.

[5] COWIE, R.H., LOGAN, J.W.M., WOOD, T.G Termite (Isoptera) damag controd an e tropican i l l forestry with special referenc Africo et d Indo-Malaysiaaan review:a . Bull. ent. Res. 79 (1989) 173-184. [6] WOOD, T.G., PEARCE, M.J. Termites in Africa: The environmental impact of control measures and damage to crops, trees, rangelan rurad an d l buildings. Sociobiol9 1 . 1 (1991) 221-234. ] HARRIS[7 , W.V. "Termites their recognitio d controlnan d 2n " edition Longman, London. ] LOGAN[8 , J.W.M.; BUCKLEY, D.S. Subterranean termite contron i l buildings. Pesticide Outlook, 2 1 (1991) 33-37.

] RACKE[9 , K.D., LUBINSKI, R.N., FONTAINE, D.D al.,t e . Comparative fat chlorpyrifof eo s insecticid urban i ed nan agricultural environments. "Pesticide urbae th nn si environment; fate and significance." ACS Symposium Series 522. Racke, K.D Leslied .an , A.R. (Eds) American Chemical Society, Washington C (1993,D ) 70-85.

[10] LOGAN, J.W.M., RAJAGOPAL, D. WIGHTMAN, J.A., PEARCE, M.J. Contro termitef lo othed san r soil pest groundnutf so s with special reference to controlled release formulations of non- persistent insecticides in India and Sudan. Bull. ent. Res. 82 (1992) 57-66.

308 [11] JOHNSON, R.A., GUMMEL, M.H. Termite damag d croean p loss studies in Nigeria - the incidence of termite scarified groundnut pods and resulting kernel contamination in field and market samples. Trop. Pest Manage. 27 (1981) 343-350. [12] JOHNSON, R.A., LAMB, R.W., WOOD. T.G. Termite damage and crop loss studies in Nigeria - a survey of damage to groundnuts. Trop. Pest Manage. 27 (1981) 325-342. [13] MITCHELL, M.R. Compariso non-persistenf no t insecticiden i s control release granules with a persistent organochlorine insecticide for the control of termites in young Eucalyptus plantations in Zimbabwe. Comm. For. Rev. 68 4 (1989) 281- 294. [14] ATKINSON, P.R. Controlled release insecticide granules compared with other soil insecticides for use against the termite, Macrotermes natalensis Havilande th n i , establishment of Eucalyptus plantations. Crop Protect. 8 (1989) 387-396. [15] CANTY, C.P. Controlled release technology protects forest trees from termite attack. Proc Intd .3r . Conf. Plant Prot. in the Tropics. (001, P.A.C., LIM, G.S. AND TENG, P.S. Eds.) Kuala Lumpur, Malaysia, Malaysian Plant Protection Society p 182-18,p 6

[16] CHILEMA, C.Z. Termite contro younn li g Eucalyptus plantation Malawn si i using controlled release insecticides. Comm. For. Rev. 70 4 (1991) 237-247. [17] ATUAHENE CHILEMA, ,S. , C.Z yeae .On f termit ro e contron li Malawi using Marshall Suscon. Proc Workshot .1s Termitn po e Researc Controld han , August 17-19 1992 Jacaranda Hotel, Nairobi Kenya. Danish Technological Institute, Taastrup, Denmark (1992) 87. [18] LOGAN, J.W.M. Report on a visit to southern Africa to assess the progres f trialso s establishe e INCITEy db th r Cfo contro f termitelo forestryn i s , 11-27 November 1987 Unpublished report, Overseas Development Natural Resources Institute, London (1987p p 7 )2 [19] LOGAN, J.W.M. Report on a visit to Kenya, Tanzania, Ethiopia and Malaw asseso t i progrese sth f trialso s establishey db INCITEC for the control of termites as forestry pests, 7-28 July 1988. Unpublished report, Overseas Development Natural Resources Institute, Chatham (1988) 44 pp [20] VALVASORI, M. Incitée Ltd., Morningside Australia, personal commueition. [21] HARTLEY KIDDD AN . "Th,H . , D e agrochemicals handbook" second edition Royae .Th l Societ Chemistryf o y . Nottingham (1987)

Next page(s) left blank 309 CONTRO INSECF LO T PESTS USING SLOW RELEASE PHEROMONE CONTAINING DEVICES

P.S. RANKIN AgriSense-BCS Ltd, Pontypridd, Mid-Glamorgan, United Kingdom

Abstract

A numbe f sloo r w release device e bein ar sr hav o g e been develope d commercialisean d e detectioth r fo f do insecn t pestse for monitorinf th o mn i , g r theiluresfo rd an ,control , by lure and kill or mating disruption techniques. The devices are based upon matrix-type polymer formulations with pheromone or attractant distributed therein. Aerylate based thermoset, and proprietary thermoplastic compositions have been used as e polymeth r matrix d variouan , s parameters, including polymer composition, pheromone loading d devican , e shap d sizan ee have been used to optimise product performance and user friendliness. The development of such materials involve, as appropriate, laboratory and/or field weathering testd an s field bioevaluations. Release profile f o devices s were determine s Chromatographiga y b d e analysi f o pheromons e residin e devicesth a n functioi gs a , f timeo n ; release rates were then derived, a functioals s a o f time o nd compare an , d with bioefficacy result f fielo s d testse loweTh .r rate limit, consistent with mating disruption, can be determined, and wil e appropriatb l d dependenan , to e t upoe fielth n d test condition g e temperatures , wind conditions, point source density, insect pressure e presenc,th bénéficiaisf o e e th d an , influenc f otheo e r attractants suc s plana h t volatiles. Such an approach has been taken in the development of products for Pectinophora gossypiella (Pink Bollworm), Chilo suppressalis (Rice Stem Borer) Lymantria dispar (Gypsy Moth), Ceratitis capitata (Mediterranean Fruit Fly), Rhyacionia buoliana (European Pine Shoot Moth), and Keiferia lycopersicella (Tomato Pinworm) s essentiai t I . l thae cos th f tpheromono t e be minimised in order to maximise the possibility of successful product development. To this end, the metathesis rout s beeeha n found usefu somn i l er Pincasefo k g e sBollwor m pheromone (50/50) Z,E/Z,Z-7,11-hexadecadienyl acetate.

1. INTRODUCTION The term 'pheromone 1 has been proposed by Karlson and Luscher, 1959, [1] for the naturally occurring chemical(s) use e purposey insectb dth r f communicationfo so s . Thee ar y species specific and mediate a wide variety of behaviour such as aggregation, alarm, trail following, and dispersive, but the most widely documented type is the sex pheromone, used to increas probabilite eth matingf o y . Historically, insect responses have been observer fo d ovehundree on r d e firsyearsth tt bu ,successfu l investigation of pheromone o assigt s n structur d determinan e e responss wa e

311 not until 1940, by the USDA, for Bombyx mori . The next several decades saw great strides-forward in analytical method development, essentia r structurfo l e elucidation; alsn i o polymer development w provinno , g essentiae efficienth r fo l t use of these chemicals. Excellent reviews of pheromone structure are given by Jones [2] and Morgan and Mandava [3]. Of particular e interespheromoneth e ar tf lepidopterao s . These are alkenyl compounds (10 to 20 carbon atoms) with one, sometimes two, and occasionally three centres of unsaturation, and terminal functionality (alcohol, aldehyde or acetate) . Pheromon e use b n conjunctioi luredn ca s n wita h suitabl e edetectio th trap r d fo monitorin,an n f inseco g t pests. Insect controe achieveb n a mass-trappingca l vi d r o , attracticide (usin a contacg t insecticide) techniques, again using pheromone lures. Alternatively, mating disruption ca n be used for control; here a device releases pheromone above a critical rate so as to confuse the insect, and significantly diminis chance hth matingf eo . e controlleTh d releas f o suce h biologically active

agents has been amply reviewed by Baker [4] . Membrane or 'reservoir 1 systems exhibit constant release (zero ordery b ) diffusion. Solid matri r 'monolithico x ' system e morar se complex with an ever decaying release rate. This paper describes the development of matrix devices containing pheromone r attractanto s f selecteo s d specief o s dipter d lepidopteraan a r purposefo , f o monitorins d an g control.

2 . SAMPLE PREPARATION

Two types of polymer matrix have been evaluated: (a) Methacry late-based cross-linked polymers containing pheromone have been prepare y b fred e radical polymerisation of selected monomers. Chain extension is achieved using monomer type th e f so CH2= C (CH3) COOR, with cross-linking derived from monomer type th e f so

CH2=C(CH3)COO or CH2=C (CH3 ) COO

CH2=C(CH3)OXr CH2=C(CH3)COO— R"

CH2=C(CH3)COO Some common examples are hexyl, lauryl, and phenoxyethyl methacrylate; ethylene glycol, tetraethylene glycol dimethacrylat d trimethylolpropanean e trimethacrylate. The mole ratio of monomers used, and their weight ratio to attractant and any other incorporated release rate- modifying component is critical in determining the final release profile. Screening tests were initially performed, whereby chosen monomer e mixesar d with attractan d frean te radical initiator suc t-butys ha l peroctoate, deoxygenatedd an ,

312 heated to effect polymerisation. Release rate-modifying oils and solids may also be incorporated into such formulations. Once polymerisatio completes nwa , product s inspectewa r it'fo ds suitability with respeco t t end application; critical parameters observed included frangibility, hardness, eas cuttingf eo , solubilitf o y attractan e polymeth n i rt matrix d crackinan , n o g aging. Product was then be prepared in the form of solid rod r withi,o cupna e e ,affixeb th whic n o t ca dh substrate via an attachable clip-type faceplate, or as a sprayable a suspensiovi , n polmerisation. These techniques are more fully described by Abrutyn [5,6] and Weiss, Jacobson and Rankin [7]. In this way, devices were conveniently prepared containin0 7 o t p gu wt% attractant.

(b) Thermoplastic devices were prepared by doctoring proprietery formulations ont continuousloa y moving belt passing thoug oven ha effeco nt t fusion n I addition. , other compositions were extruded into the form of bands, tubes, and string, as appropriate to end application. Up attractan% wt 5 t1 o s incorporatedtwa .

3. RELEASE PROFILING Samples were aged either within a field biocontrol study, a field weathering test, or in the laboratory using an environmental chambe n i whicr h light, temperatured an , humidit e adjusteb n cycled ca y an d o mimit d c field conditions. Appropriate numbers of samples were removed at appropriate times for analysis. Each sample was extracted using hexane/aceton e solutioth d an ne then s analysega y b d Chromatographie analysis y differenceB . e amounth , t released s calculatewa a functio s a d f time o nd releas an , e rates determine y drawinb d g tangente best-fith o t s t plote Th . release profil s thewa en compared y wittraan hp catcd an h crop damage data generated from bioevaluation e fieldth n i .s A 'lower rat ee attractanth limit f o 1 t e frodeviceth m , consistent with producing the desired bioeffect, could then be assesseparticulae th r fo d r condition teste th .f so For the insect Pectinophora gossypiella (Pink Bollworm), stake devices have been prepared containing entrapped pheromone 'Gossyplure' (a 50/50 mixture of Z,E/Z,Z-7,11- hexadecadienyl acetate), and release profiles measured in the environmental chamber [Figure 1] . A field study was also undertake o observt n matiny an e g disruption activity a travi ,p catch depression data [Figurcrod an p ] damage2 e assessment.

4. PHEROMONE SYNTHESIS There is no general synthetic route for the preparation of pheromones of lepidoptera. Indeed, routes have to be sought which involve low-cost raw materials, since it is the pheromone which is the major cost-contributor to the final cost of application. To this end the metathesis route has

313 Mt d(Mt)/dt (mg/device) (mg/day/device)

100 r r 2.5

80 - •- 2.0

d(Mt)/dt 60 - 5 1. ••

- • 0 4 0 1. ••

20 •- 5 0. ••

TIME 0 7 (days 0 )6 0 5 0 4 0 3 0 2 0 1 0

FIGURE 1 RELEASE OF Z,E/Z,Z-7,11-HEXADECADIENYL ACETATE FROM STAKE DEVICES; ENVIRONMENTAL CHAMBER

TRAP CATCH (moths/day)

20 T

15 -

10

5 -•

G AU L JU N JU Y MA R AP

FIGURE 2 TRAP CATCHES OF PBW STAKE 75mg

been abl o offet e r significant cost e synthesisavingth r fo s s of some insect pheromones [8,9] r exampleFo . e synthesith , s of Gossyplur s beeha e n reporte o involvt d s man a es eigh a y t a metathesistagesvi t bu ,s reduce i sa two-pot o t d , three- stage synthesis [Figure 3].

314 + CH3(CH2)3CH=CH2

Mo03/SiO2

) (Z (Z/E) CH3 (CH2) 3CH=CH(CH2) 2CH=CH(CH2) 2CH=CH2

1 CH3CH2CH2MgCl 2 EtO/Cu2Br2 AcCl

) (Z (Z/E) CH3 (CH2) 3CH=CH(CH2)2CH=CH(CH2)6-OCOCH3

FIGURE 3 SYNTHESIS OF Z,E/Z,Z-7,11-HEXADECADIENYL ACETATEE METHESITH A VI ,S ROUTE

The process, based on catalysed olefinic rearrangement, produces mixed isomers, so that the pheromone must either involve those e ratiisomer th o prepareds on e i sth r o , unwanted component merely a diluenact s a s t wit o bioeffecn h t on the insect in question.

5. DISCUSSION

A number of matrix-type devices have been prepared and found to be economically viable and bioeffective. For example, a stake produc r Pinfo tk Bollworm, containing 75mf o g pheromone was found to effect mating disruption for 60 days when applied at a rate of 30 gm of the active per hectare. The release rats measure wa e 1.5b 0o t dmg/day/devic e (da) 1 y which droppe o 0.4t d 1 mg/day/device (day e 60)Figurse , . 1 e

Trap catch data (Figure 2) indicates a biolongevity of 60 days with a 'lower response limit 1 of about 0.40 mg/day under the conditions of test. A further test, utilising samples returned fro a fielm d test, showe a similad r profil o that e t environmentae deriveth r fo d l chamber test. In developin e devicesth g , various parameters were carefully investigated so as to maintain the most efficient e th attractant f o e , us consistent with application requirements. Thus monomer ratio, device shape, and addition of rate modifying agents were evaluate o obtait d e desireth n d longevit practicaa r fo y l device, with minimum cos minimue i t m amoun attractantf to . In this way, devices have been, and are being developed for the monitoring and control of a wide variety of insect pests. Figure 4 summarises some of the more important devices develope dateo t d .

315 INSECT DEVICE LONGEVITY APPLICATION RATE (days (gin ai minimum) /ha/appn)

Ceratitis capitata PLUG (L) 56 na (MEDFLY)

Chilo suppressalis TUBE (MD)120 40 (RICE STEM BORER)

Dacus oleae SPRAY (LK) 7 0.4 (OLIVE FRUIT FLY)

Keiferia ly copers i eel la CLIP (MD)60 - 90 88 (TOMATO PINWORM)

Lobes botranaia CLIP (MD) 85 160 (GRAPE BERRY MOTH)

Lymantriapar dis POWDER (MD) 14 15 - 37* (GYPSY MOTH)

Musca domestica POWDER (L) 14 na (HOUSEFLY)

Pectinophora gossypiella POWDER (MD) 14 13 (PINK BOLLWORM) STAKE (MD) 60 85 BAND (MD)150 80

Rhyacionia buoliana POWDER (MD) 14 15* (EUROPEAN PINE SHOOT MOTH)

L monitoring lure LK lure & kill MD mating disruption * two application seasor spe n

FIGURE 4 PERFORMANCE OF SLOW-RELEASE DEVICES FOR THE CONTRO F SOMO L E INSECT PESTS

It mus e noteb t d thae releasth t e profiles generated, and their correspondence with bioeffectiveness in the field can only be made for the conditions employed. Factors such as point source density (for a given weight/area of applied attractant), humidity, temperature and wind speed can have drastic effect on release rate and bioeffectiveness; also, insect pressure e presencth , f bénéficiaio e d attractivan s e plant volatiles will all have a bearing on performance and the 'lower response limit l interpretationAl 1. s mus e carefullb t y made bearing these factor mindn si .

6. CONCLUSIONS n concludca e W e e thachoseth t n methacrylid an c thermoplastic polymer systems incorporating insect pheromones

316 or attractants can be formulated so as to be economical, bioeffective materials for the monitoring and control of a number of insect pests, particularly diptera and lepidoptera. Control can be season long, but external factors must be carefully assessed. The metathesis route has been found to b ee synthesi viablth r fo esomf so e pheromones.

REFERENCES

[1] CARLSONLUSCHERd an , , 'PheromonesP. ,,M. Terw Ne m a for a Class of Biologicall1 y Active Substances, Nature (London), 183, 55 (1959). [2] HUTSON, D.H. ROBERTSd an , , T.R. (eds), Insecticides, Pub WileJ . Son& y s Ltd., (1985); JONES, O.T., Chemical Mediatio Insecf no t Behaviour, Chapte. 8 r [3] MORGAN, D.E. MANDAVAd an , , N.B. (eds), Handboof ko Natural Pesticides, VolPheromones, PartB , & IV . sA , PubPressC CR . , (1988). [4] BAKER, R., (ed), Controlled Release of Biologically Active Agents, Pub. Wiley-Interscience (1987). [5] ABRUTYN, E.S., SCARFO, L, and CHROMECEK, R, Lattice- Entrapped Composition, U.S. Pat. 4,855,127 (1989). [6] ABRUTYN, E.S., A Solid Entrapped Emollient-Moisturiser Composition and the Use Thereof, Eur. Pat. 0 061 701 (1986). [7] WEISS, L.L., JACOBSON, L.R. RANKINd an , , F.S., Lightweight Easily Attachable Dispensing Device having Interchangeable Container Holdinr sfo Chemicaa g e b o lt Dispensed, U.S. Pat. 5,115,976 (1992). [8] BANASIAK, D.S., Insect Pheromones from Olefin Metathesis, J. of Mol. Catalysis, 28., 107-115 (1985) . [9] BANASIAK, D.S., Olefin Disproportionation using Neutral Carbene Comple Tungstef xo n Molybdenum, Titanium, Rhenium or Chromium with Metal Halide or Oxyhalide, U.S. Pat. 4,269,780 (1981)d ,an BANASIAK, D.S., Olefin Disproportionation Catalyst producing Non-polymeric Olefin(s) Comprising Metal (Oxy)halide and Neutral Carbene Complex, U.S. Pat. 4,331,559 (1982).

Next page(s7 )31 left blank USE OF IONIZING RADIATION IN THE MANUFACTUR MATRIF EO X MATERIALS FOR DRUG CONTROLLED-RELEASE SYSTEMS SELECTED AN , D APPLICATIONS (Abstract)

E.E. SMOLKO Comisiön Naciona Energfe d l a Atömica, Gerenci Aplicacionese ad , Centre Atomico Ezeiza, Buenos Aires, Argentina

Radiation polymerization procedures have been used for the obtention of adequate materials for controlled-release systems, Co-60 radiation was used for the polymerization and cross/inking of HEM A monomers alone, HE M A in combination with other monomers and natural rubber latex. Irradiations were carriedsolidin out water solutions with inclusion drugsof medicaments.and The work included the immobilization of antibiotics (ampiciilin, gentamicin); antineoplasic drugs (ciclofosf amide); hormones (medroxiprogesterone); pheromones (trimedlure); pesticides (fenbendazole), and beta-ad rensrgfcs (propranolol). Kinetics swellingof theof polymeric matrix aloneand studies of the rate of release of ampiciilin and propranolol givenare compared with traditional tablets of the same préparâtes. The swelling of PHEMA was anaiized and was fitted to a second order kinetics. Accumulated release ratesfor ampiciilin and propranoloi was demonstrated to be proportional squaredthe to root time.of in vitro and in vivo systems were selected for the kinetic studies. In vitro release studies comprise, drug content in the eiution medium measured s pect r o photometrically and drug re/ease kinetics by the agar difussion method using Microcoeeus luteos bacteria/ preparations. The biological activity antibioticsof released from the matrix from orally administred preparations were measured in dog plasma.

Next page(s9 )31 left blank LIST OF PARTICIPANTS

Albores-Velasco, M.E. Faculta Qufmice dd a Universidad Nacional Autönoma Circuito Interior, Ciudad Universitaria Coyoacän 04510, Mexico City MEXICO

Allan, G.G. Department of Chemical Engineering College of Forest Resources Universit f Washingtoyo n Seattle A W ,98119 5 UNITED STATE AMERICF SO A

Alsop, N.J. Hoechst/Roussel Uclaf Xüarlton', Meadow Lane Houghton, Huntington, Cambs. PE1P 72B UNITED KINGDOM

Bajet, C. National Crop Protection Center University of the Philippines at Los Bafios College, Laguna 4031, PHILIPPINES

Bauer. B , Vector Control Unit C.R.T.A. B.P. 454 Bobo-Dioulass , BURKINo01 A FASO

Bennettovä, B.L. Institute of Entomology Czech Academy of Sciences Branisovska 31 Ceské Budejovice 370 05, CZECH REPUBLIC

Benyakoff. J , Makhteshim Chemicals Industrial Zone 0 6 x Bo P.O. Beer Sheva 84100, ISRAEL

Biernacka, E. Warsaw Agricultural University Nowoursynowska Street 166 PL-020776 Warsaw, POLAND

Bosshardt, H.P. Eidgen. Forschungsanstalr fü t Obst-, Wein-und Gartenbau CH-8820 Wädensvil, SWITZERLAND

Carr, M.E. National Cente Agriculturafor r l Utilization Research USDA-ARS . Universit181N 5 . ySt Peoria, IL 61604 UNITED STATES OF AMERICA

321 Carroll, J.P. Department of Chemical Engeering and College of Forest Resources University of Washington Seattle 9819A W , 5 UNITED STATES OF AMERICA

Chadenga. V , Tsetse and Trypanosomiasis Control Branch P.O. Box 8283 Causeway Harare, ZIMBABWE

Dombovari, J. Research Institut r Irrigatioefo n Divisio f Agrochemistrno Radiology& y Szabadsag U. 2 H-5540 Szarvas, HUNGARY

Flynn. A , Daratecd Lt y hPt Victorian Institut r Drylanefo d Agriculture Private Bag 260 Horsham, Victoria 3401 AUSTRALIA

Gerstl. Z , Institute of Soils & Water, ARO Volcane Th i Center P.O. Box 6 Bet Dagan 50-250, ISRAEL

Gish, T.J. Hydrology Laboratory Agricultural Research Service U.S. Department of Agriculture Beltsville 2070D M , 5 UNITED STATE AMERICF SO A

Gurfinkel, E. Makhtehim Chemicals Industrial Zone P.O. Box 60 Beer Sheva 84100, ISRAEL

Hall, D.R. Natural Resources Institute Central Avenue Chatham Maritime Chatham, Kent ME4 4TB UNITED KINGDOM

Hickman, M.V. USDA-ARS Botany Plant Pathology Department Purdue University West Lafayette 47907-115N I , 5 UNITED STATES OF AMERICA

Holmes, L.J.E. Pesticide Safety Directorate Ministr Agriculturef yo , Fisherie Food san d Rothamsted HarpendenS , 2S Herts S AL . UNITED KINGDOM

322 Jelic. A , Faculty of Agriculture University in Osijek 7 11 x P.OBo . 54000 Osijek, CROATIA

Krause, H.P. Hoechst AG Abteilung G 813 D-65926 Frankfurt am Main, GERMANY

Langley, P.A. Insect Investigations School of Veterinary Science University of Bristol Churchill Building, Roo1 m23 Langford, Bristol BS1D 87E UNITED KINGDOM

Logan, J.W.M. Natural Resources Institute Central Avenue Chatham Maritime Chatham, Kent ME4 4TB UNITED KINGDOM

Matechi, H.T. Tropical Pesticides Research Institute P.O. Box 3024 Arusha, UNITED REPUBLIC OF TANZANIA

McGuire, M.R. Plant Polymer Research Agricultural Research Service U.S. Departmen f Agriculturo t e . Universit181N 5 . ySt Peoria, IL 61604 UNITED STATES OF AMERICA

Menschel, G. Biologische Bundesanstalt für Land-und Fortswirtschaft Fachgruppe für Chemische Mittelprüfung Messeweg 11/12 D-3300 Braunschweig, GERMANY Neuenschwander, E. Ciba-GeigG yA CH-4333 Münchwilen SWITZERLAND

Oh, Byung-Youl Agricultural Chemicals Research Institute Rural Development Administration # 249 Seodungdong, Kwonsunku Suweon 441-707 REPUBLIC OF KOREA Omar. D , Departmen Planf o t t Protection Universiti Pertanian Malaysia 4340 SerdangM 0UP , Selangor MALAYSIA

Opiyo, E.A. Kenya Trypanosomiasis Research Institute 2 36 x P.OBo . Kikuyu, KENYA

323 Phillemon-Motsu, T. K. Tsetse Control Division Department of Animal Health & Production P.O. Box 14 Maun, BOTSWANA

Qureshi, J.M. Nuclear Institut r Agriculturefo Biologd ean y 8 12 x P.OBo . Faisalabad, PAKISTAN

Rabia, A.K. Ministry of Agriculture, Livestock and Natural Sources 9 15 x P.OBo . Zanzibar, UNITED REPUBLIC OF TANZANIA

Rajagopalan. N , Polymer Chemistry Division National Chemical Laboratory Pune411 008, INDIA

Rankin, F.S. AgriSence-BCS Ltd. Taff'1 s Mead Road Treforest Industrial Estate Pontypridd Mid-Glamorgan CF37 5SU UNITED KINGDOM

Reich. H , Bundesanstalt für Pflanzenschutz Trunrierstrasse5 A-1021 Vienna, AUSTRIA

Rodrigues, J. Centre Nacional de Proteçao da Produçao Agricola Quint Marqueo ad s 2780 Oeiras, PORTUGAL

Schreiber, M.M. Department of Botany & Plant Pathology USDA-ARS-MWA-IWRC Purdue University West Lafayette, IN 47907-1155 UNITED STATE AMERICF SO A

Semanhyia. C , National Nuclear Research Institute Ghana Atomic Energy Commission 0 8 x P.OBo . Legon, GHANA

Smolko, E.E. Comisiön Naciona Energfe d l a Atömica Avda. del Libertador 8250 142 Bueno- 9 s Aires ARGENTINA

Soerjani, M. Cente r Researcrfo f Humaho n Resources and Environment Universit f Indonesiyo a Jalan Salemba4 Jakarta 10430, INDONESIA

324 Soldan. T , Institut f Entomologeo y Czech Academ f Scienceyo s Branisovskâ 31 Ceské Budejovic5 0 0 e37 CZECH REPUBLIC

Soldanova. Z , Institut Entomologf eo y Czech Academ f Scienceyo s Branisovsk1 â3 Ceské Budejovice 370 05 CZECH REPUBLIC

Soltan, H.R. Department of Plant Protection Universit f Alexandriyo a El-Shatby, Alexandria EGYPT

Starer, M.S. Werne Pfleidere& r r 663 E. Crescent Ave. Ramsey, NJ 07446 UNITED STATE AMERICF SO A

Sun Jinhe Institute of Nuclear Agricultural Sciences Zhejiang Agricultural University Hangzhou 310029, CHINA

Szejtli, J. CYCLOLAB Cyclodextrin Researc Developmend han t Laboratory Ltd. 5 43 P.Ox Bo . H-1525 Budapest, HUNGARY

Szente, L. CYCLOLAB Cyclodextrin Researc Developmend han t Laboratory Ltd. 5 43 x P.OBo . H-1525 Budapest, HUNGARY

Vimond-Laboudigue, A. Institut Nationa Rechercha l e d l e Agronomique Ministère de l'Environnement Station de Science du Sol Route de Saint Cyr F-78026 Versailles, FRANCE

Wang Fujun Laboratory for Application of Nuclear Techniques Beijing Agricultural University Beijing 100094, CHINA

Wilson, A. Cooper Zimbabwe (1992) (Pvt) Ltd. Box 2699 Harare, ZIMBABWE

Yamin, B.M. Jabatan Kimia Fakulti Sains Fizi Gunaas& n Universiti Kebangsaan Malaysia 43600 UKM Bangi, Selangor D.E. MALAYSIA

325 Yousefzadeh Faal Deghati, P. Gamma Irradiation Center Atomic Energy Organization of Iran P.O. Box 11365-8486 Tehran, ISLAMIC REPUBLI IRAF CO N

Zerjav, M. Agricultural Institute Hacquetova 2 61000 Ljubljana, SLOVENIA

International Organizations

Hance, R.J. Joint FAO/IAEA Divisio f Nucleano r Techniques in Foo Agriculturd dan e International Atomic Energy Agency 0 10 P.Ox Bo . A-1400 Vienna AUSTRIA

Hussain. M , Joint FAO/IAEA Division of Nuclear Techniques (Scientific Secretary) in Food and Agriculture International Atomic Energy Agency P.O. Box 100 A-1400 Vienna AUSTRIA

Slingenbergh, J. Foo Agriculturad dan l Organizatioe th f no United Nations Viale délie Term i Caracalled a 1-00100 Rome ITALY

Sugavanam. B , Agrochemical Industries Unit Division of Industrial Operations United Nations Industrial Development Organization P.O. Box 300 A-1400 Vienna AUSTRIA

326