fate and transport of sulfadimethoxine and prim in two southeastern united states soils

8/2/2019 Fate and Transport of Sulfadimethoxine and prim in Two Southeastern United States Soils

http://slidepdf.com/reader/full/fate-and-transport-of-sulfadimethoxine-and-prim-in-two-southeastern-united 1/10

www.vadosezonejournal.org · Vol. 8, No. 1, February 2009 32

A are introduced into the environment

primarily through human and animal wastes. Tese com-

pounds are designed to be toxic at low concentrations; however,

their bioactivity may remain after excretion, causing adverse effects

on the ecosystem (Wollenberger et al., 2000). Several studies have

found bacterial-resistant genes in the environment from overex-

posure to antimicrobials (e.g., Costanzo et al., 2005; Oliveira

et al., 2006; Hamscher et al., 2004). In addition to causing the

development of antimicrobial resistance, these bacteriostatic or

bactericidal compounds may alter microbial community struc-

tures and inhibit the growth of important microorganisms

(Westergaard et al., 2001), thereby affecting essential microbially

mediated ecosystem functions.

One target route of antimicrobial contamination is animalhusbandry. Current agricultural practices in the United Statesrequire the use of confined animal feeding operations, where itis often necessary to provide antimicrobials for the treatmentand prevention of diseases. In animal agriculture, antimicrobials

may make their way into the environment through land-appliedanimal waste used as organic fertilizers, or aquaculture ponds

where medicated feed is applied in the water. Te poultry andaquaculture industries both rely on antimicrobials, such as sul-fadimethoxine (SDM) and ormetoprim (OMP), to maintainhealthy animals. Both SDM and OMP are sold in combination asthe U.S. Food and Drug Administration approved drugs Romet30 and Rofenaid 40 for aquaculture and poultry, respectively.Recent antimicrobial environmental occurrence studies havedetected SDM and OMP in pond water, and several others havedetected SDM in receiving waters from agricultural sources. Forexample, Dietze et al. (2005) detected both SDM (maximum 36μg L−1) and OMP (maximum 12 μg L−1) in pond water where

Romet 30 had been applied. Another pond analysis found SDMin >25% of the water samples at concentrations up to 0.88 μg L−1,but only found sediment contamination in <8% of the samples(7.7 μg L−1) (Hamscher et al., 2006). Te USGS found multipleantimicrobial contaminants in streams across the United States,including low concentrations of SDM (0.06 μg L−1) (Kolpinet al., 2002). Poultry farms were also shown to contribute tostream surface contamination in a study by Campagnolo et al.(2002), who found SDM in receiving waters proximal to poultry farms. Other studies (e.g., Kreuzig et al., 2005) have reportedsulfonamide concentrations up to 703 μg L−1 in runoff from

Fate and Transport of Sulfadimethoxineand Ormetoprim in Two SoutheasternUnited States Soils

P. Srivastava,* S. M. Sanders, J. H. Dane, Y. Feng, J. Basile, and M. O. Barne

P. Srivastava and S.M. Sanders, Biosystems Engineering, 200 Corley Bldg.,

Auburn Univ., Auburn, AL 36849-5417; J.H. Dane and Y. Feng, Agronomy

and Soils Dep., 201 Funchess Hall, Auburn Univ., Auburn, AL 36849-5412;

J. Basile, Alabama Dep. of Agriculture and Industries, 1445 Federal Dr.,

Montgomery, AL 36107; and M.O. Barne, Civil Engineering Dep., 208 Har-

bert Engineering Center, Auburn Univ., Auburn, AL 36849-5337. Received

20 Dec. 2007. *Corresponding author ([email protected]).

Vadose Zone J. 8:32–41

doi:10.2136/vzj2007.0186

© Soil Science Society of America677 S. Segoe Rd. Madison, WI 53711 USA.All rights reserved. No part of this periodical may be reproduced or transmi ed

in any form or by any means, electronic or mechanical, including photocopying,

recording, or any informaon storage and retrieval system, without permission

in wring from the publisher.

A: OMP, ormetoprim; SDM, sulfadimethoxine.

O R

There have been increasing concerns regarding anmicrobial contaminants in the environment. To date, limited research

exists regarding the fate and transport of these compounds in the environment once they have been discharged in

human and animal wastes. The fate and transport of two anmicrobials, sulfadimethoxine (SDM) and ormetoprim

(OMP), were invesgated in two soils and a sand using miscible pulse and step displacement column studies. Because

OMP and SDM are oen administered in combinaon, their fate and transport were invesgated individually as single

solutes as well as in combinaon as cosolutes. The transport of SDM and OMP was modeled by the chemical nonequi-

librium model of the convecon–dispersion equaon. Sorpon of SDM in soils and sand was found to be weak (i.e., it

readily desorbed); OMP was sorbed more strongly and required more me for desorpon. Results for the pulse input

columns yielded mass recoveries >0.90 for the sand and two soils with SDM and for the sand with OMP. Mass recoveries

of OMP in the two soils were 0.56 and 0.55, respecvely, indicang irreversible sorpon or chemical transformaon.

Comparisons of single-solute and cosolute column studies of OMP and SDM indicate that sorpon and transport of

these compounds in mixtures were not considerably diff erent from their individual sorpon and transport. Overall, the

results from this study indicate that both compounds have the potenal to move through soils, contaminang nearby

surface waters and groundwater.




agricultural fields, indicating that land-applied antimicrobials areoften at risk of surface transport. Limited studies have assessedthe occurrences of SDM and OMP in groundwater, but a few have found SDM and other sulfonamides in groundwater atenvironmentally relevant concentrations (Batt et al., 2006;Holm et al., 1995).

A thorough understanding of antimicrobial fate and trans-port in the environment is necessary for adequate environmentalrisk assessments of these compounds. Little information is avail-able, however, on SDM and OMP fate and transport in soils.Te Environmental Impact Analysis Report for the approvalof Romet 30 to date is the only known study to evaluate thesorption and mobility of both SDM and OMP (U.S. Foodand Drug Administration, 1984). Tis report, however, did notevaluate the necessary fate and transport parameters nor did itpossess the analytical capabilities to analyze environmentally relevant concentrations (<0.05 mg L−1). More specific fate andtransport information than given by the U.S. Food and Drug

Administration (1984) is necessary to adequately determine andpredict environmental concentrations.

Terefore, the objective of this study was to study the fateand transport of SDM and OMP individually and in combina-

tion in two soils and a pure sand from the southeastern UnitedStates using miscible displacement column experiments.

Materials and Methods

Selected Soils

wo soils representative of agricul-ture and aquaculture in the southeasternUnited States were chosen for this study.One, a Coastal Plain soil (loamy sand),

was collected from Geneva County, AL, and the other, a ennessee Valley soil (loam), was collected from Sevier

County, N. A pure sand, Ottawa 4.0sand, was used as a reference. Bothsoils were air dried, ground, and sieved(≤2-mm diameter). Te sand was bakedat 550°C for 4 h to eliminate any pos-sible organic matter. Soil physical andchemical properties (able 1) were deter-mined at the Soil esting Laboratory at

Auburn University. Te sand and soils were irradiated using a 60Co source at0.05 MGy to eliminate microorgan-isms that could potentially biodegradethe antimicrobials. Plate counts using

half-strength nutrient agar (4 g nutri-ent broth L−1) confirmed the sterility of the two soils and sand (zero colo-nies on plates after 14 d). Accordingto McNamara et al. (2003), 60Coirradiation is highly effective as a soilsterilization method, causing minimalchanges to soil physical and chemicalproperties.

Anmicrobialshe antimicrobials OMP [2,4-diamino-5-(4,5-dime-

thoxy-2-methylbenzyl) pyrimidine] and SDM [N ′-(2,6dimethoxy-4-pyrimidinyl) sulfanilamide], were obtained fromChem Service, Inc. (West Chester, PA) and Sigma Aldrich (St.Louis, MO), respectively. Molecular structures and selected physi-cochemical properties of SDM and OMP are shown in able2. Te stock solutions for both antimicrobials were initially dis-solved in methanol (high-performance liquid chromatography grade, Fisher Scientific International, Hampton, NH) such thatthe final working solutions contained <0.2% methanol.

Miscible Displacement ExperimentsBoth soils and the sand (able 1) were each uniformly packed

in individual glass columns. wo different columns were used,one 10 cm long with 5-cm i.d. (for step input experiments) andthe other 4 cm long with 5-cm i.d. (for pulse input experiments).For each experiment, soil columns were packed with new soil tothe same bulk density. Both ends of the soil columns were covered

with several layers of cheesecloth, a paper filter, and eflon endcaps to retain the soil during the experiments.

Te apparatus included a precision constant-volume pump

(Masterflex Quick Load, Cole Parmer, Vernon Hills, IL), a frac-tion collector (Spectra/Chrom CF-1 [Spectrum Chromatography,Houston, X] or ISCO Retriever II [eledyne Isco, Lincoln, NE])

T 1. Physical and chemical properes of the selected soils and sand.

Soil Descripon Order pHOrganic

maerSand Silt Clay eCEC†

Mineralogical

composion of clay

fracon

————————% —————— cmolc kg−1

Loamy

sand

Plinthic

Kandiudult

Ulsol 5.03 1.5 81.5 13.5 5 3.19 interlayer vermiculite,

kaolinite, gibbsite,

quartz, goethite

Loam Typic

Eutrudept

Incepsol 4.66 2.07 52 38 10 6.64 vermiculite, illite,

kaolinite, quartz,

interstrafied

vermiculite/mica

Sand Oawa 4.0‡ 6.97 0 100 0 0 0.33

† Eff ecve caon exchange capacity.

‡ Blasng sand from Oawa, IL, with a parcle size distribuon of 0.076% 0–106 μm, 0.856% 106–250 μm,

7.883% 250–50 0 μm, 90.47% 500–840 μm, and 0.443% 840–2000 μm.

T 2. Physiochemical characteriscs of the anmicrobials ormetoprim and sulfadimethoxine(structures and physiochemical properes retrieved from Naonal Library of Medicine [2006]).

Anmicrobial Key properes†

Ormetoprim structure class: diaminopyrimidine

molecular formula: C14H18N4O2molecular weight: 274.32

water solubility: 1540 mg L−1

log K ow = 1.23

Weak base, pK a ~ 7vapor pressure: 3.04 × 10−6 N m−2

atmospheric OH rate constant: 6.34 × 10−11 cm3 molecule−1 s−1

Sulfadimethoxine structure class: sulfonamide

molecular formula: C12H14N4O4S

molecular weight: 310.33

water solubility: 343 mg L−1

log K ow: 1.63

pK a1/pK a2: 2.4/6.0

vapor pressure: 2.12 × 10−7 N m−2

atmospheric OH rate constant: 2.02 × 10−10 cm3 molecule−1 s−1

† K ow, octanol–water paroning coefficient; pK a, negave log of the acid dissociaon constant K a, sub-

scripts 1 and 2 denote two pK a values for this compound.




with 15-mL polypropylene test tubes, Masterflex BioPharm Plusplatinum silicone tubing, and polypropylene volumetric flasks tostore resident, tracer, and antimicrobial solutions. Polypropylene

was determined to be chemically inert to SDM and OMP for theduration of these experiments and yielded preferred recoveries overglassware (Sanders, 2007). Because polypropylene columns werenot readily available, however, glass columns were acid washedand used for these experiments. All tubing, test tubes, pipettes,filters, and flasks were checked for SDM and OMP compatibility.

Additionally, column controls (solution only) with SDM and OMP were performed for the duration of the column experiments. Teinfluent and effl uent concentrations were regularly checked andremained within 95% of the initial concentration.

Te experiments were conducted under saturated flow condi-tions because it provided a well-defined and repeatable startingpoint and the desired initial chemical conditions in the soil couldbe established quickly (Skaggs et al., 2002). Each soil was wetted,from the bottom up, with a 0.01 mol L−1 CaCl2 resident solu-tion and the flow (3 mL min−1) was maintained for a minimumof 24 h to remove entrapped air and create steady-state saturatedflow conditions. Te Ca2+ in the resident solution minimizedclay dispersion and served as an exchangeable cation of known

charge and sorption affi nity (Skaggs et al., 2002). Even thoughpast studies (e.g., Tiele- Bruhn et al., 2003; U.S. Food andDrug Administration, 1984) suggested that photodegradationis negligible for sulfonamides and is probably not important forOMP (U.S. Food and Drug Administration, 1984), a few recentstudies (e.g., Boreen et al., 2004, 2005) have suggested that pho-tochemical degradation is expected to have an impact on theenvironmental fate of the sulfa drugs. o ensure that photodeg-radation was not a factor, the experiments were performed in thedark by covering the tubes, column, and flasks with Al foil.

Once steady-state pore water velocity was achieved, theresident solution was displaced with a step feed (for 10-cm-longcolumns) or pulse feed (for 4-cm-long columns) of Br− tracer

solution (0.01 mol L−1 CaBr2) and fractions of the effl uent werecollected. Te Br concentration of each fraction was measuredusing an ion chromatograph (Dionex DX-120, Dionex Corp.,Sunnyvale, CA) and a breakthrough curve was determined.Calcium chloride (0.01 mol L−1) was then flushed through thecolumn for several pore volumes. Following this, each antimicro-bial or antimicrobial combination (100 μg L−1) was applied as astep feed (for the 10-cm-long columns) or pulse feed (for 4-cm-long columns). For the 4-cm columns, the antimicrobial pulse

was followed with the CaCl2 solution. Te column effl uent wascollected in polypropylene test tubes with increments of 15 or 9mL for the 10- and 4-cm columns, respectively. Following collec-tion, samples were immediately prepared for analysis by filtration

with 0.45-μm polytetrafluoroethylene (eflon) membrane filters.Te filtrates were placed into 750-μL polypropylene autosamplervials and were acidified with formic acid for sample preservationso that the final amount of acid was <1% of the total volume.

Chemical Analysis Antimicrobial samples were analyzed in cooperation with

the Food and Drug Laboratory at the Alabama Department of Agriculture and Industries, Montgomery. All samples were storedat 4°C until they were transported on ice (2–4°C) to the Foodand Drug Laboratory. Te samples were analyzed using a triple-

quadrupole liquid chromatograph mass spectrometer (LC/MS/MS). A sample run consisted of seven standard solutions followedby an initial calibration verification and an initial calibrationblank. Tis was followed by a repeating sequence of 10 samplesand subsequently by a continuing calibration verification andcontinuing calibration blank.

Te analytical method uses a gradient separation with thefollowing solvents: acetonitrile and 0.1% formic acid in water.

Analytes were separated chromatographically using a Phenomenex(orrance, CA) Gemini C18 column (5 μm, 150 by 2.00 mm) andsamples were analyzed using a Termo Finnigan SQ QuantumLC/MS/MS system (Termo Fisher Scientific, Bellefonte, PA).Compounds of interest were ionized using atmospheric pressurechemical ionization in positive mode generating an M+H ion of the parent compound or electrospray ionization. A minimum of three daughter ions were generated from the parent compoundusing selective ion monitoring mode. Compounds were identifiedby the presence and ratio of all daughter ions and the retentiontime of the compound compared with the calibration standards.Quantification was based on the area of the peak of the mostabundant daughter ion of the analyte and regressed against theseven-point standard curve.

Fate and Transport ModelMiscible displacement solute transport experiments have

frequently been used to assess the possible contamination of soil and groundwater (Porro and Wierenga, 1993; Singh andKanwar, 1991; ipton et al., 2003) as they yield more realis-tic solute transport parameters than traditional batch sorptionequilibrium experiments. raditional batch sorption experimentsallow estimation of only a sorption or distribution coeffi cient,

whereas miscible displacement column studies account for addi-tional solute transport parameters (e.g., diffusion–dispersioncoeffi cient, retardation factor, distribution coeffi cient, kinetic ratecoeffi cients, and degradation or transformation rate constants).

Models, such as CXFI (Version 2.1) developed by oride et al.(1999), allow analysis of column breakthrough curves and subse-quent determination of these transport parameters. Te CXFImodel uses inverse modeling to fit mathematical solutions of theoretical transport models to the experimental data (oride etal., 1999). Several transport models based on the convection–dispersion equation are included in CXFI. Tese consist of local sorption equilibrium, chemical nonequilibrium, and physi-cal nonequilibrium.

Te one-dimensional convection–dispersion transport equa-tion for reactive solutes under steady-state flow in a homogenoussoil is given by

2b

l s2C C C R D v C S t x x

∂ ∂ ∂ ρ= − − μ − μ∂ ∂ θ∂

[1]

where R is a dimensionless retardation factor, C is the soluteconcentration in the liquid phase [M L−3], t is time [], D is thesolute diffusion–dispersion coeffi cient [L2 −1], x is the distancein the direction of flow [L], v is the average pore velocity [L −1],μl and μs are the liquid and solid first-order decay coeffi cients[−1], ρb is the bulk density of the soil [M L−3], θ is the volu-metric water content [L3 L−3], and S is the mass of solute in thesorbed phase per mass of solid [M3 M−3]. Te retardation factoris determined by




b1S

R C

ρ ∂= +

θ ∂ [2]

where ∂S /∂C is the slope of the sorptionisotherm. For a linear sorption isotherm,the slope is represented by K d, the dis-tribution coeffi cient [L3 M−1].

Chemical nonequilibrium trans-port is often used for modeling reactivesolutes that exhibit kinetic sorption(oride et al., 1999). A number of otherfactors can cause chemical nonequilib-rium transport, including nonlinearsorption, desorption hysteresis, andthe presence of physical and chemicalheterogeneity in the soil (Skaggs andLeij, 2002). Nonreactive solutes suchas Br− are not retarded in the soil andthe decay coeffi cients are zero (Casey etal., 2003).

Chemical nonequilibrium trans-port is often described by two sorption sites (ype 1 and ype

2 sorption sites) (Selim et al., 1976, 1977). In a two-site model,ype 1 sorption sites, S e, are subject to instantaneous sorption, while ype 2 sorption sites, S k , obey a kinetic rate law (Skaggsand Leij, 2002):

e k S S S = + [3]

Using a linear equilibrium sorption isotherm, S e is represented by

edS K C = [4]

and the ype 2 sorption sites are given by

( )k

k k k d s1

S f K C S S

t

∂ ⎡ ⎤= α − − − μ

⎢ ⎥⎣ ⎦∂

[5]

where α is a first-order kinetic sorption coeffi cient [−1], f is thefraction of exchange sites always in equilibrium with the solu-tion phase, and μs

k is the solid-phase first-order kinetic decay coeffi cient [−1].

Results and Discussion

Major physical and flow parameters for each column arelisted in able 3. Since for each experiment soil columns werecarefully packed with new soil (or sand) to the same bulk density,the volumetric water contents and the pore water velocities weresimilar for each soil and sand (able 3).

Bromide breakthrough curves were symmetrical and fit well with the equilibrium transport model of the convection–disper-sion equation, indicating that Br− transport was nearly ideal andphysical nonequilibrium was probably not occurring in the soilcolumns. Te hydraulic residence time (R) for the Br− in thecolumn was calculated (able 3). Since Br− is a nonreactive solute,the R was calculated by dividing the length of the column by the pore water velocity. Tis provided another line of evidence

that each column was similarly packed. Te diffusion–dispersioncoeffi cient, D, was determined and is shown for the 10-cm-longcolumns in able 4 and for the 4-cm columns in able 5. Tediffusion–dispersion coeffi cients determined from the Br− (non-reactive) breakthrough curves were assumed to be the same for

T 3. Physical properes of repacked soil and sand columns to measure the displacement of ormetoprim (OMP) and sulfadimethoxine (SDM).

Sorbent Column†Bulk

density

Volumetric

water

content

Pore water

velocity

Pore

volume

(PV)

Pulse input

(relave

PV)‡

Residence

me

g cm−3 m3 m−3 cm min−1 cm3 h

Sand 4-cm cosolute§ 1.78 0.327 0.47 25.7 84.0 0.14

10-cm cosolute 1.80 0.320 0.48 62.8 n/a 0.35

10-cm OMP single solute 1.78 0.329 0.47 64.6 n/a 0.36

10-cm SDM single solute 1.80 0.321 0.48 63.0 n/a 0.35

Loamy sand 4-cm cosolute 1.49 0.438 0.35 34.4 121.3 0.19

10-cm cosolute 1.45 0.454 0.34 89.2 n/a 0.5010-cm OMP single solute 1.37 0.485 0.32 95.2 n/a 0.53


Loam 4-cm cosolute 1.26 0.524 0.29 41.2 132.9 0.23

10-cm cosolute 1.23 0.537 0.29 105.4 n/a 0.59

10-cm OMP single solute 1.16 0.561 0.27 110.1 n/a 0.61


† 4-cm columns used only with SDM and OMP cosolute, 10-cm columns used with SDM and OMP cosolute

as well as OMP and SDM as single solutes.

‡ Relave pore volume of anmicrobial pulse input for 4-cm columns. Desorpon began aer pulse.

§ Cosolute, SDM and OMP administered together; single solute, OMP or SDM administered separately.

T 4. Transport parameters† for 10-cm-long columns with step input; 95% confidence intervals in parentheses.

Sorbent Solute‡ D R K d f α μ R2

cm2 min−1 L kg−1 ————min−1 ————

Sand OMP single solute 0.093 (0.062–0.123) 4.26 (3.17–5.34) 0.60 0.40 0.0016 4.6 × 10−9 0.97

OMP cosolute 0.075 (0.065–0.086) 3.36 (2.95–3.77) 0.42 0.53 0.0394 4.8 × 10−9 0.93

SDM single solute 0.136 (0.084–0.188) 1.51 (1.44–1.57) 0.09 0.30 0.0055 0.0012 0.98

SDM cosolute 0.075 (0.065–0.086) 1.52 (1.42–1.62) 0.09 0.51 5.5 × 10−7 0.0037 0.98

Loamy sand OMP single solute 0.140 (0.093–0.186) 28.26 (25.56–30.96) 9.69 0.09 0.0051 0.0229 0.98OMP cosolute 0.143 (0.093–0.193) 26.34 (23.03–29.66) 7.95 0.13 0.0044 0.0252 0.99


SDM cosolute 0.143 (0.093–0.193) 14.97 (10.53–19.41) 4.38 0.04 0.0037 0.0018 0.96

Loam OMP single solute 0.049 (0.019–0.078) 84.91 (83.07–86.75) 40.44 0.33 0.0005 0.0094 0.99

OMP cosolute 0.146 (0.083–0.209) 91.83 (50.19–133.5) 39.75 0.60 0.0013 0.0133 0.99


SDM cosolute 0.146 (0.0829–0.2087) 11.21 (9.19–13.22) 4.47 0.35 0.0171 0.0047 0.74

† D, solute diff usion–dispersion coefficient; R, dimensionless retardaon factor; K d, distribuon coefficient; f , fracon of exchange sites always in equilib-

rium with the soluon phase; α, first-order kinec sorpon coefficient; μ, degradaon rate coefficient. Linear sorpon coefficients were calculated

from column experiments.

‡ Single solute, the anmicrobial ormetoprim (OMP) or sulfadimethoxine (SDM) was administered individually; cosolute, OMP and SDM were administered

in combinaon.




the OMP or SDM (reactive) breakthrough curves. Te valueof D was therefore used as a fixed parameter when modelingthe antimicrobial breakthrough curves, which allowed a morereliable estimation of additional transport parameters (oride etal., 1999).

Step Input Columns

A total of nine columns used the step input procedure fortracer and antimicrobial solutions (Fig.1 and 2). Both single-solute experiments(containing SDM or OMP individually;Fig. 1) and cosolute experiments (con-taining both SDM and OMP together;Fig. 2) were performed for each soil andthe sand.

he antimicrobial breakthroughcurves (Fig. 1 and 2) were best mod-eled by assuming a two-site chemicalnonequilibrium model with degrada-tion being equal in the solid and liquid

phases. Te breakthrough curve for OMPin the loam soil (Fig. 1c) showed strongretardation, as it took >13 pore volumesbefore the relative concentration (C /C 0)started to increase from its initial value of zero. Even after about 100 pore volumes,the C /C 0 values had only increased toabout 0.5. A similar picture was observedfor the C /C 0 values of OMP when it

was applied in combination with SDM(Fig. 2c), except that approximately 30pore volumes were needed for C /C 0 toincrease from zero. For the loamy sandsoil, the C /C 0 for OMP, applied eitherby itself (Fig. 1b) or in combination withSDM (Fig. 2b), also approached valuesbetween 0.4 and 0.5. In these two cases,however, the breakthrough of OMP wasmuch more rapid. Te breakthrough of OMP in sand, when applied either by itself (Fig. 1a) or in combination withSDM (Fig. 2a), showed lower retarda-tion of the antimicrobials compared withthe two soils. Te OMP breakthrough

curves (Fig. 1b and 1c) in the two soils exemplify strong sorptionof OMP, where the C /C 0 never reached 1. Upon first glance, itmay seem that the low C /C 0 for OMP in the two soils should beattributed to degradation alone; however, no significant degrada-tion was occurring in the sand column, so it may be an indicationof irreversible sorption in the two soils.

For the columns exposed to SDM, either by itself or in com-

bination with OMP, the breakthrough curves for the sand (Fig.

T 5. Transport parameters† for 4-cm-long columns with pulse input; 95% confidence intervals in parentheses.

Sorbent Solute‡ D R K d f α μ R2 Mass recovery§

cm2 min−1 L kg−1 ————min−1 ————

Sand OMP cosolute 0.158 (0.052–0.263) 3.38 (2.89–3.86) 0.44 0.70 0.0895 0.0051 0.96 0.95 (0.96)

SDM cosolute 0.158 (0.052–0.263) 2.26 (1.99–2.53) 0.23 0.52 0.1951 0.0074 0.98 0.94 (0.94)

Loamy sand OMP cosolute 0.202 (0.021–0.384) 35.00 (34.12–35.87) 10.01 0.38 0.0065 0.0586 0.99 0.56 (0.56)

SDM cosolute 0.202 (0.021–0.384) 8.58 (6.93–10.23) 2.23 0.66 0.0060 0.0073 0.99 0.91 (0.92)

Loam OMP cosolute 0.047 (−0.004–0.097) 85.86 (82.03–89.69) 35.26 0.27 0.0057 0.0348 0.99 0.55 (0.63)

SDM cosolute 0.047 (−0.004–0.097) 13.72 (12.09–15.34) 5.29 0.32 0.0153 0.0075 0.97 0.90 (0.90)

† D, solute diff usion–dispersion coefficient; R, dimensionless retardaon factor; K d, distribuon coefficient; f , fracon of exchange sites always in equilib-

rium with the soluon phase; α, first-order kinec sorpon coefficient; μ, degradaon rate coefficient. Linear sorpon coefficients were calculatedfrom column experiments.

‡ The anmicrobials ormetoprim (OMP) and sulfadimethoxine (SDM) were administered in combinaon.

§ Mass recovery was calculated from model results.

F. 1. Step input breakthrough curves from ormetoprim (OMP) or sulfadimethoxine (SDM)single-solute column experiments (10-cm-long columns). Relave concentraon (C /C 0) as afuncon of pore volume.




1d and 2d), the loamy sand soil (Fig. 1eand 2e), and the loam soil (Fig. 1f and 2f)looked very much alike. As for the OMP,the sand column showed lower retarda-tion compared with the two soils, while C /C 0 values for the SDM approached 0.8 to0.9. Te latter indicated less retardation forSDM than OMP.

o further investigate the sorption of SDM and OMP in sand, an extraction forFe, Fe oxide, Mn, and Al was performedon the sand. Tis procedure was chosenbecause the sand had indications (a slightred color) of the presence of Fe or Mn aftermuffl ing. Additionally, other antimicrobi-als have been shown to sorb to Fe oxides(Zhang and Huang, 2007; Figueroa andMackay, 2005). Results from the extrac-tion procedure, however, revealed thatthe sand contained <0.01% Fe oxide orFe, <0.0001% Mn, and no detectable Al,leading to the conclusion that these sand

fractions were not likely contributors to theantimicrobial sorption. Sorption of SDMin sand was also noted by Tiele-Bruhn etal. (2004), where SDM and other sulfon-amides actually sorbed more to sand than tothree other soils investigated. A similar phe-nomenon to the sorption in silica sand wasillustrated in the incompatibilities of silicaglassware with these antimicrobials, wherethere was an affi nity to the silica active sites(Sanders, 2007). Overall, sorption of OMPand SDM in sand was relatively low com-pared with sorption in the loamy sand and

loam soils.Results from the single-solute and cosolute sorption experi-

ments indicate that there was little difference in SDM and OMPsorption or mobility when administered alone as single solutesor in combination as cosolutes. Tis was evident not only inthe breakthrough curves of the cosolute columns and the indi-vidual columns (Fig. 1 and 2), but also in the relevant transportparameters (able 4). Retardation factors for single-solute andcosolute systems were similar, indicating similar mobility of these compounds when administered alone and in combination.

Additionally, there was <5% difference in distribution coeffi cients(K d) in single-solute and cosolute systems for SDM in sand andOMP in the loam soil. Although the single and cosolute distribu-

tion coeffi cients for OMP in sand, OMP in loamy sand, SDMin loamy sand, and SDM in loam represent a 20 to 36% differ-ence, in relation to other mobile compounds this difference israther small. Te distribution coeffi cients for OMP and SDMare much less than the distribution coeffi cients found for highly sorbing classes of antimicrobials such as tetracyclines. Sassmanand Lee (2005) found linear distribution coeffi cients for tetracy-cline, oxytetracycline, and chlortetracycline ranging from 1229to 352,911 L kg−1. Sorption of OMP falls in a similar range assorption of macrolide antimicrobials, such as tylosin. Rabølle andSpliid (2000) and olls (2001) reported tylosin linear sorption

coeffi cients in sandy loam and clay loam between 8.3 and 128 Lkg−1. Overall, the sequence of sorption for OMP and SDM wassand < loamy sand < loam. Te increased sorption in the two soilsfollows the pattern of increasing organic matter, cation exchangecapacity, and clay content and decreasing soil pH.

Although all SDM C /C 0 ratios reached at least 0.9 at somepoint, the effl uent concentrations often showed substantial varia-tion throughout the experiment. Tese large variations did notoccur in all of the SDM experiments, which raises many questionsas to the cause of this phenomenon. In the loam soil experiment

with SDM in cosolute with OMP, after 41 pore volumes the C /C 0 began to have large fluctuations that did not occur early inthe experiment (Fig. 2f). Tis was contrary to the SDM singlesolute in sandy loam (Fig. 1e), where the fluctuations began early in the experiment.

One cause for the fluctuations in SDM (Fig. 1e and 2f)might have been degradation by microorganisms, although thesoils were gamma irradiated and initially sterile. o check thesterility of the soils after the completed SDM and OMP columnexperiments, microbial plate counts were performed on each of the soils using half-strength nutrient agar. After incubation atroom temperature for 7 d, the plates were examined and foundto contain too many colonies to count. Tis was not surprisingbecause of the presence of microorganisms in the air and the

F. 2. Step input breakthrough curves for ormetoprim (OMP) or sulfadimethoxine (SDM)combinaon column experiments (10-cm-long columns). Relave concentraon (C /C 0) as afuncon of pore volume.




fact that it was not possible to completely sterilize the entirecolumn setup. Te addition of 100 or 500 μg L−1 of SDM andOMP to the agar media did not inhibit the growth of thesebacteria. Tis is consistent with findings by Halling-Sørensenand Ingerslev (2000), who showed that microbial growth wasnot inhibited in the presence of multiple sulfonamides and thatthe sulfonamides were not readily biodegradable. Additionally,studies have shown that sulfonamides are stable in the presenceof light (U.S. Food and Drug Administration, 1984; Halling-Sørensen et al., 2003) and others have shown that the half-life of OMP and SDM exceeds 1 yr in an aquatic environment (Bakaland Stoskopf, 2001).

Furthermore, fluctuations in column effl uent concentra-tions have been reported by others working with pharmaceuticals(Scheytt et al., 2004; Scheytt et al., 2006). Propyphenazone con-centration fluctuations in the study of Scheytt et al. (2004) werepartially attributed to degradation and partially to analyticaluncertainty. Analytical uncertainty could explain some of theSDM fluctuations.

Antimicrobial sorption is often affected as the aqueousspeciation changes in various pH environments (olls, 2001).Because pH can strongly influence antimicrobial sorption, the

effl uent pH of SDM and OMP cosolutes in the loamy sand andloam soils was evaluated. Te effl uent pH of the loamy sand andloam soils varied between 4.75 and 5.01 and between 4.65 and5.15, respectively. No correlation between pH and SDM effl u-ent concentration was found, indicating thatthe change in pH did not correspond to thefluctuations in SDM effl uent concentration.

Although no covariance was found betweenthe SDM effl uent concentration and pH, thepH fluctuations are probably important. Evena slight change in pH can affect the specia-tion of antimicrobials, which in turn affectsthe affi nity for the surface. Gao and Pedersen

(2005) found that as the pH dropped below 5,marked increases in sulfonamide sorption wereoften observed.

Te OMP pH-dependent sorption wouldbe similar to that found for another diamin-opyrimidine, trimetoprim (e.g., Bekçi et al.,2006). Bekçi et al. (2006) found that at low pH, all trimetoprim was in the protonatedform (cationic species), and that at pH valuesnear neutral, the weak base was near the nega-tive log of its acid dissociation constant, pK a,and was consequently a neutral species. Moretrimetoprim sorption was found between pH4 to 6 than above and below this range. Teresearchers found that above pH 6, the neutralspecies dominated and had little attraction forthe negatively charged soil surface. At low pH,the protonated trimetoprim was in competi-tion with the decreasing H+ ions in solutionand little sorption occurred.

It should be noted that even a slight changein pH could affect the speciation of SDM andOMP and could therefore cause more sorp-tion or desorption to occur. Although soil has

a natural buffering capacity, it may not buffer it enough to pre-vent slight sorption and desorption changes in antimicrobials,particularly for the SDM observed here. Te pH changes in thenatural environment would play an important role in the sorp-tion and desorption of these compounds.

Pulse Input Columns

After modeling the step input for the 10-cm-long col-umns, it was deemed necessary to perform pulse input columnexperiments (e.g., square wave) with a shorter column length.Performing the experiments with a shorter column would reveal whether the SDM variations were occurring only because of theextended experiment time or whether the variations would alsooccur when the hydraulic residence time (shorter column) wasreduced. Additionally, the shorter columns would be subjected toa pulse input, which would evaluate desorption and mass recovery.Because the differences between single solute and cosolute weredetermined to be minimal, the pulse input columns were only run with SDM and OMP in combination.

Breakthrough curves for the pulse input columns are illus-trated in Fig. 3. Te same sequence of sorption for OMP andSDM was observed in the pulse input columns as in the step

input columns, sand < loamy sand < loam. Te increased sorptionin the two soils again followed the pattern of increasing organicmatter, cation exchange capacity, and clay content and decreasingsoil pH. As described above, the decreasing pH would probably

F. 3. Pulse input breakthrough curves for ormetoprim (OMP) or sulfadimethoxine(SDM) cosolute column experiments (column length = 4 cm). Relave concentraon (C /C 0) as a funcon of pore volume.




change the speciation of OMP and SDM to more cationic species,thereby enhancing sorption.

Te SDM was more readily desorbed than OMP, as indicatedin the model comparisons (Fig. 3 and 4). Figure 4 is composed of breakthrough curves redrawn from Fig. 3. Experimentally, OMP

was not fully desorbed in the loam soil, but based on the resultsof the model it appeared to approach zero. Even with completedesorption, however, the mass recovery would not have reached1. Mass recoveries (able 5) were mostly >0.90 with the exceptionof the two soil columns with OMP. Here the mass recoveries inthe column effl uent were 0.56 and 0.55 for the OMP cosolutein the loamy sand soil and the OMP cosolute in the loam soil,

respectively. Low recoveries indicate irreversible sorption, chemi-cal transformations, or degradation. Degradation rate coeffi cients(μ), which account for degradation, chemical transformations,and irreversible sorption, are given in able 4. Te OMP in thetwo soils has the highest μ of all the other combinations.

he influent antimicrobial concentration was regularly checked to ensure no loss of antimicrobial at the inlet (relativedifference <10%). Furthermore, to check the effl uent variationsfor possible metabolites, several of the low-effl uent SDM samples

were evaluated in a full ion scan on the LC/MS/MS in mass spec-trometry mode. Sulfonamide metabolites of particular interest

were those from hydroxylation or acetylation because they havebeen detected in chickens, pigs, manure, and milk (Furusawa,2006; Furusawa and Mukai, 1994; Haller et al., 2002; Kishidaand Furusawa, 2004). None of the molecular weights of theexpected metabolites were identified for SDM; unknown metabo-lites may be present, but based on the peaks in the total ion scan,they would represent <1% of the total analytes.

Te loam SDM cosolute (Fig. 3f ) pulse input in a 4-cm-longcolumn showed a similar pattern of SDM fluctuations as in the10-cm-long columns. Tis indicated that the shorter hydraulicresidence time did not eliminate the SDM fluctuations. Te SDM

mass recovery for all three SDM columns with fluctuations was>0.90. For the pulse input columns, the loam SDM cosolute showedthe largest concentration variation but was modeled well (Fig. 3f).

Model parameters for the pulse input columns yielded excel-lent results with R 2 > 0.96. Te OMP linear sorption coeffi cients(K d) ranged from 0.44 to 35.26 L kg−1 and followed the samesequence of sorption in the sand and the two soils as in the stepinput columns. Halling-Sørensen and Ingerslev (2000) found thattrimetoprim, another diaminopyrimidine with similar structureand properties to OMP, had a distribution coeffi cient of 76 Lkg−1 in sludge. Te trimetoprim distribution coeffi cient is higherthan the largest distribution coeffi cient of OMP in the loam soil;however, the sewage sludge would contain a greater percentage

of organic matter for sorption than the loam soil (2.07% organicmatter). Te SDM sorption also followed the same sequence of sorption and yielded sorption coeffi cients from 0.23 to 5.29 Lkg−1. Te weak sorption of SDM in column studies was similarto sulfonamide sorption in soils found by others. Boxall et al.(2002), Tiele-Bruhn and Aust (2004), and Tiele-Bruhn (2003)reported sulfonamide sorption coeffi cients from 0.3 to 10 L kg−1 in a variety of soils from sand to clay. Gao and Pedersen (2005)found higher K d values for sulfonamides in clay minerals, rangingfrom 2.3 to 22.2 L kg−1.

ConclusionsTis study revealed new information on the fate and trans-

port of SDM and OMP in the environment. Both SDM andOMP are relatively mobile in sand, but illustrate more retarda-tion in the two southeastern United States soils. Comparisonsof single-solute and cosolute column studies of OMP and SDMindicate that sorption of these compounds in mixture is not con-siderably different from their individual sorption.

Te relative mobility of OMP is much greater than that of tetracycline antimicrobials observed by Sassman and Lee (2005),yet less than sulfonamide antimicrobials such as SDM. Sorptionof SDM from column transport studies in soil and sand wasfound to be very weak, as noted by other researchers studying

F. 4. Modeled breakthrough curve comparisons for sulfadime-thoxine (SDM) and ormetoprim (OMP) cosolute in (a) sand, (b)loamy sand, and (c) loam soils. Relave concentraon (C /C 0) as afuncon of pore volume; vercal line denotes end of pulse.




sulfonamides (Boxall et al., 2002; Tiele-Bruhn and Aust, 2004;Göbel et al., 2005; Tiele-Bruhn, 2003).

ransport of SDM and OMP was modeled well by the chem-ical nonequilibrium model of the convection–dispersion equation.Neither SDM nor OMP reached sorption equilibrium in the soilcolumns ( f < 1) and were therefore both modeled by rate-limitedsorption and first-order kinetics. Results for the pulse input col-umns yielded mass recoveries >0.90 for the sand and the two soils

with SDM and for the sand with OMP. Te mass recoveries forOMP in the two soils were lower, at values of 0.56 and 0.55 inthe loamy sand and loam soils, respectively, indicating irreversiblesorption or chemical transformation.

Results from this study indicate that both SDM and OMPhave the potential to move through soils, contaminating nearby surface waters and groundwater. Te SDM demonstrates somesorption capacity but is readily desorbed, whereas OMP isadsorbed more strongly and requires more time for desorption.Te pH may be an important factor affecting sorption in thenatural environment. Tis study revealed that SDM and OMPsorption increased and relative mobility decreased in soils withincreasing cation exchange capacity and soil organic matter, anddecreasing pH.

A We wish to acknowledge the assistance provided by Dr. LaurentBahaminyakamwe, Department of Agronomy and Soils, Auburn Uni-versity, with the soil column experiments, and Dr. H.M. Selim, Schoolof Plant, Environmental and Soil Sciences, Louisiana State University,Baton Rouge, LA, with the data analysis. We also wish to acknowledgefunding provided by an EPA-SAR fellowship to S.M. Sanders; Ala-bama Water Resources Institute and USGS; and Environmental Associ-ates at the Academy of Natural Sciences of Philadelphia.

ReferencesBakal, R.S., and M.K. Stoskopf. 2001. In vitro studies of the fate of sulfadi-

methoxine and ormetoprim in the aquatic environment. Aquaculture195:95–102.

Batt, A.L., D.D. Snow, and D.S. Aga. 2006. Occurrence of sulfonamide an-timicrobials in private water wells in Washington County, Idaho, USA.Chemosphere 64:1963–1971.

Bekçi, Z., S. Yoldaş, and M.K. Yurdakoç. 2006. Equilibrium studies fortrimethoprim adsorption on montmorillonite KSF. J. Hazard. Mater. B133:233–242.

Boreen, A.L., W.A. Arnold, and K. McNeill. 2004. Photochemical fate of sulfadrugs in the aquatic environment: Sulfa drugs containing five-memberedheterocyclic groups. Environ. Sci. echnol. 38:3933–3940.

Boreen, A.L., W.A. Arnold, and K. McNeill. 2005. riplet-sensitized photo-degradation of sulfa drugs containing six-membered heterocyclic groups:Identification of an SO2 extrusion photoproduct. Environ. Sci. echnol.39:3630–3638.

Boxall, A.B.A., P. Blackwell, R. Cavallo, P. Kay, and J. olls. 2002. Te sorptionand transport of a sulphonamide antibiotic in soil systems. oxicol. Lett.131:19–28.

Campagnolo, E.R., K.R. Johnson, A. Karpati, C.S. Rubin, D.W. Kolpin, M..

Meyer, J.E. Esteban, R.W. Currier, K. Smith, K.M. Tu, and M. McGee-hin. 2002. Antimicrobial residues in animal waste and water resourcesproximal to large-scale swine and poultry feeding operations. Sci. otalEnviron. 299:89–95.

Casey, F.X.M., G.L. Larsen, H. Hakk, and J. Šimůnek. 2003. Fate and transport of 17B-estradiol in soil–water systems. Environ. Sci. echnol. 37:2400–2409.

Costanzo, S.D., J. Murby, and J. Bates. 2005. Ecosystem response to antibioticsentering the aquatic environment. Mar. Pollut. Bull. 51:218–223.

Dietze, J.E., E.A. Scribner, M.. Meyer, and D.W. Kolpin. 2005. Occurrence of antibiotics in water from 13 fish hatcheries, 2001–2003. Int. J. Environ.

Anal. Chem. 85:1141–1152.Figueroa, R.A., and A.A. Mackay. 2005. Sorption of oxytetracycline to iron ox-

ides and iron oxide-rich soils. Environ. Sci. echnol. 39:6664–6671.

Furusawa, N. 2006. Determining sulphamonomethoxine and its acetyl/hydroxylmetabolites in chicken plasma under organic solvent-free conditions. Anal.Bioanal. Chem. 385:1570–1574.

Furusawa, N., and . Mukai. 1994. Simultaneous high-performance liquidchromatographic determination of residual sulphamonomethoxine, sulph-adimethoxine and their N4–acetyl metabolites in foods of animal origin. J.Chromatogr. A. 677:81–85.

Gao, J., and J.A. Pedersen. 2005. Adsorption of sulfonamide antimicrobialagents to clay materials. Environ. Sci. echnol. 39:9509–9516.

Göbel, A., A. Tomsen, C.S. McArdell, A. Joss, and W. Giger. 2005. Occurrenceand sorption behavior of sulfonamides, macrolides, and trimethoprim inactivated sludge treatment. Environ. Sci. echnol. 39:3981–3989.

Haller, M.Y., S.R. Muller, C.S. McArdell, A.C. Alder, and M.J.-F. Suter. 2002.Quantification of veterinary antibiotics (sulfonamides and trimethoprim)in animal manure by liquid chromatography-mass spectrometry. J. Chro-motogr. A 952:111–120.

Halling-Sørensen, B., and F. Ingerslev. 2000. Biodegradability properties of sul-fonamides in activated sludge. Environ. oxicol. Chem. 19:2467–2473.

Halling-Sørensen, B., G. Sengelov, F. Ingerslev, and L.B. Jensen. 2003. Reducedantimicrobial potencies of oxytetracycline, tylosin, sulfadiazin, streptomy-cin, ciprofloxacin, and olaquindox due to environmental processes. Arch.Environ. Contam. oxicol. 44:7–16.

Hamscher, C., H.. Pawelzick, H. Hoper, and H. Nau. 2004. Antibiotics in soil:Routes of entry, environmental concentrations, fate and possible effects.p. 139–147. In K. Kummer (ed.) Pharmaceuticals in the environment:Sources, fate and risk. 2nd ed. Springer-Verlag, Berlin.

Hamscher, G., B. Priess, and H. Nau. 2006. A survey of the occurrence of vari-

ous sulfonamides and tetracyclines in water and sediment samples originat-ing from aquaculture systems in northern Germany in summer 2005. Arch.Lebensmittelhyg. 57:97–101.

Holm, J.V., K. Rügge, P.L. Bjerg, and .H. Christensen. 1995. Occurrence anddistribution of pharmaceutical organic compounds in the groundwaterdowngradient of a landfill (Grindsted, Denmark). Environ. Sci. echnol.29:1415–1420.

Kishida, K., and N. Furusawa. 2004. Application of shielded column liquidchromatography for determination of sulfamonomethoxine, sulfadi-methoxine, and their N4–acetyl metabolites in milk. J. Chromatogr. A.1028:175–177.

Kolpin, D.W., E.. Furlong, M.. Meyer, E.M. Turman, S.D. Zuagg, L.B.Barber, and H.. Buxton. 2002. Pharmaceuticals, hormones, and otherorganic wastewater contaminants in U.S. streams, 1999–2000: A nationalreconnaissance. Environ. Sci. echnol. 36:1202–1211.

Kreuzig, R., S. Holtge, J. Brunotte, N. Berenzen, J. Wogram, and R. Schulz.2005. est-plot studies on runoff of sulfonamides from manured soils aftersprinkler irrigation. Environ. oxicol. Chem. 24:777–781.

McNamara, N.P., H.I.J. Black, N.A. Beresford, and N.R. Parekh. 2003. Effectsof acute gamma irradiation on chemical, physical and biological propertiesof soils. Appl. Soil Ecol. 24:117–132.

National Library of Medicine. 2006. ChemIDplus Lite. Available at http://chem2.sis.nlm.nih.gov/chemidplus/ (accessed 17 Mar. 2006; verified 12Nov. 2008). Natl. Library of Medicine, Specialized Inf. Serv. Div., Bethes-da, MD.

Oliveira, K.A.D.M., R.C.S. Mendonça, G.V.D. Oliveira, and A.F. Sodré. 2006. Antibiotic resistance of Campylolobacter from automated broiler farms. J.Food Safety 26:82–91.

Porro, I., and P.J. Wierenga. 1993. ransient and steady-state solute transportthrough a large unsaturated soil column. Ground Water 31:193–200.

Rabølle, M., and N.H. Spliid. 2000. Sorption and mobility of metronidazole,olaquindox, oxytetracycline and tylosin in soil. Chemosphere 40:715–722.

Sanders, S.M. 2007. Te fate and transport of the antimicrobials sulfadime-thoxine and ormetoprim in the environment. M.S. thesis. Auburn Univ.,

Auburn, AL.Sassman, S.A., and L.S. Lee. 2005. Sorption of three tetracyclines by several

soils: Assessing the role of pH and cation exchange. Environ. Sci. echnol.39:7452–7459.

Scheytt, .J., P. Mersmann, and . Heberer. 2006. Mobility of pharmaceuticalscarbamazepine, diclofenac, ibuprofen, and propyphenazone in miscible-displacement experiments. J. Contam. Hydrol. 83:53–69.

Scheytt, .J., P. Mersmann, M. Leidig, A. Pekdeger, and . Heberer. 2004. rans-port of pharmaceutically active compounds in saturated laboratory col-




umns. Ground Water 42:767–773.Selim, H.M., J.M. Davidson, and R.S. Mansell. 1976. Evaluation of a two-site

adsorption–desorption model for describing solute transport in soils. p.444–448. In Proc. Summer Comput. Simulat. Conf., Washington, DC.12–14 July 1976. Soc. Comput. Simulat., La Jolla, CA.

Selim, H.M., J.M. Davidson, and P.S.C. Rao. 1977. ransport of reactive solutesthrough multilayered soils. Soil Sci. Soc. Am. J. 41:3–10.

Singh, P., and R.S. Kanwar. 1991. Preferential solute transport through ma-cropores in large undisturbed saturated soil columns. J. Environ. Qual.20:295–300.

Skaggs, .H., and F.J. Leij. 2002. Solute transport: Teoretical background. p.1353–1380. In J.H. Dane and G.C. opp (ed.) Methods of soil analysis.Part 4. Physical methods. SSSA Book Ser. 5. SSSA, Madison, WI.

Skaggs, .H., G.V. Wilson, P.J. Shouse, and F.J. Leij. 2002. Solute transport:Experimental methods. p. 1381–1402. In J.H. Dane and G.C. opp (ed.)Methods of soil analysis. Part 4. Physical methods. SSSA Book Ser. 5.SSSA, Madison, WI.

Tiele-Bruhn, S. 2003. Pharmaceutical antibiotic compounds in soils: A review. J. Plant Nutr. Soil Sci. 166:145–167.

Tiele-Bruhn, S., and M.O. Aust. 2004. Effects of pig slurry on the sorption of sulfonamide antibiotics in soil. Arch. Environ. Contam. oxicol. 47:31–39.

Tiele-Bruhn, S., P. Leinweber, and B. Halling-Sørensen. 2003. Photodegrada-tion and ageing of antibiotics pharmaceuticals on soil surfaces. Poster pre-sented at ENVIRPHARMA, Eur. Conf. on Human and Veterinary Phar-maceuticals in the Environment, Lyon, France. 14–16 Apr. 2003.

Tiele-Bruhn, S., . Seibicke, H.-R. Schulten, and P. Leinweber. 2004. Sorptionof sulfanamide pharmaceutical antibiotics on whole soils and particle-size

fractions. J. Environ. Qual. 33:1331–1342.ipton, D.K., D.E. Rolston, and K.M. Scow. 2003. ransport and biodegrada-

tion of perchlorate in soils. J. Environ. Qual. 32:40–46.olls, J. 2001. Sorption of veterinary pharmaceuticals in soils: A review. Environ.

Sci. echnol. 35:3397–3406.oride, N., F.J. Leij, and M.T. van Genuchten. 1999. Te CXFI code for

estimating tranport parameters from laboratory or field tracer experiment.Version 2.1. U.S. Salinity Lab., Riverside, CA.

U.S. Food and Drug Administration. 1984. Environmental impact analysisreport, Romet-30 (sulfadimethoxine + ormetoprim) medicated premixfor control of certain salmonid diseases. AMD4–72, 01AMD4. USFDA,Rockville, MD.

Westergaard, K., A.K. Müller, S. Christensen, J. Bloem, and S.J. Sørensen. 2001.Effects of tylosin as a disturbance on the soil microbial community. SoilBiol. Biochem. 33:2061–2071.

Wollenberger, L., B. Halling-Sørensen, and K.O. Kusk. 2000. Acute andchronic toxicity of veterinary antibiotics to Daphnia magna. Chemosphere40:723–730.

Zhang, H., and C.H. Huang. 2007. Adsorption and oxidation of fluoroqui-nolone antibacterial agents and structurally related amines with goethite.Chemosphere 66:1502–1512.

fate and transport of sulfadimethoxine and prim in two southeastern united states soils

Documents