Ao

MI

a

ARRAA

KAFCC

1

thntudtpga

eccleAta[

0d

Journal of Hazardous Materials 190 (2011) 993–1000

Contents lists available at ScienceDirect

Journal of Hazardous Materials

journa l homepage: www.e lsev ier .com/ locate / jhazmat

ssessment of the generation of chlorinated byproducts upon Fenton-likexidation of chlorophenols at different conditions

acarena Munoz ∗, Zahara M. de Pedro, Jose A. Casas, Juan J. Rodriguezngenieria Quimica, Universidad Autonoma de Madrid, Crta. Colmenar km 15, 28049 Madrid, Spain

r t i c l e i n f o

rticle history:eceived 20 December 2010eceived in revised form 5 April 2011ccepted 9 April 2011vailable online 15 April 2011

eywords:

a b s t r a c t

Homogeneous Fenton-like (H2O2/Fe3+) oxidation proved to be highly efficient in the degradation ofmonochlorophenols but some important issues need to be considered depending on the operating con-ditions. When using the stoichiometric amount of H2O2 and a dose of Fe3+ in the range of 10–20 mg/L,complete breakdown of 4-CP up to CO2 and short-chain acids was achieved. Nevertheless, when substo-ichiometric amounts of H2O2 or lower concentrations of iron were used, significant differences betweenthe TOC measured and the calculated from the identified species were found. These differences were

dvanced oxidationenton-like processhlorophenolsondensation byproducts

attributed to condensation byproducts, including chlorinated species, formed by oxidative couplingreactions. PCBs, dioxins and dichlorodiphenyl ethers were identified. A solid residue was also formedconsisting mainly in carbon, oxygen and chlorine including also Fe. The occurrence of these highly toxicspecies must be carefully considered in the application of Fenton oxidation to wastewaters containingchlorophenols. The possibility of reducing costs by lowering the H2O2 dose below the stoichiometric oneneeds to tke this into account.

. Introduction

Chlorophenols (CPs) are a group of priority pollutants listed byhe European Decision 2455/2001/EC [1] because most of them areighly toxic and poorly biodegradable, and therefore they have aegative potential impact on the environment. However, due toheir antimicrobial properties, these compounds have been widelysed as wood, paints, fibers and leather preservatives, as well asisinfectants. They are also commonly used in a diversity of indus-rial processes leading to the manufacture of herbicides, fungicides,esticides, insecticides, pharmaceuticals and paints. CPs can also beenerated in waste incineration and chlorine pulp bleaching [2]. Asresult, they can be found in wastewaters, groundwaters and soils.

During the last decade there has been an increasing inter-st in the development of effective methods for the removal ofhlorophenols from wastewaters. Some authors have reported onhlorophenols degradation by biological processes [3,4]. Neverthe-ess, the low biodegradability of these compounds and their toxicffects on the active biomass, demands long acclimation stages.dvanced oxidation processes (AOPs) can be regarded as a poten-

ial solution [5,6]. Specifically, Fenton and Fenton-like oxidationre being increasingly used for the abatement of water pollutants7–11].

∗ Corresponding author. Tel.: +34 91497 3991; fax: +34 91497 3516.E-mail address: macarena.munnoz@uam.es (M. Munoz).

304-3894/$ – see front matter © 2011 Elsevier B.V. All rights reserved.oi:10.1016/j.jhazmat.2011.04.038

© 2011 Elsevier B.V. All rights reserved.

Studies on CPs oxidation by Fentonıs reagent have been com-monly focused on the analysis of operating conditions, such as thehydrogen peroxide and iron doses, pH and the presence of scav-engers [12–14]. Usually, the main goal of these studies has been themineralization of the organochlorinated compound to carbon diox-ide, water and hydrogen chloride. However, complete depletionof organic carbon requires high amounts of oxidant and/or largeresidence times involving high operating costs [8].

When lower amounts of oxidant or shorter residence timesare used, oxidation of chlorophenols can lead to the formationof byproducts which needs to be investigated in depth for a bet-ter understanding of this process. Intermediates detected in theoxidation of chlorophenols are usually benzenediols and short-chain organic acids. Among benzenediols, chlorocatechol appearsin greater concentration, while benzoquinone and hydroquinoneare formed to a lower extent [13,15]. Detomaso et al. [9] alsosuggested the formation of aromatic intermediates other thanbenzenediols when working with substoichiometric H2O2 doses.Sedlak and Andren [16] proposed a reaction pathway in the oxi-dation of chlorobenzenes, which involves the formation of dimers(such as chlorinated biphenyls and hydroxychlorobiphenyls) andcolored aromatic polymers, which can be oxidized by subsequent•OH attack.

The toxicity of the byproducts obtained upon incomplete degra-dation of CPs was measured by Pérez-Moya et al. [17], and theyreported fairly similar values to those of the starting compounds.Identification of the aromatic byproducts appears crucial for a

9 rdous

bccit

Fu4tlopawodrs

2

2

tgtaw

decrwcdataoattt

2

tthlw1s2aaucwid

94 M. Munoz et al. / Journal of Haza

etter understanding, design and optimization of the Fenton pro-ess [18]. To the best of our knowledge, the formation of theseompounds at low [H2O2]0/[CP]0 molar ratios has been scarcelynvestigated and thus there is a lack of information on that impor-ant issue.

This work reports on the degradation of monochlorophenols byenton-like oxidation and identification of the resulting byprod-cts. The effect of Fe3+ and H2O2 doses on the oxidation of-chlorophenol (4-CP) was evaluated in terms of products distribu-ion and TOC balance. Although most of the Fenton studies in theiterature work with H2O2 excess, the extended implementationf this technology in the future requires a better knowledge of itserformance at as low as possible reagent consumption since this isn important feature regarding the economy of the process. In thisay, Fenton oxidation of the three monochlorophenols in aque-

us solution was tested within a wide range of hydrogen peroxideoses covering from fairly substoichiometric up to the stoichiomet-ic ratio in order to learn on the evolution of highly toxic chlorinatedpecies and the ecotoxicity of the effluent under these conditions.

. Materials and methods

.1. Fenton-like experiments

All the chemicals used were analytical grade reagents. Oxida-ion runs were carried out at atmospheric pressure and 50 ◦C in alass batch reactor, equipped with a magnetic stirrer (700 rpm), andemperature control. The initial pH value was adjusted with nitriccid to 3.0 in all the experiments and 500 mL of reaction solutionas always used.

In the first set of runs the influence of iron catalyst and H2O2oses was evaluated. Unless otherwise indicated, the oxidationxperiments were carried out using 100 mg/L 4-CP as startingoncentration which, according to several authors [19,20], is a rep-esentative concentration of chlorophenols in several industrialastewaters (e.g. chlorine pulp bleaching, manufacture of herbi-

ides, fungicides, pharmatheuticals and paints, among other). Theoses of Fe3+ and H2O2 were varied in the ranges of 20–1 mg/Lnd 344–17.2 mg/L, respectively. The H2O2 doses tested correspondo 5%, 10%, 20%, 50% and 100% of the theoretical stoichiometricmount to completely oxidize 4-CP up to CO2, H2O and HCl. A sec-nd set of experiments was carried out with 2-CP, 3-CP and 4-CPt higher initial concentration (2000 mg/L) in order to better iden-ify the condensation byproducts formed during the reaction. Inhis case different H2O2 doses covering from 20% to 100% of theheoretical stoichiometric amount and 10 mg/L Fe3+ were used.

.2. Analytical methods

In each run, the progress of the reaction was followed byaking and analyzing periodically liquid samples from the reac-or outlet. Monochlorophenols, chlorocatechols, benzoquinone andydroquinone were quantified by means of high-performance

iquid chromatography (HPLC; Varian Pro-Start 325) equippedith a UV detector and a Microsorb C18 5 �m column (MV 100,

5 cm length, 4.6 mm diameter) as the stationary phase. Analy-is of monochlorophenols and chlorocatechols was carried out at70 nm using a 30/70% (v/v) mixture of acetonitrile and aceticcid aqueous solution (75 mM) as the mobile phase. Benzoquinonend hydroquinone were analyzed at 250 and 210 nm, respectivelysing an aqueous sulfuric solution (4 mM) as mobile phase. Short-

hain acids and chloride were analyzed by ion chromatographyith chemical suppression (Metrohm 790 IC) using a conductiv-

ty detector. A Metrosep A supp 5-250 column (25 cm length, 4 mmiameter) was used as stationary phase and an aqueous solution

Materials 190 (2011) 993–1000

3.2 mM Na2CO3 as the mobile phase. TOC was measured using a TOCanalyzer (Shimadzu, mod. TOC, VSCH) and residual H2O2 concen-tration was determined by colorimetric titration using the titaniumsulfate method Eisenberg [21].

In order to analyze the condensation products, samples fromthe reactor at the end of each run were extracted in n-hexane andconcentrated. These samples were analyzed by Gas Chromatog-raphy (MS detector) with an electron impact ionization source(Saturn 2100T). The capillary column used was a Factor Four, Varian(30 m length, 0.25 mm internal diameter). To confirm the structuralassignment of the identified compounds, analytical standards wereused, as well as the NIST 05 library.

The dark brown solid formed upon Fenton-like oxidation of thehigh-concentration solutions of monochlorophenols (2000 mg/L)was separated using a nylon membrane filter 0.45 �m pore size(Sigma–Aldrich). After drying, the solid was characterized by ele-mental analysis (LECO elemental analyzer) and total reflectionX-ray fluorescence (Chemical Analysis Seifert EXTRA-II spectrom-eter).

2.3. Ecotoxicity tests

The ecotoxicity of monochlorophenols and reaction sampleswas evaluated by the Microtox toxicity test (ISO 11348-3, 1998).The bioluminescence was measured in a photomultiplier M500Microtox Analyzer (Azur Environmental). The test was conductedat 15 ◦C, adjusting the osmotic pressure close to 2% NaCl and pHbetween 6 and 8. The EC50 is used to quantify the ecotoxicity ofa compound. It is defined as the effective nominal concentration(mg/L) of the compound that reduces the intensity of light emis-sion by 50% after 15 min contact time. For complex samples, IC50 isused, defined as the dilution ratio of the sample that yields this 50%light emission reduction. The EC50 and IC50 values are inversely pro-portional to the ecotoxicity expressed in toxicity units (TU). Theyare related by Eq. (1), where C0 corresponds to the concentrationof the compound used to obtain its EC50 value.

TU = C0(mg/L)EC50

= 100IC50

(1)

3. Results and discussion

3.1. 4-Chlorophenol oxidation by Fenton-like process

4-CP can be oxidized by the Fenton process [22], which, as iswell known, uses H2O2 and Fe2+. Iron can be fed as Fe3+ insteadof Fe2+ giving rise to a Fenton-like process of similar character-istics. The rate of this process increases with the H2O2 and Fe3+

concentrations. Nevertheless, high H2O2/Fe molar ratios are usuallyemployed in the literature and even in many cases in real indus-trial applications due to the non-selective nature of the •OH radicals[23], which reacts with any oxidizable matter and not only with thetarget compounds. Moreover, the use of low concentrations of ironprevent the uncontrolled generation of radicals which would favorscavenging reactions [14] leading to a less efficient consumption ofhydrogen peroxide, a critical factor on the economy of the process.In the same way, the use of substoichometric hydrogen peroxideconcentrations favors the economic viability of the Fenton process.

A previous set of experiments was carried out using the theo-retical stoichiometric dose of H2O2 to completely oxidize 100 mg/L4-CP (which corresponds to 13 H2O2/4-CP molar ratio). Fe3+ wastested within the range of 1–20 mg/L (565–28 H2O2/Fe3+ molar

ratio). 4-CP disappeared in less than 5 min at all the iron dosestested although a slower rate was observed, as expected, at thelower iron concentration (1 mg/L). The oxidation byproducts weremainly short-chain acids and 4-chlorocatechol (4-CCA), the only

M. Munoz et al. / Journal of Hazardous Materials 190 (2011) 993–1000 995

Fig. 1. Measured and calculated organic carbon and chlorine balance at dif-f[

atamtrFtwpo

ul(

a4Tdtsac(bfics

Fig. 2. Measured and calculated organic carbon and chlorine balance at different

erent Fe3+ doses (A: 20 mg/L; B: 10 mg/L; C: 1 mg/L). [4-CP]0 = 100 mg/L;H2O2]0 = stoichiometric.

romatic compound detected, which indicates that hydroxylationakes place in the ortho position. The distribution of the organiccids depends on the iron concentration used. Fumaric, maleic andalonic acids were not detected at the highest iron concentration

ested. The concentration of oxalic acid was always higher than theest of the acids and it increased with reaction time in all cases (seeigure S1 in the Supplementary data for the evolution of oxida-ion byproducts upon reaction time). These results are consistentith those reported by Zazo et al. [23] from Fenton oxidation ofhenol concluding that oxalic acid is highly resistant to Fentonxidation.

The evolution of hydrogen peroxide depended on the iron dosesed. As expected, slower decomposition rates and significantly

ower conversion values were obtained as the iron dose decreasedsee Figure S2 in the Supplementary data).

As can be observed in Fig. 1, the carbon and chlorine bal-nces could be closed at almost 100% after a reaction time ofh only at the highest iron dose tested (20 mg/L). A slowerOC degradation rate was obtained when the iron dose wasecreased. Under the operating conditions of Fig. 1 the forma-ion of unidentified chlorine-containing condensation byproductseems to be favored since the difference between the measurednd calculated organic carbon increased and the chlorine balancelosure was decreased up to around 50% at the lower Fe2+ dose1 mg/L). In Fenton oxidation of phenol, the differences observedetween the TOC measured and the calculated from the identi-

ed compounds have been explained by the formation of aromaticondensation species [24]. These compounds can be oxidized tohort-chain acids (formic and oxalic) under more severe con-

H2O2 doses (A: 100%; B: 50%; C: 20%; D: 10%; E: 5% of the stoichiometric dose) with1 mg/L Fe3+. [4-CP]0 = 100 mg/L.

ditions. In the case of 4-CP, similar types of compounds butincluding chlorine-containing species must be formed at low irondoses.

The effect of varying the hydrogen peroxide dose was testedwithin a wide range of 100–5% of the stoichiometric ratio (13–0.65H2O2/4-CP molar ratios), at a constant iron dose of 1 mg/L (565–28H2O2/Fe3+ molar ratio). Complete 4-CP conversion to different

aromatic byproducts and short-chain acids was achieved at the sto-ichiometric dose but also at 50% of that H2O2 amount, whereasconversion values of 93%, 74% and 42% were obtained at 20%,

996 M. Munoz et al. / Journal of Hazardous Materials 190 (2011) 993–1000

ton-li

1F

((Hu4rrtmwr

Fig. 3. Reaction pathway for Fen

0% and 5% of the stoichiometric H2O2 dose, respectively (seeigure S3 in the Supplementary data).

The only aromatic byproducts detected were p-benzoquinoneBQ) and 4-chlorocatechol (4-CCA). Low concentrations of BQ<2 mg/L) were detected in the initial stages of reaction when low2O2 doses were used, being this compound completely oxidizedpon reaction time (4 h). On the other hand, high concentrations of-CCA (>15 mg/L) were always obtained at the initial stage of theeaction (5 min) and the complete disappearance of this byproductequired at least the stoichiometric dose of H2O2 and 4 h reac-

ion time. The low-weight organic acids identified were fumaric,

aleic, malonic, acetic, oxalic and formic. Among them, oxalicas only detected at H2O2 doses above 20% of the stoichiomet-

ic dose, and its concentration increased with the H2O2 dose (see

ke oxidation of 4-chlorophenol.

Figure S4 in the Supplementary data for the evolution of byproductsupon reaction time).

More than 90% conversion of H2O2 was reached in all the experi-ments carried out at substoichiometric H2O2 doses, but when usingthe stoichiometric amount less than 50% conversion was observedafter 4 h. This can be due to the low concentration of iron usedin all the experiments, which showed not to be sufficient for pro-moting a complete decomposition of H2O2 at the above mentionedstoichiometric dose.

Although high conversions of 4-CP were achieved even with low

relative amounts of H2O2, mineralization was negligible at H2O2doses below 20% of the stoichiometric dose, as can be seen in Fig. 2.Zimbron and Reardon [25] also reported negligible mineralizationupon Fenton oxidation of pentachlorophenol, even at higher H2O2

rdous Materials 190 (2011) 993–1000 997

dottutciodOspcTwqcddcivc(m

i

3s

iiiierm

meadiwrr

aFcoiCdltcw

woCC

M. Munoz et al. / Journal of Haza

oses. Obviously, the efficiency of the process cannot be evaluatednly in terms of 4-CP disappearance but following the evolution ofhe organic carbon and chlorine. The highest difference betweenhe TOC measured and the calculated from the identified byprod-cts and the unconverted 4-CP occurred at a dose of H2O2 20% ofhe stoichiometric amount (Fig. 2). Under that condition a high 4-CPonversion is already achieved but with almost negligible mineral-zation. These results indicate the existence of important amountsf unidentified oxidation byproducts which may correspond to con-ensation species not analyzed by HPLC or ion chromatography.n the other hand, the chlorine balance was closed at about 80%

uggesting that although a significant partial dechlorination takeslace prior to condensation reactions, part of the non-identifiedondensation products must include chlorine in their composition.he differences between the measured and calculated TOC decreasehen increasing the H2O2 dose. Although these differences are still

uite significant even at the stoichiometric dose, the closure of thehlorine balance remains around the same percentage indepen-ently of the H2O2 dose being even lower at the stoichiometricose. Thus, the aforementioned condensation products, includinghlorine-containing compounds, must be formed also under a sto-chiometric amount of H2O2 if the Fe3+ dose is below a thresholdalue around 10 mg/L under the experimental conditions used (thatorresponds to 56.5 H2O2/Fe3+ molar ratio). At very low H2O2 doses10% of the stoichiometric dose and below) the differences become

uch lower but because of the low 4-CP conversion.Based on the results presented here the reaction scheme of Fig. 3

s proposed for the Fenton-like oxidation of 4-chlorophenol.

.2. Fenton-like oxidation of monochlorophenols underubstoichiometric conditions

The formation of aromatic condensation byproducts can be anmportant drawback for the application of the Fenton process tondustrial effluents containing chlorophenols due to the high tox-city of those compounds even more when they include chlorinen their compostion. Pérez-Moya et al. [17] demonstrated thatven 90% TOC removal was insufficient for achieving an acceptableeduction of ecotoxicity when treating 2-CP by Fenton oxidation atoderate reagent loads.To gain further insights into the Fenton oxidation of

onochlorophenols under substoichiometric H2O2 doses, severalxperiments were performed with the three monochlorophenolst a relatively high starting concentration (2000 mg/L) and a H2O2ose substantially below the stoichiometric one (20% of the sto-

chiometric dose which corresponds to 2.6 H2O2/CP molar ratio)ith the aim of easily identify the byproducts obtained during the

eaction. The Fe3+ dose was always 10 mg/L (230 H2O2/Fe3+ molaratio).

Under the operating conditions employed, the disappear-nce of the three starting chlorophenols was very fast (seeigure S5 in the Supplementary data for the evolution of the threehlorophenols upon reaction time). Around 90% conversion wasbtained for 2-CP and 4-CP and close to 95% for 3-CP in 5 min. Its well known that OH and Cl are ortho and para directors [26]. 3-P has three ortho and para positions activated by both OH and Clirectors and thus, it can be expected to be more easily hydroxy-

ated than 2-CP and 4-CP. On the other hand, in the case of 2-CPhere are some steric hindrance and inductive effects due to theloser location of OH and Cl on the aromatic ring. Accordingly, 2-CPill be less reactive to hydroxylation.

The byproducts detected by HPLC and ion chromatography

ere chlorocatechols and short-chain acids (Fig. 4). Formation

f chlorocatechols occurs by hydroxylation of monochlorophenol.onsequently, 4-CP led to 4-CCA, 3-CCA was produced from 2-P and 3-CCA and 4-CCA from 3-CP. As can be observed similar

Fig. 4. Time-evolution of byproducts from Fenton-like oxidation of the threemonochlorophenols ([CP]0 = 2000 mg/L; [Fe3+]0 = 10 mg/L; [H2O2]0 = 20% of the sto-ichiometric dose).

concentrations of 3-CCA and 4-CCA were obtained from 3-CP oxi-dation, which indicates that the hydroxyl radical attack was equallyfavored on both the Cl and OH ortho positions. Maleic, fumaric, mal-onic and formic acids increased moderately their concentrations asoxidation proceeds. On the other hand, although the H2O2 conver-sion was complete in the three cases, a higher H2O2 amount seemsto be consumed in 3-CP hydroxylation since higher concentrationof CCA and lower of fumaric and malonic were obtained from thatcompound.

Dramatic differences were observed between the measured andcalculated TOC in the cases of the three monochlorophenols (Fig. 5),confirming that in addition to the oxidation byproducts of Fig. 4important amounts of other unidentified species must be formed.Several authors [16,18] attributed these differences to aromaticcondensation byproducts including chlorinated species. Formationof that type of species is supported by the development of a brown-ish color in the reaction medium and the existence of a dark solidresidue after evaporation. The clearly unbalanced chlorine in Fig. 5suggests that these species must contain that element in relativelysignificant amounts.

3.3. Condensation byproducts

The analysis of the oxidation byproducts not identified byHPLC or ion chromathography was carried out by GC/MS after

998 M. Munoz et al. / Journal of Hazardous Materials 190 (2011) 993–1000

Table 1Chlorinated by-products identified by GC/MS from Fenton-like oxidation of monochlorophenols (pH0 = 3, T = 50 ◦C, [Fe3+]0 = 10 mg/L; [H2O2]0 = 20% of the stoichiometric dose).

2-CP 3-CP 4-CP

O

Cl Cl

OHHO

O

OH

Cl

Cl

O

OCl Cl

OH OH

Cl Cl

OH OH

Cl

O

OHHO

Cl Cl

O Cl

HO

O

Cl Cl

HOOH

O

Cl Cl

HO

OH OH

eacb

3b

dwatatutm2iodHteb[oiorpa

xtraction with n-hexane of the reaction samples. The resultsre collected in Table 1. Several two-ring aromatic chlorinatedompounds were detected including chlorinated hydroxylatediphenyls, dichlorodiphenyl ethers and dichlorodibenzodioxins.

.4. Effect of the H2O2 dose on the formation of condensationyproducts

With the aim of elucidating if these compounds are also pro-uced when the H2O2 dose is increased, the dose was variedithin the range of 50–100% of the stoichiometric one. The CPs

nd Fe3+ starting concentrations were 2000 and 10 mg/L respec-ively. These conditions correspond to 6.6–13 H2O2/CP molar rationd 575–1150 H2O2/Fe3+ molar ratio. Fig. 6 shows the concen-ration of the species of Table 1 depending on the H2O2 dosessed. As can be seen, at the lowest H2O2 dose, significant concen-rations of dichlorodiphenyl ethers were obtained from the three

onochlorophenols, whereas biphenyls were only detected from-CP and 4-CP, and a low concentration of dichlorodibenzodiox-

ns were detected from 2-CP. These byproducts were effectivelyxidized by increasing the H2O2 dose, and only trace amounts ofichlorodiphenyl ethers were detected when the stoichiometric2O2 dose was employed. Sedlak and Andren [16] investigated

he oxidation of chlorobenzenes by Fentonıs reagent providingvidences of the formation of dimers (specifically hydroxychloro-iphenyls) as a possible reaction pathway. Poerschmann et al.18] reported similar type of compounds upon Fenton oxidationf 2-chlorophenol at substoichiometric H2O2. Hirvonen et al. [27]dentified some byproducts of similar molecular weights from

zonation and UV/H2O2 oxidation of monochlorophenols. Chlo-inated hydroxylated diphenyl ethers have been recognized asrecursors for the formation of dioxins [28] which are highly toxicnd well-known carcinogenic compounds.

Cl Cl

From dimers, oxidative coupling reactions of radicals mightresult in the generation of trimers, tetramers, pentamers, etc.Alternatively, the addition of radicals onto the dimers could beresponsible for the formation of species of higher molecularweights. Both processes can ultimately result in the generationof polymers which can precipitate from the aqueous media. Thiswas visually supported when using 20% of the stoichiometric H2O2dose by the formation of a dark solid residue which representeda low but significant relative amount of the starting concentra-tion of 2-CP, 3-CP and 4-CP (2.0%, 1.4% and 5.7% w/w, respectively;see Table S1 in the Supplementary data). The elemental composi-tion of the solid residue obtained during the oxidation of 4-CP byFentonıs reagent correspond to C (46.4%), H (2.6%), Cl (12.6%), Fe(3.6%) and O (34.8%, calculated by difference). From these resultsthe following empirical formula (C12H8ClO7)5Fe can be proposed,which suggests the presence of condensed aromatic rings associ-ated to chlorine and iron. Moreover, that solid residue was insolubleunder the experimental conditions (pH 3) but soluble in 1 M aque-ous NaOH solution which is a common feature of humic acid-likespecies. This type of compounds typically contains a complex mix-ture of many different acids with carboxyl and phenolate groupswhich are able to form complexes with different ions, such as Fe2+

and Fe3+. Kwon et al. [13] gave the following elemental analysis:C (40%), H (2.7%), O (27%) and Fe (6.6%) from the solid residue iso-lated from Fenton oxidation of 4-CP. The higher Fe content of thissolid compared with ours can be due to the significantly higher Feto 4-CP ratio used by these authors.

3.5. Effect of the H2O2 dose on the evolution of ecotoxicity

The ecotoxicity values of the final effluents were measuredand compared with those of the initial monochlorophenol solu-

M. Munoz et al. / Journal of Hazardous Materials 190 (2011) 993–1000 999

Fig. 5. Measured and calculated TOC and chlorine balance from Fenton-like oxi-dation of monochlorophenols (CP]0 = 2000 mg/L; [Fe3+]0 = 10 mg/L; [H2O2]0 = 20% ofthe stoichiometric dose).

Fm

tmsuebNit

Table 2Ecotoxicity values (T.U.) after Fenton-like oxidation of monochlorophenols upon 4 hat different H2O2 doses (A: 20%; B: 50%; C: 75%; D: 100% of the stoichiometric H2O2

dose). [Fe3+]0 = 10 mg/L; [CP]0 = 2000 mg/L.

Chlorophenol Initial sample A B C D

2-CP 102.4 256.4 25.0 21.0 <1.0

ig. 6. Condensation byproducts obtained from Fenton-like oxidation ofonochlorophenols ([CP]0 = 2000 mg/L; [Fe3+]0 = 10 mg/L; reaction time = 4 h).

ions. As can be seen in Table 2, despite the high conversion ofonochlorophenols, the final effluents obtained with 20% of the

toichiometric H2O2 dose show significantly higher ecotoxicity val-es than the starting solutions except in the case of 4-CP. Thatcotoxicity must be mainly due to the chlorinated condensation

yproducts since the organic acids have very low ecotoxicity values.evertheless, increasing the H2O2 dose, ecotoxicity was reduced

n a significant amount, reaching negligible ecotoxicity values withhe stoichiometric H2O2 dose.

3-CP 114.2 571.4 30.6 22.2 <1.04-CP 1052.6 384.6 59.2 31.7 <1.0

4. Conclusions

In this work it has been reported that Fenton-like oxidation of4-CP at 50 ◦C with the stoichiometric amount of H2O2 and a dose ofFe3+ around 20 mg/L led to complete conversion of that compoundto CO2 and short-chain organic acids. Nevertheless, when substoi-chiometric amounts of H2O2 or lower concentrations of iron wereused, great differences between the measured TOC and the cal-culated from the identified species were found. These differenceswere attributed to condensation compounds formed by oxidativecoupling reactions. These compounds include chlorine-containingspecies.

Oxidation of monochlorophenols at relatively high concentra-tions with substoichiometric H2O2 and low iron doses allowedidentifying PCBs, dioxins and dichlorodiphenyl ethers in the reac-tion samples. The solid residue formed consisted mainly in carbon,oxygen and chlorine including also Fe.

The effluents from Fenton-like oxidation of chlorophenols wereincreasingly toxic as the H2O2 dose used decreased below the sto-ichiometric one. Therefore, the formation of highly toxic speciesunder substoichiometric H2O2 doses and low iron concentrations isan important issue to be considered regarding the potential applica-tion of Fenton oxidation to wastewaters containing chlorophenolssince it limits the possibilities of lowering the hydrogen peroxideand iron doses which is an important consideration for the economyof the process.

Acknowledgments

This research has been supported by the Spanish MICINNthrough the project CTQ2010-14807 and by the CM through theproject S-2009/AMB-1588. M. Munoz thanks the Spanish Ministryof Education for a research grant.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.jhazmat.2011.04.038.

References

[1] EC Decision 2455/2001/EC of the European Parliament and the Council ofNovember 20, 2001 establishing the list of priority substances in the field ofwater policy and amending Directive 2000/60/EC.

[2] U.G. Ahlborg, T.M. Thunberg, Chlorinated phenols: occurrence, toxicity,metabolism and environmental impact, Crit. Rev. Toxicol. (1980) 1–35.

[3] A. Chiavola, R. Baciocchi, R.L. Irvine, R. Gasvaci, P. Sirini, Aerobic biodegradationof 3-chlorophenol in a sequencing batch reactor. Effect of co metabolism, WaterSci. Technol. 50 (2004) 235–242.

[4] V.M. Monsalvo, A.F. Mohedano, J.A. Casas, J.J. Rodriguez, Cometabolicbiodegradation of 4-chlorophenol by sequencing batch reactors at differenttemperatures, Bioresour. Technol. 100 (2009) 4572–4578.

[5] C. Comninellis, A. Kaplaka, S. Malato, S.A. Parsons, I. Poulios, D. Mantzavinos,Advanced oxidation processes for water treatment: advances and trends forR&D, J. Chem. Technol. Biotechnol. 83 (2008) 769–776.

[6] R. Andreozzi, V. Caprio, A. Insola, R. Marotta, Advanced oxidation processes forwater purification and recovery, Catal. Today 53 (1999) 51–59.

[7] P. Bautista, A.F. Mohedano, M.A. Gilarranz, J.A. Casas, J.J. Rodriguez, Applicationof Fenton oxidation to cosmetic wastewaters treatment, J. Hazard. Mater. 143(2007) 128–134.

1 rdous

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

000 M. Munoz et al. / Journal of Haza

[8] E. Chamarro, A. Marco, S. Esplugas, Use of Fenton reagent to improve chemicalbiodegradability, Water Res. 35 (2001) 1047–1051.

[9] A. Detomaso, A. Lopez, G. Lovecchio, G. Mascolo, R. Curci, Practical applica-tions of the Fenton reaction to the removal of chlorinated aromatic pollutants.Oxidative degradation of 2,4-dichlorophenol, Environ. Sci. Pollut. Res. 10 (2003)379–384.

10] M. Pérez, F. Torrades, X. Domènech, J. Peral, Fenton and photo-Fenton oxidationof textile effluents, Water Res. 36 (2002) 2703–2710.

11] M.J. Liou, M.C. Lu, J.N. Chen, Oxidation of explosives by Fenton and photo-Fentonprocesses, Water Res. 37 (2003) 3172–3179.

12] E. Neyens, J. Bayens, A review of classic Fenton’s peroxidation as an advancedoxidation technique, J. Hazard. Mater. B98 (2003) 33–50.

13] B.G. Kwon, D.S. Lee, N. Kang, J. Yoon, Characteristics of p-chlorophenol oxidationby Fenton’s reagent, Water Res. 33 (1999) 2110–2118.

14] N. Kang, D.S. Lee, J. Yoon, Kinetic modeling of Fenton oxidation of phenol andmonochlorophenols, Chemosphere 47 (2002) 915–924.

15] Y. Du, M. Zhou, L. Lei, Role of the intermediates in the degradation ofphenolic compounds by Fenton-like process, J. Hazard. Mater. B136 (2006)859–865.

16] D.L. Sedlak, A.W. Andren, Oxidation of chlorobenzenes with Fenton’s reagent,Environ. Sci. Technol. 25 (1991) 777–782.

17] M. Pérez-Moya, M. Graells, L.J. Valle, E. Centelles, H.D. Mansilla, Fenton and

photo-Fenton degradation of 2-chlorophenol: multivariate analysis and toxic-ity monitoring, Catal. Today 124 (2007) 163–171.

18] J. Poerschmann, U. Trommler, T. Gorecki, F. Kopinke, Formation of chlorinatedbiphenyls, diphenyl ethers and benzofurans as a result of Fenton-driven oxi-dation of 2-chlorphenol, Chemosphere 75 (2009) 772–780.

[

[

Materials 190 (2011) 993–1000

19] Y.G Ghaly, G. Hartel, R. Mayer, R. Haseneder, Photochemical oxidation of p-chlorophenol by UV/H2O2 and photo-Fenton process. A comparative study,Waste Manage. 21 (2001) 41–47.

20] S.W. Maloney, J. Manem, J. Mallevialle, F. Fiessinge, Transformation of traceorganic compounds in drinking water by enzymic oxidative coupling, Environ.Sci. Technol. 20 (1986) 249–253.

21] G.M. Eisenberg, Colorimetric determination of hydrogen peroxide, Ind. Eng.Chem. Res. 15 (1943) 327–328.

22] M. Pera-Titus, V. Garcia-Molina, M.A. Banos, J. Gimeneza, S. Esplugas, Degra-dation of chlorophenols by means of advanced oxidation processes: a generalreview, Appl. Catal., B 47 (2004) 219–256.

23] J.A. Zazo, J.A. Casas, A.F. Mohedano, M.A. Gilarranz, J.J. Rodriguez, Chemicalpathway and kinetics of phenol oxidation by Fenton’s reagent, Environ. Sci.Technol. 39 (2005) 9295–9302.

24] J.A. Zazo, J.A. Casas, C.B. Molina, A. Quintanilla, J.J. Rodriguez, Evolution of eco-toxicity upon Fenton’s oxidation of phenol in water, Environ. Sci. Technol. 41(2007) 7164–7170.

25] J.A. Zimbron, K.F. Reardon, Fenton’s oxidation of pentachlorophenol, Water Res.43 (2009) 1831–1840.

26] W.Z. Tang, C.P. Huang, The effect of chlorine position of chlorinated phenolson their dechlorination kinetics by Fenton’s reagent, Waste Manage. 15 (1995)615–622.

27] A. Hirvonen, M. Trapido, J. Hentunen, J. Tarhanen, Formation of hydroxylatedand dimeric intermediates during oxidation of chlorinated phenols in aqueoussolution, Chemosphere 41 (2000) 1211–1218.

28] J.C. Yu, T.Y. Kwong, Q. Luo, C. Zongwei, Photocatalytic oxidation of triclosan,Chemosphere 65 (2006) 390–399.