Photodegradation of roxarsone in poultry litter leachates

The Science of the Total Environment 302(2003) 237–245

0048-9697/03/$ - see front matter� 2002 Elsevier Science B.V. All rights reserved.PII: S0048-9697Ž02.00322-4

Photodegradation of roxarsone in poultry litter leachates

A.J. Bednar , J.R. Garbarino *, I. Ferrer , D.W. Rutherford , R.L. Wershaw ,a,b b, a,b c c

J.F. Ranville , T.R. Wildemana a

Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, CO 80401, USAa

National Water Quality Laboratory, US Geological Survey, Denver, CO 80225, USAb

National Research Program, US Geological Survey, Denver, CO 80225, USAc

Received 28 April 2002; accepted 2 August 2002

Abstract

Arsenic compounds have been used extensively in agriculture in the US for applications ranging from cottonherbicides to animal feed supplements. Roxarsone(3-nitro-4-hydroxyphenylarsonic acid), in particular, is used widelyin poultry production to control coccidial intestinal parasites. It is excreted unchanged in the manure and introducedinto the environment when litter is applied to farmland as fertilizer. Although the toxicity of roxarsone is less thanthat of inorganic arsenic, roxarsone can degrade, biotically and abiotically, to produce more toxic inorganic forms ofarsenic, such as arsenite and arsenate. Experiments were conducted on aqueous litter leachates to test the stability ofroxarsone under different conditions. Laboratory experiments have shown that arsenite can be cleaved photolyticallyfrom the roxarsone moiety at pH 4–8 and that the degradation rate increases with increasing pH. Furthermore, therate of photodegradation increases with nitrate and natural organic matter concentration, reactants that are commonlyfound in poultry-litter-water leachates. Additional photochemical reactions rapidly oxidize the cleaved arsenite toarsenate. The formation of arsenate is not entirely undesirable, because it is less mobile in soil systems and less toxicthan arsenite. A possible mechanism for the degradation of roxarsone in poultry litter leachates is proposed. Theresults suggest that poultry litter storage and field application practices could affect the degradation of roxarsone andsubsequent mobilization of inorganic arsenic species.� 2002 Elsevier Science B.V. All rights reserved.

Keywords: Photodegradation; Roxarsone; Leachates

1. Introduction

Various arsenicals have been used in agriculturein the US for more than 100 years(Onken andHossner, 1996). Historically, inorganic arseniccompounds were used as pesticides, and methyl-ated arsenicals are still used widely as herbicides

*Corresponding author. Tel.:q1-303-236-3945; fax:q1-303-236-3499.

E-mail address: [email protected](J.R. Garbarino).

for cotton production(Perkins and Brushwood,1991; Hiltbold et al., 1974; Marin et al., 1993).More complex organoarsenical species, often hav-ing substituted phenyl rings in their structure, arebeing used to control intestinal parasites in poultryand swine(Moore et al., 1998). These compoundsare nontoxic to domestic livestock, and are excret-ed unchanged from the animals(Moore et al.,1998; Morrison, 1969; Abdo, 1989). However,once these compounds are introduced into the

238 A.J. Bednar et al. / The Science of the Total Environment 302 (2003) 237–245

environment through litter storage in weather-exposed windrows or field application as a fertil-izer, degradation reactions, biotic and abiotic,potentially can convert them into more mobile andtoxic arsenic species. These arsenical degradatescan then contaminate local ground and surfacewater used for drinking-water supplies. Roxarsone(3-nitro-4-hydroxyphenylarsonic acid) is a com-mon arylarsenical used in broiler poultry produc-tion that can be a source of arsenic to theenvironment through field application of litter.

In 2000, more than 8.2 billion broilers wereraised in the US(US Department of Agriculture,2001). Using a common feed formulation, eachbird will excrete;150 mg of roxarsone during its7-week growth period(Alpharma, personal com-munication, 1999). These numbers translate intomore than 350 000 kg of arsenic being introducedinto the environment in the US each year as aresult of applying nearly all the poultry litter asfertilizer. This is an upper-limit estimate becausesome broilers may not be fed roxarsone, or maybe fed roxarsone at a reduced rate(Chapman andJohnson, 2002). Land application of poultry litteris localized because of the expense of transportingthe litter. This results in a large amount of arsenicbeing applied to a small area. Primary areas ofinterest in the US include the Delmarva peninsulaof Delaware, Maryland and Virginia; Ozark plateauof Arkansas, Oklahoma, and Missouri; as well asGeorgia and the Carolinas(US Department ofAgriculture, 2001). These areas represent histori-cally high-volume poultry production regions, and,therefore, will be the most affected by the repeateduse of poultry litter as fertilizer.

Understanding the degradation of roxarsone iscritical to understanding the fate of arsenic-con-taining byproducts that have different toxicitiesand mobilities in the environment. Roxarsone isfound unchanged in fresh, dry litter, yet is notfound in litter-storage windrows that have beenexposed to precipitation and sunlight, in soilsamended with litter, or in local surface and ground-water (Garbarino et al., unpublished research,2002). These findings suggest that the degradationof roxarsone is rapid in the environment andinfluenced by poultry litter management practices.Biotic processes associated with composting most

likely will dominate the degradation of roxarsonewhen the litter is stored in uncovered windrowsexposed to precipitation. Composting experimentshave been performed that suggest roxarsone deg-radation in litter compost and water-saturated litteris controlled by anaerobic microorganisms(Gar-barino et al., 2001). Abiotic hydrolysis reactionsalso might occur; however, such reactions seem tobe slow in many composting experiments thathave been investigated.

Many regions of the US that are high poultryproduction areas(particularly in the Southeast)also receive large amounts of rainfall. This canproduce pools of leachate at the base of theuncovered windrows. Arsenic, and other watersoluble constituents in the litter, accumulate inthese pools, where abiotic reactions, particularlyphotoinduced ones, might occur. Similar processesalso could occur in irrigated fields amended withpoultry litter.

The purpose of the present study is to investigatethe photodegradation processes that could degraderoxarsone in poultry litter leachates, the pH rangeunder which such processes can occur, and theeffects of concomitant constituents present in thelitter leachates. In addition, the photodegradationbyproducts were identified and used to suggest apossible mechanism for the degradation.

2. Experimental details

All chemicals used in this study were reagentgrade or higher purity and used without furtherpurification; the deionized water used had a resis-tivity of 18.3 MV cm. Arsenic speciation in thephotodegradation solutions and the poultry litterleachates was determined using two recently devel-oped HPLC-ICP-MS techniques. A version of themethod of Jackson and Bertsch(2001) was usedfor identification of five arsenic species, whereasa method developed by the present authors for therapid determination of arsenite, arsenate, and rox-arsone was used for the photolysis kinetic experi-ments(Bednar et al., 2001, 2002). Both methodshave detection limits of less than 1mg arsenicylfor each species.

Organic photodegradation byproducts were iden-tified using a Bruker Daltonics Esquire Electro-

239A.J. Bednar et al. / The Science of the Total Environment 302 (2003) 237–245

Table 1Concentrations of selected water soluble constituents in poultry litter leachates

wNO xy3 wNO xy

2 wNH x3 wAsx pH

Laboratory leachate of roxarsone-containing litter 1–40 1–20 60–150 1100 7–8Laboratory leachate of nonroxarsone-containing litter 1–40 1–20 60–150 30 7–8Natural leachate collected in the field from roxarsone-containing litter 10 10 75 81 6–7

Concentrations of nitrogen species given in mgyl; concentrations of arsenic given inmgyl.

spray Ionization-Ion Trap-Tandem Mass Spectro-meter (ESI-MS-MS) operated in negative ionmode(Ferrer et al., 2002). The sample(in deion-ized water or carbonate buffer) and mobile phase(acetonitrile) were introduced into the ion trap bya Hewlett Packard 1100 HPLC at a flow rate of0.2 mlymin. The MS was operated in full-scanmode (myz 50–300), and the ionization sourcepotential was optimized for fragmentation andsignal intensity.

Poultry litter leachates were prepared in thelaboratory by shaking 20 g of dry, roxarsone-freelitter from a local pullet farm in 200 ml ofdeionized water, followed by filtration to 0.45mm.The litter leachate was spiked with roxarsone(500mgyl arsenic as roxarsone–aliquots were dilutedfor analysis) prior to photolysis experiments toprevent contamination from degradation byprod-ucts in a naturally roxarsone-containing litter. Typ-ical chemical concentrations and pH for thelaboratory leachates are compared to a naturalleachate collected in the field and listed in Table1. Nitrate, nitrite, and ammonia concentrationswere determined in the leachates using standardcolorimetric and cadmium reduction techniques(Fishman, 1993).

Photolysis experiments were conducted in aRayonet photochemical reactor with the capacityof up to 16 lightsource tubes, each having 20-Woutput and a 300-nm maximum emission line. Sixtubes were used for the kinetic experiments, whileall 16 tubes were installed for the organic byprod-uct identification experiment. All samples wereirradiated in quartz vessels open to the atmosphere.Aliquots were collected at 1-h intervals during thephotolysis experiments, diluted in brown glassvials, and analyzed immediately for arsenite, arse-nate, and roxarsone.

Solutions with only roxarsone(the control) orroxarsone with nitrate were prepared in reagentwater to eliminate the problems associated withidentifying degradation byproducts in the complexlitter leachate. All the test solutions for the kineticstudies contained roxarsone at 500mg arsenicyl,with pH adjusted from 4 to 8 using a sodiumcarbonate buffer(roxarsone stock solutions arenaturally acidic, pH;3–4). Nitrate concentrationwas varied from 1 to 30 mgyl using sodium nitrate.Natural organic matter(NOM) in the litter leachateexperiments was diluted between 10 and 1000 mgcarbonyl. Higher concentrations of roxarsone(;60 mg roxarsoneyl) and nitrate(300 mgyl), aslisted in Table 2, and longer irradiation times(upto 15 h) were required to produce enough organicdegradation byproduct for analysis by ESI-MS-MS.

3. Results and discussion

In deionized water matrices containing onlyroxarsone, the disappearance of roxarsone and theappearance of arsenite and arsenate indicated thatphotolytic degradation was taking place. Further-more, there was some evidence of denitration(lossof the nitro group) from roxarsone leading toformation of phenol as a major organic degradationproduct.

Several reagent-water matrices containing rox-arsone with nitrate at environmentally relevantconcentrations were prepared and irradiated toestablish possible abiotic reaction mechanisms forthe photodegradation of roxarsone. A possiblemechanism for the photodegradation of roxarsoneis based on the photolytic reductive cleavage ofthe As–C bond. Ionization of the phenolic groupproduces, through resonance, an electron-rich car-


Table 2Concentrations of nitrate(in mgyl) and roxarsone(as arsenic inmgyl) used during the degradation studies

wNO xy3 wAsx pH

Photodegradation experiments 1–30 500 4–8Organic moiety identification experiments 300 17 000 8–9

Scheme 1. Proposed photolytic degradation mechanism of roxarsone in the presence of nitrate.

bon bonded to the arsenic atom, which promotesthe bond cleavage by photoactivation of the ring(Scheme 1). The formation of an electron-richAs–C bond promotes the elimination of arsenitefrom the ring that is subsequently oxidized toarsenate by photolytically produced hydroxyl freeradicals(Emett and Khoe, 2001; Hug et al., 2001).The presence of the electron-rich As–C bond andring photoactivation is supported by the fact thata colorless solution of roxarsone becomes yellowat pH)6 indicating visible light absorbance bythe activated ring. Photoactivation of the ring isrequired for the proposed reaction mechanism,because the As–C bond does not cleave whenshielded from light, even at high pH()12), or inthe presence of other matrix components, such ascarbonate, nitrate, nitrite, or NOM.

The effect of pH on the rate of roxarsonephotodegradation in deionized water was investi-gated because of the dependence of the proposed

mechanism on the ionization of the phenolic group.Experiments conducted at pH 4–8 indicated thathigher pH increased the degradation reaction rateby deprotonating the phenolic group on roxarsone(pK s5.8). Rates of photodegradation increaseda

by a factor of 2 over the pH range 4–8(4=10 –y7

8=10 mol roxarsoneylyh).y7

The photodegradation of roxarsone at pH;6with three different concentrations of nitrate isshown in Fig. 1. The rate of reaction was approx-imately 1=10 mol roxarsoneylyh (more than 3y6

times faster than the rate when no nitrate waspresent) for the lowest nitrate concentration of 3mgyl. Increasing the nitrate concentration to 30mgyl increased the degradation rate to 5=10y6

mol roxarsoneylyh. The data(Fig. 1) also showthat the degradation of roxarsone was faster thanthe oxidation of arsenite at low nitrate concentra-tion. Nitrate photolysis has two major roles, asshown in Scheme 1. The formation of nitrite is


Fig. 1. Roxarsone photolytic degradation in deionized water showing the effect of various nitrate concentrations. Nitrate concentra-tions: as3 mgyl, bs15 mgyl, cs30 mgyl.

required to replace the arsenite leaving group onthe ring, and arsenite is oxidized by hydroxy freeradicals formed during nitrate photolytic reduction(Hug et al., 2001; Sharpless and Linden, 2001).The photoreduction of nitrate has been simplifiedhere, but actually involves a complex series ofreactions including the oxidation of water tohydroxyl free radical(Dzengel et al., 1999; Sharp-less and Linden, 2001). At the higher nitrateconcentration of 30 mgyl, substantial concentra-tions of arsenite are seen only initially when thereis a large amount of roxarsone producing arsenitefaster than the nitrate photolysis reactions canoxidize.

Substitution of nitrite for the arsenite leavinggroup on the organic moiety creates 2,4-dinitro-phenol as the primary organic degradate in nitrate-containing solutions, as shown in Scheme 1. AnESI-MS-MS spectrum, used to identify the organicmoieties in nitrate-containing solutions, is shownin Fig. 2. The ions(in negative ion mode) at myz183 and 167 represent 2,4-dinitrophenol and

meta-dinitrobenzene, respectively. The small peakat myz 93 corresponding to phenol was morepronounced in nonnitrate-containing solutions, thusindicating the denitration of the ring during pho-tolysis. The lack of a substantial peak atmyz 154indicates that hydroquinone structures were notformed from direct or photoinduced hydrolysis ofthe As–C bond. The presence ofmeta-dinitroben-zene does not necessarily indicate photolyticallyinduced loss of the phenolic group of roxarsonebut rather is an ionization fragment of 2,4-dinitro-phenol formed in the MS source. However, thespectrum is further evidence of addition of nitriteto the ring during the photolysis experiments. Thepeak atmyz 147 was caused by an unidentifiedcontaminant in the sodium carbonate buffer usedin the photolysis experiments.

Aqueous solutions of roxarsone, with and with-out NOM, were irradiated to investigate the effectof NOM on the photochemical degradation ofroxarsone. The photodegradation of roxarsone in1:10 and 1:1000 dilutions of the laboratory poultry


Fig. 2. ESI-MS spectrum showing possible photolytic degradation products of roxarsone. Dinitro-organic moieties are the primarydegradation products.

litter leachate(;1–100 mg carbonyl) is shownin Fig. 3. The degradation rate was about equal inboth leachate dilutions, yet the addition of nitrate(15 mgyl) substantially increased the degradationrate in the more dilute leachate. This effect sug-gests that the NOM concentration in the leachatemoderates the rate of reaction. The decrease inphotodegradation might have resulted from thescavenging of reactive species by the NOM or bythe attenuation of the light intensity in the quartzreaction tube by the NOM(Boule et al., 1999;Thorn, 2002). Use of Saint Francis River water(surface water from an agricultural region ofNortheastern Arkansas that has 5–7 mg carbonyl)slightly increases the rate of photodegradation overthe roxarsone control shown in Fig. 4. The variableeffect of NOM concentration likely will play arole in natural systems that are influenced byagricultural practices because dilution of nitrateand NOM by irrigation or precipitation could affectthe roxarsone degradation process. Identificationof the organic moiety degradates in the litterleachate and river water solutions by ESI-MS-MS

was not possible because of the high concentrationof NOM, however, arsenic speciation results byHPLC-ICP-MS support the previously describedmechanism.

Analytical results for samples collected frompools of natural leachate adjacent to uncoveredpoultry-litter-storage windrows in the ArkansasOzark Plateau region might also support the pro-posed roxarsone degradation mechanism. The totalarsenic concentration in the leachate was greaterthan 80mgyl arsenic(Table 1). The sample wascollected during the winter when the air tempera-ture was below 08C; the low temperature mostlikely inhibited microbial processes. Approximate-ly 50% of the arsenic in the natural leachate wasassociated with unidentified organoarsenic com-pounds. However, more than one-third of thearsenic was present as arsenite, with the remainingarsenic being arsenate and monomethylarsonate.The presence of substantial amounts of arsenite,which is a minor product in most laboratorypoultry-litter composting experiments(Garbarinoet al., unpublished research, 2002), possibly could


Fig. 3. Roxarsone photolytic degradation showing the effects of poultry litter-derived NOM and added nitrate. Concentrations As1:10 dilution of leachateq15 mgyl nitrate, AAs1:10 dilution of leachate; Bs1:1000 dilution of leachateq15 mgyl, BBs1:1000dilution of leachate.

be explained by photolysis reactions similar tothose proposed, which have been shown to producearsenite as an intermediate.

4. Conclusions

A possible mechanism for the degradation ofroxarsone in poultry litter leachates has beendescribed involving photoinduced cleavage of theAs–C bond and elimination of arsenite from thering moiety. The reaction pathways, intermediates,and degradation products have been investigated

and confirmed using HPLC-ICP-MS and ESI-MS-MS. Reactants, such as nitrate and NOM, that arenaturally present in litter leachates, have beenshown to be important in the photodegradationprocess. The rate of photodegradation was directlyproportional to the nitrate concentration. The pho-todegradation rate also was increased in the pres-ence of low concentrations of NOM, similar tothose found in rivers of agricultural regions of theSoutheastern US(;1–10 mg organic carbonyl;Leenheer, 1982). However, at higher concentra-tions of NOM (;1000 mg carbonyl), similar to


Fig. 4. Roxarsone photolytic degradation showing the effects of NOM derived from a river in Northeastern Arkansas. asroxarsonein Saint Francis river water; bsroxarsone in reagent water(control).

those found in the poultry litter leachates, thephotodegradation was extremely slow. Based onthese findings, it is likely that practices used forlitter storage and application to fields will impactthe rate of degradation. While biotic processesmost certainly will dominate roxarsone degradationin composting litter, abiotic photolysis might besignificant in litter leachates associated with litterstorage. In such cases, photodegradation could beimportant in the ultimate fate of roxarsone in theenvironment.

Acknowledgments

The authors wish to thank Colleen Rostad ofthe US Geological Survey, National Research Pro-gram, for providing the Rayonet Photoreactor usedfor the photodegradation studies and Billy Justusof the US Geological Survey, Arkansas District,for assistance in collecting field samples. Thisresearch is funded by the US Environmental Pro-tection Agency, the US Geological Survey Nation-

al Water Quality Laboratory, and a Sigma XiGrant-in-Aid of Research from the National Acad-emy of Sciences. The use of trade, product, orfirm names in this report is for descriptive purposesonly and does not imply endorsement by the USGovernment.

References

Abdo K. Toxicology and carcinogenesis studies of roxarsonein F344yN rats and B6C3F1 mice. Research Triangle Park,National Toxicology Program 1989; TR 345:198.

Bednar AJ, Ranville JF, Wildeman TR, Garbarino JR, LamothePJ, Smith KS, McCleskey BR, Ball JW, Nordstrom DK.Analytical speciation methods for dissolved inorganic andorganic arsenic species. In: Div. Envir. Chem. Extend. Abst.Preprints. 221st Annual meeting of the American ChemicalSociety, vol. 41. San Diego, CA: 2001. p. 248–252.

Bednar AJ, Garbarino JR, Ranville JF, Wildeman TR. Analysisand occurrence of organoarsenicals used in agriculture. In:Div. Agrochem. Oral Presentation. 223rd Annual meetingof the American Chemical Society, Orlando, FL: 2002.

Boule P, Bolte M, Richard C. Phototransformations induced inaquatic media by NOyNO , Fe(III ) and humic substances.y y

3 2


In: Hutzinger O, editor. The handbook of environmentalchemistry, vol. 2, Springer-Verlag, 1999.

Chapman HD, Johnson ZB. Use of antibiotics and roxarsonein broiler chickens in the USA: Analysis for the years 1995to 2000. Pollut Sci 2002;81:356–364.

Dzengel J, Theurich J, Bahnemann DW. Formation of nitro-aromatic compounds in advanced oxidation processes: Pho-tolysis versus photocatalysis. Environ Sci Technol1999;33:294–300.

Emett MT, Khoe GH. Photochemical oxidation of arsenic byoxygen and iron in acidic solutions. Water Res2001;35:649–656.

Ferrer I, Furlong ET, Thurman ET. Ion-trap LCyMSyMS as aresearch tool for the identification of environmentalunknowns. In: Div. Environ. Chem. Extend. Abst. Preprints.223rd Annual meeting of the American Chemical Society,vol. 42. Orlando, FL: 2002. p. 352–358.

Fishman MJ. Methods of analysis by the U.S. GeologicalSurvey National Water Quality Laboratory—Determinationof inorganic and organic constituents in water and fluvialsediments. Open-File Rep 1993;93–125:119–170.

Garbarino JR, Rutherford DW, Wershaw RL. Degradation ofroxarsone in poultry litter. In: Arsenic in the environment,Proceedings of the US Geological Survey Workshop.(Den-ver, CO) February 21–22, 2001, http:yywww.brr.cr.usgs.govyarsenic.

Hiltbold AE, Hajek BF, Buchanan GA. Distribution of arsenicin soil profiles after repeated applications of MSMA. WeedSci 1974;22:272–275.

Hug SJ, Canonica L, Wegelin M, Gechter D, VonGunten U.Solar oxidation and removal of arsenic at circumneutralpH in iron containing waters. Environ Sci Technol2001;35:2114–2121.

Jackson BP, Bertsch PM. Determination of arsenic speciationin poultry wastes by IC-ICP-MS. Environ Sci Technol2001;35:4868–4873.

Leenheer J. United States Geological Survey Data InformationService. 1982;1:355–356.

Marin AR, Pezeshki SR, Masschelen PH, Choi HS. Effect ofdimethylarsenic acid(DMAA ) on growth, tissue arsenic,and photosynthesis of rice plants. J Plant Nutr 1993;16:865–880.

Moore PA, Daniel TC, Gilmour JT, Shreve BR, Edwards DR,Wood BH. Decreasing metal runoff from poultry litter withaluminum sulfate. J Environ Qual 1998;27:92–99.

Morrison JL. Distribution of arsenic from poultry litter inbroiler chickens, soil, and crops. J Agric Food Chem1969;17:1288–1290.

Onken BM, Hossner LR. Determination of arsenic species insoil solution under flooded conditions. Soil Sci Soc Am J1996;60:1385–1392.

Perkins HH, Brushwood DE. Arsenic acid desiccant residuesin cotton. Trans ASAE 1991;34:1629–1632.

Sharpless CM, Linden KG. UV photolysis of nitrate: effectsof natural organic matter and dissolved inorganic carbonand implications for UV water disinfection. Environ SciTechnol 2001;35:2949–2955.

Thorn KA. Reduction and incorporation of nitrate nitrogeninto aquatic NOM upon UV irradiation examined by N-15NMR. In: Div. Envir. Chem. Extend. Abst. Preprints. 223rdAnnual meeting of the American Chemical Society, vol. 42.Orlando, FL: 2002. p. 537–542.

US Department of Agriculture National Agricultural StatisticsService. Broiler production by States. 2001; http:yywww.usda.govynass.

Documents

Photodegradation of roxarsone in poultry litter leachates