Adsorption of Sulfonamide Antimicrobial Agents to Clay Minerals

Adsorption of SulfonamideAntimicrobial Agents to ClayMineralsJ U A N G A O † A N D J O E L A . P E D E R S E N * , † , ‡

Department of Soil Science, and Environmental Chemistryand Technology Program, University of Wisconsin, Madison,Wisconsin 53706

Adsorption of three sulfonamide antimicrobials to clayminerals was investigated as a function of pH, ionic strength,and type of exchangeable cation. Sulfonamide antimicrobialadsorption exhibited pronounced pH dependenceconsistent with sorbate speciation and clay properties.Sulfonamide antimicrobials did not intercalate intomontmorillonite, and surface charge density influencedsorption by determining adsorption domain size. Adsorptionedge data were best fit to a model including terms forthe cationic and uncharged species. Adsorption of unchargedsulfamethazine to montmorillonite was relatively insensitiveto pH, ionic strength, and type of exchangeable cation,while that to kaolinite was highly sensitive to ionic strength.Adsorption of cationic sulfamethazine to montmorilloniteexceeded that of the neutral species by 1-2 orders ofmagnitude, but was unimportant for kaolinite at the pH valuesexamined. Cation exchange appeared to contribute tosorption of cationic sulfonamide species to montmorillonite.Anionic sulfamethazine adsorption was negligible. Thenature of the sulfonamide R group influenced the degreeof adsorption of cationic and neutral species. Our resultshighlight the importance of considering sulfonamidespeciation and clay surface charge density in predictingthe transport of these antimicrobials.

IntroductionSulfonamide antimicrobials comprise a class of synthetic,primarily bacteriostatic, sulfanilamide derivatives and finduse in human therapy, livestock production, and aquaculture.Treated individuals excrete a fraction of the administereddose as the unaltered parent compound or an acetylatedmetabolite (1, 2) susceptible to reactivation by bacterialdeacetylation (3). Sulfonamides enter the environmentthrough disposal of domestic and hospital waste and viarunoff and infiltration from confined animal feeding opera-tions and fields treated with animal manure. Althoughsulfonamide antimicrobials are susceptible to biodegradationby sewage sludge microorganisms (4, 5), hydraulic residencetimes in biological treatment processes are often insufficientfor their elimination from treated effluent (5). Sulfonamideantimicrobials ranked among the most frequently detectedpharmaceuticals in U.S. streams (6) and have been found ingroundwater (7), landfill leachate (8), and runoff from andsoils beneath effluent-irrigated lands (9, 10).

The primary concern with introducing antimicrobialagents into soil and water environments is the spread ofantibiotic resistance in response to increased selectivepressure, potentially leading to proliferation of resistantpathogens. Schmitt et al. (11) reported that exposure of soilmicrocosms to 7.3 mg‚kg-1 sulfachloropyridazine resultedin increased bacterial community tolerance to this sulfon-amide. Because sulfonamide antimicrobials may be presentin groundwater and surface water resources used as drinkingwater supplies, and can cause allergic reactions in sensitiveindividuals (12), human exposure to these compounds mayalso warrant concern.

Understanding sulfonamide antimicrobial sorption to soiland sediment particles is essential for assessing their potentialto leach into groundwater and to be transported in aquifersand surface waters. Because sulfonamides must enter bacte-rial cells to interfere with folate metabolism, sorption toenvironmental particles impacts the selective pressureexerted by these compounds. Previous studies of sulfonamidesorption focused on whole and size-fractionated soils, ratherthan specific well-defined sorbent phases (13-15). Sulfon-amides sorbed weakly to bulk soil, with greater sorption tofine silt and clay particle-size fractions (13). Thiele-Bruhn etal. (13) emphasized the importance of natural organic matterin sulfonamide sorption to soils. To date, clay mineralconstituents of the soil clay fraction have not been examinedas sorbents for sulfonamides, although smectites are oftenimportant sorbents for polar and ionizable organic com-pounds, including antimicrobial agents from other classes(16-22).

The objectives of this study were to investigate theadsorption of three sulfonamide antimicrobial agents tocommon phyllosilicate minerals and to derive sorptionparameters applicable over a range of solution conditions.Reference montmorillonite and kaolinite clays were selectedso that we could examine sulfonamide antimicrobial inter-action with these important soil minerals without interferencefrom organic matter coatings. We obtained clay adsorptionedges under a range of solution conditions by separatelyvarying proton activity, ionic strength (I), and type ofexchangeable cation.

Materials and MethodsChemicals. The chemicals used, suppliers, and purities aredescribed in the Supporting Information (SI). Tables 1 andS1 present structures and physicochemical properties of thesulfonamide antimicrobials studied. Sulfonamide antimi-crobials possess two ionizable functional groups consideredrelevant to the environmental pH range: the anilinic amineand the amide moieties. The cationic species (SA+) dominatesat low pH values; the neutral form (SA0) is the principal speciesat pH values between pKa,1 and pKa,2; and the anionic species(SA-) is the main form at higher pH values (Figure S1). Aminor zwitterionic species (SA() is in tautomeric equilibriumwith SA0 (contribution to overall speciation <0.2%) (23).Tautomeric constants (Kt ) [SA0]/[SA(]) for sulfamethazine(SMZ) and sulfapyridine (SPY) are 102.80 and 103.12 (23).

Sorbents. Reference montmorillonite (SWy-2 and SAz-1)and kaolinite (KGa-1b) clays were obtained from the ClayMinerals Society Source Clays Repository (West Lafayette,IN). Clay preparation, homoionization, permanent chargereduction, and characterization are described in the SI. Table2 presents characteristics of the clay minerals.

Batch Sorption Experiments. We conducted sorptionedge experiments to investigate the pH-dependent adsorp-

* Corresponding author phone: (608) 263-4971; fax: (608) 265-2595; e-mail: [email protected].

† Department of Soil Science.‡ Environmental Chemistry and Technology Program.

Environ. Sci. Technol. 2005, 39, 9509-9516

10.1021/es050644c CCC: $30.25 2005 American Chemical Society VOL. 39, NO. 24, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 9509Published on Web 11/10/2005

tion of SMZ, sulfamethoxazole (SMX), and SPY to nearhomoionic clays. Isotherm experiments were conducted toexamine SMZ adsorption as a function of sorbate concen-tration. We refer to the clays in these experiments as nearhomoionic because at pH values less than∼6, protons occupysome cation exchange sites. Experimental details are providedin the SI. Briefly, SMZ (3H-labeled + unlabeled), SPY(unlabeled), or SMX (unlabeled) were equilibrated with nearhomoionic clays at room temperature for 2.5 h in the dark.Adsorption edge experiments were conducted in 10 mMNaHCO3/Na2CO3; adsorption isotherm experiments wereconducted in 10 mM acetate buffer. Phase separation wasachieved by 10-min centrifugation at 9200g. Supernatantswere analyzed by liquid scintillation counting (for SMZ) orhigh-performance liquid chromatography with UV detection(for SMX and SPY). Desorption of SMZ from clays with a pH10 background solution allowed closing of mass balances(97 ( 2%, n ) 18). We therefore calculated the amount sorbedby difference after accounting for losses in the blanks (seeSI). The intercalation of sulfonamide antimicrobials intoSWy-2 was investigated by X-ray diffractometry (see SI).

Results and DiscussionX-ray Diffraction Analysis. Neutral and cationic sulfonamidespecies did not intercalate into SWy-2, but interactedprimarily with external surfaces (see SI). We cannot excludethat part of the molecule (e.g., the anilinium group) was ableto enter the interlayer space at the edges of smectite particles.That SMZ did not intercalate into SWy-2 is not surprisinggiven the pronounced nonplanarity of the molecule (TableS1).

Adsorption Edge Experiments. Sulfonamide antimicro-bial adsorption to phyllosilicate minerals exhibited pro-nounced pH dependence consistent with sorbate speciationand clay properties. We examined adsorption between pH3.5 and 9.3, a range encompassing proton activities found inmost natural environments. Proton activities higher than10-3.5 M were not investigated because smectite claysspontaneously decompose at low pH values (30). Figure 1displays SMZ adsorption edges for sodium-saturated mont-morillonite and kaolinite clays. Sulfamethazine adsorptionto montmorillonite clays (SWy-2 and SAz-1) exhibited onlyslight pH-dependence between pH ∼5 and 7. Decreases inpH below ∼5 resulted in a marked increase in SMZ adsorptionto Na-SWy-2. Sulfamethazine adsorption to Na-SWy-2exceeded that to Na-SAz-1 below pH ∼8. As pH increasedbeyond this point, adsorption approached zero for bothsmectites.

In contrast to the smectites, SMZ adsorption to thekaolinite clay displayed no pH-dependence between pH 4.3and 7, proton activities at which the neutral sulfamethazinespecies (SMZ0) dominated solution speciation (Figure S1).The lack of pH dependence for SMZ0 adsorption suggeststhat the variable charge edge sites of kaolinite did notcontribute substantially to overall sorption. Increases in pHabove 6.5 resulted in a decrease in adsorption to a minimumvalue at pH ∼9. Between pH 5.5 and 7, experimentaladsorption coefficients for Na-KGa-1b and Na-SWy-2 didnot differ significantly on a sorbent mass basis (p > 0.05).Here, and throughout the paper we employ Student’s t test

TABLE 1. Selected Physicochemical Properties of the Sulfonamide Antimicrobials Studieda

a Abbreviations: Kow, n-octanol-water partition coefficient; Mi, molar mass of the neutral species; MSA, molecular surface area. b Ref 24. c Ref25. d Ref 26. e Ref 27. f Ref 28. g Ref 29.

TABLE 2. Clay Mineral Characteristics

surface area

claytotala

(m2‚g-1)externalb

(m2‚g-1)CECc

(µmolc‚g-1)

Homoionic ClaysNa-SWy-2 749.5 ( 18.9 31.8 760 ( 30Na-SAz-1 997.5 ( 15.5 93.1 1070 ( 10Na-KGa-1b 84.6 ( 7.5 11.7 15 ( 1

Heat-Treated and Charge-Reduced ClaysNa-SAz-1 d d 1020 ( 600.7Na:0.3Li-SAz-1 1158.8 ( 87.3 d 790 ( 200.3Na:0.7Li-SAz-1 d d 210 ( 10

a Surface area accessible to ethylene glycol monoethyl ether; mean( standard deviation. b Surface area accessible to N2, Brunauer-Emmett-Teller (BET) method. c Cation exchange capacity; mean (standard deviation. d Not determined.

FIGURE 1. Sulfamethazine adsorption to near homoionic Na-SWy-2, Na-SAz-1, and Na-KGa-1b as a function of pH (I ) 10 mM).Initial SMZ concentration, C0 ) 3.5 µM. Symbols represent meansof triplicate measurements; error bars indicate one standarddeviation. Trendlines represent best fits to eq 2.

9510 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 24, 2005

to determine the significance of differences between treat-ments.

To examine the contributions of individual sulfonamidespecies to overall sorption, we followed the approach ofFigueroa et al. (22) to calculate species-specific equilibriumadsorption coefficients. The overall adsorption coefficientat a given pH value can be represented as the sum of thecontributions of the individual species (31):

where Kd (L‚kg-1) is the overall equilibrium adsorptioncoefficient and is defined as the ratio of the sorbed anddissolved concentrations; K d

SA+, K d

SA0, K d

SA(, and K d

SA-are the

adsorption coefficients of the cationic, uncharged, zwitter-ionic, and anionic species, respectively; and RSA+, RSA0, RSA(,and RSA- represent the mass fractions of these species inbulk solution. The model formulation did not explicitlyconsider possible altered sulfonamide speciation at the claysurface. Experimental data were fit to eq 1 using the GeneralLinear Model (SYSTAT 10.2, SYSTAT Software, Point Rich-mond, CA). We evaluated each sorption term to determinewhether model fits were improved by its inclusion. Exceptin a minority of cases (indicated below), inclusion of theanionic species did not improve model fits to the experi-mental data (p > 0.05). Desorption of sulfonamide antimi-crobials from smectite surfaces at pH 10 provided additionalevidence that anionic species generally did not participatein sorption interactions. In no case were model fits improvedby inclusion of a term for the zwitterion (p > 0.05). Therefore,in most cases adsorption edge data were modeled using termsfor only cationic and uncharged species:

For SMZ adsorption to Na-SWy-2 and Na-SAz-1, theempirical model was best fit with K d

SA+values of 2620 ( 250

L‚kg-1 and 771 ( 8 L‚kg-1 and K dSA0

values of 12.0 ( 0.7 L‚kg-1

and 2.3 ( 0.6 L‚kg-1 (Table 3). For Na-KGa-1b, the best fitto the empirical model was obtained using a K d

SA0value of

14.8 ( 0.6 L‚kg-1; inclusion of K dSA+

did not improve themodel fit (p > 0.05).

The empirical model results allowed evaluation of therelative contributions of SMZ species to overall adsorption.Between pH 2.3 and 7.4, SMZ0 was the main solution species.Nevertheless, the model fit to the experimental data indicatedthat cationic sulfamethazine (SMZ+) dominated adsorptionto Na-SWy-2 up to pH 4.6 (a proton activity at which SMZ+

comprised only 0.45% of solution species). Between pH 5and 7, SMZ0 dominated solution speciation (99-72%) andcontributed the most to overall adsorption based on themodel-derived sorption parameters (72-99%). Over this pHrange adsorption to Na-SWy-2 exhibited relatively little pH

dependence. As pH increased above 7.4, the anionic species(SMZ-) assumed dominance in solution, and the adsorptioncoefficient approached a minimum value, consistent withincreasing electrostatic repulsion of SMZ- by the negativelycharged smectite surface as proton activity declined. Sorptionparameters obtained in this study are strictly valid for onlythe buffer system employed. However, distribution coef-ficients from experiments using unbuffered, bicarbonate-buffered, and acetate-buffered solutions did not differsignificantly (p > 0.05; data not shown).

Consideration of the nature of the clay surfaces and SMZsolution speciation allows rationalization of the observedtrends in adsorption with pH. Due to isomorphic substitutionof primarily Mg2+ for Al3+ in the dioctahedral layer (32) andvariable charge edge sites, the smectites carry net negativecharge over the pH range examined (33), are cation ex-changers, and exhibit negative adsorption of anions (34).Electrostatic forces favor SMZ+ association with the siloxanesurface and disfavor adsorption of the anion. At higher protonactivities, SMZ adsorption to Na-SWy-2 did not parallelchanges in bulk solution speciation. Instead, SMZ+ domi-nated adsorption ∼2.5 pH units above that expected frombulk solution speciation. This effect may be attributed toenhanced dissociation of water solvating the exchange ionson the smectite surface (35) and the increasing replacementof exchangeable Na+ by protons as pH decreases below ∼6.Kaolinite carries a net negative charge at pH values above4.7 (36). However, due to the low charge density and higherpHpzc of kaolinite, SMZ+ adsorption was unimportant at thepH values examined. Dissociation of water at kaolinitesurfaces is not significantly enhanced over that of bulk waterdue to the very low density of exchange sites.

Because sulfonamides appeared to interact with externalsmectite surfaces, we normalized modeled species-specificadsorption coefficients by the clay BET surface areas (Table3). On a surface area basis, SMZ0 adsorption to Na-SWy-2exceeded that to Na-SAz-1 by a factor of 15.2 ( 1.4. Thesurface-area-normalized K d

SMZ+value for Na-SWy-2 was

larger than that for Na-SAz-1 by a factor of 9.9 ( 1.0. Overthe pH range at which SMZ0 dominated solution speciation,surface-area-normalized experimental Kd values were higherfor Na-KGa-1b than for the montmorillonites (p < 0.05).

Effect of Surface Charge Density. Montmorillonite surfacecharge density appeared to influence SMZ sorption. On amass basis, SAz-1 surface charge as measured by CEC wasconsiderably higher than that of SWy-2 (Table 2). Sulfa-methazine sorption to the low-charge SWy-2 substantiallyexceeded that to the high-charge SAz-1 at proton activitiesat which SMZ0 dominated solution speciation, especiallybelow pH ∼5 (Figure 1). On the basis of the fit of theexperimental data to eq 2, SMZ0 and SMZ+ adsorption toNa-SWy-2 was larger than that to Na-SAz-1 by factors of5.2 ( 1.4 and 3.4 ( 0.3 on a sorbent mass basis (Table 3).Normalization of K d

SMZ+by the CEC of the external clay

TABLE 3. Distribution Coefficients for SMZ Adsorption to Near Homoionic Claysa

sorbentK d

SMZ+

(L‚kg-1)K d

SMZ0

(L‚kg-1) r2K d

SMZ+/Aextsurf

(L‚m-2 × 10-3)K d

SMZ0

/Aextsurf(L‚m-2 × 10-3)

K dSMZ+

/CECextsurf(L‚molc

-1 × 103)

Li-SWy-2 3240 ( 460 15.4 ( 0.8 0.97 109 ( 16 0.52 ( 0.03 108 ( 16Na-SWy-2 2620 ( 250 12.0 ( 0.7 0.96 82.4 ( 7.9 0.38 ( 0.02 81 ( 9Mg-SWy-2 b 17.1 ( 0.8 0.95 b 0.66 ( 0.03 bK-SWy-2 1070 ( 44 16.2 ( 0.8 0.99 44.0 ( 1.8 0.67 ( 0.03 44 ( 3Ca-SWy-2 728 ( 88 22.2 ( 0.6 0.98 26.7 ( 3.2 0.81 ( 0.02 26 ( 3Na-SAz-1 771 ( 8 2.3 ( 0.6 >0.99 8.3 ( 0.1 0.02 ( 0.01 7.7 ( 0.2Na-KGa-1b b 14.8 ( 0.6 0.97 b 1.26 ( 0.05 ba Ionic strength ) 10 mM; values represent best fits of experimental data to eq 2 (mean ( standard error). Abbreviations: Aextsurf, external surface

area (cf. Table 2); CECextsurf, cation exchange capacity of external surfaces (see Table S2). b Model fit not improved by including a term for the cation.

Kd ) K dSA+

‚RSA+ + K dSA0

‚RSA0 + K dSA(

‚RSA( + K dSA-

‚RSA- (1)

Kd ) K dSA+

‚RSA+ + K dSA0

‚RSA0 (2)

VOL. 39, NO. 24, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 9511

surfaces indicated that SMZ+ adsorption to Na-SWy-2surpassed that to Na-SAz-1 by a factor of 11 ( 1. Usingmeasured CEC values (Table 2), a surface area per unit cell(excluding edge surfaces) of 0.93 nm2/unit cell, and the layersilicate molecular mass of 746.21 g‚mol-1 (37), and assumingregular spacing of negative charges on siloxane surfaces,average distances between adjacent sodium ions on externalsurfaces were calculated to be 0.90 nm for Na-SWy-2 and0.67 nm for Na-SAz-1. The average adsorption domain areaof the low-charge SWy-2 exceeded that of the high-chargeSAz-1 and apparently influenced the degree of SMZ sorption.This argument is not free from ambiguity because isomorphicsubstitution by Al3+ for Si4+ in the tetrahedral layer of SWy-2but not in SAz-1 (32) and more extensive ordering of chargedeficit sites in the octahedral sheet of SWy-2 (38) may havecontributed to differences in SMZ adsorption.

To eliminate the ambiguity associated with comparingsmectites differing in tetrahedral charge and ordering ofcharge deficit sites and to further investigate the influenceof surface charge density on SMZ adsorption, we conductedsorption edge experiments with charge-reduced SAz-1 usingheat-treated Na-SAz-1 as a control. In charge-reduced clays,heat treatment caused Li+ to migrate into the smectiteoctahedral sheet and compensate part of the permanentnegative charge (39). This treatment permanently reducessurface charge density, spreads the remaining excess negativecharge over a larger number of surface oxygen atoms, andincreases the size of adsorption domains between exchange-able cations (35, 40). Model fits to adsorption edge dataindicated that charge reduction significantly enhancedadsorption of SMZ0 and SMZ+ to SAz-1 (p < 0.05) comparedto the heat-treated control sample (Table 4). Mixing charge-reduced SAz-1 at a 0.7:0.3 Na-to-Li ratio (22.7% reduction incharge density) increased K d

SMZ0by a factor of 4.3 ( 0.4 and

K dSMZ+

by a factor of 1.9 ( 0.5; mixing at a 0.3:0.7 Na-to-Liratio (79.5% reduction in charge density) resulted in K d

SMZ0

and K dSMZ+

increasing by factors of 4.6 ( 0.4 and 3.8 ( 0.7.These results demonstrate the sensitivity of SMZ adsorp-

tion to reduction in clay surface charge density; the size ofavailable adsorption domains on the siloxane surface (i.e.,areas unoccupied by exchangeable cations) was apparentlyimportant for sulfonamide sorption. On the basis of themeasured CEC values for the charge-reduced clays (Table 2)and the assumptions used above, the average spacingbetween adjacent Na+ atoms on external surfaces of Na-SAz-1 and charge-reduced clays were 0.70 nm for heat-treatedNa-SAz-1, 0.87 nm for 0.7Na:0.3Li-SAz-1, and 2.21 nm for0.3Na:0.7Li-SAz-1. Reduction in the number of exchangeablecations and concomitant increase in sorption domain sizeenhanced SMZ adsorption. This phenomenon has beenobserved for other polar sorbates (40, 41). The ability of SMZ+

to displace exchangeable cations (see below) appears to haveresulted in the availability of larger adsorption domains,enhancing adsorption of the cationic species over that ofSMZ0. Charge reduction also increased the overall Lewis base

softness of the siloxane surface (35), increasing the ease withwhich Na+ (a hard Lewis acid) can be replaced and presum-ably enhancing the stability of complexes with the SMZanilinium cation (a soft Lewis acid).

Ionic Strength Effects. Solution ionic strength influencesthe double layer thickness of clays and the aqueous activityof sulfonamide species. We probed ionic strength effects onSMZ adsorption to Na-SWy-2 using three salt concentrations(10 mM NaHCO3, 10 mM NaHCO3 + 100 mM NaCl, and 10mM NaHCO3 + 300 mM NaCl) (Figure 2). Species-specificKd values were calculated at each ionic strength by fittingexperimental data to eq 2 (Table 5) applying the Daviesequation (37) to correct sulfonamide speciation for ionicstrength in solutions with I ) 110 mM and 310 mM. ForNa-SWy-2, little change in sorption was apparent betweenpH ∼5.5 and ∼7. Below pH 5.5, experimental Kd values weresignificantly lower at ionic strength of 310 mM than thoseat 10 mM (p < 0.05); significant differences in Kd valuesbetween ionic strengths of 10 mM and 110 mM becameapparent at pH values below ∼5 (p < 0.05). The largedecreases in K d

SMZ+observed as ionic strength increased

from 10 to 310 mM suggests that cation exchange contributesto SMZ+ interaction with the smectite surface. At pH values>8.3, no SMZ sorption to Na-SWy-2 was apparent at 10 mMionic strength, while some sorption occurred at higher ionicstrength (p < 0.05). The adsorption coefficient of SMZ-

increased with ionic strength (p < 0.05) as expected fromdouble layer compression and reduction of anion repulsion.For the higher ionic strength values, model fits to theexperimental data were improved by including a term forSMZ- sorption (p < 0.05):

The K dSMZ+

values obtained from the empirical model fitsare valid for only the ionic strength values at which theywere determined. To facilitate extension of these results to

TABLE 4. Distribution Coefficients for SMZ Adsorption toCharge-Reduced Smectitea

sorbentK d

SMZ+

(L‚kg-1)K d

SMZ0

(L‚kg-1) r2

Na-SAz-1 771 ( 8 2.3 ( 0.6 >0.99heat-treated Na-SAz-1 1120 ( 190 6.8 ( 0.5 0.93

charge-reduced SAz-10.3 Na: 0.7 Li-SAz-1 2110 ( 380 29.4 ( 1.7 0.970.7 Na: 0.3 Li-SAz-1 4270 ( 300 31.4 ( 1.1 >0.99

a Values represent best fits of experimental data to eq 2 (mean (standard error).

FIGURE 2. Effect of ionic strength on SMZ adsorption to Na-SWy-2. Initial SMZ concentration, C0 ) 3.5 µM. Solutions were 10 mMNaHCO3; the desired (nominal) ionic strength was achieved byaddition of NaCl prior to pH adjustment. Symbols represent meansof triplicate measurements; error bars indicate one standarddeviation. Trendlines represent best fits to eq 2 or 3.

TABLE 5. Effect of Ionic Strength on SMZ Adsorption toNa-SWy-2a

I (mM)K d

SMZ+

(L‚kg-1)K d

SMZ0

(L‚kg-1)K d

SMZ-

(L‚kg-1) r2

10 2620 ( 250 12.0 ( 0.7 0.96110 1380 ( 150 13.2 ( 0.2 7.4 ( 0.3 0.99310 750 ( 64 10.3 ( 0.3 8.6 ( 0.6 0.96a Values represent best fits of experimental data to eq 2 or 3 (mean

( standard error).

Kd ) K dSA+

‚RSA+ + K dSA0

‚RSA0 + K dSA-

‚RSA- (3)


other ionic strength conditions, we calculated a stoichio-metric selectivity coefficient for SMZ+ in the presence ofcompeting Na+ (Ks,SMZ:Na) using the following expression (22,42):

where Ks,SMZ:Na carries units of equiv/equiv. Linear regressionof our data using this expression yielded a slope of -0.35 anda Ks,SMZ:Na value of 190 equiv/equiv (r2 ) 0.96). We note thatour slope deviated from that expected (viz. - 1). The changein K d

SMZ+observed as ionic strength increased was less than

that predicted by eq 4.Ionic strength affected SMZ adsorption to Na-KGa-1b in

a different manner (Figure S2). At I ) 110 mM, distributioncoefficients became independent of pH and did not differsignificantly from zero. We were therefore unable to deter-mine species-specific Kd values at I ) 110 mM for Na-KGa-1b.

Effect of Exchangeable Cation. To examine the influenceof the type of exchangeable cation on SMZ association withmontmorillonite clays, we conducted pH adsorption edgeexperiments using SWy-2 initially homoionic in Li+, Na+, K+,Ca2+ and Mg2+. The trend in Kd with pH discussed aboveheld regardless of the exchangeable cation (Figure 3). Thetype of exchangeable cation exerted relatively little influenceon experimental Kd values over the range of pH valuesexamined. Between pH 5.5 and 8, Kd values for Ca-SWy-2were somewhat larger than those for SWy-2 saturated withother exchangeable cations (p < 0.05). The fit of eq 2 to theexperimental data indicated that the type of exchangeablecation had the largest effect on K d

SMZ+with SMZ+ adsorption

to initially homoionic SWy-2 decreasing in the followingorder: Li+ ≈ Na+ > K+ > Ca2+ (Table 3). We were unable torank Mg2+ because the fit of the experimental data to eq 2was not improved by inclusion of a term for SMZ+ (p > 0.05).For the monovalent metal cations, K d

SMZ+decreased ap-

proximately as the stability of their inner-sphere complexeswith the siloxane ditrigonal cavity increased (35) consistentwith cation exchange contributing to SMZ+ interaction withthe smectite surface. Based on the model fit to the experi-mental data, the type of exchangeable cation generally exertedlittle influence on K d

SMZ0(Table 3). Calcium and sodium

represented exceptions: relative to the other exchangeablecations examined, SMZ0 adsorption was slightly lower forNa+- and higher for Ca2+-saturated SWy-2 (p < 0.05).

Implications for Mechanisms of SMZ Adsorption to ClayMinerals. Between pH 5 and 7, SMZ0 dominated adsorptionto smectite surfaces. Ionic strength and the nature of theexchangeable cation exerted little influence on SMZ0 sorption,although adsorption to Ca-SWy-2 exceeded that to Na- and

Li-SWy-2. The size of available adsorption domains on thesiloxane surface was an important determinant of SMZ0

sorption. Formation of π-complexes between basal planeoxygens and either the aniline or 4,6-dimethyl-pyrimidinerings of SMZ does not appear important because thetetrahedral -SO2- group prevents close, coplanar orientationof either aromatic moiety with the siloxane surface. Possiblemechanisms of SMZ0 interaction with the smectite surfaceinclude water and cation bridging. For the exchangeablecations examined in this study, aromatic amines such asaniline (pKa ) 4.6) interact with smectite surfaces primarilyvia water bridging (43). Because the anilinic amines ofsulfonamide antimicrobials (pKa,1 e 2.3) are weaker basesthan aniline, water-bridging rather than surface protonationrepresents a possible mechanism of interaction with mont-morillonite between pH 5 and 7. Consistent with a waterbridging mechanism via the aniline group, Ca2+ and Mg2+

saturation of the smectite enhanced SMZ0 adsorption oversaturation with Na+, and adsorption to Ca-SWy-2 exceededthat to K-SWy-2 (43). Alternatively, SMZ0 could complexexchangeable cations through a pyrimidine N and/or the-SO2- group. Pyrimidine forms inner and outer spherecomplexes with exchangeable metal cations on montmo-rillonite surfaces (44). Bidentate coordination by a pyrimidineN and the -SO2- group has been proposed for the solutioncomplexation of divalent transition metal cations by SMZand sulfadiamidine (45, 46). Further evidence from spec-troscopic studies is needed to distinguish among thesehypothesized mechanisms of interaction with smectitesurfaces.

Cationic SMZ dominated adsorption to smectite surfacesat bulk solution pH values ∼1.5-2.5 units higher thanexpected based on pKa,1 depending on the nature of theexchangeable cation. This was apparently due to increasedoccupation of exchange sites by protons as solution pHdecreased below ∼6 and enhanced dissociation of watersolvating the exchange ion (35), leading to protonation ofthe aromatic amine at the smectite surface. Sulfamethazineadsorption to montmorillonites at low pH may have beenfurther enhanced by protonation of the pyrimidine R group.The dissociation constant of the heterocyclic moiety appearsto be <1.9 (47) and to our knowledge, has not beendetermined experimentally. Thus, adsorption of a SMZdication may have contributed to the high Kd values observedat low pH. As noted above, the effects of ionic strength andthe nature of the exchangeable cation were consistent witha contribution of a cation exchange mechanism to SMZ+

adsorption to smectites. Adsorption domain size also influ-enced the degree of SMZ+ sorption to smectites.

The data on SMZ adsorption to kaolinite do not allow usto confidently posit a mechanism of interaction. However,a number of observations can be made. Kaolinite particlespresent three types of surfaces to solution: a siloxane surfacewith a small amount of permanent negative charge, a gibbsiticsurface exposing aluminol groups, and edge sites displayingpH-dependent charge due to the protonation/deprotonationof silanol and aluminol groups. Only the neutral SMZ speciesappeared to interact with kaolinite. That SMZ+ did notcontribute to adsorption at the pH values examined isconsistent with the low permanent negative charge of KGa-1b, the dissociation constants of ionizable kaolinite surfacefunctional groups, and the properties of water associatedwith kaolinite surfaces. Sulfamethazine adsorption to kao-linite was nearly invariant over a broad range of protonactivities (pH 4.3-7), suggesting little, if any, interaction ofSMZ0 with surfaces exhibiting variable charge over this pHrange (i.e., edge sites). Surface-area-normalized Kd valuesfor Na-KGa-1b exceeded those for the montmorillonites overthe pH range where SMZ0 dominated solution speciation (p< 0.05), consistent with the notion that lower permanent

FIGURE 3. Adsorption edge for SMZ sorption to SWy-2 initiallyhomoionic in Li+, Na+, K+, Mg2+, and Ca2+ (I ) 10 mM). Symbolsrepresent means of triplicate measurements; error bars indicateone standard deviation. Trendlines represent best fits to eq 2.

log(K dSMZ+

/CECextsurf) ) -log[Na+] + log Ks,SMZ:Na (4)


negative charge density on kaolinite siloxane surfacesprovided larger domains for SMZ0 adsorption. We find itinteresting that increased ionic strength resulted in a dramaticdecrease in sulfamethazine sorption, a result that contrastswith the limited influence of ionic strength on the sorptionof SMZ0 to montmorillonite siloxane surfaces. From ourexperimental data, we cannot exclude SMZ0 association withthe gibbsitic surface of kaolinite. The manner in whichincreased NaCl concentration prevented interaction withkaolinite is not obvious. Clearly, further experimentation willbe required to propose a compelling mechanism of sul-fonamide interaction with kaolinite.

Influence of Sulfonamide Structure on Adsorption. Toinvestigate the influence of the sulfonamide antimicrobial Rgroup on adsorption to smectite surfaces, we conducted pHadsorption edge experiments with three sulfa drugs (viz. SMZ,SMX, and SPY) and fitted the experimental data to eq 2 toobtain species-specific Kd values (Table 6). The sulfanilamideportion of these molecules is identical; the nature of theheterocyclic R group confers differences in physicochemicalproperties (Table 1). All three sulfonamides exhibited quali-tatively similar trends in pH-dependent adsorption to Na-SWy-2 (Figure 4). The pKa values of SMX are much lowerthan those of SMZ and SPY, resulting in little sorption overmost of the pH range examined. On the basis of fits of eq 2to the experimental data, K d

SA+. K d

SA0for all three sulfon-

amides (Table 6). The K dSA0

value was largest for SMZ; thosefor SPY and SMX were 3- and 4-fold smaller. The Kd valuesfor SMZ+ and SPY+ adsorption to Na-SWy-2 were statisticallyindistinguishable (p > 0.05) due to the high variance in theSPY data; that of SMX+ was more than an order of magnitudesmaller than those of SMZ+ and SPY+ (p < 0.05).

No clear relationship was apparent between modeledspecies-specific adsorption coefficients and compoundhydrophobicity (as assessed by Kow), but K d

SA0values ap-

peared to increase with molecular surface area (Table 1).Mechanisms of SPY and SMX interaction with the smectitesurface are likely similar to those of SMZ. As was the case

with SMZ, the -SO2- groups of these sulfonamides restrictinteraction of the π-electron systems of aromatic moietieswith the siloxane surface. Water and/or cation bridgingrepresent likely mechanisms of interaction with montmo-rillonite. The trend in the magnitude of K d

SA0paralleled that

reported for stability constants of these compounds withdivalent transition metal cations (45). The existence of adicationic species is possible for both SMX and SPY. We wereunable to find a study reporting a pKa value for the SPYpyridinium group, but work on structurally related meth-anesulfonamidopyridines (48) suggests that this moietypossesses a dissociation constant below that of the aniliniumcation of sulfapyridine. As was the case for SMZ, adsorptionof a SPY dication may have contributed to the large K d

SA+

value for this compound. For SMX, protonation of theisoxazole group likely occurs only at very low pH, below thedissociation constant for the anilinium group (pKa,1 ) 1.8).The relatively low K d

SMX+value may therefore reflect negli-

gible contribution by the SMX dication at the proton activitiesexamined.

Several previous investigators examined SMZ and SPYsorption to whole soils and soil clay fractions. Sulfamethazinesorption to homoionic SWy-2 and kaolinite exceeded thatreported by Thiele-Bruhn et al. (13) to whole soils and soilclay fractions (Kd ) 2.4 and 2.7 L‚kg-1) and by Langhammer(49) to whole soils (Kd ) 0.88-3.47 L‚kg-1); the Kd values forSMZ reported by these authors were similar to those weobtained for Na-SAz-1. Sulfapyridine sorption coefficientsfor whole soils and soil clay fractions (13, 50) were of thesame order as those for Na-SWy-2.

Adsorption Isotherms. We examined SMZ adsorption tonear homoionic SWy-2 as a function of sorbate concentrationat pH 5.2 ( 0.1, a proton activity at which the neutral formcomprised >99% of solution species and was the maincontributor to adsorption (76-83%). Initial solute concen-trations spanned 6 orders of magnitude (3.61 × 10-10 to 3.61× 10-4 M). Isotherms for SMZ adsorption to SWy-2 nearhomoionic in Na+, K+, Ca2+, and Mg2+ (Figure S3) were fitwith the van Bemmelen-Freundlich equation:

where Cs (mol‚kg-1) and Cw (mol‚L-1) are the sorbed anddissolved concentrations; Kf is the Freundlich constant, ameasure of sorption capacity at a specific aqueous concen-tration; and the Freundlich exponent n provides a measureof the heterogeneity of adsorption site energies (51). TheFreundlich coefficient is related to Kd by the followingequation:

When n ) 1, the isotherm is linear, sorption free energiesat all sorbate concentrations are constant (51), and Kd ) Kf.Table 7 summarizes Freundlich parameters for SMZ adsorp-tion to initially homoionic SWy-2. As expected from pHsorption edge experiments, the type of exchangeable cationexerted little effect on SMZ adsorption. Adsorption isotherms

TABLE 6. Effect of the Sulfonamide R Group on Adsorption toNa-SWy-2a

sorbateK d

SA+

(L‚kg-1)K d

SA0

(L‚kg-1) r2

SMZ 2620 ( 250 12.0 ( 0.7 0.96SMX 160 ( 20 2.9 ( 0.1 0.99SPY 3350 ( 1420 4.0 ( 0.3 0.95

a Values represent best fits of experimental data to eq 2 (mean (standard error).

FIGURE 4. Adsorption edges of three sulfonamide antimicrobials(initial concentration, C0 ) 3.5 µM) to Na-SWy-2 (I ) 10 mM).Symbols represent means of triplicate measurements; error barsindicate one standard deviation. Trendlines represent best fits toeq 2.

TABLE 7. Freundlich Parameters for SMZ Adsorption to NearHomoionic Smectite and Kaolinitea

sorbent pH n Kf r2

Na-SWy-2 5.05 1.01 ( 0.00 13.7 ( 1.0 >0.99Mg-SWy-2 5.05 0.96 ( 0.01 15.6 ( 1.1 >0.99K-SWy-2 5.15 0.97 ( 0.01 10.2 ( 1.1 >0.99Ca-SWy-2 5.10 1.01 ( 0.01 10.7 ( 1.0 >0.99Na-KGa-1b 5.02 0.85 ( 0.01 11.1 ( 1.0 0.98

a Values reported as mean ( standard error.

logCs ) n‚logCw + logKf (5)

Kd ) Kf‚C wn-1 (6)


for SWy-2 were linear, or nearly so, over the concentrationrange examined (i.e., n ) 0.96-1.01), indicating that SMZdid not saturate available adsorption sites. Solubility limita-tions prevented investigation of higher SMZ concentrations.

The SMZ adsorption isotherm for Na-KGa-1b departedfrom linearity at high sorbate concentrations (Figure S4) witha Freundlich exponent of 0.85 (Table 7). The isotherm wasnot well fit by a Langmuir model, and the relatively lowaqueous solubility of SMZ prevented extension of theisotherm to higher concentrations to examine approach tosorption capacity. The highest solution concentration ex-amined resulted in <5% coverage of the kaolinite surfaceassuming monolayer coverage.

The linearity of the SWy-2 isotherms contrasts with thepronounced nonlinearity observed by some investigatorsexamining sulfonamide sorption to whole soils and particle-size fractions. Thiele-Bruhn et al. (13) obtained markedlynonlinear isotherms for SPY sorption to clay fractions of loessChernozem soils (n ) 0.81-0.84) and attributed this tospecific interaction with organic matter functional groups.These Freundlich coefficients are similar to those we obtainedfor SMZ adsorption to kaolinite. SPY sorption to whole soilsexhibited a range of values for the Freundlich exponent (0.54-1.21), attributed by the authors to soil organic materproperties (13, 15, 50). In contrast to this, Boxall et al. (15)obtained Freundlich coefficients of 0.97 and 0.91 for thesorption of sulfachloropyridazine to clay loam and sandyloam. The soil mineralogy was not reported.

Environmental Implications. Accurate prediction ofsulfonamide antimicrobial mobility and bioavailability in soilsand subsurface environments requires a quantitative de-scription of sorption as a function of aqueous phasecomposition. Our adsorption edge experiments and modelingof species-specific distribution coefficients highlight theimportance of considering sulfonamide speciation in envi-ronmental fate modeling and enable estimation of adsorptionto clay minerals over a range of pH and ionic strengthconditions. Sulfonamide sorption to clay minerals is relativelyinsensitive to pH over the range of proton activities at whichthe neutral species dominates solution speciation. At higherand lower pH values, proton activity has a pronounced affecton sorption as the anionic and cationic species assume greaterimportance to overall sorption. Extrapolation of our resultsto real soils and sediments must be done with caution, asclays in such systems are often coated with organic matterand metal oxides. Our results demonstrate the importanceof surface charge density for sulfonamide sorption to clayminerals. The surface charge densities of soil and sedimentclays likely differ from those of the reference clays weemployed. Sorption to soil and sediment clays is expectedto vary from that observed for the reference clays in a mannerthat can be rationalized from this study.

AcknowledgmentsThis research was funded by USDA CSREES project WIS04621,Wisconsin Groundwater Coordinating Council Grant 04-CTP-02, and the Wisconsin Small Scale Waste ManagementProject. We thank William F. Bleam and Philip W. Barak forreviewing an earlier draft of this paper. We appreciate theassistance of Mauricio Avila, Kennedy F. Rubert, IV, and WaltZeltner with CEC and surface area measurements, and thelaboratory assistance of Anthony Racke. We thank fouranonymous reviewers for constructive comments that im-proved the quality of this manuscript.

Supporting Information AvailableChemical purity and supplier information; sulfonamideproperties and speciation; clay preparation and character-ization; batch adsorption experiments; high-performance

liquid chromatography method; liquid scintillation countingmethod; X-ray diffractometry method; smectite d001 spacingsafter sulfonamide adsorption; SMZ adsorption to kaolinite;SMZ adsorption isotherms. This material is available free ofcharge via the Internet at http://pubs.acs.org.

Literature Cited(1) Mandell, G. L.; Sande, M. A. Antimicrobial agents: Sulfonamides,

trimethoprim - sulfamethoxazole, quinolones and agents forurinary tract infections. In Goodman & Gilman’s The Pharma-cological Basis of Therapeutics, 8th ed.; Gilman, A. G., Rall, T.W., Nies, A. S., Taylor, P., Eds.; McGraw-Hill: New York, 1990;pp 1047-1064.

(2) Hirsch, R.; Ternes, T.; Haberer, K.; Kratz, K. L. Occurrence ofantibiotics in the aquatic environment. Sci. Total Environ. 1999,225, 109-118.

(3) Berger, K. V.; Petersen, B.; Bunning-Pfaue, H. Persistenz vonGulle-Arzneistoffe in der Nahrungskette. Archiv. Lebensmittel-hyg. 1986, 37, 85-108.

(4) Ingerslev, F.; Halling-Sørensen, B. Biodegradability propertiesof sulfonamides in activated sludge. Environ. Toxicol. Chem.2000, 19, 2467-2473;

(5) Perez, S.; Eichhorn, P.; Aga, D. S. Evaluating the biodegradabilityof sulfamethazine, sulfamethoxazole, sulfathiazole and trime-thoprim at different stages of sewage treatment. Environ. Toxicol.Chem. 2005, 24, 1361-1367.

(6) Kolpin, D. W.; Furlong, E. T.; Meyer, M. T.; Thurman, E. M.;Zaugg, S. D.; Barber, L. B.; Buxton, H. T. Pharmaceuticals,hormones, and other organic wastewater contaminants in U.S. Streams, 1999-2000: A national reconnaissance. Environ.Sci. Technol. 2002, 36, 1202-1211.

(7) Karthikeyan, K. G.; Meyer, M. T. Occurrence of antibiotics inwastewater treatment facilities Wisconsin, USA. Sci. TotalEnviron. (in press).

(8) Holm, J. V.; Rugge, K.; Bjerg, P. L.; Christensen, T. H. Occurrenceand distribution of pharmaceutical organic compounds in thegroundwater downgradient of a landfill (Grindsted, Denmark).Environ. Sci. Technol. 1995, 29, 1415-1420.

(9) Pedersen, J. A.; Soliman, M. A.; Suffet, I. H. Human pharma-ceuticals, hormones, and personal care product ingredients inrunoff from agricultural fields irrigated with treated wastewater.J. Agric. Food Chem. 2005, 53, 1625-1632.

(10) Kinney, C. A.; Furlong, E. T.; Werner, S. L.; Cahill, J. D. Occurrenceand behavior of wastewater-derived pharmaceuticals in soilsirrigated by reclaimed wastewater. Abstr. Papers Am. Chem. Soc.,227, 112-ENVR, Part 1, March 28-April 1, 2004.

(11) Schmitt, H.; Beelen, P. V.; Tolls, J.; Leeuwen, C. L. V. Pollution-induced community tolerance of soil microbial communitiescaused by the antibiotic sulfachloropyridazine. Environ. Sci.Technol. 2004, 38, 1148-1153.

(12) Cribb, A. E.; Miller, M.; Tesoro, A.; Spielberg, S. P. Peroxidase-dependent oxidation of sulfonamides by monocytes and neu-trophile from humans and dogs. Mol. Pharmacol. 1990, 38, 744-751.

(13) Thiele-Bruhn, S.; Seibicke, T.; Schulten, H.-R.; Leinweber, P.Sorption of sulfonamide pharmaceutical antibiotics on wholesoils and particle-size fractions. J. Environ. Qual. 2004, 33, 1331-1342.

(14) Tolls, J. Sorption of veterinary pharmaceuticals in soils: A review.Environ. Sci. Technol. 2001, 35, 3397-3406.

(15) Boxall, A. B. A.; Blackwell, P.; Cavallo, R.; Kay, P.; Tolls, J. Thesorption and transport of a sulphonamide antibiotic in soilsystems. Toxicol. Lett. 2002, 131, 19-28.

(16) Pinck, L. A.; Holton, W. F.; Allison, F. E. Antibiotics in soils. I.Physicochemical studies of antibiotic-clay complexes. Soil Sci.1961, 91, 22-28.

(17) Pinck, L. A.; Soulides, D. A.; Allison, F. E. Antibiotics in Soils. II.Extent and mechanism of release. Soil Sci. 1961, 91, 94-99.

(18) Pinck, L. A.; Soulides, D. A.; Allison, F. E. Antibiotics in soils. IV.Polypeptides and macrolides. Soil Sci. 1962, 94, 129-131.

(19) Bewick, M. W. M. Adsorption and release of tylosin by clays andsoils. Plant Soil 1979, 51, 363-372.

(20) Sithole, B. B.; Guy, R. D. Models for tetracycline in aquaticenvironment. 1. Interaction with bentonite clay systems. WaterAir Soil Pollut. 1987, 32, 303-314.

(21) Nowara, A.; Burhenne, J.; Spiteller, M. Binding of fluoroquinolonecarboxylic acid derivatives to clay minerals. J. Agric. Food Chem.1997, 45, 1459-1463.

(22) Figueroa, R. A.; Leonard, A.; MacKay, A. A. Modeling tetracyclineantibiotic sorption to clays. Environ. Sci. Technol. 2004, 38, 476-483.


(23) Sahurai, H.; Ishimitsu, T. Microionization constants of sul-fonamides. Talanta 1980, 27, 293-298

(24) Martınez, F.; AÄ vila, C. M.; Gomez, A. Thermodynamic study ofsolubility of some sulfonamides in cyclohexane. J. Braz. Chem.Soc. 2003, 14, 803-808.

(25) Advanced Chemistry Development (ACD) Software Solaris V4.67;ACD/Labs: Toronto, ON, 2004.

(26) Lin, C. E.; Chang, C. C.; Lin, W. C. Migration behavior andseparation of sulfonamides in capillary zone electrophoresisIII. Citrate buffer as a background electrolyte. J. Chromatogr.A 1997, 768, 105-112.

(27) Thiele-Bruhn, S. Pharmaceutical antibiotic compounds in soil.J. Plant Nutr. Soil Sci. 2003, 166, 145-167.

(28) Lin, C. E.; Chang, C. C.; Lin, W. C. Migration behavior andseparation of sulfonamides in capillary zone electrophoresis. 2.Positively charged species at low pH. J. Chromatogr. A 1997,759, 203-209.

(29) Kan, C. A.; Petz, M. Residues of veterinary drugs in eggs andtheir distribution between yolk and white. J. Agric. Food Chem.2000, 48, 6397-6403.

(30) Zysset, M.; Schindler, P. W. The proton promoted dissolutionkinetics of K-montmorillonite. Geochim. Comsochim. Acta 1996,60, 921-931.

(31) Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M.Environmental Organic Chemistry, 2nd ed.; John Wiley & Sons:New York, 2002.

(32) Olphena, H. V.; Fripiat, J. J. Data Handbook for Clay Mineralsand Other Nonmetallic Minerals; Oxford: New York, 1979.

(33) Stumm, W. Chemistry of the Solid-Water Interface; John Wiley& Sons: New York, 1992

(34) Bolt, G. H. Adsorption of anions by soil. In Soil Chemistry A.Basic Element; Bolt, G H., Briggenwert, M. G. M., Eds.; Elsevier:New York, 1976; pp 91-95.

(35) Sposito, G. The Surface Chemistry of Soils; Oxford UniversityPress: New York, 1984.

(36) Sverjensky, D. A.; Sahai, N. Theoretical prediction of single-sitesurface-protonation equilibrium constants for oxides andsilicates in water. Geochim. Cosmochim. Acta 1996, 20, 3773-3797.

(37) Essington, M. E. Soil and Water Chemistry: An IntegrativeApproach; Wiley-Interscience: New York, 2003.

(38) Vantelon, D.; Montarges-Pelletier, E.; Michot, L. J.; Briois, V.;Pelletier, M.; Thomas, F. Iron distribution in the octahedralsheet of dioctahedral smectites. An Fe K-edge X-ray adsorptionspectroscopy study. Phys. Chem. Miner. 2003, 30, 44-53.

(39) Jaynes, W. F.; Bigham, J. M. Charge reduction, octahedral charge,and lithium retention in heated, Li-saturated smectites. ClaysClay Miner. 1987, 35, 440-448.

(40) Sheng, G.; Johnston, C. T.; Teppen, B. J.; Boyd, S. A. Adsorptionof dinitrophenol herbicides from water by montmorillonite.Clays Clay Miner. 2002, 50, 25-34.

(41) Weissmahr, K. W.; Haderlein, S. B.; Schwarzenbach, R. P.Complex formation of soil minerals with nitroaromatic explo-sives. Soil Sci. Soc. Am. J. 1998, 62, 369-378.

(42) Bradbury, M. H.; Baeyens, B. A mechanistic description of Niand Zn sorption on Na-montmorillonite. 2. Modeling. J. Contam.Hydrol. 1997, 27, 223-248.

(43) Yariv, S.; Heller, L.; Kaufherr, N. Effect of acidity in montmo-rillonite interlayers on the sorption of aniline derivatives. ClaysClay Miner. 1969, 17, 301-308.

(44) Akyuz, S.; Akyuz, T. FT-IR spectroscopic investigation of adsorp-tion of pyrimidine on sepiolite and montmorillonite from Ana-tolia. J. Inclusion Phenom. Macrocyclic Chem. 2003, 46, 51-55.

(45) Mukherjee, G. N.; Basu, S. Mixed-ligand complex formationequilibria of cobalt-, nickel-, copper- and zinc(II) with sulfa-pyridine, sulfadiazine, and sulfathiazole as primary ligands andglycine, â-alanine and dl-methionine as secondary ligands. J.Indian Chem. Soc. 1999, 76, 288-291.

(46) Hosny, W. M. Complexes of sulphadiamidine. Part II. Equilib-rium and thermodynamic studies of iron(II) and iron(IIII) withsulphadiamidine. Synth. React. Inorg. Met.-Org. Chem. 1999,29, 361-375.

(47) Amrallah, A. H.; Aboul-Kasim, E.; Abdallah, N. A.; Ghandour,M. A.; Farghaly, O. A. Spectral behaviour and acidity constantsof some sulfa drugs. Aswan Sci. Technol. Bull. 1996, 17, 89-103.

(48) Jones, R. A.; Katritzky, A. R. Potentially tautomeric pyridines.III. 2-, 3- and 4-methanesulfonamidopyridine. J. Chem. Soc.1961, 378-384.

(49) Langhammer, J. P. Ph.D. Thesis, Rheinische Friedrich-WilhelmsUniversitat, Bonn, 1989.

(50) Thiele, S. Adsorption of the antibiotic pharmaceutical compoundsulfapyridine by a long-term differently fertilized loess Cher-nozem. J. Plant Nutr. Soil Sci. 2000, 163, 589-594.

(51) Weber, W. J., Jr. Environmental Systems and Processes: Principles,Modeling, and Design; Wiley-Interscience: New York, 2001.

Received for review April 4, 2005. Revised manuscript re-ceived September 29, 2005. Accepted September 30, 2005.

ES050644C


Documents

Adsorption of Sulfonamide Antimicrobial Agents to Clay Minerals