Mono Disperse Alcohol Ethoxylates

8/3/2019 Mono Disperse Alcohol Ethoxylates

1/9

J, Agric. Food Chem. 1995, 43, 1067-1075 1067

Sorption of Monodisperse Alcohol Ethoxylates and Their Effects onthe Mobility of 2,4-D in Isolated Plant CuticlesMarkus Riederer,* Markus Burghard t, Sabine Mayer, a nd H ans Obermeier

Julius-von-Sachs-Institut fur Biowissenschaften, Universitat Wurzburg, Mittlerer Dallenbergweg 64,D-97082 Wurzburg, GermanyJorg Schonherr

Institut fur Obstbau und Baumschule, Universitat Hannover, Am Steinberg 3, D-31157 Sarstedt, Germany

The sorption of homologous monodisperse alcohol ethoxylates in cuticles isolated from the leaves ofbitter orange (Ci trus aurant iu m L.) and thoroughly extracted with chloroform (yielding polymermatrix membranes, MX) has been investigated. Sorption isotherms exhibited a linear phase atlow aqueous concentrations of surfactant, while above the critical micelle concentrations (cmc) heconcentration sorbed in the MX was independent of external concentration. W w a t e r partitioncoefficients (25-148 000, depending on the homologue), maximum concentrations in theMX (91-318 mmolkg), and cmc were derived from the isotherms. Quantitative structure-propertyrelationships were established for estimating the W w a t e r partition coefficients, cmc, and maximumcuticular concentrations of alcohol ethoxylates from the number of carbon atoms and oxyethylenegroups of their alcohol and poly(oxyethy1ene)moieties, respectively. These relationships were usedfor analyzing the effects of monodisperse alcohol ethoxylates and primary alcohols on the mobilityof (2,4-dichlorophenoxy)acetic cid in isolated cuticular membranes from bitter orange leaves.

Keywords: Plant cuticles; foliar penetration; surfactants; adjuva nts; Citrus aur an tiu mINTRODUCTION

Nonionic surfactants of the poly(oxyethy1ene) ype arewidely used in pesticidal formulations. Some enhancethe activity of many foliage-applied pesticides by in-creasing the rates of uptake of active ingredients fromdroplets deposited on leaf surfaces (Stock and Holloway,1993; Kirkwood, 1993; Stevens et al., 1991; Gaskin andHolloway, 1992; Stock et al., 1992; Knoche and Bukovac,1992; Holloway and Edgerton, 1992). The cuticle,representing the main barrier to foliar uptake, has beenthe object of further studies performed under rigorouslycontrolled experimental conditions with the objectiveofa mechanistic understanding of the processes involvedduring the enhancement of activity. Selected nonionicsurfactants effectively increased the permeability ofisolated cuticles for water (Geyer and Schonherr, 1989;Riederer and Schonherr, 1990; Schonherr and Bauer,1992) and organic solutes (Schonherr and Bauer, 1992;Schonherr et al,, 1991; Tan and Crabtree, 1992; Chamelet al., 1992; Knoche and Bukovac, 1993). Enhancedcuticular permeability in the presence of alcohol ethoxyl-ates can be attributed in part t o an increase in themobility of the penetrant in the cuticle (Schonherr,1993a,b). Adjuvants having this property have beentermed accelerators (Schonherr, 1993b).

It is an essential prerequisite for accelerating themovement of the active ingredient across the plantsurface that sufficient quantities of surfactant arepresent in the cuticle during permeation of the activeingredient. Supposing a linear dose-effect relationship,the efficacy of an accelerator should depend on itsconcentration in the cuticle, which in turn will directly*Author t o whom correspondence should be ad-dressed (fax+ 49 931 888 6234; e-mail rie der eab ota n-ik.uni-wuerzburg.de).

0021 8561 95/1443-1067$09.00/0

depend on the accelerators activity in the outsidemedium. Such dat a are needed t o interpret the widevariation of efficiency between different classes of non-ionic surfactants or among homologous series.The sorption of nonionic surfactants from aqueoussolutions in isolated plant cuticles has been studiedrepeatedly (Schonherr et al., 1991; Shafer et al., 1989;Bukovac and Petracek, 1993). These experiments havebeen performed using a technical polydisperse surfac-tant preparation containing a large number of indi-vidual homologues. Due t o their complex and unknowncomposition, such polydisperse surfactants are not suit-able fo r critical studies on the effect of chemical struc-ture of a poly(oxyethy1ene) surfactant on it s sorption byplant cuticles.Therefore, the sorption of monodisperse alcohol ethox-ylates has been investigated in this study. The objectiveof our work was t o study sorption by standardizedcuticular material as a function of concentration in theaqueous phase for a series of alcohol ethoxylates varyingin chemical structure. Our findings will help elucidatethe general principles governing the cuticular sorption

of alcohol ethoxylates and thus surfactant concentra-tions in the cuticle during the permeation of surfactantsand active ingredients. The results and the quantitativerelationships derived will also be used to analyze theeffectsof monodisperse alcohol ethoxylates and primaryalcohols on the mobility of 2,4-dichlorophenoxyaceticacid (2,4-D) in isolated cuticles (Schonherr, 1993a,b).EXPERIMENTAL PROCEDURES

Cuticular Membranes. Adaxial cuticular membraneswere enzymatically isolated (Schonherr and Riederer, 1986)from mature leaves of small bitter orange (Ci trusaurantiumL.) trees grown in growth chambers as described elsewhere(Geyer and Schonherr, 1990). The isolated cuticles were0 995 American Chemical Society


2/9

1068 J. Agric. Food Chem., Vol. 43, No. 4, 1995Table 1. Molar Weights ( M W )and Characteristic Molecular Volumes (V,) of the Alcohol Ethoxylates Used in thePresent Study

Riederer et al.

name abbrev MW Vza cm3/mol)diethylene glycol monobutyl ether C4EZ 162.23 141.19triethylene glycol monohexyl ether 234.34 203.42diethylene glycol monooctyl ether CsEz 218.34 197.55triethylene glycol monooctyl ether CsE3 262.39 231.60tetraethylene glycol monooctyl ether CsE4 306.45 265.65pentaethylene glycol monooctyl ether CsE5 350.50 299.70triethylene glycol monodecyl ether C10E3 290.45 259.78tetraethylene glycol monodecyl ether C10E4 334.50 293.83pentaethylene glycol monodecyl ether ClOE5 378.56 327.88hexaethylene glycol monodecyl ether ClOE6 422.61 361.93heptaethylene glycol monodecyl ether 466.66 395.98octaethylene glycol monodecyl ether CioEs 510.72 430.03diethylene glycol monododecyl ether CizEz 274.45 253.91triethylene glycol monododecyl ether C12E3 318.50 287.96tetraethylene glycol monododecyl ether C12E4 362.56 322.01pentaethylene glycol monododecyl ether Cl2E5 406.61 356.06hexaethylene glycol monododecyl ether ClZE6 450.66 390.11heptaethylene glycol monododecyl ether CizE7 494.72 424.16octaethylene glycol monododecyl ether CizEs 538.77 458.21triethylene glycol monotetradecyl ether C14E3 346.56 316.14pentaethylene glycol monotetradecyl ether C14E5 434.66 384.24heptaethylene glycol monotetradecyl ether C14E7 522.77 452.34octaethylene glycol monotetradecyl ether C14& 566.83 486.39octaethylene glycol monohexadecyl ether C16E8 594.88 514.57

a Calculated according to the method of Abraham and McGowan (1987).exhaustively extracted (16h) with chloroform (Riedel-de Haen,Seelze, Germany) in a Soxhlet apparatus to remove cuticularwaxes. The resulting cuticular material will be referred t o aspolymer matrix membranes (MX). This treatment was neces-sary to avoid problems during the gas chromatographicquantification of the amounts of alcohol ethoxylates sorbed inthe cuticle since wax constituents and surfac tants had similarretention times. MX consisting mainly of cutin are goodmodels for the overall sorptive properties of plant cuticles forlipophilic compounds. It has been shown experimentally th atthe fundamental processes governing sorption are identical forextracted and nonextracted cuticles (Riederer and Schonherr,1984, 1986; Kerler and Schonherr, 1988).Monodisperse Alcohol Ethoxylates. Monodisperse al-cohol ethoxylates were obtained from Fluka (Neu-Ulm, Ger-many) and their identity and purity confirmed by capillarygas chromatography/mass spectrometry. Twenty-four homo-logues were selected on the basis of purity (97-99%), struc-tur al diversity, and size (Table 1).Sorption Isotherms. Stock solutions (80-150 mmol/L) ofthe different homologues were prepared in chloroform. Vari-ous volumes of these solutions were added to 100-mL glassvials, the solvent was evaporated under a gentle stream of Nz,and the surfactant was redissolved in 100 mL of waterbidistilled over quartz. Sodium azide (1mM) was added tothe solutions t o prevent growth of microorganisms. For eachhomologue, 12- 14 different concentrations were prepared suchthat they covered a range of concentrat ions from 1 o 2 ordersof magnitude below t o about 1 order of magnitude above thecritical micelle concentration (cmc).The sorption experiments were started by adding polymermatrix membranes (ca. 6 mg) to the solution that had beenbrought t o 25 "C before. The vials were closed with Teflon-lined septum caps and slightly agitated for20 h in an incubator(Memmert , Schwabach, Germany) a t a temperature of 25 & 1"C. Preliminary experiments had shown that after this periodof time the concentrations of the alcohol ethoxylates in theMX were invariant with time. The variation of the amount ofMX added to the surfactant solution had no influence onK m ,indicating that, in the concentration range studied, surfaceadsorption was negligible with respect to absorption by thecutin.Subsequently, the MX were carefully removed from thesolution, gently blotted with soft tissue paper t o removeadhering solution, and air-dried for a t least 24 h. Internalstandard (an alkane between hexadecane and nonacosanedepending on the retention time of the respective alcohol

ethoxylate) was added to the dry MX in a 1-mL Reactivial(Wheaton, Millville, NJ). After extraction of the MX with 1mL of chloroform a t 60 "C (preliminary tests had shown tha ta 10-min treatment under these conditions was sufficient forcomplete extraction), he cuticular materia l was removed andthe resulting solution was reduced to dryness under a streamof Nz. These samples were subsequently analyzed by gaschromatography, and the concentration of surfactant sorbedin the MX was calculated.The surfactant concentration in the aqueous phase wasdetermined by taking aliquots of various volumes and evapo-rating the water under a gentle stream of clean air from anair generator (Chrompack, Middelburg, The Netherlands) at40 "C. Preliminary study established that volatility of the

surfactants was not a problem. After an appropriate amountof internal standa rd was added, the samples were analyzedby gas chromatography.Gas Chromatographic Analysis of Alcohol Ethoxyl-ates. Forgas chromatographic analysis, the terminal hydroxylgroups of the alcohol ethoxylates were transformed into thecorresponding trimethylsiloxyl ethers by reaction with N,N-bis(trimethylsily1)trifluoroacetamide (Macherey & Nagel, Dii-ren, Germany) in d ry pyridine (Merck, Darmstadt, Germany).Analyses were performed with capillary gas chromatographs(Fractovap 4200, Carlo Erba Strumentazione, Milano, Italy;GC-l4A, Shimadzu, Kyoto, Japan; HP 5890 11, Hewlett-Packard, Millville, NJ) equipped with on-column injectors andflame ionization detectors and operated with Hz as carrier gas.Separation was achieved on a 25 m x 0.32 mm fused silicaWCOT capillary column (df 0.12 pm; CP-Si1 5 CB, Chrom-

pack) using temperature programs adjusted t o the retentionof the respective homologue.For quantification, peak areas were multiplied with specificcorrection factors determined for each homologue separatelyby analyzing solutions of known concentrations of the alcoholethoxylate and the internal standard. Correction factors weredetermined daily for each of the chromatographs and homo-logues used.Tensiometry. The surface tension of aqueous solutions ofmonodisperse alcohol ethoxylates was measured a t 25 "C as afunction of concentration using an interfacial tensiometer (K8,Kriiss, Hamburg, Germany) with a thermostated sampleholder. The solutions containing 1 mM sodium azide wereprepared by successive dilution of a stock solution. Care wastaken tha t only the equilibrium values of surface tension wererecorded (up to 60 min of equilibration time).


3/9

Sorption of Alcohol Ethoxylates in Plant CuticlesCritical micelle concentrations (cmc)of monodisperse alcoholethoxylates were determined in the conventional way byplotting the surface tensions us the logarithm of the surfactantconcentration in the solution. The cmc was obtained by solvingthe regression equation fitted to the descending branch of thegraph for the constant value of surface tension measured athigh surfactant concentrations.The regression equations describing the change in surfacetension with concentrat ion were also used to measure partitioncoefficients of alcohol ethoxylates between the aqueous phaseand MX. For this purpose, MX were added to a surfactant

solution with a concentration below the cmc and exactly knownsurface tension. After incubation a t 25 f.1 "C for at leas t 20h, the surface tension of the supernatant was measured againand the corresponding concentration estimated using theregression equation. The equilibrium concentration of sur-factant sorbed in the MX was computed from the concentrationdifference, the volume of the aqueous phase, and the mass ofthe MX used in the experiment. The W w a t e r partitioncoefficient K M ~as calculated according to

A

J. Agric. Food Chem.,Vol. 43, No. 4, 1995 1069

50

with CD.IXnd CW eferring t o the equilibrium concentration(molkg) of the surfactant in the MX and the aqueous super-natant, respectively.To apply this method for determining W w a t e r partitioncoefficients to the maximum range of homologues, the massratio of MX to aqueous phase was adjusted for each homologuesuch that the concentration difference due to sorption wasmeasurable and did not exceed the validity of the regressionequations.Effects on the Mobility of 2,4-D in Isolated PlantCuticles. Data on the effects of monodisperse alcohol ethoxyl-ates on the mobility of (2,4-dichlorophenoxy)aceticcid (2,4-D) in isolated cuticular membranes from leaves of bitter orangewere taken from two previous publications (Schonhem, 1993a,b).Mobilities were deduced from the kinetics of unilateral desorp-tion of 2,4-D from the outer surface of isolated cuticles. Effectswere standardized to a common initial rate constant k, f l/kL= 15 x l o 5 s. With compounds exhibiting time-dependenteffects on 2,4-D mobility, only the standardized maximumvalues (i.e. maximum effect) were used. For details on thematerials and methods see Schonherr (1993a,b).Statistics. Data were statistically evaluated using SPSSfo r Windows (SPSS Inc., Chicago, IL). Unless stated other-wise, the coefficients of the regression equations (eqs 4-8) weresignificantly different from zero on a level ofp < 0.0005. Thecoefficients are given together with their 95% confidenceintervals.

-

-

RESULTS AND DISCUSSIONSorption Isotherms. Monodisp erse alcohol ethoxyl-ates in aqueou s solution were readily sorbed in cuticular

polymer matrix materials isolated from bitter orangeleaves. The equilibrium concentrations in t he MX (CMX)varied with those in the aqueous supernatant (CW)(Figur e 1A).

The isotherms can be fully characterized by threedistinctive features: (1) linear initial slopes (Figure l B),(2) relatively sh ar p points of inflection intermediat ebetween th e portions with linear slopes and (3) thosewith zero slopes due to constan t values of CMX Figure1A). This general shape has been found with allhomologues studied even though t he absol ute values ofinitial slopes, the locations of th e inflection, and th eplat eau values vari ed significantly and systematicallyamong t he different compounds studied.

Similar isotherms h ave been observed for the adsorp-tion of nonionic sur fac tan ts on th e surfaces of solids likesilica, coal, or clay (Clunie and Ingram, 1983; Furlongand Aston, 1982; Celik and Yoon, 1991). Those adsorp-tion isoth erms could be treated according to th e Lang-

200-0S 150EE0

ug 100

50

0" ' . ' ' ' ' I ' '0.0 0.1 0. 2 0.3 0.4 0 .5

mn

150 1-n0E 100EI30

/ Y C 1 2 4

"0.00 0.02 0.04 0.06C, [mmollkg]

Figure 1. Isotherms (25 "C ) fo r the sorption of a homologousseries of alcohol ethoxylates (ClzE,) in cuticular polymer matrixmembranes from bitter orange leaves: (A ) isotherms over thewhole concentration range studied; (B ) initial linear portionsof these isotherms. Concentrations in the polymer matr ix( CMX)nd the aqueous phase (CW)were determined by the gaschromatographic method.mui r formalism (Adamson, 1982; Clunie and Ingram,1983; Rosen, 1989). Langmui r adsorption isotherm sapproach a plateau which is due to the sa tura t ion of alimited num ber of sorption sites on t he surface of th esubstrate. However, attem pts to fit the Langmuirequation to th e data describing the sorption of mono-disperse alcohol ethoxylates in MX failed. Regressionanalyses showed that t he initial portions of the iso-therms were strictly linear (Figure 1B). Thus, t h eisothe rms observed in this stud y consist of two linearcomponents (the initial slope and the plateau) and ashort transitional region in between (Figure 2).

The isotherms found in the present s tudy (a t aqueousconcentrations 1-2 orders of magnitude below and 1order of magnitude above cmc) can be classified accord-ing to Giles et al. as type C2 (Giles et al., 1974). Thelinear is otherm s are indicative for conditions where th enumber of sorption sites remains constant throughouta wide range of solute concentrations. Such behavioris characteristic for amorphous solids which, at lowconcentrations, ac t as quasi-liquid phases (Giles et al.,1974). Thus, at low concentrati ons of surfac tan ts in t heaqueou s solution, the dist ribution of alcohol ethoxylatesbetween the cuticular matrix and th e aqueous solutionclosely resembles parti tionin g between two liquid phases.Constant partitioning isotherms have also been de-scribed for monodisperse in th e system heptane/


4/9

1070 J. Agric. Food Chem.,Vol. 43, No. 4, 1995 Riederer et al.

1501 ,J

01 ' , I I I0. 0 0.1 0.2 0.3 0.4

C, [mmollkg]Figure 2. Isotherm (2 5 "C ) for the sorption of ClzEs inpolymer matrix membranes from bitter orange leaves. Thearrows indicate how the critical micelle concentration (cmc)and the maximum concentration in the MX were deduced. TheW w a t e r partition coefficientK was obtained from theslope of the initial linear portion of the isotherm.water (Aveyard et al., 1990). Linear sorption isothermshave already been described for the sorption of severallipophilic organic compounds in the cuticles isolatedfrom various plant species (Riederer and Schonherr,1984, 1986).The intercepts of the initial linear portions of thesorption isotherms measured in this study are notsignificantly different from the origin. This fact under-lines that a constant partitioning phenomenon (obeyingHenry's law) is taking place at low concentrations ofalcohol ethoxylates in the aqueous phase. The linearportions can therefore be described by

where the slope KMxlw is the W w a t e r partition coef-ficient.The plateaus of the sorption isotherms (Figure 1)found in this study could, theoretically, be attributedt o two different causes: (1) he filling of the sorptivecapacity of the MX or (2) a dramatic concentration-dependent change of the activity of the solute in theaqueous phase. The validity of the first hypothesis canbe assessed by estimating the volume fraction occupiedby the alcohol ethoxylates in the MX in the plateauregion. The characteristic molecular volumes of thedifferent homologues were computed according t o themethod of Abraham and McGowan (1987),and a specificmass of 1100 kg/m3 was assumed for the MX (Schreiberand Schonherr, 1990). The resulting volume fractionsvary from 4.7 vol % for to 9.7 vol % for C12E2.Such a large variation (by almost a factor of 2) is notcompatible with the hypotheses that the plateaus aredue to attainment of a maximum sorption capacitybecause this capacity should be equal for all homologues.In addition, the volume fractions occupied by the alcoholethoxylates are more than a factor of 2 lower than themaximum sorption capacity of approximately 21 vol %determined fo r 4-nitrophenol (Riederer and Schonherr,1986).For these reasons, the a lternative explanationof theplateaus of the sorption isotherms is favored. Indepen-dent of the treatment of micellization in aqueous solu-tions of nonionic surfactants (Moroi, 1992; Nagarajan

and Ruckenstein, 1977, 1991; Pu wa da and Blank-schtein, 1990; Ruckenstein and Nagarajan, 1975,1981;Tanford, 1973) t has been stated that, above the criticalmicelle concentration (cmc), any added surfactant mol-ecules will not increase the concentration of the freemonomers in solution. Instead, they will form micellaraggregates. Therefore, the activity of free monomerswill increase linearly with the bulk concentration in theaqueous phase up to a concentration close to the cmc.Above the cmc, the activity of free monomers of mono-disperse surfactants will remain nearly constant eventhough the bulk concentration still rises (Levitz, 1991).The sorption of monodisperse alcohol ethoxylates inplant cuticular matrix closely follows the concentrationdependence of the activity of the free monomer in theaqueous phase as predicted by a theoretical model(Levitz, 1991). This shows that the species taking par tin the partitioning between the aqueous solution andthe MX is only the free monomer and not those ag-gregated in micelles. The micelles act as a reservoir,keeping the concentration of the free monomers at aconstant level even when sorption by the MX removespart of them. Consequently, the plateau portionsof theisotherms represent the maximum concentrations(CgF) that can be reached by sorption from aqueoussolution at a given temperature. The maximum con-centration is in equilibrium with a constant concentra-tion of the free monomers in the aqueous phase whichis practically equal to the cmc. According t o this model,it should be possible t o estimate cmc from sorptionisotherms by solving eq 2 for C M X= C'$z using KMNW(Figure 2). Thus, complete sorption isotherms as de-termined by the gas chromatographic method providethe three basic parameters ( KM X W ,mc, CE?) neces-sary fo r a description of the concentration-dependentsorption of alcohol ethoxylates in plant MX.This interpretation of the sorption isotherms of mono-disperse alcohol ethoxylates can be tested by measuringW w a t e r partition coefficients and cmc by an indepen-dent method and by estimating CEF according to

In this study, tensiometry has been used as an inde-pendent method for measuring cmc and KMxlw. Theresults obtained by both methods will be compared anddiscussed in the following sections.MX/Water Partition Coefficients. The partitioncoefficients of a series of monodisperse alcohol ethoxyl-ates determined in the system C. aurantium leafWwater (Table 2) ranged from 25 (C8E5) to 148 000(ClsEs). The gas chromatographic and tensiometricmethods produced results identical on the 95% confi-dence level (Table 2).For monodisperse alcohol ethoxylates with variousalkyl chain lengths and degreesof ethoxylation, the MX !water partition coefficients depended systematically onchemical structure. Within a homologous series havingan equal number of oxyethylene groups but differing inthe length of the alkyl chain, the values of KMXWincreased markedly with the numbers of methylenegroups in the alcohol moiety. For instance, partitioncoefficients measured fo r the series C,E3 increased from69 (CsE3) o 69 200 (C14E3) with C10E3 and C12E3 havingintermediate values of 740 and 7590, respectively.Homologous series with a constant alkyl chain lengthbut a varying degree of ethoxylation exhibited a com-parable, even though much less pronounced, dependence


5/9

Sorption of Alcohol Ethoxylates in Plant CuticlesTable 2. W a t e r Partition Coefficients K m ) ,Critical Micelle Concentrations (cmc), and MaximumCuticular Concentrations (CEy)for MonodisperseAlcohol Ethoxylates

J. Agric. food Chem., Vol. 43, No. 4, 995 1071

cmc C Elog KMxiw (mmol/kg) mmoy g/TM a GCb TMa GCb kg kg

C4E2 537.0C6E3 81.0C8E2 2.02 (1.97-2.06FC8E3 1.84 (1.81-1.86)C8E4 1.60 (1.58-1.63) 6.58C8E5 1.40 (1.39-1.42)2.87 (2.83-2.90) 0.42 318 922.68 (2.67-2.70)2.58 2.50 (2.45-2.55)2.39 (2.31-2.46)2.22 (2.15-2.27)2.00 1.95 (1.90-2.00)4.06 4.21 (4.15-4.26)3.88 3.88 (3.86-3.90)3.74 3.72 (3.68-3.75)3.56 3.49 (3.47-3.50)3.39 3.39 (3.34-3.43)3.19 3.16 (3.12-3.20)3.08 2.95 (2.91-2.98)4.84 (4.78-4.891

4.44 (4.39-4.49)4.224.06 4.08 (4.01-4.15)5.05 5.17 (5.13-5.20)

0.69

1.10.0310.0310.0420.0530.0650.0850.089

0.00700.00880.00073

0.680.620.791.10.0290.0370.0530.0630.0790.0890.00270.00400.0090

217 82148 63133 62107 55259 82231 84169 67140 63109 5497 52234 81182 7991 52

a Determined by tensiometry. Wwa t e r partition coefficientsdetermined by this method are exact to approximately 10.1 ogunit. Determined by capillary gas chromatography. Upper andlower limits of the 95% confidence intervals.of W w a t e r partition coefficients on chemical structure.Partition coefficients in the series from C12E2 to C12E8steadily decreased from 16 200 t o 890, respectively(Table 2).The dependence of the W w a t e r partition coefficientson the chemical structure of the monodisperse alcoholethoxylates can be expressed by a simple multiple-regression equation using the number of carbon atoms( C ) and the number of oxyethylene groups ( E ) as de-scriptors for molecular structure:lOgKMx,, = -1.78(&0.11) + 0.52(&0.01)C-

0.17(kO.Ol)E (4)N = 33 , adjusted r2= 0.997, F = 5357

The data obtained using both methods for measuringpartition coefficients were combined for establishing thisquantitative structure-property relationship (QSPR).Values ofK m stimated from this equation are in goodagreement with the experimental results (Figure 3).Equations with coefficients not significantly differentfrom those in eq 4 were obtained when the results fromboth methods were treated separately.Equation 4 can be used to derive general rules fo r thedependenceof W w a t e r partition coefficients of alcoholethoxylates on chemical structure. This QSPR predictsthat increasing the alkyl chain length by two methylenegroups will increase the W w a t e r partition coefficientby almost exactly 1order of magnitude. A decrease ofthe same extent would result from an elongation of th epoly(oxyethy1ene)part of the molecule by six units.Critical Micelle Concentrations. Critical micelleconcentrations give the upper limit of the concentrationrange where partition coefficients can serve as validdescriptors fo r the sorption of alcohol ethoxylates in

prediction...... ......... 'A'

Q.Q..' C . . Q .

A-.&...0 TM

l o 0 I . I I l I 10 2 4 6 8E

Figure 3. Plots of the Wwater partition coefficients (KMXN)of a homologous series of alcohol ethoxylates us the numberof ethylene oxide units (E ) in the poly(oxyethy1ene) parts ofthe molecules. Data obtainedby the gas chromatographic(GC)and the tensiometric methods (TM) are shown together withthe prediction calculated from eq 4.plant cuticular material (Figures 1 and 2). The cmc ofthe monodisperse alcohol ethoxylates used in this studyvaried from 0.00073 (C16E8) to 537 (C4E2) mmoVkg(Table 2). Both the tensiometric and the gas chromato-graphic methods used for determining cmc producedresults identical within the limits of error. This sup-ports the interpretation of the shape of the sorptionisotherms outlined above. Reaching the cmc in theaqueous phase causes the inflection between the ini tiallinear portion with a constant slope to the plateau ofthe sorption isotherms (Figures 1A and 2).Critical micelle concentrations of homologues varyingin alkyl chain length and ethoxylation also dependedsystematically on chemical structures. Within thehomologous series of C,E3, the cmc decreased from 0.68(C10E3) o 0.0027 mmoLkg (C14E3) while it increased forthe series C12E, from 0.029 (C12E3)to 0.089 mmoVkg(C12E8). The values of cmc measured in this study arein good agreement with data in the lite rature (Meguroet al., 1987). Using the results obtained from bothmethods, the following multiple regression equation wasestablishedlog cmc = 4.75(&0.09)- 0.54(fO.O1)C+

0.09(&0.01)E ( 5 )N = 29, adjusted r2= 0.998,F = 9114

describing the dependence of the critical micelle con-centration (in mmoVkg) on the number of the carbonatoms in the alkyl ( C ) and those of' the oxyethylenegroups (E) n the poly(oxyethy1ene)chains, respectively.This equation successfully predicts the cmc valuesdetermined experimentally (Figure 4). The regressionequations obtained for the results from the two inde-pendent methods were not significantly different fromeq 5 .This QSPR shows that the cmc of the homologousmonodisperse alcohol ethoxylates used in this studyagain depend markedly on the chain lengths of thealcohol moieties. Increasing the length of the alkylchain by two methylene groups decreases the cmc by afactor of approximately 10. The same decrease of cmcis predicted by eq 5 for a shortening of the poly-(oxyethylene) chain by 11 ethylene oxide units. The


6/9

1072 J. Agric. Food Chem., Vol. 43, No. 4, 1995

300-EEE 2002E

E-U

mEO" 100

Riederer et al.

-

-

-

1 0 5 ~ , I 9 , I I

- A .......@ . . . * . . . k . - 4 -

........ A ........o . -@................ .o -c1 6 .......... . I . I

c 1 0 ....-- c12 o . . . ~ . . . e . . a . - ~ . - - Q . . O-- C14 ....

A GC0 TM04 1 ..... prediction

................ . . . . . . . . . . .

1 0 2I -31040 2 4 6 0

EFigure 4. Plots of the critical micelle concentrations (cmc)ofa homologous series of alcohol ethoxylates us the number ofthe ethylene oxide units (E) in the poly(oxyethy1ene)parts ofthe molecules. Data obtained by the gas chromatographic (GC)and the tensiometric methods (TM) are shown together withthe prediction calculated from eq 5 .coefficients fo r C and E in eqs 4 and 5 have oppositesigns, respectively, showing that, fo r instance, a struc-tural change leading to a lower cmc will simultaneouslyincrease the W w a t e r partition coefficient. As far asvariations of the alkyl chain length are concerned, theywill even result in equal absolute changes of cmc andK M X ~ Wut in opposite directions.Maximum Concentrations in the MX. When theconcentration of a monodisperse alcohol ethoxylate inaqueous solution exceeds the cmc, its equilibriumconcentration in the MX phase remains constant (Fig-ures 1 and 2). As the activity of the free monomer inan alcohol ethoxylate solution remains practically con-stant a t concentrations above the cmc, the correspond-ing equilibrium concentration in the MX is the maxi-mum concentration (CGy) attainable in an Wa q u e -ous solution system. Using the gas chromatographicmethod, values of CGy were determined for somehomologues (Table 2). The experimental values ofCr$ ranged from 91 (C14E8) to 318 mmoykg (C10E3).Generally, variation of Czy as a consequenceof chem-ical structure was small ( at the maximum, by a factorof 3.5) compared t o that observed with KMXW nd cmc,which varied over several orders of magnitude. Withinhomologous series, Czy decreased with increasingnumbers of both C and E (Table 2).This dependence of the experimentally determinedvalues of CGy (mmoykg) on C and E can be expressedby the multiple regression equationlog CzF = 2.96(f0.16)- O.OZ(&O.Ol)C-

0.09(fO.Ol)E (6)N = 14, djusted r2= 0.969,F = 206.3

In contrast to the QSPRs for KMXIWnd cmc (eqs4 nd5, respectively), the coefficients for the independentvariables have both negative signs and the dependenceof CgF on C is fairly small but, nevertheless, signifi-cantly different from zero ( p = 0.0026). Here, only avariation of the alkyl chain length by 50 methylenegroups would produce a change of CGF by a factor of10, while in the cases of KMXIWnd cmc a variation by

I I I

100 200 300Coma'predicted [mmol/kg]

Figure 5. Plot of the experimentally determined maximumconcentrations of various homologous alcohol ethoxylates inbitter orange MX (C;l;x")us values predicted from eq 6. Theslope of the graph is not significantly different from 1.0(adjusted rz = 0.969).two carbon atoms would have the same effect. Also incontrast t o K M X ~nd cmc, C E s affected to a muchhigher extent by a variation of the number of theethylene oxide units than by the carbon number of thealcohol moiety. Elongating the poly(oxyethy1ene)chainby 11 units would decrease CGy by 1 order of magni-tude.A different approach leads t o an equivalent quantita-tive relationship between Cp$ and C and E, respec-tively. Based on eq 3, eqs 4 and 5 can be combined t ogive (Czy again in mmoykg)

log CzF = 2.97 - 0.02C - 0.08E (7)This combination of the QSPRs fo r KMXIWnd cmcillustrates how the drastic effect of C on both singleparameters is reduced such that it exerts a fairly smalleffect on CG?. This is due t o almost identical absolutevalues of the coefficient for the C variable and t o it sopposite signs in eqs 4 and 5. On the other hand, thecoefficients of the E variable in eqs 4 and 5 do notcounteract completely, resulting in a slight negativedependence of CZY on the degree of ethoxylation. Thisdependence is dominated by that ofKMxw on the lengthof the poly(oxyethy1ene)chain.The multiple regression equation derived from ex-perimental data (eq 6) and tha t obtained by combiningeqs 4 and 5 (eq 7) are not significantly different. Thisfact supports the interpretation of the sorption iso-therms of monodisperse alcohol ethoxylates in plantcuticular matrix as outlined above. Both equations canbe used t o predict maximum concentrations in the MXbecause the estimates agree well with the experimentaldata (Figure 5).It can be expected that similar relationships betweenthe numbers of C and E and KMX~W,he cmc, andC E xist for other groups of ethoxylated surfactants.The dependence on the number of oxyethylene groupsmay even be identical, while the quantitative structure-property relationships will have intercepts varying withthe nature of the hydrophobic part of the surfactantmolecule.Analysis of the Effectsof Alcohol Ethoxylates onthe Mobility of 2,4-D in Isolated Cuticles. Most of


7/9

Sorption of Alcohol Ethoxylates in Plant CuticlesTable 3. MX/Water Partition Coefficients ( K m ) ,Critical Micelle Concentrations (cmc),EquilibriumConcentrations(Cm),nd Volume Fractions in the MXas well as the Maximum Effects of the Compounds onthe Mobility of 2,4-D in Isolated Bitter Orange LeafCuticles 30Y#f 2 0 -

2E'R

I O

J. Agric. Food Cbem., Vol. 43, No . 4, 1995 1073

-

-

compda K M X ~72240162

11074.150.133.979426301780120081355037225117011528840131808910407027501860126020410151400

cmcC(mmolf'kg)

3.34.15.06.27.6

0.280.340.420.510.630.780.951.170.0280.0350.0520.0650.0790.0980.00660.00068

cM.8(mmol/kg)1066102451042034028022077666047038031025020017013011045635028019015012010011083.2

vol maxfractionC effectf0.135 38.80.146 24.90.092 25.40.091 21.00.086 17.60.081 15.60.074 13.40.123 29.80.115 22.80.099 16.90.094 16.10.088 14.90.081 18.20.074 17.50.066 12.00.059 8.40.052 5.40.093 15.50.097 16.80.089 15.40.073 14.10.065 12.30.057 8.40.050 8.670.056 9.120.047 5.3

a C,E,, monodisperse alcohol ethoxylates; C,Eo, 1-alkanols.Estimated according to eq 4. Estimated according to eq 5 .Estimated according to eq 6 except for compounds marked by g.e Equilibrium volume fractions occupied by the compounds sorbedin the MX were calculated from the corresponding C M X , hecharacteristic molecular volumes (V,) (Abraham and McGowan,19871, and a specific mass of 1100 kg/m3of the MX (Schreiber andSchbnherr, 1990). Data from Schonherr (1993a,b). Effects arestandardized to an initial rate constant of desorption (k,) f l l k , =15 x s. g Equilibrium concentrations of 1-alkanols in the MXswere estimated from their aqueous solubilities SW according toCM X = KMX//WSW.queous solubilities used: (Yalkowsky andValvani, 1980) 1-heptanol, 15 mm ohg; 1-octanol, 4.3 mmolkg;1-nonanol, 0.98 mmolkg; 1-decanol, 0.25 mmomg; 1-dodecanol,0.016 mmolf'kg.the monodisperse alcohol ethoxylates used in this studyhave also been tested for their effects on the mobilityof 2,4-D in isolated cuticular membranes from leavesof C. urantium (Table 3) (Schonherr, 1993a,b). Inaddition, the effectsof some primary alcohols have beeninvestigated. The method used fo r measuring adjuvanteffects on 2,4-D mobility was unilateral desorption fromthe outer surface (Schonherr, 1993a,b). This methodis especially suitable for this kind of study as it allowsfo r the equilibration of the adjuvant between the de-sorption medium and the cuticle. Thus, the equilibriumapproach used in the present study for describing thepartitioning of alcohol ethoxylates between an aqueousphase and the cuticular matrix is applicable. Therefore,data and concepts from the present study can be usedto analyze quantitatively the adjuvant effects on 2,4-Dmobility by relating them to the concentrations effectivewithin the cuticle during the experiments. A furtherprerequisite fo r this approach is a linear relationshipbetween the partition coefficients of the surfactants inthe systems W w a t e r and wd wa te r. Only under thiscondition is C$F a valid relative measure for thesurfactant concentration in the cuticular wax represent-ing the actual transport barrier. The validity of this

40m10.03 0.06 0.09 0.12 t

Volume fraction15

Figure 6. Dose-effect curve for alcohol ethoxylates andprimary alcohols on the mobility of 2,4-D in isolated cuticularmembranes from bitter orange (data from Table 3) .assumption has recently been shown experimentally(Riederer and Schreiber, 1995; Burghardt and Riederer,unpublished results, 1995).

For the different alcohol ethoxylates and alcohols usedin the mobility studies, W w a t e r partition coefficients,critical micelle concentrations (where appropriate), andequilibrium concentrations in the cuticle can be esti-mated using the QSPRsdeveloped in the present work(Table 3). Only those homologues of alcohol ethoxylateswere included that had been used in the desorptionexperiments at concentrations significantly exceedingtheir cmc as estimated from eq 5 .A linear dose-effect relationship was obtained whenthe maximum effects on the mobilityof 2,443 in isolatedcuticles were plotted us the volume fractions of thealcohol ethoxylates o r alcohols in the cuticle (Figure6 ) .The relationship can be described bymaximum effect = -6 .5( f4 .1) +272(f 4 8 . 5 volume fraction (8 )

N = 25 , adjusted r2= 0.847, F = 134.1Equilibrium volume fractions occupied by the com-pounds in the cuticle were calculated from the corre-sponding equilibrium concentrations in the cuticle, thecharacteristic molecular volumes (Abraham and McGow-an, 19871, and the specific mass of the cuticle (Schreiberand Schonherr, 1990). Concentrations in the cuticlewere estimated either according to eq 7 (alcohol ethoxyl-ates) or from the product of the cuticle/water partitioncoefficient (Table 3) and the aqueous solubilities(Yalkowsky and Valvani, 1980) (alcohols). In the latte rcase concentrations in the aqueous solutions used asdesorption medium always exceeded aqueous solubili-ties. A somewhat smaller r2 was obtained when theequilibrium concentration in the cuticle was used in-stead of the volume fraction.The type of dose-effect relationship found here (eq8) may help to advance the understanding of the modeof action of aliphatic alcohols and their poly(oxyethy1-ene) derivatives on the cuticular transport properties.It suggests that their effects can be expectedto linearlydepend on the volume fraction o r concentration ofsurfactants in the transport limiting barrier of thecuticle. The fact that all homologuesof alcohol ethoxyl-ates studied and even the free alcohols share one


8/9

1074 J. Agrk Food Chem., Vol. 43, No. 4, 1995common linear dose-effect relationship further suggeststh at the individual dose-effect curves of each of thesecompounds are also linear. The slopesof these (hypo-thetical) individual plots must be equal because onlythen do effects measured a t widely different volumefractions combine t o form a common dose-effect curve(Figure 6). This further implies that all compoundsincluded in this study have a common intrinsic activity(Le. slopeof the individual dose-effect relationship) andthat , within the range of validity of eq 8, differences inthe chemical and physical properties of the compoundsincluded in this study do not affect the acceleration of2,4-D diffusion in isolated cuticular membranes.The concepts and hypotheses put forward here mustbe tested and expanded t o be useful from a practicalpoint of view. First, the partitioning of technical poly-disperse surfactants between cuticular material andaqueous solutions should be investigated to assess thevalidity of the proposed QSPRsfo r real-world situations.Second, the sorption of surfactants in cuticular waxesand their effects on the mobility of active ingredientsin the wax barrier of the cuticle have t o be studied toevaluate the assumptions underlying some of the con-clusions drawn in this paper. Both subjects are cur-rently under investigation in the authors laboratories.

Riederer et al.

ACKNOWLEDGMENT

valuable help in improving this paper.We are indebted to three anonymous reviewers for

LITERATURE CITEDAbraham, M. H.; Mcgowan, J. C. The Use of CharacteristicVolumes to Measure Cavity Terms in Reversed PhaseLiquid Chromatography. Chromatographia 1987,23,243-246.Adamson, A. W. Physical Chem istry of Surfaces; Wiley: NewYork, 1982.Aveyard, R.; Binks, B. P.; Clark, S.; Fletcher, P. D. I. Effectsof Temperature on the Partitioning and Adsorption of C12Ebin Heptane-Water Mixtures. J . Chem . SOC., araday Tran s.1990,86,3111-3115.Bukovac, M. J.; Petracek, P. D. Characterizing Pesticide andSurfactant Penetration With Isolated Plant Cuticles. Pestic.Sc i . 1993, 37, 179-194.Celik, M. S.; Yoon, R. H. Adsorption of Poly(oxyethy1ene)-nonylphenol Homologues on a Low-Ash Coal. Langmui rChamel, A.; Gambonnet, B.; Coret, J. Effects of two Ethoxy-lated Nonylphenols on Sorption and Penetration of [l4C1-Isoproturon Through Isolated Plant Cuticles. Plant Physiol.Biochem. 1992,30, 713-721.Clunie, J. S.;ngram, B. T. Adsorption of Nonionic Surfactants.In Adsorption from Solution at the SolidlLiquid Interface;Parfitt, G . D., Rochester, C. H., Eds.; Academic Press:London, 1983.Furlong, D. N.; Aston, J. R. Adsorption of PolyoxyethylatedNonyl Phenols at SilicdAqueous Solution Interfaces. Col-loids Surfaces 1982, 4, 121-129.Gaskin, R. E.; Holloway, P. J. Some Physicochemical FactorsInfluencing Foliar Uptake Enhancement of Glyphosate-Mono(isopropy1ammonium) y Polyoxyethylene Surfactants.Pestic. Sci. 1992, 34, 195-206.Geyer, U.; Schonherr ,J. In Vitro Test for Effect of Surfactantsand Formulat ion on Permeabil ity of Plant Cuticles. InPesticide Formulations: Innovations and Developments;Cross, B., Scher, H. B., Eds.; American Chemical Society:Washington, 1989.Geyer, U.; Schonherr,J. The Effect of the Environment on thePermeability and Composition of Citrus Leaf Cuticles. I.Water Permeability of Isolated Cuticular Membranes. Plan-tu 1990,180,147-152.

1991, 7, 1770-1774.

Giles, C. H.; Smith, D.; Huitson, A. A General Trea tment andClassification of the Solute Adsorption Isotherm. Pa rt I.Theoretical. J . Colloid Interface Sci. 1974, 47, 755-765.Holloway, P. J.; Edgerton, B. M. Effects of Formulation WithDifferent Adjuvants on Foliar Uptake of Difenzoquat and2,4-D-Model Experiments with Wild Oat and Field Bean.Weed Res. 1992,32, 183-195.Kirkwood, R. C. Use and Mode of Action of Adjuvants forHerbicides-A Review of Some Cur rent Work. Pestic. Sci.1993,38, 93-102.Knoche, M.; Bukovac, M. J. Surfactants Influence Foliar

Absorption of Gibberellic Acid by Sour Cherry Leaves. J .Am. SOC . ort ic . Sc i . 1992, 117, 80-84.Knoche, M.; Bukovac, M. J. Studies on Octylphenoxy Surfac-tants: XI. Effect on NAA Diffusion through the IsolatedTomato Fruit Cuticular Membrane. Pestic. Sci. 1993, 38,211-217.Levitz, P.Aggregative Adsorption of Nonionic Surfactants ontoHydrophilic SoliWate r Interface-Relation With Bulk Mi-cellization. Langmui r 1991, 7, 1595-1608.Meguro,K.; Ueno, M.; Esumi, K. Micelle Formation in AqueousMedia. InNonionic Surfactants. Physical Chemistry; Schick,M. J., Ed.; Dekker: New York, 1987.Moroi, Y. Micelles; Plenum Press: New York, 1992.Nagarajan, R.; Ruckenstein, E. Critical Micelle Concentra-tion: A Transition Point for Micellar Size Distribution. J .

Colloid Interface Sci. 1977, 60, 221-231.Nagarajan, R.; Ruckenstein, E. Theory of Surfactant Self-Assembly-A Predictive Molecular Thermodynamic Ap-proach. Langmui r 1991, 7, 2934-2969.Puwada,S.; lankschtein, D. Molecular-Thermodynamic Ap-proach t o Predict Micellization, Phase Behavior and PhaseSeparation of Micellar Solutions. I. Application to NonionicSurfactants. J . Phys . Chem. 1990,94,3710-3724.Riederer, M.; Schonherr, J. Accumulation and Transport of(2,4-Dichlorophenoxy)Acetic cid in Plant Cuticles: I. Sorp-tion in the Cuticular Membrane and its Components.Ecotoxicol. Environ. Sa f. 1984, 8, 236-247.Riederer, M.; Schonherr, J. Thermodynamic Analysis of Non-electrolyte Sorption in Plant Cuticles: The Effects ofConcentration and Temperature on Sorption of 4-Nitrophe-nol. Planta 1986, 169, 69-80.Riederer, M.; Schonherr, J. Effects of Surfactants on WaterPermeability of Isolated Cuticles and on the Compositionof thei r Cuticular Waxes. Pestic. Sci. 1990, 29, 85-94.Riederer, M.; Schreiber, L. Waxes-the Transport Barr iers ofPlant Cuticles. In Waxes;Hamilton, R. J. ,Ed.; Oily Press:West Ferry, Scotland, 1995.Rosen, M. J. Surfactants and Interfacial Phenomena; Wiley:New York, 1989.Ruckenstein, E.; Nagarajan, R. Critical Micelle Concentration.A Transit ion Point for Micellar Size Distribution. J . Phys.Chem. 1976, 79, 2622-2626.Ruckenstein, E.; Nagarajan, R. Critical Micelle Concentrationand the Transition Point of Micellar Size Distribution. J .Phys. Chem. 1981,85,3010-3014.Schonherr,J. Effects of Monodisperse Alcohol Ethoxylates on

Mobility of 2,4-D in Isolated Plant Cuticles. Pestic. Sci.1993a, 38, 155-164.Schonherr, J. Effects of Alcohols, Glycols and MonodisperseEthoxylated Alcohols on Mobility of 2,4-D in Isolated PlantCuticles. Pestic. Sci . 1993b, 39, 213-223.Schonherr,J.;Bauer, H. Analysis of Effects of Surfactants onPermeabil ity of Plant Cuticles. In Adjuvan t s and Ag-richemicals; Foy, C . L., Ed.; CRC Press: Boca Raton, FL,1992; Vol. 2.Schonherr, J. ; Riederer, M. Plant Cuticles Sorb LipophilicCompounds dur ing Enzymatic Isolation. Plant Cell Envi-ron. 1986,9, 459-466.Schonherr, J.; Riederer, M.; Schreiber, L.; Bauer, H. FoliarUptake of Pesticides and its Activation by Adjuvants:Theories and Methods for Optimization. In Pesticide Chem-istry; Frehse, H., Ed.; VCH Verlag: Weinheim, 1991.


9/9

Sorption of Alcohol Ethoxylates in Plant CuticlesSchreiber, L.; Schonherr, J. Phase Transitions and Thermal

Expansion Coefficients of Plant Cuticles. Planta 1990,182,186-193.Shafer, W. E.; Bukovac, M. J.; Fader, R. G. Studies onOctylphenoxy Surfactants. IV.Their Sorption and Effectson NAA Partitioning into Pla nt Cuticles. In Adjuvants andAgrochemicals, Recent Development, Applicatio n, and Bi b-liography of Agroadjuvants ; Chow, P. N. P., Grant, C. A.,Hinshalwood, A. M., Simundsson, E., Eds.; CRC Press: BocaRaton, FL, 1989; Vol. 2.Stevens, P. J. G.; Gaskin, R. E.; Hong, S.0.;Zabkiewicz,J. A.Contributions of Stomatal Infiltration and Cuticular Pen-etration to Enhancements of Foliar Uptake by Surfactants.Pestic. Sci. 1991, 3 , 371-382.Stock, D.; Holloway,P. J. Possible Mechanisms for Surfactant-Induced Foliar Uptake of Agrochemicals. Pestic. Sci. 1993,

Stock, D.; Edgerton, B. M.; Gaskin, R. E.; Holloway, P. J.Surfactant-Enhanced Foliar Uptake of Some Organic Com-pounds: Interactions with two Model PolyoxyethyleneAliphatic Alcohols. Pestic. Sci. 1992, 4 , 233-242.

38 , 165-177.

J. Agric. Food Chem., Vol. 43, No. 4, 995 1075Tan, S.Y.; Crabtree, G. D. Effects of Nonionic Surfactants onCuticular Sorption and Penetration of 2,4-DichlorophenoxyAcetic Acid. Pestic. Sci. 1992, 5 , 299-303.Tanford, C. The Hydrophobic E fec t ; Wiley: New York, 1973.Yalkowsky, S. H.; Valvani, S. C. Solubility and Partitioning I:Solubility of Nonelectrolytes in Water. J . Pharm. Sc i . 1980,69, 912-922.Received fo r review August 15 , 1994. Accepted Jan uary 25,1995.@This study has been supported in part by a grant fromthe Forschungszentrum Julich GmbH. Par ts of the experi-mental work reported here have been performed at the Insti-tu t fur Botanik und Mikrobiologie, Technische UniversitatMiinchen, and the Fachbereich Biologie, UniversitatKaiserslautern.JF940469B@Abstract published inAdvanceAC S Abstracts, March1, 1995.

Documents

Mono Disperse Alcohol Ethoxylates