chemical oxidation of dissolved organic matter by chlorine dioxide,

10
Chemical Oxidation of Dissolved Organic Matter by Chlorine Dioxide, Chlorine, And Ozone: Eects on Its Optical and Antioxidant Properties Jannis Wenk, ,,Michael Aeschbacher, Elisabeth Salhi, Silvio Canonica, Urs von Gunten, ,,§, * and Michael Sander , * Eawag, Swiss Federal Institute of Aquatic Science and Technology CH-8600, Dü bendorf, Switzerland Institute of Biogeochemistry and Pollutant Dynamics, ETH Zü rich, CH-8092, Zü rich, Switzerland § School of Architecture, Civil and Environmental Engineering (ENAC), Ecole Polytechnique Fé de ́ rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland * S Supporting Information ABSTRACT: In water treatment dissolved organic matter (DOM) is typically the major sink for chemical oxidants. The resulting changes in DOM, such as its optical properties have been measured to follow the oxidation processes. However, such measurements contain only limited information on the changes in the oxidation states of and the reactive moieties in the DOM. In this study, we used mediated electrochemical oxidation to quantify changes in the electron donating capacities (EDCs), and hence the redox states, of three dierent types of DOM during oxidation with chlorine dioxide (ClO 2 ), chlorine (as HOCl/OCl ), and ozone (O 3 ). Treat- ment with ClO 2 and HOCl resulted in comparable and prominent decreases in EDCs, while the UV light absorbances of the DOM decreased only slightly. Conversely, ozonation resulted in only small decreases of the EDCs but pronounced absorbance losses of the DOM. These results suggest that ClO 2 and HOCl primarily reacted as oxidants by accepting electrons from electron-rich phenolic and hydroquinone moieties in the DOM, while O 3 reacted via electrophilic addition to aromatic moieties, followed by ring cleavage. This study highlights the potential of combined EDC-UV measurements to monitor chemical oxidation of DOM, to assess the nature of the reactive moieties and to study the underlying reaction pathways. INTRODUCTION Drinking water and wastewater treatment facilities often have a chemical oxidation step for disinfection, the removal of organic micropollutants, color removal, and taste and odor control. Among the most commonly used oxidants are chlorine dioxide (ClO 2 ), chlorine (as hypochlorous acid, HOCl and OCl ), and ozone (O 3 ). 1 For a number of reasons, the eciency of the oxidation step and the quality of the treated water largely depend on the reaction of the chemical oxidant with dissolved organic matter (DOM). First, DOM is a major contributor to drinking water color, which negatively aects the acceptance of the water among consumers. 2 Second, the reaction of DOM with the chemical oxidants accelerates their consumption and, thus, may reduce the eciency of the oxidation step for disinfection and micropollutant oxidation. 3,4 Third, the reaction of the oxidants with DOM may result in the formation of potentially harmful disinfection/oxidation byproducts. 57 Fourth, chemical DOM oxidation results in the generation of low molecular weight assimilable organic carbon (AOC). 8,9 Following the oxidation step, the AOC needs to be removed by biological ltration to improve the biological stability of drinking waters. 4,10,11 For these reasons, information on the DOM concentration and its reactivity is indispensable to nd the appropriate dose of an oxidant to meet the various requirements of oxidative water treatment processes and to avoid underperformance, higher costs, and undesired byproduct formation during the oxidation step. As a consequence, there is considerable interest in simple and readily measurable parameters that provide information on the concentration and reactivity of the DOM in the water. 12,13 Two commonly measured parameters are the dissolved organic carbon (DOC) content, which captures the concentration of DOM, and the specic UV absorbance of the water at the wavelength of 254 nm (SUVA 254 , expressed in L mgC 1 m 1 ), which is a proxy for DOM aromaticity. 14 Previous work showed that both the consumption of chemical oxidants by DOM and the occurrence of some disinfection/oxidation byproducts are Received: June 7, 2013 Revised: August 22, 2013 Accepted: August 26, 2013 Published: August 26, 2013 Article pubs.acs.org/est © 2013 American Chemical Society 11147 dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 1114711156

Upload: dchy

Post on 07-Nov-2015

19 views

Category:

Documents


3 download

DESCRIPTION

article

TRANSCRIPT

  • Chemical Oxidation of Dissolved Organic Matter by Chlorine Dioxide,Chlorine, And Ozone: Eects on Its Optical and AntioxidantPropertiesJannis Wenk,,, Michael Aeschbacher, Elisabeth Salhi, Silvio Canonica, Urs von Gunten,,,,*and Michael Sander,*Eawag, Swiss Federal Institute of Aquatic Science and Technology CH-8600, Dubendorf, SwitzerlandInstitute of Biogeochemistry and Pollutant Dynamics, ETH Zurich, CH-8092, Zurich, SwitzerlandSchool of Architecture, Civil and Environmental Engineering (ENAC), Ecole Polytechnique Federale de Lausanne (EPFL),CH-1015 Lausanne, Switzerland

    *S Supporting Information

    ABSTRACT: In water treatment dissolved organic matter(DOM) is typically the major sink for chemical oxidants. Theresulting changes in DOM, such as its optical properties havebeen measured to follow the oxidation processes. However,such measurements contain only limited information on thechanges in the oxidation states of and the reactive moieties inthe DOM. In this study, we used mediated electrochemicaloxidation to quantify changes in the electron donatingcapacities (EDCs), and hence the redox states, of threedierent types of DOM during oxidation with chlorine dioxide(ClO2), chlorine (as HOCl/OCl

    ), and ozone (O3). Treat-ment with ClO2 and HOCl resulted in comparable andprominent decreases in EDCs, while the UV light absorbancesof the DOM decreased only slightly. Conversely, ozonation resulted in only small decreases of the EDCs but pronouncedabsorbance losses of the DOM. These results suggest that ClO2 and HOCl primarily reacted as oxidants by accepting electronsfrom electron-rich phenolic and hydroquinone moieties in the DOM, while O3 reacted via electrophilic addition to aromaticmoieties, followed by ring cleavage. This study highlights the potential of combined EDC-UV measurements to monitor chemicaloxidation of DOM, to assess the nature of the reactive moieties and to study the underlying reaction pathways.

    INTRODUCTIONDrinking water and wastewater treatment facilities often have achemical oxidation step for disinfection, the removal of organicmicropollutants, color removal, and taste and odor control.Among the most commonly used oxidants are chlorine dioxide(ClO2), chlorine (as hypochlorous acid, HOCl and OCl

    ), andozone (O3).

    1 For a number of reasons, the eciency of theoxidation step and the quality of the treated water largelydepend on the reaction of the chemical oxidant with dissolvedorganic matter (DOM). First, DOM is a major contributor todrinking water color, which negatively aects the acceptance ofthe water among consumers.2 Second, the reaction of DOMwith the chemical oxidants accelerates their consumption and,thus, may reduce the eciency of the oxidation step fordisinfection and micropollutant oxidation.3,4 Third, the reactionof the oxidants with DOM may result in the formation ofpotentially harmful disinfection/oxidation byproducts.57

    Fourth, chemical DOM oxidation results in the generation oflow molecular weight assimilable organic carbon (AOC).8,9

    Following the oxidation step, the AOC needs to be removed bybiological ltration to improve the biological stability of

    drinking waters.4,10,11 For these reasons, information on theDOM concentration and its reactivity is indispensable to ndthe appropriate dose of an oxidant to meet the variousrequirements of oxidative water treatment processes and toavoid underperformance, higher costs, and undesired byproductformation during the oxidation step.As a consequence, there is considerable interest in simple and

    readily measurable parameters that provide information on theconcentration and reactivity of the DOM in the water.12,13 Twocommonly measured parameters are the dissolved organiccarbon (DOC) content, which captures the concentration ofDOM, and the specic UV absorbance of the water at thewavelength of 254 nm (SUVA254, expressed in L mgC

    1 m1),which is a proxy for DOM aromaticity.14 Previous work showedthat both the consumption of chemical oxidants by DOM andthe occurrence of some disinfection/oxidation byproducts are

    Received: June 7, 2013Revised: August 22, 2013Accepted: August 26, 2013Published: August 26, 2013

    Article

    pubs.acs.org/est

    2013 American Chemical Society 11147 dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 1114711156

  • positively correlated to SUVA254.1519 These correlations

    suggest activated aromatic moieties as major oxidizablefunctional groups in DOM, consistent with the high reactivityof low-molecular weight activated aromatic moieties, includingphenols, methoxybenzenes, and anilines, with ClO2, chlorine,and O3.

    2029 However, despite the positive correlations withchemical oxidant consumption, SUVA254 alone was found to bea relatively poor predictor of DOM reactivity and disinfectionbyproduct formation with chlorine.14,30 Other methods thathave been used to determine the concentration and reactivity ofoxidizable moieties in DOM are dicult to adapt for routinewater analysis or provide only indirect information on the redoxstates of DOM.3137 Therefore, an analytical method isdesirable that allows for a direct quantication of changes inDOM oxidation states caused by reaction with chemicaloxidants.38

    Mediated electrochemical oxidation (MEO), an analyticaltechnique recently developed in our research group, fulllsthese requirements. MEO allows for a fast and reliablequantication of the electron donating capacities (EDCs)(i.e., the number of electrons that are donated by a givenamount of DOM) of dilute DOM samples in electrochemicalcells with well-controlled pH and Eh conditions.

    39,40 Wepreviously demonstrated that MEO quanties activatedphenolic moieties in DOM: EDC values of a set of chemicallydiverse humic substances (HS) were positively correlated withtheir titrated phenol contents and showed dependencies on Ehand pH comparable to those of low molecular weight phenolsand hydroquinones.40 We expect that chemical oxidants oxidizethese activated phenolic moieties in DOM, resulting indecreasing EDC values of the DOM during treatment. MEOmay therefore be a powerful technique to quantify DOMreactivity with chemical oxidants and to directly monitorchanges in DOM oxidation states during chemical oxidation inwater treatment.The goal of this study was to explore the potential of

    combined MEO and UVvisible absorbance measurements toselectively quantify the oxidation states of DOM duringchemical oxidation and to elucidate the underlying oxidant-dependent reaction pathways. We measured the UVvisabsorbance spectra and the EDC values of three HS (SuwanneeRiver Humic and Fulvic Acids (SRHA and SRFA) and PonyLake Fulvic Acid (PLFA)) during dose-dependent treatmentwith ClO2, chlorine, and O3. HS, in general, make up the majorfraction of DOM. We specically chose SRHA, SRFA, andPLFA because these materials are commercially available andhave been used in previous oxidation studies,18,41 and their keyphysicochemical properties are known. Furthermore, SRHA/FA and PLFA represent allochthonous and autochthonousaquatic HS with terrestrial higher plant-derived and withmicrobially derived precursor materials, respectively. This studyaddresses fundamental questions on the changes in DOMantioxidant properties and reactivities during chemicaloxidation, and, in the Implication section, highlights thepotential of combining MEO and SUVA254 measurements tomonitor chemical oxidant demand in water treatment facilities.

    MATERIALS AND METHODSChemicals. All chemicals were from commercial sources

    and used as received: tert-butanol (t-BuOH) (99.7%), 2,2-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammoniumsalt (ABTS) (>99%), potassium peroxodisulfate (99%),sodium chlorite (NaClO2) (puriss. p.a. 80%), sodium chlorate

    (NaClO3) (99%), ortho-phosphoric acid (85%), sodiumdihydrogen phosphate dihydrate (99%), disodium hydrogenphosphate dodecahydrate (98.0%) and hypochlorite solution614% were from Sigma-Aldrich, sodium dihydrogen phos-phate monohydrate (99102%) was from Merck.Humic Substances. Suwannee River Humic Acid Standard

    II (SRHA; catalogue number: 2S101H), Suwannee River FulvicAcid Standard II (SRFA; 2S101F), and Pony Lake Fulvic AcidReference (PLFA; 1R109F) were obtained from the Interna-tional Humic Substances Society (IHSS, St. Paul, MN) andused as received. Selected physicochemical properties of theHS, including elemental compositions, aromaticities andtitrated phenol contents, are provided in Table S1 in theSupporting Information (SI).Preparation of Aqueous Solutions. Aqueous solutions

    were prepared using deionized water either from Milli-Q(Millipore) or Barnsteadt water purication systems. HS stocksolutions (100 mg C L1) were prepared in 5 mM phosphatebuer (pH 8) or in deionized water. The DOC of the HS stocksolutions was determined after 25-fold dilution on a ShimadzuV-CPH TOC analyzer (Kyoto, Japan) and used to calculatespecic UV absorbance values and carbon-normalized EDCvalues.Chlorine dioxide (ClO2) stock solutions (10 mM) were

    produced by mixing potassium peroxodisulfate (K2S2O8, 2 g in50 mL water) with sodium chlorite (NaClO2, 4 g in 50 mL).

    42

    The stock solution of chlorine (Cl2; 10 mM) was prepared bydiluting a sodium hypochlorite solution with water. Ozone(O3) stock solutions (1.3 to 1.5 mM) were prepared bysparging ozone gas through water cooled in an ice bath.43 TheO3 gas was formed from pure oxygen with an Apaco CMG 33ozone generator (Grellingen, Switzerland). The exact concen-trations of oxidants in the stock solutions were quantiedspectrophotometrically using molar absorption coecients of = 1200 M1 cm1 at = 359 nm for ClO2,

    44 = 350 M1 cm1

    at = 290 nm for chlorine (as ClO),45 and = 3000 M1

    cm1 at = 258 nm for ozone.46

    ClO2, Chlorine, and O3 Oxidation of DOM. Oxidationexperiments were carried out in a series of identical glassreaction vessels (50 or 100 mL) (Schott, Germany). Thevessels contained either DOM solutions (nominal concen-trations of 0.83 mmol C L1 (=10 mg C L1) after reagentmixing) or DOM-free blank solutions at pH 7 (50 mMphosphate buer). Oxidant stock solutions were added to thevessels under vigorous mixing on a magnetic stirrer plate. Theemployed oxidant doses were in the range of 00.36 mmolClO2/mmol C, 00.85 mmol chlorine/mmol C, and 01.12mmol O3/mmol C, which cover the ranges commonly used forwater treatment.28,47,48 Ozonation experiments were performedin the presence (5 mM) and absence of t-BuOH as a scavengerfor formed hydroxyl radicals (OH). After oxidant addition, thevessels were closed, removed from the stirrer and stored at 22C for 12h for chlorine dioxide, 3 days for chlorine, and 2 h forozone. Subsequently, unreacted ClO2 and O3 were removedfrom the solution by gently purging with helium for 20 min. Inselected experiments, residual chlorine (max. 0.5 M) wasmeasured using the DPD colorimetric method.42 The ozoneexposure in the DOM-containing systems in the presence andabsence of t-BuOH was measured according to previouslydescribed methods.49,50 Control experiments in which t-BuOHwas added to the solutions after depletion of ozone showed thatt-BuOH did not aect UVvisible absorption and EDCmeasurements.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611148

  • UV/Visible Light Absorbance Measurements. Absorb-ance spectra of untreated and oxidized HS were collected onUvikon 940 (Kontron Instruments) or Varian Cary 100(Agilent Technologies) spectrophotometers in quartz glasscuvettes (Hellma) (10 or 100 mm path lengths). All samplespectra were corrected for the spectrum of the HS-freephosphate buer (pH 7). The carbon-specic absorptioncoecients of untreated and treated HS, a() [L/(mg Cm)],were calculated according to eq 1, where A() is the sampleabsorbance at a given wavelength , b [m] is the path length,and CHS [mg C/L] is the organic carbon concentration of theuntreated HS.

    =

    aAb C

    ( )( )

    HS (1)

    The values a(254 nm) and a(280 nm) are referred to asSUVA254 and SUVA280, respectively. The spectral slopecoecients of the HS absorbance spectra, S [1/nm], wereobtained by nonlinear least-squares tting of DOM absorptiondata from = 300 to 600 nm with a single exponential decayfunction,51 where a(ref) is the specic absorption coecient atthe reference wavelength of ref = 350 nm.

    52

    = a a S( ) ( ) exp[ ( )]ref ref (2)

    The parameter S describes the steepness of DOM absorbancespectra on a logarithmic scale: The relative decrease inabsorbance with increasing wavelength becomes steeper as Sincreases. Changes in the spectral slopes were also determinedover narrower wavelength ranges (i.e., from 275 to 295 nm,S275295, and from 350 to 385 nm, S350385) following theapproach suggested by Helms and co-workers.53 Data ttingand integrations were performed using Origin 8.0 software(OriginLab).Quantication of Electron Donating Capacities. EDC

    values of untreated and oxidant-treated HS solutions werequantied by MEO using 2,2-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) as electron transfer mediator.39,40

    MEO measurements were conducted in an electrochemical cellcontaining a reticulated vitreous carbon working electrode(WE), a Pt counter electrode, and an Ag/AgCl referenceelectrode. The electrochemical cells were rst lled with 6065mL of buer solution (0.1 M KCl, 0.1 M phosphate, pH 7) andthe WE was polarized to an oxidizing potential of Eh = +0.725V vs the standard hydrogen electrode (SHE), controlled by apotentiostat (either an Autolab PG302 (EcoChemie B.V.) or a630C instrument (CH Instruments)). A volume of 2 mL of anaqueous ABTS solution (5 mM) was added to the cell, resultingin an oxidative current peak due to the oxidation of ABTS to itsradical cation ABTS+ (standard reduction potentialEh

    0(ABTS+/ABTS) = 0.68 V vs SHE54). Upon attainment ofredox equilibrium between ABTS+/ABTS and the WE (andhence stable current readings), HS samples (57 mL) weresuccessively spiked to the cell. Oxidation of electron donatingmoieties in the added HS by ABTS+ resulted in the formationof reduced ABTS, which was subsequently reoxidized at theWE to ABTS+ to re-establish redox equilibrium. The resultingoxidative current peak was integrated to yield the EDC valuesof the added HS:

    =

    t

    mEDC

    dIF

    HS (Eq. 3)

    where I [A] is the baseline-corrected current and F (=96485 sA/mole) is the Faraday constant, and mHS [mgC or mmolC ] isthe mass/amount of HS analyzed. Most HS samples wereanalyzed in triplicates and some in duplicates with t = 50 minbetween replicate analysis to ensure baseline-separation ofindividual current peaks.

    RESULTS AND DISCUSSIONEects of Oxidant Treatments on DOM Optical

    Properties. The specic absorption coecients of theuntreated samples decreased in the order SRHA>SRFA>PLFAover the entire measured wavelength range from 220 to 600 nm(SI Figures S1 and S2). The trend in the absorption coecientsfollows the decrease in HS aromaticity55 from 31% for SRHAto 22% for SRFA and 12% for PLFA (SI Table S1).56 Theabsorbance spectrum of untreated SRHA extended further intothe red than the spectra of both SRFA and PLFA, which isreected by the smaller S values for SRHA than for SRFA andPLFA. Longer wavelength absorbance of HS has been ascribedto charge transfer complexes between electron donor andacceptor pairs in HS,55,57,58 which may be more abundant inHA than FA.Treatment of the HS with all oxidants resulted in decreasing

    specic absorption coecients at all collected wavelengths(Figure 1ac) and increasing S values (Figure 1df) withincreasing oxidant doses, consistent with previous re-ports.8,37,59,60 The absorbance spectra, the dierential spectra,and the spectral slopes S275295 and S350385 of untreated andoxidant-treated HS are shown in SI Figures S1S5. Overall, thedecreases in the specic absorption coecients suggest adecrease in aromaticity of the treated HS. The increase in Svalues with increasing oxidant dose indicates that moieties/complexes absorbing at longer wavelengths were preferentiallyremoved and/or transformed into shorter wavelength-absorb-ing components. The increase in S and S275295 values withincreasing oxidant doses may also reect decreases in theaverage molecular weights of the DOM upon reaction with thechemical oxidants, as detailed in the SI. Consistent withprevious observations,30,61 the dierential spectra for HOCltreated HS show a local maximum in absorption loss at around270272 nm, indicating a selective removal of chromophoresin this wavelength region by reaction with chlorine under theassumption that no new chromophores are formed. A similarmaximum loss in absorbance around 270 nm was also observedfor ClO2-treated PLFA. This feature was absent from thedierential spectra of ClO2-treated SRHA and SRFA as well asof the O3-treated SRHA, SRFA, and PLFA both in absence andpresence of t-BuOH.A detailed analysis of the absorbance and dierential

    absorbance spectra revealed that ClO2 and HOCl treatmentshad dierent eects on DOM optical properties than the O3treatments. The SUVA254 and SUVA280 values of all three HSdecreased linearly with increasing doses of ClO2 and HOCl andfollowed similar dose-dependencies for the two oxidants(Figure 1a-c and SI Figure S6, respectively). ClO2 and HOCltreatment of SRHA and SRFA also resulted in comparableincreases in S with increasing oxidant doses, whereas PLFAshowed larger increases in S upon treatment with ClO2 thanHOCl at the same specic molar oxidant doses (Figure 1df).In comparison to the ClO2 and HOCl treatments, ozonationresulted in much larger decreases in the specic absorptioncoecients of the HS, both in the absence and presence of t-BuOH (Figure 1ac and SI Figure S6df). The larger

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611149

  • decreases in SUVA254 and SUVA280 upon treatment with O3than ClO2 and HOCl at the same specic molar oxidant dosesdemonstrates that UV-light absorbing aromatic moieties in theHS were more eciently removed (or transformed to lesseciently absorbing moieties) by O3 than by both ClO2 andHOCl. Note that narrower dose ranges were used for ClO2 andHOCl than for O3 based on the eects of the three oxidants onthe antioxidant properties of the HS, as detailed below.Ozonation in the presence of t-BuOH resulted in larger

    losses in HS absorbance at wavelengths >315 nm and largerincreases in S than in the absence of t-BuOH. These eects of t-BuOH can be ascribed to two factors. First, t-BuOH scavengesOH which are formed by DOMozone reactions and whichcatalytically degrade O3.

    16,62 Quenching of OH by t-BuOHtherefore enhanced O3 lifetimes and, hence, resulted in higherO3 exposures of the HS. Enhanced O3 exposure in the presencecompared to the absence of t-BuOH was veried exper-imentally with PLFA solutions (see SI Figures S7, S8). Second,by scavenging OH, t-BuOH shifted the overall oxidationpathway from unselective, diusion-controlled OH additions,H abstraction (and electron transfer reactions),62 to moreselective, direct reactions of O3 with moieties such as olens,activated aromatics, and amines in the DOM.62 The presence oft-BuOH therefore enhanced O3-reaction induced cleavage oflight absorbing olenic and aromatic systems,63 resulting inlarger changes in HS optical properties than in the absence of t-BuOH. The abatement of SUVA254 by ozone (Figure 1ac)was more ecient for SRHA in presence than in absence of t-BuOH, whereas the opposite trend was observed for specicozone doses >0.2 (mmolozone (mmolc)

    1) for SRFA and PLFA.

    This can be explained by the higher content of ozone-reactivechromophores in SRHA than in SRFA and PLFA. For SRFAand PLFA, these moieties are depleted for specic ozone doses>0.2 (mmolozone (mmolc)

    1) in the presence of t-BuOH, andozone-resistant SUVA254-contributing chromophores can thenonly be further oxidized by OH, which are present duringozonation in absence of t-BuOH.Eects of Oxidant Treatments on DOM Antioxidant

    Properties. In a rst set of experiments we evaluated thesensitivity of MEO to detect oxidant-induced changes in theEDCs of HS by quantifying the kinetics of PLFA oxidation byO3 at a constant initial dose of 0.5 mmol O3/ mmol C. Wechose PLFA because it has the lowest EDC values of severalDOMs previously tested.40 Figure 2a shows the evolution ofthe oxidative current responses in MEO for PLFA samples afterreaction with O3 for various reaction times. The correspondingEDC values, obtained by integration of the oxidative currentpeaks (Eq. 3), show fast oxidation of the electron donatingmoieties in PLFA by O3 (Figure 2b): Within one minute and12 minutes of reaction, the EDC of PLFA decreased toapproximately 50% and 15% of its original value, respectively.The results of this experiment demonstrate the suitability ofMEO to quantify changes in the oxidation states of HS duringtreatments with chemical oxidants. Based on the reactionkinetics, the dose-dependent ozonation experiments were runfor 2 h to guarantee completion of HS-O3 reactions.Figure 3 shows that the EDCs of SRHA, SRFA, and PLFA

    decreased with increasing doses of ClO2, HOCl and O3 (bothin the presence and absence of t-BuOH) and, hence, dose-dependent removal of electron donating moieties in the HS for

    Figure 1. Changes in the optical properties of Suwannee River Humic Acid (SRHA), Suwannee River Fulvic Acid (SRFA), and Pony Lake FulvicAcid (PLFA) upon treatment with chlorine dioxide (ClO2), chlorine (as HOCl), and ozone (O3) (both in the absence and presence of t-BuOH).Panels (a)-(c): Changes in the specic UV absorption at 254 nm (i.e., SUVA254) of (a) SRHA, (b) SRFA, and (c) PLFA as a function of the specicmolar oxidant dose (mmoloxidant (mmolc)

    1). Panels (d)(f): Changes in the spectral slope S (from 300 to 600 nm) of (d) SRHA, (e) SRFA, and (f)PLFA as a function of the specic molar oxidant dose.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611150

  • all three oxidants. Normalized to the same specic molaroxidant dose, the decreases in EDC were largest for ClO2,intermediate for HOCl and O3 in the presence of t-BuOH, andsmallest for O3 in the absence of t-BuOH: Linear ts of thedecreases in EDC values of SRHA and SRFA at low specicmolar oxidant doses had the steepest slopes for ClO2 (i.e.,0.69 and 0.46 mmole(mmol ClO2)1), intermediate slopesfor HOCl (i.e., 0.38 and 0.36 mmole(mmol HOCl)1) andfor O3 in the presence of t-BuOH (i.e., 0.35 and 0.29mmole(mmol O3)

    1), and the shallowest slopes for O3 in theabsence of t-BuOH (i.e., 0.15 and 0.08 mmole(mmolO3)

    1) (SI Figure S9, Table S2)). The PLFA data did not showa linear decrease in EDC with increasing specic molar oxidantdose and could therefore not be tted.Treatments with high doses of ClO2 and HOCl resulted in

    complete loss of EDC in some of the systems, including areplicate SRFA-ClO2 experiment (SI Figure S10), whereas allHS retained some EDC during ozonation even at the highestO3 doses. The removal of electron-donating moieties in thetested HS was therefore more ecient by ClO2 and HOCltreatments than by ozonation. The larger decreases in EDCs byO3 in the presence than in the absence of t-BuOH can beassigned to OH quenching by t-BuOH and hence higher O3exposures of HS and more selective oxidations of electrondonating moieties by O3.Mechanistic Interpretation. In the following, the changes

    in the optical and the antioxidant properties of the HS will befurther explored by plotting the oxidant-induced decreases inthe SUVA254 values versus the corresponding decreases in theEDC values (Figure 4). Treatments of the HS with ClO2 andHOCl resulted in comparable SUVA254-EDC dependencies forthese two oxidants with larger relative decreases in the EDCthan in the SUVA254 values. This nding implies a moreecient removal of electron donating phenolic moieties thanUV-light absorbing aromatic moieties upon treatment of theHS with ClO2 and HOCl. Compared to the ClO2 and HOCltreatments, ozonation in the presence and absence of t-BuOHled to distinctly dierent SUVA254-EDC dependencies withlarger relative losses in the SUVA254 than in the EDC values.Ozonation therefore caused a more ecient removal of UV-light absorbing aromatic moieties than electron donatingphenolic moieties. Ozonation of SRHA in the presence andabsence of t-BuOH resulted in comparable SUVA254-EDCdependencies. Conversely, ozonation of SRFA in the presence

    of t-BuOH resulted in smaller decreases in SUVA254 and largerdecreases in the EDC values as compared to ozonation in theabsence of t-BuOH at the same initial ozone dose. We note thatthe EDC measurements of PLFA samples at high O3 doseswere close to the quantication limit of MEO. The apparentincrease in the EDC value of PLFA at the highest O3 doses(Figure 3c) therefore likely reected uncertainties in the EDCquantication.The eects on the optical and antioxidant properties of the

    HS shown in Figure 4 can be rationalized on the basis of knownmajor reaction pathways of ClO2, HOCl, and O3 with light-absorbing and electron donating phenolic moieties in the HS(Figure 5). ClO2 reacts as a one-electron transfer oxidant withlow molecular weight phenols forming chlorite and thecorresponding phenoxyl radicals.21 At circumneutral pH, thisreaction proceeds mostly via the phenolate species because ofits oxidation rate constants with ClO2 that are about 6 orders ofmagnitude higher than those for the nondissociated phenolspecies.26 Analogously to low molecular weight phenols,phenolic moieties in HS are expected to undergo one electronoxidation by ClO2.We have previously shown that HS contain electron donating

    phenolic and hydroquinone moieties with apparent oxidationpotentials40,64 much lower than the standard reductionpotential of ClO2, Eh

    0(ClO2 (aq)/ClO2) = 0.954 V.65 SRHA

    and SRFA are derived from higher-plant precursor materials,including lignin, which is rich in methoxylated phenols.66

    Generally, methoxylation activates phenols for electrophilicattack and leads to faster oxidation kinetics.26 Phenoxyl radicalsresulting from a rst one electron oxidation26,67,68 may eitherbe further oxidized by reacting with another ClO2 to formortho- or para-quinones or undergo irreversible couplingreactions. Hydroquinone moieties present in the untreatedHS are expected to be oxidized by ClO2 to semiquinoneintermediates and subsequently to the respective quinonemoieties. These reaction pathways involving ClO2 as theoxidant have in common that electron donating phenolicmoieties are oxidized, whereas their UV-light absorbingaromatic structure is preserved. In fact, based on the highermolar absorption coecient of benzoquinone than hydro-quinone at 254 nm, the oxidation of hydroquinone to quinonemoieties in the DOM may have resulted in higher SUVA254values than measured if no hydroquinone moieties had beenoxidized. These pathways are therefore fully consistent with the

    Figure 2. Ozonation of Pony Lake Fulvic Acid (PLFA). Eects of the reaction time on (a) the oxidative current responses in mediatedelectrochemical oxidation (MEO) and (b) the corresponding electron donating capacities (EDC) of PLFA. Experimental conditions: 0.5 mmol O3/mmol C; 5 mM t-butanol; 50 mM PO4-buer, pH 7.0. The samples were quenched with 1 mM maleic acid at selected reaction times.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611151

  • pronounced decreases in the EDC and the relatively smalllosses in SUVA254 values observed for HS treatment with ClO2.Phenolic moieties in HS may react with HOCl in an

    electrophilic substitution reaction (Figure 5). At circumneutralpH, this reaction proceeds via the phenolate due to its muchhigher reactivity compared to the phenol.28 In this reaction,HOCl attacks at the ortho and para positions to the hydroxylsubstituent, resulting in the formation of (poly)-chlorinatedphenols. Such an initial chlorination should not lead to adecrease in the electron donating capacities of the phenolicmoieties.69,70 The reaction of low molecular weight phenolswith HOCl has been demonstrated to proceed via polychlori-nated phenols which ultimately undergo ring cleavage to formnonaromatic, chlorinated products (Figure 5).28 However, thesmall changes in the SUVA254 values of HS upon HOCltreatment do not support ring cleavage as a signicant reactionpathway for phenolic, or more general, aromatic moietiespresent in HS. Alternatively, the smaller relative decreases inthe SUVA254 than EDC values upon HOCl treatment areconsistent with the two-electron oxidation of hydroquinoneand/or catechol moieties by HOCl to form the respectivequinone moieties and chloride. These reactions are thermody-namically favorable given that the standard reduction potentialsfor the two electron reductions of HOCl and OCl

    (pKa(HOCl) = 7.54 at 25 C) (i.e., HOCl + H+ + 2e

    Cl + H2O: Eh0 = 1.48 V71 and OCl + H2O + 2e

    Cl + 2OH; Eh

    0 = 0.84 V71) are much higher than the oxidationpotentials of hydroquinones. This is in agreement with the highsecond order rate constants for the reaction of HOCl with

    hydroxyphenols.72 This reaction pathway may therefore resultin similar changes in the optical and antioxidant properties asClO2, which acts almost exclusively by an electron transfermechanism.The reaction of phenolic moieties with O3 at circumneutral

    pH is dominated by phenolate and initiated by an ozoneadduct, which may react further by (i) loss of ozonide, O3

    , toform a phenoxyl radical, (ii) loss of H2O2 to form an orthobenzoquinone, (iii) loss of singlet oxygen, 1O2, to form acatechol-type compound, and (iv) a Criegee-type reaction witha cleavage of the aromatic ring.23,24,62 The formations ofphenoxyl radicals (pathway (i)) and catechols (pathway (ii))are important but minor pathways for the oxidation of phenolwith ozone.23 If these would be the dominant reactionpathways of phenolic moieties during ozonation, this wouldlead to comparatively large decreases in the EDC and smalldecreases in the SUVA254 values, whereas the opposite eectwas observed experimentally (Figure 4). Instead, thepronounced decreases in SUVA254 support ring cleavage ofphenols and hydroquinones via the Criegee mechanism(pathway (iv)) to form muconic-type compounds andeventually aliphatic aldehydes (Figure 5). Ring cleavagereactions may have involved nonphenolic aromatic moietiessuch as anisoles and polymethoxybenzenes, as demonstrated forlow-molecular weight methoxylated compounds.22 The loss ofthese moieties would have resulted in decreasing SUVA254without aecting the EDC values of the HS, as both the targetcompounds and products would not be oxidizable in MEO.

    Figure 3. Dependencies of the electron donating capacities (EDCs) of (a) Suwannee River Humic Acid (SRHA), (b) Suwannee River Fulvic Acid(SRFA), and (c) Pony Lake Fulvic Acid (PLFA) on the specic molar doses of the chemical oxidants chlorine dioxide (ClO2), chlorine (as HOCl),and ozone (in the absence and presence of t-BuOH).

    Figure 4. Eect of chemical oxidant treatments on the specic UV absorbances (SUVA254) and the electron donating capacities (EDCs) of (a)Suwanee River Humic Acid (SRHA), (b) Suwannee River Fulvic Acid (SRFA), and (c) Pony Lake Fulvic Acid (PLFA). The chemical oxidants usedwere chlorine dioxide (ClO2), chlorine (as HOCl), and ozone (O3; in the absence and presence of tertiary butanol (t-BuOH)). The chemical oxidantdose increased in the directions indicated by the grey arrows.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611152

  • IMPLICATIONSThis study establishes that the EDC of DOM is a parameterthat directly relates to the DOM redox state. The EDC is highlysensitive to changes in DOM occurring during chemicaloxidation processes and can be readily quantied by mediatedelectrochemical oxidation (MEO). If combined with measure-ments of complementary optical parameters, such as SUVA254,the changes in the EDC values provide information on thekinetics and the dose-dependent oxidation of electron donatingmoieties in DOM. The combined analysis of optical andantioxidant properties also provides insight into which moietiesin the DOM react with the chemical oxidants and helps

    identifying the major oxidant-dependent reaction pathways ofDOM.In addition to advancing the fundamental understanding of

    chemical DOM oxidation, the results from this study are alsorelevant from a more applied, water treatment perspective.MEO has potential to be used in water treatment facilities tomonitor DOM oxidation during a chemical oxidation step.Combined determination of changes in the EDC and SUVA254(or other suitable optical parameters) in close to real-time canbe used to control chemical oxidant doses. The resultingrened dosing operation can minimize overdosing which mayhave negative impacts on water quality, such as the enhancedformation of disinfection/oxidation byproducts. Future work

    Figure 5. Proposed reaction pathways of phenolic moieties in the humic substances during reaction with chlorine dioxide (ClO2), chlorine (HOCl),and ozone (O3).

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611153

  • needs to assess the potential of EDC-SUVA254 measurements asa new tool to advance the understanding of and the capabilityto predict other important processes occurring during chemicaloxidation of DOM, such as the formation of disinfectionbyproducts, the generation of assimilable carbon, and theeciency of disinfection.

    ASSOCIATED CONTENT*S Supporting InformationAdditional information as noted in the text. This material isavailable free of charge via the Internet at http://pubs.acs.org.

    AUTHOR INFORMATIONCorresponding Authors*(U.v.G.) Phone: +41-(0)58 765 5270; fax: +41-(0)58 7655210; e-mail: [email protected].*(M.S.) Phone: +41-(0)44 6328314; fax: +41 (0)44 633 1122;e-mail: [email protected].

    Present Address (J.W.) Department of Civil & Environmental Engineering,University of California at Berkeley, Berkeley, California 94720,United States and ReNUWIt Engineering Research CenterNotesThe authors declare no competing nancial interest.

    ACKNOWLEDGMENTSJ.W. and M.A. contributed equally to the work. This work wassupported by the Swiss National Science Foundation (Beitrage200021-117911 and 200020-134801). We thank Marcel Burgerfor help on measuring absorbance spectra and Hans-UlrichLaubscher for technical support.

    REFERENCES(1) Edzwald, J. K., Water Quality & Treatment: A Handbook onDrinking Water, 6th ed.; McGraw-Hill Professional: New York, 2011.(2) Secondary Drinking Water Regulations: Guidance for NuisanceChemicals, 816-F-10-079; U.S. Environmental Protection Agency:Washington, DC.(3) Wert, E. C.; Rosario-Ortiz, F. L.; Snyder, S. A. Effect of ozoneexposure on the oxidation of trace organic contaminants in wastewater.Water Res. 2009, 43 (4), 10051014.(4) von Gunten, U. Ozonation of drinking water: Part I. Oxidationkinetics and product formation. Water Res. 2003, 37 (7), 14431467.(5) Richardson, S. D.; Plewa, M. J.; Wagner, E. D.; Schoeny, R.;DeMarini, D. M. Occurrence, genotoxicity, and carcinogenicity ofregulated and emerging disinfection by-products in drinking water: Areview and roadmap for research. Mutat. Res., Rev. Mutat. Res. 2007,636 (13), 178242.(6) Sedlak, D. L.; von Gunten, U. The chlorine dilemma. Science2011, 331 (6013), 4243.(7) Lavonen, E. E.; Gonsier, M.; Tranvik, L. J.; Schmitt-Kopplin, P.;Kohler, S. J. Selective chlorination of natural organic matterIdentification of perviously unknown disinfection by-products. Environ.Sci. Technol. 2013, 47 (5), 22642271.(8) Swietlik, J.; Dabrowska, A.; Raczyk-Stanislawiak, U.; Nawrocki, J.Reactivity of natural organic matter fractions with chlorine dioxide andozone. Water Res. 2004, 38 (3), 547558.(9) Ramseier, M. K.; Peter, A.; Traber, J.; von Gunten, U. Formationof assimilable organic carbon during oxidation of natural waters withozone, chlorine dioxide, chlorine, permanganate, and ferrate. WaterRes. 2011, 45 (5), 20022010.(10) Camel, V.; Bermond, A. The use of ozone and associatedoxidation processes in drinking water treatment. Water Res. 1998, 32(11), 32083222.

    (11) Hammes, F.; Salhi, E.; Koster, O.; Kaiser, H. P.; Egli, T.; vonGunten, U. Mechanistic and kinetic evaluation of organic disinfectionby-product and assimilable organic carbon (AOC) formation duringthe ozonation of drinking water. Water Res. 2006, 40 (12), 22752286.(12) Bourgeois, W.; Burgess, J. E.; Stuetz, R. M. On-line monitoringof wastewater quality: A review. J. Chem. Technol. Biotechnol. 2001, 76(4), 337348.(13) Olsson, G. ICA and meA subjective review. Water Res. 2012,46 (6), 15851624.(14) Weishaar, J. L.; Aiken, G. R.; Bergamaschi, B. A.; Fram, M. S.;Fujii, R.; Mopper, K. Evaluation of specific ultraviolet absorbance as anindicator of the chemical composition and reactivity of dissolvedorganic carbon. Environ. Sci. Technol. 2003, 37 (20), 47024708.(15) Kitis, M.; Karanfil, T.; Wigton, A.; Kilduff, J. E. Probingreactivity of dissolved organic matter for disinfection by-productformation using XAD-8 resin adsorption and ultrafiltration fractiona-tion. Water Res. 2002, 36 (15), 38343848.(16) Buffle, M. O.; Schumacher, J.; Meylan, S.; Jekel, M.; vonGunten, U. Ozonation and advanced oxidation of wastewater: Effect ofO3 dose, pH, DOM, and HO center dot-scavengers on ozonedecomposition and HO center dot generation. Ozone: Sci. Eng. 2006,28 (4), 247259.(17) Westerhoff, P.; Aiken, G.; Amy, G.; Debroux, J. Relationshipsbetween the structure of natural organic matter and its reactivitytowards molecular ozone and hydroxyl radicals. Water Res. 1999, 33(10), 22652276.(18) Elovitz, M. S.; Von Gunten, U.; Kaiser, H., The inuence ofdissolved organic matter character on ozone decomposition rates andRct. In Natural Organic Matter and Disinfection By-Products: Character-ization and Control in Drinking Water; Barrett, S. E., Krasner, S. W.;Amy, G. L., Eds.; ACS Publications, 2000; Vol. 761, pp 248269.(19) Reckhow, D. A.; Singer, P. C.; Malcolm, R. L. Chlorination ofhumic materials: Byproduct formation and chemical interpretations.Environ. Sci. Technol. 1990, 24 (11), 16551664.(20) Hoigne, J.; Bader, H. Rate constants of reactions of ozone withorganic and inorganic compounds in water. 1. Non-dissociatingorganic compounds. Water Res. 1983, 17 (2), 173183.(21) Hoigne, J.; Bader, H. Rate constants of reactions of ozone withorganic and inorganic compounds in water. 2. Dissociating organiccompounds. Water Res. 1983, 17 (2), 185194.(22) Mvula, E.; Naumov, S.; von Sonntag, C. Ozonolysis of ligninmodels in aqueous solution: Anisole, 1,2-dimethoxybenzene, 1,4-dimethoxybenzene, and 1,3,5-trimethoxybenzene. Environ. Sci. Technol.2009, 43 (16), 62756282.(23) Mvula, E.; von Sonntag, C. Ozonolysis of phenols in aqueoussolution. Org. Biomol. Chem. 2003, 1 (10), 17491756.(24) Ramseier, M. K.; von Gunten, U. Mechanisms of phenolozonation-kinetics of formation of primary and secondary reactionproducts. Ozone: Sci. Eng. 2009, 31 (3), 201215.(25) Singer, P. C.; Gurol, M. D. Dynamics of the ozonation ofphenol. 1. Experimental observations. Water Res. 1983, 17 (9), 11631171.(26) Tratnyek, P. G.; Hoigne, J. Kinetics of reactions of chlorinedioxide (OClO) in water. 2. Quantitative structure activity relation-ships for phenolic compounds. Water Res. 1994, 28 (1), 5766.(27) Norwood, D. L.; Johnson, J. D.; Christman, R. F.; Hass, J. R.;Bobenrieth, M. J. Reactions of chlorine with selected aromatic modelsof aquatic humic material. Environ. Sci. Technol. 1980, 14 (2), 187190.(28) Gallard, H.; von Gunten, U. Chlorination of phenols: Kineticsand formation of chloroform. Environ. Sci. Technol. 2002, 36 (5), 884890.(29) Lee, Y.; von Gunten, U. Oxidative transformation ofmicropollutants during municipal wastewater treatment: Comparisonof kinetic aspects of selective (chlorine, chlorine dioxide, ferrate(VI),and ozone) and non-selective oxidants (hydroxyl radical). Water Res.2010, 44 (2), 555566.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611154

  • (30) Li, C. W.; Benjamin, M. M.; Korshin, G. V. The relationshipbetween TOX formation and spectral changes accompanyingchlorination of pre-concentrated or fractionated NOM. Water Res.2002, 36 (13), 32653272.(31) Abbt-Braun, G.; Lankes, U.; Frimmel, F. H. Structuralcharacterization of aquatic humic substancesThe need for a multiplemethod approach. Aquatic Sciences 2004, 66 (2), 151170.(32) Westerhoff, P.; Mezyk, S. P.; Cooper, W. J.; Minakata, D.Electron pulse radiolysis determination of hydroxyl radical rateconstants with Suwannee river fulvic acid and other dissolved organicmatter isolates. Environ. Sci. Technol. 2007, 41 (13), 46404646.(33) Westerhoff, P.; Debroux, J.; Aiken, G.; Amy, G. Ozone-inducedchanges in natural organic matter (NOM) structure. Ozone: Sci. Eng.1999, 21 (6), 551570.(34) These, A.; Reemtsma, T. Structure-dependent reactivity of lowmolecular weight fulvic acid molecules during ozonation. Environ. Sci.Technol. 2005, 39 (21), 83828387.(35) Pelekani, C.; Newcombe, G.; Snoeyink, V. L.; Hepplewhite, C.;Assemi, S.; Beckett, R. Characterization of natural organic matter usinghigh performance size exclusion chromatography. Environ. Sci. Technol.1999, 33 (16), 28072813.(36) Collins, M. R.; Amy, G. L.; Steelink, C. Molecular weightdistribution, carboxylic acidity, and humic substance content of aquaticorganic matter - Implications for removal during water treatment.Environ. Sci. Technol. 1986, 20 (10), 10281032.(37) Korshin, G. V.; Benjamin, M. M.; Chang, H. S.; Gallard, H.Examination of NOM chlorination reactions by conventional and stop-flow differential absorbance spectroscopy. Environ. Sci. Technol. 2007,41 (8), 27762781.(38) Macalady, D. L.; Walton-Day, K., Redox chemistry and naturalorganic matter (NOM): Geochemists dream, analytical chemistsnightmare. In Aquatic Redox Chemistry; Tatnyek, P. G., Grundl, T. J.,Haderlein, S. B., Eds.; American Chemical Society: Washington, DC,2011.(39) Aeschbacher, M.; Sander, M.; Schwarzenbach, R. P. Novelelectrochemical approach to assess the redox properties of humicsubstances. Environ. Sci. Technol. 2010, 44 (1), 8793.(40) Aeschbacher, M.; Graf, C.; Schwarzenbach, R. P.; Sander, M.Antioxidant properties of humic substances. Environ. Sci. Technol.2012, 46 (9), 49164925.(41) Zhang, X. R.; Minear, R. A.; Barrett, S. E. Characterization ofhigh molecular weight disinfection byproducts from chlorination ofhumic substances with/without coagulation pretreatment using UF-SEC-ESl-MS/MS. Environ. Sci. Technol. 2005, 39 (4), 963972.(42) Gates, D. The Chlorine Dioxide Handbook; American WaterWorks Association: Denver, CO, 1998.(43) Bader, H.; Hoigne, J. Determination of ozone in water by theindigo method. Water Res. 1981, 15 (4), 449456.(44) Hoigne, J.; Bader, H. Kinetics of reactions of chlorine dioxide(OClO) in water. 1. Rate constants for inorganic and organiccompounds. Water Res. 1994, 28 (1), 4555.(45) Soulard, M.; Bloc, F.; Hatterer, A. Diagrams of existence ofchloramines and bromamines in aqueous solution. J. Chem. Soc., DaltonTransactions 1981, 12, 23002310.(46) Liu, Q.; Schurter, L. M.; Muller, C. E.; Aloisio, S.; Francisco, J.S.; Margerum, D. W. Kinetics and mechanisms of aqueous ozonereactions with bromide, sulfite, hydrogen sulfite, iodide, and nitriteions. Inorg. Chem. 2001, 40 (17), 44364442.(47) Huber, M. M.; Korhonen, S.; Ternes, T. A.; von Gunten, U.Oxidation of pharmaceuticals during water treatment with chlorinedioxide. Water Res. 2005, 39 (15), 36073617.(48) Zimmermann, S. G.; Wittenwiler, M.; Hollender, J.; Krauss, M.;Ort, C.; Siegrist, H.; von Gunten, U. Kinetic assessment and modelingof an ozonation step for full-scale municipal wastewater treatment:Micropollutant oxidation, by-product formation and disinfection.Water Res. 2011, 45 (2), 605617.(49) Hoigne, J.; Bader, H. Characterization of water quality criteriafor ozonation processes. 2. Lifetime of added ozone. Ozone: Sci. Eng.1994, 16 (2), 121134.

    (50) von Gunten, U.; Holgne, J. Bromate formation during ozonationof bromide containing watersInteraction of ozone and hydroxylradical reactions. Environ. Sci. Technol. 1994, 28 (7), 12341242.(51) Twardowski, M. S.; Boss, E.; Sullivan, J. M.; Donaghay, P. L.Modeling the spectral shape of absorption by chromophoric dissolvedorganic matter. Mar. Chem. 2004, 89 (14), 6988.(52) Golanoski, K. S.; Fang, S.; Del Vecchio, R.; Blough, N. V.Investigating the mechanism of phenol photooxidation by humicsubstances. Environ. Sci. Technol. 2012, 46 (7), 39123920.(53) Helms, J. R.; Stubbins, A.; Ritchie, J. D.; Minor, E. C.; Kieber, D.J.; Mopper, K. Absorption spectral slopes and slope ratios as indicatorsof molecular weight, source, and photobleaching of chromophoricdissolved organic matter. Limnol. Oceanogr. 2008, 53 (3), 955969.(54) Scott, S. L.; Chen, W. J.; Bakac, A.; Espenson, J. H.Spectroscopic parameters, electrode potentials, acid ionizationconstants, and electron exchange rates of the 2,2-azinobis(3-ethylbenzothiazoline-6-sulfonate) radicals and ions. J. Phys. Chem.1993, 97 (25), 67106714.(55) Chin, Y. P.; Aiken, G.; Oloughlin, E. Molecular weight,polydispersity, and spectroscopic properties of aquatic humicsubstances. Environ. Sci. Technol. 1994, 28 (11), 18531858.(56) Thorn, K. A.; Folan, D. W.; MacCarthy, P. Characterization ofthe International Humic Substances Society Standard and Reference Fulvicand Humic Acids by Solution State Carbon-13 (13C) and Hydrogen-1(1H) Nuclear Magnetic Resonance Spectrometry; U.S. Geological Survey:Denver, CO, 1989.(57) Boyle, E. S.; Guerriero, N.; Thiallet, A.; Del Vecchio, R.; Blough,N. V. Optical properties of humic substances and CDOM: Relation tostructure. Environ. Sci. Technol. 2009, 43 (7), 22622268.(58) Del Vecchio, R.; Blough, N. V. On the origin of the opticalproperties of humic substances. Environ. Sci. Technol. 2004, 38 (14),38853891.(59) Korshin, G. V.; Li, C. W.; Benjamin, M. M. Monitoring theproperties of natural organic matter through UV spectroscopy: Aconsistent theory. Water Res. 1997, 31 (7), 17871795.(60) Wert, E. C.; Rosario-Ortiz, F. L.; Snyder, S. A. Using ultravioletabsorbance and color to assess pharmaceutical oxidation duringozonation of wastewater. Environ. Sci. Technol. 2009, 43 (13), 48584863.(61) Li, C. W.; Benjamin, M. M.; Korshin, G. V. Use of UVspectroscopy to characterize the reaction between NOM and freechlorine. Environ. Sci. Technol. 2000, 34 (12), 25702575.(62) von Sonntag, C.; von Gunten, U. Chemistry of Ozone in Waterand Wastewater Treatment - From Basic Principles to Applications; IWAPublishing, 2012; p 320.(63) Criegee, R. Mechanism of ozonolysis. Angew. Chem., Int. Ed.Engl. 1975, 14 (11), 745752.(64) Gulkowska, A.; Sander, M.; Hollender, J.; Krauss, M. Covalentbinding of sulfamethazine to natural and synthetic humic acids:Assessing laccase catalysis and covalent bond stability. Environ. Sci.Technol. 2013, 47 (13), 69166924.(65) Wardman, P. Reduction potentials of one-electron couplesinvolving free radicals in aqueous solution. J. Phys. Chem. Ref. Data1989, 18 (4), 16371755.(66) Reale, S.; Di Tullio, A.; Spreti, N.; De Angelis, F. Massspectrometry in the biosynthetic and structural investigation of lignins.Mass Spectrom. Rev. 2004, 23 (2), 87126.(67) Ravacha, C. The reactions of chlorine dioxide with aquaticorganic materials and their health effects. Water Res. 1984, 18 (11),13291341.(68) Wajon, J. E.; Rosenblatt, D. H.; Burrows, E. P. Oxidation ofphenol and hydroquinone by chlorine dioxide. Environ. Sci. Technol.1982, 16 (7), 396402.(69) Lind, J.; Shen, X.; Eriksen, T. E.; Merenyi, G. The one-electronreduction potential of 4-substituted phenoxyl radicals in water. J. Am.Chem. Soc. 1990, 112 (2), 479482.(70) Li, C.; Hoffman, M. Z. One-electron redox potentials of phenolsin aqueous solution. J. Phys. Chem. B 1999, 103 (32), 66536656.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611155

  • (71) CRC Handbook of Chemistry and Physics, 70th ed.; CRC Press,Inc.: Boca Raton, FL, 1989.(72) Rebenne, L. M.; Gonzalez, A. C.; Olson, T. M. Aqueouschlorination kinetics and mechanism of substituted dihydroxyben-zenes. Environ. Sci. Technol. 1996, 30 (7), 22352242.

    Environmental Science & Technology Article

    dx.doi.org/10.1021/es402516b | Environ. Sci. Technol. 2013, 47, 111471115611156