Hydroxyl radical-induced degradation of fenuron

RESEARCH ARTICLE

Hydroxyl radical-induced degradation of fenuron in pulse and gamma radiolysis:kinetics and product analysis

Krisztina Kovács &Viktoria Mile &Tamás Csay &Erzsébet Takács &LászlóWojnárovits

Received:30January 2014/Accepted:11June 2014/Published online:26June 2014#Springer-Verlag Berlin Heidelberg 2014

Abstract Radiolytic reactions of phenylureas were studied in detail with fenuron model compound in dilute aqueous solu-tions using pulse radiolysis for detection of the intermediates,gamma radiolysis with UV –Vis and HPLC-MS techniques for analysis of the final products.The kinetics of oxidation was followed by COD,TOC and toxicity measurements.During radiolysis of aerated solutions hydroxyl radical (?OH),e aq ?,H ?and O 2??/HO 2?reactive intermediates are produced,the deg-radation of solute takes place practically entirely through ?OH reactions.Therefore,the product distribution is similar to the distributions reported in other advanced oxidation processes with ?OH as main reactant.?OH mainly reacts with the aro-matic ring,forming cyclohexadienyl radical as an intermedi-ate.This radical in pulse radiolysis has a wide absorption band in the 310–390nm wavelength range with a maximum at 350nm.Cyclohexadienyl radical reacts with dissolved O 2with a rate coefficient of ～4×108mol ?1dm 3s ?1forming peroxy radical.The latter may eliminate HO 2?giving phenols or undergoes fragmentation.The one-electron oxidant ?OH on average induces more than two-electron oxidations.The toxicity first increases with absorbed dose,then decreases.This increase is partly due to phenols formed during the first degradation period.

Keywords Fenuron .Hydroxyl radical .Hydrated electron .Advanced oxidation processes .Degradation .Radiation technology

Introduction

A large variety of techniques,with a common name “ad-vanced oxidation processes ”(AOP)have been developed to degrade harmful,badly biodegradable organic contaminants in water.Most of these processes were applied only in labo-ratory experiments,just few of them were tested also on larger scale,e.g.radiation technology was used at the level of 10,000m 3/day (Radiation Processing 2007).

In the case of irradiation technology,the decomposition of water supplies the reactive intermediates,no additives are needed.Three reactive intermediates are produced during irradiation of water,namely hydroxyl radical (?OH),hydrated electron (e aq ?)and hydrogen atom (H ?)(Reaction (1))(Buxton et al.1988;Spinks and Woods 1990).The yields (G -values)are well-known:in pure water the yields of ?OH,e aq ?and H ?are 2.8×10?7,2.8×10?7and 0.6×10?7mol J ?1,respectively.Under practical conditions,when dissolved O 2is present (aerated solution)e aq ?and H ?transform to the O 2??/HO 2?pair in Reactions (2)and (3)(p K a (O 2??/HO 2?)=4.8).By systemati-cally changing the reaction conditions and by that the compo-sition of reactive radical intermediates,we may obtain infor-mation about the individual reactions selectively in laboratory experiments (Buxton et al.1988).The reactions of ?OH are often investigated in N 2O saturated solution (0.025mol dm ?3)in order to convert e aq ?to ?OH in Reaction (4).In N 2O bubbled solution,the active intermediates are ?OH and H ?,their yields being 5.6×10?7and 0.6×10?7mol J ?1,respectively.Radiolysis of N 2O saturated solutions can be regarded as the cleanest source of ?OH (Wojnárovits and Takács 2013).

e aq ?tO 2→O 2??

k ?1:9?1010mol ?1dm 3s ?1

à

á

e2T

e1T

Responsible editor:Roland Kallenborn

K.Kovács :V .Mile :T.Csay :E.Takács :L.Wojnárovits (*)Institute for Energy Security and Environmental Safety,Centre for Energy Research,Hungarian Academy of Sciences,Budapest,Hungary

e-mail:https://www.sodocs.net/doc/877028056.html,szlo@energia.mta.hu

Environ Sci Pollut Res (2014)21:12693–12700DOI 10.1007/s11356-014-3197-9

H ?tO 2→HO 2?

k ?2:1?1010mol ?1dm 3s ?1

à

á

e3Te aq ?tN 2O tH 2O →?OH tOH ?tN 2

k ?9:1?109mol ?1dm 3s ?1

àáe4T

?

OH tCH 3eT3COH →?CH 2CH 3eT2COH tH 2O k ?6?108mol ?1dm 3s ?1

àáe5T

HCO 2?t?OH →CO 2??tH 2O

k ?3?109mol ?1dm 3s ?1

à

á

e6T

CO 2??tO 2→CO 2tO 2??

k ?4:2?109mol ?1dm 3s ?1

à

á

e7T

In usual practice,the reactions of e aq ?are investigated in deoxygenated (N 2bubbled)solutions containing also 0.1?1.0mol dm ?3tert -butanol (Reaction (5)).Under such conditions,in addition to e aq ?and H ?,?CH 2(CH 3)2COH radicals are also present and may react with the substrate.O 2saturated solution containing sodium-formate offers a possibility for selective investigation of O 2??/HO 2?reac-tions (Yamasita et al.2008).In such solution,all three intermediates,e aq ?,?H (Reactions (2)and (3)),and ?OH (Reactions (6)and (7))are converted to O 2??/HO 2?.

Herbicides,such as phenylureas,are highly persistent com-pounds,they are regularly detected in surface waters in con-centrations up to several μg dm ?3(Mazellier et al.2007).The decompositions of phenylureas were investigated by direct photolysis,in order to study the degradation by sunlight (Canle L.et al.2001;Amine-Khodja et al.2004;Canle Lopez et al.2005).However,they exhibit little absorbance above 300nm,therefore,their degradation by direct solar light is slow (Brahmia et al.2002).The reactions,initiated by hy-droxyl radicals,were also studied in several AOP.In TiO 2photocatalysed,H 2O 2/UV and in different versions of the Fenton reaction (Richard and Bengana 1996;Brahmia et al.2002;Canle Lopez et al.2005;Mazellier et al.2007;Oturan et al.2010;Dao and De Laat 2011)hydroxylation,demethyl-ation and methyl group oxidation to aldehyde or carboxylic acid were observed.In principle,the products of ?OH reactions should not depend on the way of hydroxyl radical production.In practice,some differences are observed between the results obtained by different AOP techniques,probably due to reac-tions of reactive intermediates other than ?OH.

When searching for the literature,we did not find publica-tions on radiolysis of fenuron solutions,except the works of Canle Lopez et al.(2001)in which they used pulse radiolysis for oxidation and reduction by SO 4??or e aq ?,respectively.Zhang et al.(2008)published a paper on γ-radiolysis of

diuron.According to their opinion,both ?OH and e aq ?reac-tions contribute to diuron degradation.

Here on the example of fenuron by using both pulse radiolysis and γ-radiolysis,we identify the role of different water radiolysis intermediates in the degradation of the solute,showing how the intermediate radicals transform to stable final products,we also evaluate the absorbed dose dependence of oxidation and toxicity.We are convinced that these results are important not only for radiation chemis-try and radiation technology,but also have some relevance to other AOP.

Experimental

Fenuron and the other chemicals were obtained from Spectrum-3D or Carlo Erba.Pulse radiolysis investigations were carried out using 800ns pulses of accelerated electrons,dose/pulse 20Gy/pulse (J kg ?1/pulse)(Takács et al.2000).In pulse radiolysis,fast energy absorption from short pulses of accelerated electrons initiates the degradation of water.The reactive intermediates of water radiolysis are formed during the pulse and react with the solute molecules on a longer timescale.With the optical detection applied,the time depen-dence of absorption of intermediates can be followed.γ?irradiations were carried out by a 60Co facility with a constant dose rate of 11.5kGy h ?1.Because of this constant dose rate,the absorbed dose is proportional to the time of irradiation.All experiments were carried out at room temperature in unbuf-fered solutions.The solutions were saturated with appropriate gases.

UV ?Vis spectra were taken using a JASCO 550UV ?Vis spectrophotometer with 1cm cell.The products were ana-lyzed by an Agilent Technologies 6410Triple Quad LC/MS equipped with ESI source using both mass spectrometric (total,selective and product ion monitoring)and diode array (DAD)detections.Separations were made with Phenomenex XB-C18column (particle size 2.6μm,length 100mm,thermostated to 25°C)using H 2O (A)and methanol (B)eluents in isocratic condition,running time 15min,flow rate 0.25cm 3min ?1.

The chemical oxygen demand (COD)was measured ac-cording to ISO Standard 6060:1989using a Behrotest TRS 200system.The technique involves boiling of 10cm 3sam-ples at 148±3°C for 2h in 8mol dm ?3H 2SO 4solution with the introduction of K 2Cr 2O 7as an oxidizing agent,Ag 2SO 4as a catalyst and HgSO 4for removing chlorides.The non-reacted Cr 2O 72?is removed by titration with Mohr salt,using ferroin indicator.In total organic carbon content (TOC)measure-ments,Shimadzu TOC-L CSH/CSN,in Microtox acute tox-icity tests,LANGE LUMIStox 300equipment were used.The latter test measures the luminescence inhibition of Vibrio Fischeri bacteria after incubation for 30min.Hydrogen

peroxide was analyzed through spectrophotometric detection of Cu(I)/2,9-dimethyl-1,10-phenanthroline(Merck)com-plexes in Cu(II)containing solution(Kosaka et al.1998). Results and discussion

Pulse and continuousγradiolysis with optical detection Reactions of?OH,pulse radiolysis

The reactions of?OH were studied in N2O saturated solution where?OH and H?are the reactive intermediates in～9:1ratio. (H?,similarly to?OH,reacts with most organic molecules in addition or abstraction reaction).

As inset b in Fig.1shows,the absorbance signal at350nm builds-up quickly and afterwards decays very slowly.By recording the signals at many wavelengths between300–430nm and by collecting the absorbance values at5,70and 360μs after the pulse,the spectra shown in Fig.1a were obtained.In these spectra,broad absorption bands appear between310and390nm,with maxima at350nm.Below 300nm,the absorbance of fenuron(Fig.2)and fenuron depletion disturbs the measurements.The shapes of the spec-tra at5and70μs are similar(Fig.1a),indicating no change in intermediates during decay.

The first rising part(the first5–10μs)of the absorbance-time curve was used to determine the pseudo-first-order rate coefficients of the?OH+fenuron reaction(Fig.1b)by fitting to the experimental absorbance points.The pseudo-first-order rate coefficients were measured at several concentrations, and the slope of the pseudo-first-order rate coefficient-fenuron concentration plot(Fig.1c)supplied the second-order rate coefficient:(8.3±0.6)×109mol?1dm3s?1.Our value agrees with those determined in Fenton,7.4×109and 7.0×109mol?1dm3s?1(Acero et al.2002;Dao and De Laat2011),or in H2O2/UV reaction,(7±1)×109mol?1dm3s?1 (Mazellier et al.2007).These coefficients are close to the theoretical maximum,the diffusion controlled rate coefficient: 1.1×1010mol?1dm3s?1(Wojnárovits and Takács2013).

The hydroxyl radical may attack fenuron by addition to any of the carbon atoms in the ring or H-abstraction from-CH3or-NH-forming radicals shown in Scheme1.In?OH-addition to ring,hydroxycyclohexadienyl radical was suggested as an intermediate(Gallard and De Laat2001;Canle Lopez al. 2005;Mazellier et al.2007).Generally,cyclohexadienyl rad-icals have characterless absorption bands between310?390nm withεmax3,000?5,000mol?1dm3cm?1.Here,as-suming all?OH reacting with the ring,εmax4,000mol?1dm3 cm?1is calculated.The spectrum on Fig.1a is similar to that obtained in pulsed248nm photolysis where H2O2splitting supplied?OH.Canle Lopez et al.(2005)also assigned the spectrum to cyclohexadienyl intermediates.Ipso-addition(I) is probably of low importance due to thermodynamic and steric reasons(Singh et al.2001).The electron donating urea side chain is expected to direct?OH to o-and p

-positions

Fig.1Transient absorption spectra obtained in pulse radiolysis of 5×10?4mol dm?3N2O saturated solution(a),red circle5,blue triangle70and green left pointing triangle360μs after the pulse. Insets transient signals at350nm in N2O and N2O/O2saturated solution(b).Concentration dependence of the pseudo-first-order rate coefficient of absorbance build-up at350nm(c).The intensity of transient spectrum at360μs in Fig.1a around300nm is higher than expected based on the shorter time spectra.This increased absorbance,in agreement with the UVabsorption spectra ofγ-irradiated solutions(Fig.2), is due to the absorbance of end-products(hydroxylated molecules)which build-up parallel to the decay of intermediates

(II and IV).However,as three ring monohydroxylated products were observed (see Section 3.2.),m -addition (III)should also occur.?

OH may abstract H-atom from -NH-forming aminyl radical (V).Such radicals were observed in photoionization experiments and in SO 4??reaction with phenylureas.In the case of fenuron,λmax 425nm and εmax 2,000mol ?1dm 3cm ?1are suggested (Canle L.et al.2001).The absence of the band at 425nm shows low importance of aminyl radical formation reaction.Brahmia et al.(2002)suggested aminyl radical for-mation by H 2O elimination from hydroxycyclohexadienyl radical.Although similar elimination was suggested e.g.for aniline (Solar et al.1986),this possibility may be ruled out for fenuron.

The radical produced in H-abstraction from the methyl group (VI)may have light absorption below 300nm;this

absorbance is masked by fenuron absorbance.Based on the intense absorbance of cyclohexadienyl radical in the transient spectrum,on the abundance of ring-hydroxylated products and on the efficient degradation (see Section 3.2.),we assume that the preferred reaction is addition to the ring.

Pulse radiolysis measurements were repeated using the mixture of N 2O/O 25/1as the bubbling gas (Fig.1b ).In that experiment,the fenuron radicals formed in ?OH reaction reacted with dissolved oxygen and the decay of absorbance signal became faster (compare the time behaviour of curves marked by N 2O and N 2O/O 2).Peroxy radicals have light absorption below 300nm.By fitting to the decay curve at 350nm,the pseudo-first-order rate coefficient of the organic radical+O 2reaction was obtained,and using the O 2concen-tration estimated from the solubility (2.5×10?4mol dm ?3),the second-order rate coefficient of the reaction was calculated.The decay suggested a rate coefficient of ～4×108mol ?1dm 3s ?1for the reaction of intermediate with dissolved O 2.In a work of Fang et al.(1995),the rate coefficients with O 2for a large number of cyclohexadienyl radicals were determined in the (1?10)×108mol ?1dm 3s ?1range.The observed sensitivity of the transient absorbance at 350nm to dissolved O 2gives a further support to identifying it with cyclohexadienyl radical absorbance.

Reactions of e aq ?,pulse radiolysis

In water radiolysis,the e aq ?yield is equal to that of ?OH.The experiments with e aq ?were carried out in N 2saturated solutions with 5vol.%tert -butanol added in order to convert ?OH to the less reactive ?CH 2(CH 3)2COH radical (Reaction (5)).In N 2saturated tert -butanol containing solu-tions,the degradation can be attributed to the reactions of both e aq ?and ?CH 2(CH 3)2COH.In pulse radiolysis

experiments,

Fig.2UV absorption spectra of 1×10?4mol dm ?3fenuron solutions irradiated under conditions when ?OH (N 2O saturated solution)(a )or ?

OH+O 2??/HO 2?(aerated solution)(b ),respectively,were the main reactive species.Irradiations were carried out with a γ-radiation source for 0?30min,and the UV spectra were taken with a conventional

spectrophotometer

Scheme 1Radicals produced in addition or abstraction reactions

weak transient UVabsorption was observed(not shown)with a maximum around350nm.Since e aq?has very strong absor-bance in the visible-near infrared region with a maximum at 715nm,this absorbance,more precisely its decay in the pres-ence of fenuron,was used to calculate the rate coefficient of the e aq?+fenuron reaction.Our measurements suggested a rate coefficient of～1×109mol?1dm3s?1.Canle Lopez et al. (2005)determined(9.2±0.7)×108mol?1dm3s?1.Because of the much smaller rate coefficient of the e aq?+fenuron reaction than that of the e aq?+O2reaction(1.9×1010mol?1dm3s?1, Spinks and Woods1990),the e aq?+fenuron reaction practically has no contribution to the radiolytic degradation in aerated solution.

Spectrophotometric studies,gamma radiolysis

The maximum of the aromaticπ→π*absorption band is at 238nm for fenuron(Fig.2).This absorbance decreased strongly when the solution was irradiated by Co-60γ-rays. The UV absorption spectra of aeratedγ-irradiated solutions (Fig.2b)show the same dose dependence as the spectra obtained in N2O bubbled solutions(Fig.2a).However,the decrease of absorbance at238nm is somewhat slower in aerated solutions(yields of?OH in N2O and air bubbled

solutions are5.6×10?7and2.8×10?7mol J?1,respectively), than in N2O saturated solutions.

In air saturated solutions in addition to?OH,the O2??/HO2?pair is also present,representing reactive intermediate(Reac-tions(3)and(4)).In order to separate the contribution of?OH and O2??/HO2?reactions to degradation,we repeated the UV dose dependence studies in O2saturated solution containing sodium-formate(0.05mol dm?3,see Reactions(5)and(6)).In such solution,all the reactive intermediates of water radiolysis are converted to the O2??/HO2?pair.Under these conditions, the fenuron UVabsorbance did not decrease during irradiation indicating no reaction between fenuron and O2??/HO2?(Fig.3).

With the decrease of the absorbance at238nm upon irradiation in aerated(Fig.2b)and N2saturated solutions (Fig.2a),a new absorption band appears with a maximum around277nm.This band was not found in N2saturated solution when?OH was removed by tert-butanol(not shown).

Product distribution,gamma radiolysis

The same products were observed in both air or N2O saturated 1×10?3mol dm?3solutions;however,their abundances and dose dependences were different.MS identifications were carried out in both,negative and positive ionization modes. At10kGy dose fenuron was completely consumed in N2O bubbled solution,in aerated solution～10%remained(Fig.3). The molecular ion of fenuron in negative mode has m/z of 163,in positive165.The value of m/z changes to179and181in hydroxylated molecules.Fenuron may be hydroxylated in o-,m-and p-positions of the ring or on a methyl group.Since in the latter case the ring is not involved in changes,the UV spectrum of this compound must be similar to that of fenuron. For the molecules hydroxylated in the ring,the spectra should be slightly different(Mazellier et al.2007).

We do not know the UV absorption spectra of the hydroxyl-ated molecules,however,these spectra are known for aminophenols.In the UV spectra of o-,m-and p-aminophenols, there are two peaks withλmax230and281nm,231and281nm and230and295nm,respectively(Sarpal and Dogra1987).On the chromatograms in Fig.4,the compounds with elution times 4.1,5.9and8.7min are identified as singly hydroxylated versions of fenuron.They have molecular ions with m/z values of179/181in negative/positive mode.In the UV spectra taken with DAD,well separated double peaks were found,similarly to aminophenols.When the m/z179ion was fragmented in the collision chamber of MS,an ion with m/z134was found to be the most intensive fragment.The ion may have isocya-nate structure forming in the collision chamber in the reaction:?OC

6

H4NHCON(CH3)2→?OC6H4N=C=O+HN(CH3)2.Iso-cyanates are known thermal degradation products of phenylureas(Salvestrini et al.2002).In preliminary examina-tions with monuron and diuron,the mono and dichlorinated versions of the134fragment with m/z168and202were also detected.Mazellier et al.(2007)reported two hydroxylated fenuron isomers in?OH reactions(H2O2photolysis)and iden-tified them as p-and o-hydroxylated fenuron.These isomers were also detected by Richard and Bengana(1996).

The Fig.3Dose dependence of normalized absorbance at238nm in O2saturated HCOONa containing solution(fenuron concentration, 1×10?4mol dm?3,O2??/HO2?reaction)green left pointing triangle and the normalized integrated intensities of the fenuron peak in total ion chromatogram(fenuron concentration,1×10?3mol dm?3) in N2O saturated solution(?OH reaction)red circle,in air saturated solution(?OH+O2??/HO2?reaction)brown square,in N2saturated solu-tion(?OH+e aq?reaction)blue triangle,and in N2saturated solution containing0.5mol dm?3tert-butanol(e aq?+?CH2(CH3)2COH reaction) green down pointing triangle

separation of the two UV peaks is the highest for the com-pound with 4.1min elution time,based on the analogy to aminophenols,it is probably the p -isomer.In the experiments of Mazellier et al.(2007),also this isomer was eluting first.In dose dependence studies,the yields of singly hydroxylated products showed maxima at 2.5?5kGy in N 2O saturated solution,in aerated solutions the 4.1and 8.7min peaks had maxima at 1kGy.

Products with 7.1and 7.9min elution times and parent ions with m/z 195/197have UV spectra similar to that of the hydroxylated fenuron isomers.We identified them as double hydroxylated versions of fenuron.Double hydroxylation of the ring may result in six isomers.The missing variants were probably badly detectable under our conditions.The yields of double hydroxylated compounds increased with increasing dose in the dose range studied.

The peak at 5.5min with m/z 165/167and fragment ions of 134and 108in negative and 167and 110in positive ionization modes,and also with absorption spectrum similar to those of hydroxylated molecules may be a fenuron derivative hydrox-ylated in the ring but with only one methyl group on the terminal N.In electro-Fenton experiments,Oturan et al.(2010)also identified this product.

Since in radiolysis of dilute solutions the fenuron degrada-tion occurs in ?OH reactions,it is not unexpected that we identified more or less the same products as were observed also in other AOP .A difference is that at higher conversions we identified double hydroxylated products,and we found less products modified in terminal methyl group than other authors.

In Section 3.1.1.,cyclohexadienyl radicals and (in the presence of dissolved O 2)peroxy radicals were suggested as reactive radical intermediates.Peroxy radicals produced from hydroxycyclohexadienyl radicals may transform to phenols

by eliminating HO 2?:(CH 3)2NCONHC 6H 5(OH)OO ?→(CH 3)2NCONHC 6H 4(OH)+HO 2?,or they undergo fragmen-tation to smaller molecular mass acids or aldehydes (Mvula et al.2001;von Sonntag and Schuchmann 2001;Homlok et al.2013).These reactions may give explanation to the enhanced phenol formation in the presence of dissolved O 2.COD,TOC and toxicity in air saturated solutions,gamma radiolysis

In AOP ,the decrease of oxygen consumption (ΔCOD )and the decrease of the total organic carbon content (ΔTOC )are often used to follow up the oxidation.COD is expressed in mg O 2needed for total oxidation of organics in 1dm 3solution.TOC is the amount of organic carbon in 1dm 3solution expressed in mg C.The efficiency of degradation may be characterised by the ratio of the number of moles of O 2incorporated into the products and the number of moles of ?

OH injected into the solution:Efficiency =ΔCOD /(M mg ×Dose ×ρ×G (?OH))(M mg is the O 2molecular mass in mg mol ?1and ρis the density in kg dm ?3).Our COD and TOC measurements were carried out at high,1×10?3mol dm ?3concentration in order to be in the well-measureable range all through the depletion.The initial calculated COD and TO C were 315and 112.5mg dm ?3(Fig.5).When the solutions were irradiated,COD decreased almost linearly (up to 60kGy).The initial slope was 5.5×10?3mg dm ?3Gy ?1(～1.7×10?7mol O 2dm ?3Gy ?1).Since with 1Gy dose 2.8×10?7mol ?OH is injected into 1dm 3solution (G (?OH)-value),the O 2/?OH ratio is c.a.0.6.When an O 2molecule is incorpo-rated in a product,four-electron oxidation occurs.This way,the one-electron oxidant ?OH on average induced somewhat more than two-electron oxidations.In a previous publication (Homlok et al.2013),the higher than one-electron

oxidation

Fig.4Total ion chromatograms of 1×10?3mol dm ?3N 2O or air saturated solutions irradiated with 2.5kGy dose

was explained by O 2reaction with organic radicals (second oxidation).The deviation of COD from linearity above 60kGy is due to slow degradation of smaller fragments.During oxidation,the oxygen-to-carbon ratio in the products increases.This is shown by the higher decrease (in percent-age)in COD than in TOC and also by the decrease of pH:in an aerated solution with 2×10?4mol dm ?3fenuron,the pH decreased from the initial value of ～6.0to ～4.0when the solution was irradiated with 2kGy dose.

Fenuron is moderately toxic (Villa et al.2012):an un-irradiated 2×10?4mol dm ?3solution of fenuron inhibited the fluorescence of Vibrio fischeri bacteria by 15%(Fig.6).In 1kGy irradiated aerated solution,the inhibition increased to ～90%showing an increased toxicity of the products.The inhibition started to decrease above 10kGy.The higher inhi-bition may partly be due to hydrogen peroxide formation

during irradiation (Zona and Solar 2003).In a series of mea-surements,bovine catalase was added to each sample after the irradiation,but before the tests to destroy the residual H 2O 2.In such solutions,the fluorescence inhibition was much smaller;however,the initial increase was also observed.The toxicity with catalase present is mainly due to phenols that form at early stages of degradation.In Fig.6,we also show the inhibition caused by H 2O 2calculated based on the 50%inhibition (EC 50)at 3×10?4mol dm ?3concentration.In tap water,the H 2O 2effect is expected to be smaller due to H 2O 2decomposition catalyzed by traces of transition metals.Fig.6shows the toxicity changes in the 0?15kGy range.With increasing dose above 15kGy,strong toxicity decrease was found.

At the beginning of irradiation,hydrogen peroxide accu-mulation was observed with an initial yield of 1.5×10?7mol J ?1.Some part of H 2O 2may come from the termi-nation reactions of O 2??/HO 2?radicals produced mainly in Reactions (2)and (3).However,in radiolytic processes,H 2O 2also forms in deoxygenated solutions (with a yield of 0.75×10?7mol J ?1)in the reaction of two ?OH formed nearby (Spinks and Woods 1990).At the beginning of the treatment,H 2O 2concentration increases with the dose,and then,above 5kGy it decreases.At sufficiently high H 2O 2concentration,this molecule competes with O 2in reacting with e aq ?(see Reactions (2)and (8))(Spinks and Woods 1990):

e aq ?tH 2O 2→?OH tOH ?

k ?1:1?1010mol ?1dm 3s ?1

à

á

e8T

Reaction (8)through ?OH also contributes to fenuron degradation.

Hydroxylated aromatics often show higher toxicity in Microtox tests than non-hydroxylated ones (Gunatilleka and Poole 1999).E.g.benzene and toluene cause 50%decrease in luminescence (EC 50)at 4.6×10?4and 8.3×10?4mol dm ?3concentrations,the same values for phenol and o -cresol are:2.3×10?4and 1.8×10?4mol dm ?3,respectively.In catalase containing solution,the decrease of toxicity above 1kGy must be due to fenuron depletion,a source of more toxic com-pounds,and also to degradation of first formed products.

Conclusions

During irradiation in dilute,aerated,aqueous fenuron solution,the reactions of ?OH,e aq ?,H ?and O 2??/HO 2?should be con-sidered.The degradation of the organic molecule is initiated by ?OH.e aq ?,due to its low rate coefficient in reaction with fenuron,and high rate coefficient in reaction with O 2,does not play a role in fenuron degradation.The same is true for H ?because of its low yield and high rate coefficient with O

2

Fig.6Toxicity tests in air saturated 2×10?4mol dm ?3solution,without catalase black square ,with catalase red circle ,H 2O 2toxicity blue

triangle

Fig.5COD black square and TOC green circle measured in 1×10?3mol dm ?3air saturated solution

(Reaction(3)).O2??/HO2?is uncreative with fenuron.O2??/ HO2?may react with the unstable(e.g.radical)intermediates of the degradation.However,we have to mention that we did not find evidence for such reactions.The?OH-induced degra-dation of fenuron molecules in the presence of dissolved oxygen is faster than in deoxygenated solutions.It may be attributed to O2reaction with intermediate radicals. Acknowledgments The authors thank the Hungarian Science Founda-tion(OTKA,NK105802)and International Atomic Energy Agency (Contract No.16485)for support.

References

Acero JL,Benitez FJ,González M,Benitez R(2002)Kinetics of fenuron decomposition by single-chemical oxidants and combined systems.

Ind Eng Chem Res41:4225–4232

Amine-Khodja A,Boulkamh A,Boule P(2004)Photochemical behavior of phenylurea herbicides.Photochem Photobiol Sci3:145–156 Brahmia O,Boulkamh A,Sehili T,Aguer J-P,Richard C(2002)Kinetics of photocatalytic decomposition of Fenuron over TiO2in aqueous solution.Int J Photoenergy4:85–89

Buxton G,Greenstock CL,Helman WP,Ross AB(1988)Critical review of rate constants for reactions of hydrated electrons,hydrogen atoms and hydroxyl radicals(?OH/O?)in aqueous solution.J Phys Chem Ref Data17:513–886

Canle Lopez M,Fernandez MI,Rodrígez S,Santaballa JA,Steenken S, Vulliet E(2005)Mechanisms of direct and TiO2-photocatalised UV degradation of phenylurea herbicides.ChemPhysChem6:2064–2074

Canle LM,Rodrígez S,Rodrígez Vazques LF,Santaballa JA,Steenken S (2001)First stages of photodegradation of the urea herbicides Fenuron,Monuron and Diuron.J Mol Struct565–566:133–139 Dao YH,De Laat J(2011)Hydroxyl radical involvement in the decom-position of hydrogen peroxide by ferrous and ferric-nitrilotriacetate complexes at neutral pH.Water Res45:3309–2317

Fang X,Pan X,Rahmann A,Schuchmann H-P,von Sonntag C(1995) Reversibility in the reaction of cyclohexadienyl radicals with oxy-gen in aqueous solution.Chem Eur J1:423–429

Gallard H,De Laat J(2001)Kinetics of oxidation of chlorobenzenes and phenyl-ureas by Fe(II)/H2O2and Fe(III)/H2O2.Evidence of reduc-tion and oxidation reactions of intermediates by Fe(II)or Fe(III).

Chemosphere42:405–413

Gunatilleka AD,Poole CF(1999)Models for estimating the non-specific toxicity of organic compounds.Anal Chem36:235–242

Homlok R,Takács E,Wojnárovits L(2013)Degradation of organic molecules in advanced oxidation processes:relation between chem-ical structure and degradability.Chemosphere91:383–389Kosaka K,Yamada H,Matsui S,Echigo S,Shishida K(1998)Comparison among the methods for hydrogen peroxide measurements to evaluate advanced oxidation processes:application of a spectrophotometric method using copper(II)ion and2,9-dimethyl-1,10-phenanthroline.

Environ Sci Technol32:3821–3824

Mazellier P,Busset C,Delmont A,De Laat J(2007)A comparison of fenuron degradation by hydroxyl and carbonate radicals in aqueous solution.Water Res41:4585–4594

Mvula E,Schuchmann MN,von Sonntag C(2001)Reactions of phenol-OH-adduct radicals.Phenoxyl radical formation by water elimina-tion vs.oxidation by dioxygen.J Chem Soc Perkin Trans2:264–268 Oturan MA,Edelahi MC,Oturan N,El Kacemi K,Aaron J-J(2010) Kinetics of oxidative degradation/mineralization pathways of the phenylurea herbicides diuron,monuron and fenuron in water during application of the electro-Fenton process.Appl Catal B Environ97: 82–89

Radiation processing:environmental applications(2007)International Atomic Energy Agency,Vienna,ISBN92-D-100507-5

Richard C,Bengana S(1996)PH effect in the photocatalytic transforma-tion of a phenyl-urea herbicide.Chemosphere33:635–641 Salvestrini S,Di Cerbo P,Capasso S(2002)Kinetics of the chemical degradation of diuron.Chemosphere48:69–73

Sarpal RS,Dogra SK(1987)Prototropism in aminophenols and anisidines:a reinvestigation.J Photochem38:263–276

Singh TS,Gejji SP,Rao BSM,Mohan H,Mittal JP(2001)Radiation chemical oxidation of aniline derivatives.J Chem Soc Perkin Trans 2:1205–1211

Solar S,Solar W,Getoff N(1986)Resolved multisite OH-attack on aqueous aniline studied by pulse radiolysis.Radiat Phys Chem28: 229–234

Spinks JWT,Woods RJ(1990)An introduction to radiation chemistry, 3rd edn.Wiley-Interscience,New York

Takács E,Wojnárovits L,Dajka K(2000)Kinetics of the early stages of high-energy radiation initiated polymerization.Macromol Chem Phys201:2170–2175

Villa S,Migliorati S,Monti GS,Vighi M(2012)Toxicity on the lumi-nescent bacterium Vibrio fischeri(Beijerinck).II:response to com-plex mixtures of heterogeneous chemicals at low levels of individual components.Ecotoxicol Environ Saf86:93–100

von Sonntag C,Schuchmann H-P(2001)Peroxyl radicals in aqueous solution.In:Alfassi ZB(ed)Peroxy-radicals.Wiley,Chichester,pp 173–274

Wojnárovits L,Takács E(2013)Structure dependence of the rate coeffi-cients of hydroxyl radical+aromatic molecule reaction.Radiat Phys Chem87:82–87

Yamasita S,Katsumura Y,Lin M,Muroya Y,Miyazaki T,Murakami T (2008)Water radiolysis with heavy ions of energies up to28GeV.1.

Measurements of primary g values as track segment yields.Radiat Phys Chem77:439–446

Zhang J,Zheng Z,Zhao T,Zhao Y,Wang L,Zhong Y,Xu Y(2008) Radiation-induced reduction of diuron by gamma-ray irradiation.J Haz Mat151:465–472

Zona R,Solar S(2003)Oxidation of2,4-dichlorophenoxyacetic acid by ionizing radiation:degradation,detoxification and mineralization.

Radiat Phys Chem66:137–143