Chemical Degradation Methods for Wastes and Pollutants - M.a. Tarr

Chemical Degradation Methods for Wastes and Pollutants Environmental and Industrial Applications

edited by Matthew A. Tarr University of New Orleans New Orleans, Louisiana, U.S.A.

. . . . . - - .

MARCEL DEKKER, INC.

D E K K E R

NEW YORK RASEL

Although great care has been taken to provide accurate and current information,neither the author(s) nor the publisher, nor anyone else associated with this publica-tion, shall be liable for any loss, damage, or liability directly or indirectly caused oralleged to be caused by this book. The material contained herein is not intended to

provide specic advice or recommendations for any specic situation.

Trademark notice: Product or corporate names may be trademarks or registeredtrademarks and are used only for identication and explanation without intent toinfringe.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress.

ISBN: 0-8247-4307-5

This book is printed on acid-free paper.

Headquarters

Marcel Dekker, Inc., 270 Madison Avenue, New York, NY 10016, U.S.A.tel: 212-696-9000; fax: 212-685-4540

Distribution and Customer Service

Marcel Dekker, Inc., Cimarron Road, Monticello, New York 12701, U.S.A.tel: 800-228-1160; fax: 845-796-1772

Eastern Hemisphere Distribution

Marcel Dekker AG, Hutgasse 4, Postfach 812, CH-4001 Basel, Switzerland

tel: 41-61-260-6300; fax: 41-61-260-6333

World Wide Web

http://www.dekker.com

The publisher oers discounts on this book when ordered in bulk quantities. Formore information, write to Special Sales/Professional Marketing at the headquarters

address above.

Copyright nnnn 2003 by Marcel Dekker, Inc. All Rights Reserved.

Neither this book nor any part may be reproduced or transmitted in any form or byany means, electronic or mechanical, including photocopying, microlming, andrecording, or by any information storage and retrieval system, without permission

in writing from the publisher.

Current printing (last digit):

10 9 8 7 6 5 4 3 2 1

PRINTED IN THE UNITED STATES OF AMERICA

Environmental Science and Pollution Control Series

I . Toxic Metal Chemistry in Marine Environments, Muhammad Sadiq 2. Handbook of Polymer Degradation, edited by S. Halim Hamid,

Mohamed B. Amin, and Ali G. Maadhah 3. Unit Processes in Drinking Water Treatment, Wily J. Masschelein 4. Groundwater Contamination and Analysis at Hazardous Waste Sites,

edited by Suzanne Lesage and Richard E. Jackson 5. Plastics Waste Management: Disposal, Recycling, and Reuse, edited

by Nabil Mustafa 6. Hazardous Waste Site Soil Remediation: Theory and Application of

Innovative Technologies, edited by David J. Wilson and Ann N. Clarke

7. Process Engineering for Pollution Control and Waste Minimization, edited by Donald L. Wise and Debra J. Trantolo

8. Remediation of Hazardous Waste Contaminated Soils, edited by Donald L. Wise and Debra J. Trantolo

9. Water Contamination and Health: Integration of Exposure Assess- ment, Toxicology, and Risk Assessment, edited by fihoda G. M. Wang

10. Pollution Control in Fertilizer Production, edited by Charles A. Hodge and Neculai N. Popovici

1 1. Groundwater Contamination and Control, edited by Uri Zoller 12. Toxic Properties of Pesticides, Nicholas P. Cheremisinoff and John A.

King 13. Combustion and Incineration Processes: Applications in Environ-

mental Engineering, Second Edition, Revised and Expanded, Walter R. Niessen

14. Hazardous Chemicals in the Polymer Industry, Nicholas P. Chere- misinoff

15. Handbook of Highly Toxic Materials Handling and Management, edited by Stanley S. Grossel and Daniel A. Crowl

16. Separation Processes in Waste Minimization, Robert B. Long 17. Handbook of Pollution and Hazardous Materials Compliance: A

Sourcebook for Environmental Managers, Nicholas P. Cheremisinoff and Madelyn Graffia

1 8. Biosolids Treatment and Management: Processes for Beneficial Use, edited by Mark J. Girovich

19. Biological Wastewater Treatment: Second Edition, Revised and Expanded, C. P. Leslie Grady, Jr., Glen T. Daigger, and Henry C:. Lim

20. Separation Methods for Waste and Environmental Applications,, Jack S. Watson

21. Handbook of Polymer Degradation: Second Edition, Revised and Expanded, S. Halim Hamid

22. Bioremediation of Contaminated Soils, edited by Donald L. Wise, Debra J. Trantolo, Edward J. Cichon, Hilary 1. Inyang, and Ulrich Stottmeister

23. Remediation Engineering of Contaminated Soils, edited by Donald L. Wise, Debra J. Trantolo, Edward J. Cichon, Hilary 1. Inyang, and Ulrich Stottmeister

24. Handbook of Pollution Prevention Practices, Nicholas P. Cheremisinoff

25. Combustion and Incineration Processes: Third Edition, Revised and Expanded, Walter R. Niessen

26. Chemical Degradation Methods for Wastes and Pollutants: Environmental and Industrial Applications, edited by Matthew A. Tarr

Addition a 1 Volumes in Preparation

Preface

Human activities have a large and important impact on the environment.Naturally occurring elements or compounds are often concentrated andredistributed in the environment through industrial processes, power pro-duction, and consumer activity. For example, lead, which is found innaturally occurring mineral deposits, has become a major pollutant throughits use in batteries, paints, and gasoline additives. In addition, the productionof non-natural or anthropogenic substances, such as halogenated solvents,can also result in the eventual release of often toxic and biorecalcitrantsubstances into the environment. Wide-scale redistribution of pollutants byhumans dates at as far back as the ancient Greek and Roman civilizations(20002500 years ago), during which time extensive smelting activitiesresulted in signicant atmospheric pollution by heavy metals such as lead.In fact, heavy-metal contamination of Arctic and Antarctic ice has revealedevidence of global pollution from smelting and other human activities sincethese ancient times.

Most certainly the people of ancient Greek and Roman times were notaware of the extent of their pollution. In fact, only in the late twentieth centurydid widespread awareness and understanding of the degree of anthropogenicpollution begin to develop.Unfortunately, large releases of contaminants intothe environment transpired without either knowledge of or concern for theconsequences. Once contaminants have been introduced into the environ-ment, subsequent clean-up is extremely dicult, time consuming, and costly.Due to the existence of many contaminated sites, signicant research anddevelopment eorts have been expended to develop eective means ofremediating these sites. These methods must be both economically feasibleand environmentally sound. For some sites, these challenges have beensuccessfully met, while other sites remain contaminated because of lack ofacceptable (economically and/or environmentally) technologies or becausethe sites pose a low risk.

While cleaning up previous contamination is a high priority, developingnew technologies to prevent future contamination is equally important, if notmore so. Without environmentally acceptable industrial processes, powerproduction, and consumer activity, the Earths environment will continue tobe threatened. Development of inherently clean technologies as well asimplementation of eective waste stream treatment are viable routes topreventing future environmental contamination.

Chemical Degradation Methods for Wastes and Pollutants focuses onchemical methods of destroying pollutants. Chemical methods can be advan-tageous over biological methods because they are often faster, can treat highlycontaminated systems, and may be less sensitive to ambient conditions. Incontrast, bacteria can be killed by contaminants or solvents and lose viabilityoutside relatively narrow pH and temperature ranges. However, chemicalmethods are often more costly and labor-intensive than biodegradationtechnologies. Despite their limitations, both biological and chemical tech-nologies are valuable tools that can be used successfully under appropriateconditions. Furthermore, combinations of biological and chemical treatmentmethods can often provide advantages over the individual systems.

The book covers several chemical technologies for remediation or wastestream treatment of predominantly organic contaminants. Although notevery chemical technology has been included, ten common or potentiallyuseful methods are covered. Each chapter presents the fundamentals behindeach technology and covers selected applications and practical issues relevantto adaptation of the technique to real treatment systems.

Continued research into both fundamentals and applications of chem-ical treatment technologies will hopefully provide solutions to many currentpollution treatment problems, both for waste streams and for contaminatedsites. Only through cooperation among scientists, engineers, industry, gov-ernment, and consumers can we maintain a healthy and productive environ-ment for the future.

Finally, I would like to thank those who served as reviewers for eachchapter.

Matthew A. Tarr

Prefaceiv

Contents

PrefaceContributors

1. OzoneUV RadiationHydrogen Peroxide OxidationTechnologiesFernando J. Beltran

2. Photocatalytic Degradation of Pollutants in Water andAir: Basic Concepts and ApplicationsPierre Pichat

3. Supercritical Water Oxidation TechnologyIndira Jayaweera

4. Fenton and Modied Fenton Methods for PollutantDegradationMatthew A. Tarr

5. Sonochemical Degradation of PollutantsHugo Destaillats, Michael R. Homann, and Henry C.Wallace

6. Electrochemical Methods for Degradation of OrganicPollutants in Aqueous MediaEnric Brillas, Pere-Llus Cabot, and Juan Casado

7. The Electron Beam Process for the RadiolyticDegradation of PollutantsBruce J. Mincher and William J. Cooper

8. Solvated Electron Reductions: A Versatile Alternativefor Waste RemediationGerry D. Getman and Charles U. Pittman, Jr.

9. Permeable Reactive Barriers of Iron and Other Zero-Valent MetalsPaul G. Tratnyek, Michelle M. Scherer, Timothy L.Johnson, and Leah J. Matheson

10. Enzymatic Treatment of Waters and WastesJames A. Nicell

Contentsvi

Contributors

Fernando J. Beltran Departamento de Ingenieria Quimica y Energetica,Universidad de Extremadura, Badajoz, Spain

Enric Brillas Laboratori de Ciencia i Tecnologia Electroquimica deMateri-als, Departament de Quimica Fisica, Universitat de Barcelona, Barcelona,Spain

Pere-Llus Cabot Laboratori de Ciencia i Tecnologia Electroquimica deMaterials, Departament de Quimica Fisica, Universitat de Barcelona, Barce-lona, Spain

Juan Casado Departamento de Investigacion, Carburos Metalicos S.A.,Barcelona, Spain

William J. Cooper Department of Chemistry, University of North Caro-linaWilmington, Wilmington, North Carolina, U.S.A.

Hugo Destaillats Department of Environmental Science and Engineering,California Institute of Technology, Pasadena, California, U.S.A.

Gerry D. Getman Commodore Solution Technologies, Inc., Marengo,Ohio, U.S.A.

Michael R. Homann Department of Environmental Science and Engineer-ing, California Institute of Technology, Pasadena, California, U.S.A.

Indira Jayaweera SRI International, Menlo Park, California, U.S.A.

Timothy L. Johnson AMEC Earth & Environmental, Inc., Portland, Ore-gon, U.S.A.

Leah J. Matheson MSE Technology Applications, Inc., Butte, Montana,U.S.A.

Bruce J. Mincher Radiation Physics Group, Idaho National Engineering &Environmental Laboratory, Idaho Falls, Idaho, U.S.A.

James A. Nicell Department of Civil Engineering and Applied Mathe-matics, McGill University, Montreal, Quebec, Canada

Pierre Pichat Laboratoire Photocatalyse, Catalyse et Environment, EcoleCentrale de Lyon, Ecully, France

Charles U. Pittman, Jr. Department of Chemistry, Mississippi State Uni-versity, Mississippi State, Mississippi, U.S.A.

Michelle M. Scherer Department of Civil and Environmental Engineering,University of Iowa, Iowa City, Iowa, U.S.A.

Matthew A. Tarr Department of Chemistry, University of New Orleans,New Orleans, Louisiana, U.S.A.

Paul G. Tratnyek Department of Environmental and Biomolecular Sys-tems, Oregon Health and Science University, Beaverton, Oregon, U.S.A.

Henry C. Wallace Ultrasonic Energy Systems Co., Panama City, Florida,U.S.A.

Contributorsviii

1OzoneUV RadiationHydrogenPeroxide Oxidation Technologies

Fernando J. BeltranUniversidad de Extremadura, Badajoz, Spain

I. INTRODUCTION

Processes involving the use of ozone, UV radiation, and hydrogen peroxide,characterized by the generation of short-lived chemical species of highoxidation power, mainly the hydroxyl radical, are classied as advancedoxidation technologies (AOTs). Possibly, the term may be attributed toGlaze et al. [1], who pointed out that hydroxyl radical oxidation is thecommon feature of these processes. The importance of these processes is dueto the high reactivity and redox potential of this free radical that reactsnonselectively with organic matter present in water. In practical cases, theseprocesses present a high degree of exibility because they can be usedindividually or in combination depending on the problem to be solved. Forinstance, for phenols or substances with high UV molar absorption coef-cients, ozone or UV radiation can be used alone, respectively, without theneed of any additional reagent, such as hydrogen peroxide. Anotheradvantage of these AOTs is that they may be applied under mild exper-imental conditions (atmospheric ambient pressure and room temperature).

The need for the application of these AOTs is based on dierent social,industrial, environmental, and even academic reasons. The increasing aware-ness of society for the quality of drinking water has led to the establishmentof maximum contaminant levels of priority pollutants in drinking water [1,2].The preparation of ultrapure water is needed for some industrial activitiessuch as those derived from the pharmaceutical and electronic processes.

Also, the release of wastewater into natural environmental reservoirs isanother concern; recycling of wastewater is already in progress in countrieswhere the lack of water is a national problem [4]. Finally, academic interestexists because the study of these AOTs allows testing the application of somephysical and chemical laws and engineering theories (mass, energy, and/orradiation conservation equations, kinetic modeling, absorption theories, etc.)to the environmental problems of water treatment.

Because of the aforementioned reasons, the number of research worksand applications based on these AOTs in the treatment of water hasincreased considerably during the past 20 years. Numerous publicationsthat refer to dierent aspects of these processes have so far been published injournals such as Ozone Science and Engineering, Water Research, OzoneNews, IUVA News, and the Journal of Advanced Oxidation Technologies. Inaddition, several books on the subject are available, such as that edited byLanglais et al. [5] on applications and engineering aspects of ozone in watertreatment and that of Dore [6] on the chemistry of oxidants. Reviews arealso abundant, including those of Camel and Vermont [7] on ozone in-volving oxidation processes, Reynolds et al. [8] and Chiron et al. [9] on theoxidation of pesticides, Legrini et al. [10] on photochemical processes, Yue[11] on kinetic modeling of photooxidation reactors, and Scott and Ollis [12]on the integration of chemical and biological oxidation processes forwastewater treatment.

In this chapter, AOTs based on ozone, UV radiation, and hydrogenperoxide are presented with special emphasis on their fundamental andapplication aspects. Related literature of research studies and applications, es-pecially those appearing in the last decade, are also listed, and specic exam-ples of laboratory and scale-up studies are described in separate sections.

II. BACKGROUND AND FUNDAMENTALS OF O3/UV/H2O2PROCESSES

O3/UV/H2O2 processes are characterized by the application of a chemicaloxidant (ozone and/or hydrogen peroxide) and/or UV radiation. Individualdescription of properties and reactivities of these oxidation technologies isnecessary to understand their synergism when used in combination for thetreatment of specic water pollutants or wastewaters. However, becausecombined processes (O3/H2O2, UV/H2O2, or O3/UV) are usually recom-mended in real situations, a general description of the processes andfundamentals of the individual and integrated O3/UV/H2O2 technologiesis also presented in the following sections.

Beltran2

A. General Description

Ozone- or UV-radiation-based technologies (O3/UV/H2O2) are chemicaloxidation processes applied to water treatment for the degradation ofindividual pollutants or the reduction of the organic load (chemical oxygendemand, COD) and improved biodegradability of wastewaters. In addition,ozone and UV radiation alone can be used for disinfection purposes; in fact,this was their rst application in water treatment [13,14]. In addition, theseAOTs, particularly ozonation, can be used to enhance the eciency of otherprocesses such as FeMn removal [15,16], occulationcoagulationsedi-mentation [17,18], biological oxidation [12], or biological degradation oforganic carbon in granular activated carbon [1921].

O3/UV/H2O2 AOTs are suitable for the treatment of water containingorganic pollutants in concentrations not higher than some tens of milligramsper liter. However, these technologies can also be used to treat concentratedsolutions. In addition to concentration, factors such as molecular structureof pollutant, aqueous organic matrix, pH, etc. are variables that aect theeciency and applicability of O3/UV/H2O2 AOTs for practical application.For wastewater treatment, O3/UV/H2O2 AOTs are used in combinationwith biological oxidation processes because of the enhancement achieved onthe biological oxygen demand (BOD). In fact, another feature of O3/UV/H2O2 AOTs is that they steadily transform high molecular weight sub-stances into more oxygenated lower molecular weight substances, whichinvolves an increase of BOD [22,23]. Examples of studies on wastewatertreatment that give a general view of the application of O3/UV/H2O2 AOTsare those of Rice and Browning [24] and, more recently, by Rice [25] on theuse of ozonation, or Zhou and Smith [26], Rivera et al. [27], and Kos andPerkowski [28] for combined oxidation involving UV radiation.

O3/UV/H2O2 AOTs, together with other processes treated in dierentchapters (such as Fenton oxidation), can be named ambient (temperatureand pressure), advanced oxidation technologies, in contrast with otherAOTs such as hydrothermal oxidation processes that require pressuresand temperatures above 1 MPa and 150jC, respectively, and which aremore suitable for the treatment of concentrated wastewaters. It is evidentthat appropriate ranges of concentrations for the dierent oxidation tech-nologies cannot be exactly established but some recommended values havebeen reported [29]. Fig. 1 shows some possible recommended ranges ofconcentrations for these types of AOTs.

O3/UV/H2O2 AOTs generally involve two oxidation/photolysis routesto remove foreign matter present in water. Thus, ozone, hydrogen peroxide,and/or UV radiation can react individually or photolyze directly the organicin water. However, when used in combination, they can degrade pollutants by

O3/UV/H2O2 Oxidation Technologies 3

oxidation through hydroxyl free radicals generated in situ. Hydroxyl radicalshave the largest standard redox potential except for uorine (see Table 1).

In addition, they react very rapidly with almost all types of organicsubstances through reactions whose rate constants vary from 107 to 1010

M1 s1 [30]. Table 2 gives a list of rate constant values of these reactions.Because of the high and similar values of the rate constants, it is said

that these free radicals react nonselectively with the organic matter presentin water, although, as deduced from the above range of values, there arecompounds that react with them almost three orders of magnitude fasterthan others. Among the most common water pollutants, phenols and somepesticides are substances that react rapidly with hydroxyl radicals, whereassome organochlorine compounds are less reactive.

Another feature of these AOTs is that they are destructive types ofwater pollution removal processes because they eliminate compounds ratherthan transfer them to another medium. Thus, carbon adsorption or strip-ping transfers pollutants from one phase (water) to another phase such as asolid phase (carbon) or a gas phase (air). In the latter case, purication of airis required so that an additional step (i.e., carbon adsorption) is also needed,which implies higher processing costs.

Figure 1 Oxidation process advisable according to COD of water. (WAO, wet airoxidation. SCWAO, supercritical wet air oxidation).

Beltran4

Table 1 Standard Redox Potentialof Some Oxidant Species

Oxidant Ej, V

Fluorine 3.03Hydroxyl radical 2.80Atomic oxygen 2.42

Ozone 2.07Hydrogen peroxide 1.77Permanganate ion 1.67Hypochlorous acid 1.49

Chlorine 1.36Chlorine dioxide 1.27Bromine 1.09

Table 2 Rate Constants of the Reaction Between the Hydroxyl Radicaland Organic Compounds in Water

Organic compound Rate constant109, M1 s1 Reference no.Benzene 7.8 30

Nitrobenzene 2.9 312,6-Dinitrotoluene 0.75 31Naphthalene 5 32

Phenanthrene 13.4 33Phenol 11 30Phenoxide ion 9.6 30

p-Nitrophenol 3.8 30o-Chlorophenol 12 30Maleic acid 6.0 30Formic acid 0.13 30

Glyoxal 0.066 30Tetrachloroethylene 2.6 30Trichloroethylene 1.3 34

1,1,1-Trichloroethane 0.020 34Dichloromethane 0.022 35Chloroform 0.011 35

Lindane 5.8 35Atrazine 2.6 35Aldicarb 8.1 35


At rst sight, however, the main drawback of O3/UV/H2O2 AOTs isthe high processing cost, mainly because both ozone and UV radiationrequire a continuous feed of energy for process maintenance, as well as highcapital costs for ozone generators and photoreactors. However, the develop-ment of improved ozonators and UV lamp technologies has made theseprocesses more amenable in practice as can be deduced from their actualapplications (see Sec. IV).

B. Ozonation

Ozone is the basic compound for many oxidation processes included underthe general term of ozonation. In these processes, ozone may be used aloneor with other agents such as hydrogen peroxide, UV radiation, catalysts,ultrasound, activated carbon, etc. In this section, information concerningthe individual use of ozone is given, while its combined use with hydrogenperoxide or UV radiation is reported in later sections.

1. Background and Fundamentals

Ozone is an inorganic chemical molecule constituted by three oxygen atoms.It is naturally formed in the upper atmosphere from the photolysis ofdiatomic oxygen and further recombination of atomic and diatomic oxygenaccording to the following reactions:

O2!hm 2O 1O O2 ! O3 2In this way, ozone forms a stratospheric layer several kilometers wide

that protects life on earth by preventing UV-B and UV-C rays from reachingthe surface of the planet. Ozone may arise from combustion reactions inautomobile engines, resulting in pollutant gases. These gases usually containnitric oxide that is photolyzed by sunlight in the surrounding atmosphere toyield nitrous oxide and atomic oxygen. Atomic oxygen, through reaction (2),nally yields ozone. In this sense, ozone is a contaminant of breathing air;the maximum level allowed during an 8-hr exposure is only 0.1 ppm.However, despite the importance of ozone as a tropospheric pollutant, thefate of ozone in the atmosphere is beyond the scope of this chapter.

Ozone was discovered in 1840 and the structure of the molecule astriatomic oxygen was established in 1872. The rst use of ozone wasreported at the end of the 19th centuryas a disinfectant in many water-treatment plants, hospitals, and research centers such as the University ofParis where the rst doctoral thesis on ozonation was presented [36].Although the number of water-treatment plants using an ozonation step

Beltran6

increased steadily during the 20th century, it was at the end of the 1970s thatthe use of ozone signicantly increased. This increase came about when tri-halomethanes and other organohalogenated compounds were identied indrinking water as disinfection by-products arising from chlorination [37].This discovery gave rise to an enormous research eort to look foralternative oxidants to replace chlorine. Additional research aimed atdiscovering mechanisms of organochlorine compound formation estab-lished that these substances are formed from the electrophilic attack ofchlorine on nucleophilic positions of natural humic substances present insurface water [38]. Because ozone is a powerful electrophilic agent, it wasfound that, generally, the application of ozone before chlorine signicantlyreduced trihalomethane formation. Subsequent study of ozone reactions inwater led to a wide array of applications (presented in a further section) thatcan be summarized in the following: use as a disinfectant or biocide, use asan oxidant for micropollutant removal, and use as a complementary agentto improve other unit operations in drinking and industrial water andwastewater treatments (sedimentation, cooling water treatment, carbonadsorption, iron and manganese removal, biological oxidation, etc. [5]).The role of ozone in medical applications has also increased over the pasttwo decades [39]. In the mid-1980s, the need to comply with environmentalregulations on allowable levels of refractory substances such as pesticides [2]gave rise to another class of ozone water treatment for drinking water:ozone advanced oxidations. These processes are based on the combined useof ozone and hydrogen peroxide and/or UV radiation to generate hydroxylradicals as indicated above [1].

Ozone is known as a very reactive agent in both water and air. Thehigh reactivity of the ozone molecule is due to its electronic conguration.Ozone can be represented as a hybrid of four molecular resonance structures(see Fig. 2). As can be seen, these structures present negative and positivelycharged oxygen atoms, which in theory imparts to the ozone molecule thecharacteristics of an electrophilic, dipolar and, even, nucleophilic agent.

Because of this reactivity, the ozone molecule is able to react throughtwo dierent mechanisms called direct and indirect ozonation. Thus, ozonecan directly react with the organic matter through 1,3 dipolar cyclo-addition, electrophilic and, rarely, nucleophilic reactions [40,41]. In water,only the former two reactions have been identied with many organics[42]. On the contrary, the nucleophilic reaction has been proposed in onlya few cases in non-aqueous systems [43] (see examples of these mechanismsin Fig. 3).

Another group of ozone direct reactions are those with inorganicspecies such as Fe2+, Mn2+, NO2

, OH, HO2, etc. [44]. These could be

dened as redox reactions because in the overall process ozone acts as a true


oxidizing agent by taking electrons whereas the other species act as truereducing agents by losing electrons. Ozone has the highest standard redoxpotential among conventional oxidants such as chlorine, chlorine dioxide,permanganate ion, and hydrogen peroxide (see Table 1). At acid pH, theredox reaction for ozone is as follows:

O3 2H 2e!O2 H2O Ej 2:07 V 3

Figure 2 Resonance structures of the ozone molecule.

Figure 3 Direct pathways of ozone reaction with organics. (A) Criegge mechanism.(B) Electrophilic aromatic substitution and 1,3-dipolar cycloaddition. (C) Nucleo-philic substitution.

Beltran8

However, these reactions can actually be considered as electron transferor oxygen atom transfer reactions, as in the case of the ozone reactions withthe hydroxyl and hydroperoxide ions or with the nitrite ion, respectively:

O3 OH!k70 M1 s1

HO2 O2 4

O3 HO2!k2106 M1 s1

HO2 O2 5O3 NO2!k3:710

5 M1 s1NO3

O2 6Reactions (4) and (5) are extremely important because they are the initiatingsteps of the radical mechanism leading to the formation of hydroxyl radicalswhen ozone decomposes.

On the other hand, the indirect type of ozonation is due to the reactionsof free radical species, especially the hydroxyl radical, with the organic matterpresent in water. These free radicals come from reactionmechanisms of ozonedecomposition in water that can be initiated by the hydroxyl ion or, to bemore precise, by the hydroperoxide ion as shown in reactions (4) and (5).Ozone reacts very selectively through direct reactions with compounds withspecic functional groups in their molecules. Examples are unsaturated andaromatic hydrocarbons with substituents such as hydroxyl, methyl, aminegroups, etc. [45,46].

The mechanism of decomposition of ozone in water has been thesubject of numerous studies, starting from the work of Weiss [47]. Amongmore recent studies, the mechanisms of Hoigne et al. [48] and Tomiyashuet al. [49] are the most accepted in ozone water chemistry. The main con-clusion that can be drawn is that ozone stability in water is highly depen-dent on the presence of substances that initiate, promote, and/or inhibit itsdecomposition. The ozone decomposition mechanism usually assumed isgiven in Fig. 4 [50].

As observed from Fig. 4, ozone decomposition generates hydrogenperoxide that reacts with ozone [reaction (5)] to yield free radicals,initiating the propagation steps of the mechanism. It should be notedthat hydrogen peroxide has been detected during ozonation reactions inwater in the presence and absence of organics such as humic substancesor aromatic compounds [51]. From this mechanism, it is also deducedthat ozonation alone, or single ozonation, can be included under thegroup of AOTs, especially when the pH is increased. Notice that in themechanism presented in Fig. 4 other possible reactions of ozone notshown are those corresponding to the direct pathway (see later) that leadsto molecular products.

Ozone decomposition is usually a rst-order process, where theapparent pseudo rst-order rate constant depends on the concentration of


promoters, P, inhibitors, S, and initiators, I, of ozone decomposition as wasreported by Staehelin and Hoigne [48] with the equation given below:

rO3 X

kDiCMif3kiCOHX

kliClig 1P

kpiCpiPkSiCSi

CO3 7

where CIi, CPi, and CSi represent the concentrations of any species i that actsas initiator, promoter, or scavenger (see also Fig. 4); CMi is the concen-tration of any other species i present in water other than the initiators, whichreact with ozone directly to yield molecular products; ki and kIi represent therate constants of the reactions between ozone and the hydroxyl ion and anyinitiator species i, respectively; kPi and kSi represent the rate constants of thereactions between the hydroxyl radical and any promoter and inhibitor i ofozone decomposition, respectively; and kDi represents the rate constant ofthe direct reaction of ozone with any other species i present in water otherthan the initiators. As can be deduced from Eq. (7) the half-life of ozone inwater is highly dependent on the pH and matrix content of the water. Forexample, the half-life of ozone in distilled water can vary from about 102

sec at pH 12 to 105 sec at pH 2 or from 10 sec for secondary wastewatereuents to 104 sec for certain ground and surface waters as reported inthe literature [50,52].

2. Kinetics of Ozonation

The design of ozonation contactors requires knowledge of kinetic informa-tion (see later), that is, the rate at which pollutants or matter present inwater react with ozone, both directly and/or indirectly, and hence the rate ofozone absorption. Reaction rates can be calculated if rate constants of thesereactions are known. Thus, the determination of rate constants represents a

Figure 4 Scheme of ozone decomposition mechanism in water. P=promoter (e.g.,ozone, methanol). S=scavenger or inhibitor (i.e., t-butanol, carbonate ion).I=initiators (e.g., hydroxyl ion and hydroperoxide ion).

Beltran10

crucial point in contactor design. In practice, ozonation is a heterogeneousprocess involving ozone transfer from air or oxygen to the water phase andsimultaneous chemical reactions in the aqueous medium. The kinetics of thistype of processes can be established if the kinetic regime of ozone absorptionis known. This process requires knowledge of the relative importance ofboth physical and chemical rates (diusion of ozone and chemical reac-tions), which can be quantied from the dimensionless number of Hatta[53]. For any ozoneorganic substance reaction in water, second-orderirreversible reactions normally occur (rst-order with respect to ozone andcompound M) [41,4446,54]:

zO3 M! Products 8The corresponding Hatta number, Ha, then, reduces to the followingexpression:

Ha kDCMDO3

kL2

r9

The square of this number represents the ratio between the maximum re-action rate of ozone near the water interface (lm thickness) and the maxi-mum physical absorption rate (i.e., the absorption without reaction). InEq. (9), kD and kL are parameters representing the chemical reaction andphysical diusion rate constants, that is, the rate constant of the ozonecompound reaction and water phase mass transfer coecient, respectively.Their values are indicative of the importance of both the physical andchemical steps in terms of their rates. However, two additional parameters,as shown in Eq. (9), are also needed: the concentration of the compound,CM, and the diusivity of ozone in water, DO3. The ozone diusivity in watercan be calculated from empirical equations such as those of Wilke andChang [55], Matrozov et al. [56], and Johnson and Davies [57]; from theseequations, at 20jC, DO3 is found to be 1.62109, 1.25109, and1.76109 m2 s1, respectively.

The value of Ha determines the rate of the ozone reaction. Thus, forHa 3 they arefast reactions. There is also an intermediate kinetic regime dened as mod-erate, which is rather dicult to treat kinetically [53]. However, for mostcommon situations, reactions of ozone in drinking water are considered asslow reactions. This does not mean that the time needed to carry out theozonation is high (time needed to have high destruction of pollutants), butthat the mass transfer rate is faster than the chemical reaction rate. For in-stance, in most cases, ozonation of micropollutants, which are found in verylow concentrations (mg L1 or Ag L1), lies in this kinetic regime. In othercases, where the concentration of pollutants is higher (i.e., wastewaters


containing compounds that react very fast with ozone such as phenols in highconcentration), the chemical reaction rates are equal to or even much fasterthan the mass transfer rate and the kinetic regime is fast or instantaneous[58]. To distinguish between kinetic regimes of fast reactions, anotherdimensionless number, the instantaneous reaction factor, Ei, should bedetermined [53]:

Ei 1 zDMCMDO3C

O3

10

In Eq. (10) z is the stoichiometric coecient of the ozone-compound reaction[reaction (8)], DM is the diusivity of compound M in water (which can becalculated from the Wilke and Chang equation), and C*O3 is the ozonesolubility (or properly dened, the ozone concentration at the gaswaterinterface). If the parameters of Eqs. (9) and (10) are known, the kineticregime can be established, and hence the kinetics of ozonation can bedetermined. Table 3 gives the kinetic equations corresponding to dierentkinetic regimes found in ozonation processes. As can be deduced from theequations in Table 3, the rate constant, mass transfer coecients, and ozonesolubility must be previously known to establish the actual ozonationkinetics. The literature reports extensive information on research studiesdealing with kinetic parameter determination as quoted below.

Table 3 Kinetic Equations and Absorption Kinetic Regimes for Second-OrderIrreversible OzoneOrganic GasLiquid Reactionsa

Kinetic regime Kinetic equation Conditions

Very slow NO3=kLa(C*O3CO3 )=

dCO3dt +Si

ri Ha

3. Ozone Solubility, Rate Constants, and Mass TransferCoefficients

Similar to ozone decomposition, ozone solubility has been the subject ofmultiple studies. These studies usually propose an empirical equation forthe Henrys law constant as a function of pH, ionic strength, and temper-ature [59,60]. For example, Sotelo et al. [60] found the following equa-tion valid for phosphate buer aqueous solutions at temperatures between0 and 20jC, pH range of 2 to 8.5, and ionic strength varying from 103

to 101 M:

He 1:85 107exp 2119=T exp 0:961 C0:012OH kPaM1 11where T is the absolute temperature and I is the ionic strength. Theoret-ically, however, He should be dependent only on temperature and thepresence of ionic strength due to electrolytes in solution and independentof pH according to the following equation:

log He=Hej X

hiIi 12where Hej is Henrys constant in ultrapure water and h is the salting-outcoecient, a function of the dierent ionic and dissolved gas species inwater [61]. Thus, in a more recent paper, Andreozzi et al. [62] studied thisproblem and tried to develop an equation of this type. The authors did notarrive at this equation, but they concluded that the change in He with pHshould be due to the salting-out coecients of the dierent ionic speciesthat also change with pH.

For the experimental determination of He, a mass balance of ozone ina system where ozone is absorbed in ultrapure buered water in a semibatchreactor is usually applied:

kLa CO3 CO3

rO3 dCO3dt

13

where kLa is the volumetric mass transfer coecient, CO3 is the concen-tration of dissolved ozone at any time, and rO3 is the ozone decompositionreaction rate. In Eq. (13), the rst and second terms on the left side representthe contribution of ozone mass transfer and chemical reaction rates to theozone accumulation rate (right side of the equation). As can be deduced,experimental results applied to Eq. (13) allows determination of thevolumetric mass transfer coecient and the ozone solubility (see also kineticequations of Table 3). Application of Henrys law, nally, leads to thecorresponding constant, He:

PO3 HeCO3 14


Depending on the disappearance rate of the reacting compound orozone, rate constants of direct ozone reactions can be obtained from bothhomogeneous and heterogeneous ozonation systems. Thus, for very slow re-actions, homogeneous ozonation has the advantage of the absence of a masstransfer step. In these cases, the concentration of one of the reactants (ozoneor compoundM) can be considered constant throughout the reaction period,and the kinetics are determined by measuring the concentration of the othersubstance with time. When the reaction is very fast (of the order of micro-seconds or milliseconds) homogeneous ozonation can also be followed, butspecial equipment is needed to stop the reaction at very short times, forexample, with stopped ow spectrophotometers [63]. For kinetic studies inthese cases, heterogeneous ozonation reactions are recommended because thevariation of concentration with time is much slower than in homogeneousprocesses. Consequently, conventional methods, such as gas or liquid chro-matography or even classical spectrophotometry, can be used. For heteroge-neous kinetics, the equations given in Table 3 will be needed. In Table 4, a listof rate constant values for ozone direct reactions is given together with themethod of calculation. In other cases, to avoid the interferences of ozoneconsumption from by-products, the rate constants are deduced from com-petitive ozonation kinetics of two compounds: the compound whose kineticswith ozone is being determined and the reference compound. Obviously, theozone kinetics of the reference substance must be well known. In this way,Gurol and Nekouinaini [71] and Beltran et al. [72] have determined the rateconstants of ozone fast reactions with some phenolic compounds.

Ozonation processes can also be used for determination of mass trans-fer coecient. In fact, both ozone absorption in organic-free water, whichis a slow gasliquid reaction, and other ozone gasliquid reactions havebeen used for this purpose. For example, Roth and Sullivan [59] and Soteloet al. [60] determined the mass transfer coecient from ozone absorption inorganic-free water, whereas Ridgway et al. [73] and Beltran et al. [67] carriedout similar calculations from ozone absorption in water at pH 2 containingindigo and p-nitrophenol, respectively.

4. Kinetic Modeling

Kinetic models utilize a set of algebraic or dierential equations based onthe mole balances of the main species involved in the process (ozone in waterand gas phases, compounds that react with ozone, presence of promoters,inhibitors of free radical reactions, etc). Solution of these equations providestheoretical concentration proles with time of each species. Theoreticalresults can be compared with experimental results when these data areavailable. In some cases, kinetic modeling allows the determination of rateconstants by trial and error procedures that nd the best values to t the

Beltran14

experimental and calculated concentrations. Table 5 presents a list of studieswhere kinetic modeling of ozonation processes were carried out.

C. Hydrogen Peroxide Oxidation

Similar to ozone, hydrogen peroxide can react with organic matter presentin water through direct and indirect pathways. In direct mechanisms,hydrogen peroxide participates in redox reactions where it can behave asan oxidant:

H2O2 2H 2e! 2H2O Ej 1:776 V 15

Table 4 Rate Constants of the Reaction Between Ozone and OrganicCompounds in Watera

Organic compoundRate constant,

M1 s1 pH Method Reference no.

Benzene 2 1.73 AHOK 45Nitrobenzene 2.2 2 AHOK 64

2,6-Dinitrotoluene 5.7 2 AHOK 64Naphthalene 3000 2 AHOK 45Phenanthrene 2413 7 CHEK 65Phenol 1300 2 AHOK 46

2106 7 66Phenoxide ion 1.4109 10 EX 46p-Nitrophenol 4.5106 6.5 AHEK 67o-Chlorophenol 1600 2 CHEK 68

2.7106 7 68Maleic acid 1000 2 AHOK 46

Formic acid 5 24 AHOK 46100 8 46

Tetrachloroethylene

or as a reductant:

H2O2 2e! O2 2H Ej 0:7 V 16Indirect reactions are due to the oxidizing action of free radicals that areformed from the decomposition of aqueous hydrogen peroxide when itreacts with other inorganic compounds, such as ozone or Fe2+, or when itis photolyzed.

Examples of direct reactions are mainly with inorganic compoundssuch as cyanides and suldes or ozone and Fe2+. Both reactions of ozoneand Fe2+ with hydrogen peroxide represent the initiating steps of advancedoxidation processes: O3/H2O2, treated later in this chapter, and the Fentonoxidation, presented in another chapter, respectively. Hydrogen peroxide,on the other hand, does not signicantly react with most organic com-pounds, at least at appreciable rates for water treatment [6].

Hydrogen peroxide was discovered in 1818 by Tenard; the molecularstructure forms an oxygen bridge, with each oxygen bonded to one hydro-gen atom. In water, it is a weak acid, which dissociates to yield thehydroperoxide ion, HO2

:

H2O2 X HO2 pKa 11:7 17

It is the ionic form of hydrogen peroxide that reacts with ozone to yield freeradicals as indicated before [see reaction (5)].

As far as water treatment is concerned, some of the main reactions ofhydrogen peroxide are with ozone and Fe2+ or its photolysis with UV

Table 5 Studies Dealing with AOP Kinetic Modeling Involving Ozone,Hydrogen Peroxide, and UV Radiationa

Compounds treated AOP system Reacting system Reference no.

1,2-Dibromo-3-chloropropane

UV/H2O2 CMBPR 74

Chlorobutane UV/H2O2 CSTPR 75

Trichloroethene,tetrachloroethene

O3/H2O2 CBDR 76

VOCs O3/UV PFHPR 77Tri- and perchloroethene O3/H2O2 BHR 78

Atrazine UV/H2O2 CMBPR 79Acetone UV/H2O2 CMBPR 80Aromatic hydrocarbons O3/UV/H2O2 CMSBPR 81

a CMBPR=completely mixed batch photoreactor; CSTR=continuous stirred tank photo-

reactor; CBDR=continuous bubble reactor with dispersion; PFHPR=plug ow homoge-

neous photoreactor; BHR=batch homogeneous tube reactor; CMSBPR=completely mixed

semibatch photoreactor.

Beltran16

radiation, another advanced oxidation system commented on later. In thischapter, discussion of hydrogen peroxide reactions will be limited only tothose of the O3/H2O2 and UV/H2O2 systems.

D. UV Radiation

UV radiation is also the basis of several chemical oxidation technologieswhere the action of radiation and free radicals generated in the process allowfor a high degree of micropollutant degradation and/or disinfection. Similarto ozonation or hydrogen peroxide oxidation, UV radiation may act on thematter present in water in two dierent ways: direct photolysis or indirectphotolysis (e.g., free radical oxidation).


UV radiation comprises energies from about 300 kJ Einstein1 (UV-A radia-tion, 1 einstein=1 mol of photons), up to 1200 kJ Einstein1 (vacuum UV).Table 6 shows the wavelength and energy of dierent UV radiation types.

For disinfection and oxidation purposes, UV-C radiation is normallyused although the application of other types of UV radiation has also beenreported in the literature [10]. For example, the use of UV-A or even visibleradiation to treat natural organic matter present in surface water has beenreported with and without the presence of catalysts [82,83]. Concerning theutilization of UV-C radiation, the most common use is 254-nm radiationdue to the development of low-pressure vapor mercury lamps by Hewitt in1901 [13]. For this reason, in this chapter the information presented mainlyfocuses on the use of 254-nm UV-C radiation.

Similar to ozonation processes, since the discovery of the germicidaleects of solar UV radiation by Downes and Blount in 1877 [13], UV radia-tion was rst used for disinfection. The development of reaction mecha-nisms in photochemistry led to the discovery of the advantages of UVradiation as an oxidation technology. At room temperature, most molecules

Table 6 Radiation Type and Associated Energy

Radiation Wavelength range, nm Energy range, kJ Einstein1a

Infrared >780

reside in their lowest-energy electronic state, known as the ground state.When UV radiation (or any other type of radiation with sucient energyper photon) is incident upon a molecule, the radiation can be absorbed,promoting the molecule to an excited state. That is, one electron of themolecule goes to a higher-energy state or excited state. Depending on thedirection of the electron spin, for most organic molecules, the excited state iseither a singlet (all electron spins cancel) or a triplet state (two unpairedelectrons with parallel spins). Overall the most probable transition occursfrom the ground state to the singlet state. The energy dierence between theground and excited states corresponds to the absorbed energy, hm, m beingthe frequency of the absorbed radiation and h the Planck constant. Themolecule in the excited state has a very short lifetime (109 to 108 sec) [84],after which it returns to the ground state by one of several mechanisms(uorescence, phosphorescence, internal conversion, collisions, etc.) or de-composes to yield a dierent molecule; that is, it undergoes a photochemicalreaction. A simple mechanism of photochemical reaction already used insome studies [85,86] is given below:

M hm!ka M 18M!kb M 19M!kc Products 20

Nevertheless, mechanisms of UV radiation can be more complicated in thepresence of oxygen. In this case, the electron in the excited state can betransferred to one oxygen molecule in its ground state to form the super-oxide ion radical. Also, the organic molecule may rst undergo homolysis ina carbonhydrogen bond followed by reaction with oxygen to yield organicperoxyl radicals [10,87]:

M hm!ka M 18M O2!M O2 21MH hm!M H 22M O2!MO2 23Indirect photolysis mechanisms involve the excitation of an additional

compound called a photosensitizer (PS), which in its excited state candirectly oxidize the pollutant of interest. This type of mechanism wasinvestigated by Faust and Hoigne [82] using fulvic substances as photo-sensitizers of phenols in natural waters. These latter mechanisms correspondto the indirect photolysis of M. In fact, Faust and Hoigne [82] reported thatthere are four possible routes of the excited photosensitizing action:

Beltran18

reactions of the photosensitizer with any compound M, with natural oradded solvents, with itself, and unimolecular decay as shown below:

PS hm!PS 24PS M!PS Solvent! 25PS PS!PS!

Photosensitization for the removal of certain pollutants in photolyticprocesses can contribute signicantly to the degradation rate. Thus,Simmons and Zepp [88] observed increases of up to 26 times of thephotodegradation rates of nitroaromatic compounds due to the action ofnatural or commercial humic substances with solar irradiation. In anotherwork [89], the herbicide 4-chloro-2-methylphenoxyacetic acid (MCPA) wasirradiated in water with 300 nm light in the presence of dierent photo-sensitizers. This compound, which does not photolyze directly at thiswavelength, could be degraded more than 95% in 5 hr when riboavinwas used as photosensitizer.

Another possible mechanism of photolysis is through the formation ofsecondary photooxidants that can be formed from one of the photosensi-tizer routes shown above. For example, a possible mechanism with humicsubstances as photosensitizers [90] could involve the formation of hydrogenperoxide and, subsequently, hydroxyl radicals:

PS hm!PS 26PS O2!PS O2 272O2 2H!H2O2 28H2O2 hm!2HO 29PS O2 RH!PSO2HR 30PSO2H!PSO HO 31HO M!Products and radicals 32

In addition to humic substances, nitrites and nitrates usually found innatural water also act as indirect photosensitizers to produce secondaryoxidants such as hydroxyl radicals [91]. A simplied scheme of the mech-anism is as follows [92]:

NO2 hm!NO O 33NO3 hm!NO2 O 34NO3 hm!NO2 O 35


2NO2 H2O2!NO2 NO3 2H 36OH2O!2HO 37O H2O!HO OH 38HO M!Products and radicals 39The use of nitrate to improve the photodegradation rates of pollutants

has been reported. For example, Sorensen and Frimmel [92] observed thatthe rate of photolysis of xenobiotic compounds such as EDTA and somephenyl and naphthalene sulfonates was signicantly increased in the presenceof nitrates.

Finally, another possibility of photolytic reaction is due to hetero-geneous processes, that is, photocatalysis. In these processes, a metaloxide surface is irradiated to yield surface holeelectron pairs [93]. Forexample, TiO2 suspensions are often used for this aim to generate thesespecies [94]:

TiO2 hm!TiO2 h e 40

The electron and hole may react at the surface with adsorbed compounds toinitiate oxidation or reduction reactions:

e O2!O2 41h H2O!HO H 42h OH!HO 43h M!M 44

Literature reports present many examples of these photocatalytic processesthat will be described in detail in another chapter.

2. Kinetics of Photolysis

Once the basic mechanism of photolysis [reactions (18) to (20)] is established,the kinetics of the photochemical reaction can be studied. The kinetics ofphotochemical reactions is dependent on factors such as the intensity andwavelength of the incident radiation, the optical path of the radiation, and thenature of the compound irradiated and the solution in which it is present. Theperformance of UV radiation will also depend on the photoreactor design.For example, in a batch photochemical reactor, the rate of compoundremoval due to direct photolysis, assuming the mechanism of reactions (18)to (20), is as follows [95]:

rUV dCMdt 1

V

ZV

ka 1 kbkb kc

FMlqidV 45

Beltran20

where V is the reaction volume and l [96] is a function of molar absorptioncoecients of species present in water, ei, dened as follows:

A 2:303X

jCi 46and ka, kb, and kc are the rate constants of steps (18) to (20), respectively.Notice that the rst minus sign on the right side of equation (45) is due to thestoichiometric coecient of M which is 1. As a rule, stoichiometriccoecients of reacting products are negative.

In Eq. (45) qi, is the ux of incident radiation, which varies accordingto the geometrical conguration of the photoreactor and photochemicalmodel used, and F is the fraction of absorbed radiation that M absorbs:

FM MCMPjCi

47

where eM and ei are the molar absorption coecient and molar absorptivityor optical density of M and any compound i, respectively, present insolution that also absorbs radiation. The term [ka(1kb/(kb+kc)] in Eq. (45)can be dened as the quantum yield of M, /M. The quantum yield isperhaps the most important parameter in UV radiation kinetics because itmeasures the fraction of the excited molecules that are transformed intoproducts. This parameter is dened as the moles of M decomposed perEinstein absorbed (1 Einstein being 1 mol of photons, 6.0231023 photons).Substances with high quantum yields that are constant over a wide rangeof wavelengths are usually called actinometers, and are used to measure theintensity or ux of incident radiation as shown later. In any case,compounds of high quantum yield are prone to decomposition throughUV radiation. In Table 7 values of quantum yield and molar absorptioncoecients for dierent compounds and oxidants in water are shownas examples.

Equation (45) can be solved by applying dierent photoreactormodels. The literature reports several photochemical reactor models forboth homogeneous and heterogeneous reactors [11,108,109]. In practice,annular photoreactors are often used (see Fig. 5); therefore, models for thistype of reactor are considered here. For other types of reactors, attentionshould be given to other publications [109].

Here Eq. (45) is solved for three models: the linear source withemission in parallel planes to the lamp axis (LSPPM) model, the pointwith spherical emission (PSSE) model, and a semiempirical model basedon Lamberts law (LLM). The rst two models come from the solution ofa radiation balance equation throughout the photorreactor assumingdierent hypotheses.

e

e

e


Table 7 Values of Quantum Yields and Molar Absorption Coecientsfor Dierent Water Pollutants and Oxidants in Water

Pollutant or oxidante, M1 cm1

(k, nm)/, mol Einstein1

(k, nm) Reference no.

Ozone 3300 (253.7) 0.62 (253.7) 97H2O2 18.7 (253.7) 0.5

d 98,99

HO2 210 0.5d 98,99

NO3 9900 (200) 100

Phenol 516 (213400) 0.05 (213400) 1012-Chlorophenol 1920 (272) 0.03 (296) 102

2-Chlorophenolate 3760 (293) 0.30 (296) 1023-Chlorophenol 1750 (273) 0.09 (254 or 296) 1023-Chlorophenolate 3000 (292) 0.13 (254 or 296) 102

4-Chlorophenol 1650 (278) 0.25 (254 or 296) 1024-Chlorophenolate 2400 (296) 0.25 102Nitrobenzene 5564 (254) 0.007 (254) 24

2,6-Dinitrotoluene 6643 (254) 0.022 (254) 24Fluorene 16654 (254) 0.0075 (254) 73Phenanthrene 40540 (254) 0.0069 (254) 73

Acenaphthene 1333 0.0052 73Acenaphthalene 26941 0.004 1031,3-Dichlorobenzene 0.06 (213400) 1011,3,5-Trichlorobenzene 0.043 101

Trichloroethylene 18.3 (254) 0.88 (254) 95Atrazine 2486 (254) 0.05 104Simazine 2512 0.06 105

CEATa 3056 0.038 106CIATb 3211 0.035 106CAATc 3161 0.018 106

Alachlor 540 0.177 107Parathion 0.0076 (240320)e 89Parathion 0.0016f 89

a CEAT=2-chloro-4-ethylamine-1,3,5-s-triazine;b CIAT=2-chloro-4-isopropylamine-1,3,5-s-triazine;c CAAT=2-chloro-4,6-diamine-1,3,5-s-triazine.d In the presence of scavengers.e At pH 9.6.f At pH 5.7.

Beltran22

The LSPPM assumes that the lamp can be represented as a consec-utive line of points, each one emitting radiation in all directions contained ina plane perpendicular to the lamp axis. An approximate equation for qiwhen rbL is as follows [110]:

qi Roqor

exp A r Ro 48

where Ro and r are dened in Fig. 5 and qo is the ux of incident radiation atthe inside wall of the photoreactor (for r=Ro), which is calculated fromactinometry experiments. By substitution of Eq. (48) into Eq. (45) and afterintegration the photolysis rate of M becomes:

rUV dCMdt FMfM 2kRoqoL

V1 exp A r Ro 49

Figure 5 Scheme of annular photoreactor.


The PSSE model considers each point of the lamp emitting radiation in allspace directions. The expression for qi is as follows [111,112]:

qi Eo4kmzVLzV

exp AN

r2 z zV dzV 50

where r, z, and zV are dened in Fig. 5, Eo is the radiant energy of the lampper unit of length, and - is given by Eq. (51):

N r Ro

r2 z zV 2

qr

51

Substitution of Eqs. (50) and (51) in Eq. (45) leads to the photolysis rateof M:

rUV dCMdt FM EofMA

4mL0mR1RomzVLzV

exp AN r2 z zV drdzdz V 52

In this model, the solution is obtained by numerical integration of Eq.(52).

The LL model has been applied to numerous works [74,95,96,113].Because of its simplicity, this model has a great acceptance amongresearchers of advanced oxidation kinetics. The model constitutes a sim-plication of the LSPP model. The rate of photolysis is, in this case:

rUV dCMdt FMfMIo 1 exp AyL 53

where Io is the intensity of incident radiation and yL is the eective path ofradiation through the photoreactor.

In all these models, knowledge of parameters such as qo (LSPP model),Eo (PSSE model), or Io and yL (LL model) are necessary to determine thephotolysis rate of M. These parameters are determined experimentally byactinometry experiments [86]. It is noteworthy tomention that the use of thesetheoretical models (LSPP or PSSE models) implies that all radiation incidentinto the solution is absorbed without end eects, reection, or refraction. Inexperimental photoreactors, it is not usual to fulll all these assumptionsbecause of the short wall distance of the photoreactor. For instance, toaccount for such deviations, Jacob and Drano [114] introduced a correctingequation, as a function of position. Another important disadvantage is thepresence of bubbles that leads to a heterogeneous process as, for example, inthe case of O3/UVoxidation. In this case, photoreactormodels should be used[109]. This is the main reason for which the LL model is usually applied in thelaboratory for the kinetic treatment of photochemical reactions. In the LLM,

Beltran24

the eective path of radiation can be considered as the correction functionaccounting for deviations from ideality.

E. Combined Oxidations: O3/H2O2, UV/H2O2, and O3/UV

The reaction of ozone and hydrogen peroxide in its ionic form and photolysisof both oxidants constitute the initiation reactions leading to a mechanism ofhydroxyl radical formation in water. This mechanism is basically the samefor all these advanced oxidation systems, whereas the main dierences lie inthe initiating steps. These oxidation technologies have been applied for thetreatment of pollutants in water for more than two decades.


Photolysis of hydrogen peroxide was rst studied by Baxendale and Wilson[99]. They reported that the decomposition of 1 mol of hydrogen peroxideneeded one Einstein of incident 254 nm UV radiation:

H2O2!hm 2HO 54This quantum yield corresponds to the overall process where hydrogen

peroxide is not just removed by reaction (54), its direct photolysis, but alsoby parallel reactions involving the hydroxyl and hydroperoxide radicals asshown below:

HO H2O2!HO2 H2O 55HO2 H2O2!HO H2OO2 56

although the latter reaction is negligible.These authors [99] also showed that the quantum yield of both

forms of hydrogen peroxide, H2O2 and HO2, remained constant over a

wide range of concentrations and UV radiation intensity. Baxendale andWilson [99] also carried out experiments in the presence of organic sub-stances, such as acetic acid (a well-known scavenger of the hydroxylradicals) so that the measured rate of hydrogen peroxide disappearancecorresponded to the rate of its direct photolysis [reaction (54)]. From theseexperiments, they found that the rate was half of that of the process in theabsence of hydroxyl radical scavengers. Consequently, they concluded thatthe quantum yield of reaction (54) was 0.5 mol of hydrogen peroxide perEinstein. This value is called the primary quantum yield of hydrogenperoxide photolysis.

As reported by Staehelin and Holgne [115], ozone reacts only with theionic form of hydrogen peroxide, the hydroperoxide ion, HO2

. Theseauthors studied this reaction at dierent hydrogen peroxide concentrations


and in the presence of methylmercury hydroxide, another hydroxyl radicalscavenger. At pH values below the pKa of hydrogen peroxide (pKa=11.7),these authors observed that the rate of reaction (5) increased one order ofmagnitude per unit increase of pH. They found a second-order reaction rateconstant of 2.8106 M1 s1. A similar behavior can be noticed with theozone decomposition rate in organic-free water in the absence of hydrogenperoxide [reaction (4)], although the rate constant is several orders ofmagnitude lower (70 M1 s1) as found by the same authors and by Forniet al. [116].

Taube [97] studied the photolysis of aqueous ozone and postulated theformation of hydrogen peroxide, which he found to be formed with almostexact stoichiometry. Taube [97] reported a quantum yield for ozone of 0.62at 254 nm. Later, Prengle [117] claimed that ozone photolysis yields atomicoxygen, which directly leads to hydroxyl radicals. To elucidate whichmechanism is the correct one, Peyton and Glaze [118] later studied thisreaction and concluded that hydrogen peroxide is rst formed from ozonephotolysis without formation of atomic oxygen. From these studies [115118] and others reported by Staehelin and Hoigne [48] and Tomiyasu et al.[49], the mechanism of any type of advanced oxidation system involvingozone, hydrogen peroxide, and UV radiation can be established. A sim-plied scheme of this mechanism, applied to the oxidation of a potentialpollutant M, is presented in Fig. 6.

The main reactions of the mechanism are given below:

O3 H2O!hm 2HO!H2O2 57H2O2!hm 2HO 58O3 OH!HO2 O2 59H2O2!p HO2 H pK 11:7 60HO2 O3!HO2 O3 61HO2!p H O2 62O2 O3!O2 O3 63O3 H!HO3 !HO O2 64

Reactions (58) and (61) constitute the main initiation reactions of themechanism leading to the formation of hydroxyl radicals. The reactionbetween ozone and the hydroperoxide anion constitutes the faster mecha-nism as has been demonstrated in previous studies [119,120]. The reactionbetween ozone and the hydroxyl ion is negligible compared to the reactionbetween ozone and the hydroperoxide ion because the rate constants of thesereactions dier by several orders of magnitude: 70 M1 s1 for reaction (59)

Beltran26

vs. 2.8106 M1 s1 for reaction (51) [115]. The relative importance,however, is pH dependent. For the conditions usually applied in watertreatment, reaction (61) is faster. An extensive study of these reactions canbe seen in the work of Staehelin and Hoigne [115]. Reaction (61) alsopredominates against the direct photolysis of hydrogen peroxide [reaction(58)]. This is due to the low value of the molar absorption coecient ofhydrogen peroxide (see Table 7) and the very high rate constant value ofreaction (61) (see above). Furthermore, the direct photolysis of hydrogenperoxide competes, among others, with the direct photolysis of ozone, whichis also faster. This can be deduced from the values of the product betweenthe quantum yield and molar absorption coecient of each photolysisreaction. The product is 2046 L Einstein1 cm1 for the case of ozone and105 L Einstein1 cm1 for the case of hydrogen peroxide at its mostfavorable conditions (see Table 7), that is, at alkaline pH when hydrogenperoxide is present in the ionic form. A thorough study of the competitionbetween these initiation reactions can be followed from the studies ofBeltran [119,120].

For chemical structures refractory to direct reactions with ozoneand UV photolysis, free radical reactions are fundamental. Among freeradicals, the hydroxyl radical shows a high oxidizing power, and it is ge-nerally accepted as the main oxidant in these advanced oxidation systems.

Figure 6 Scheme of O3/UV/H2O2 oxidation processes. Key: 1: From directozonation. 2: From direct photolysis. 3: From free radical oxidation. 4: Intermediatepathway if M is inhibitor. 5: Intermediate pathway if M is promoter.


The reactions through which hydroxyl radicals participate are shownbelow [121]:

* Hydrogen abstraction:

HO RH!R H2O 65* Electrophilic addition:

HO PhX!HOPhX 66* Electron transfer:

HO RX!RX HO 67Organic radicals formed in these reactions may further react with oxygen(in an aerated medium as in water treatment) to yield organic peroxylradicals that can eventually react with compounds present in the medium torelease the superoxide ion radical (see route through 5 in Fig. 6; see also thework of von Sonntag and Schuchmann [122] for more details about peroxylradical reactions). In these cases, compounds that react with the hydroxylradical are known to be promoters of ozone decomposition because thesuperoxide ion radical consumes ozone at a fast rate [see reaction (63)above]. On the contrary, if the reaction between hydroxyl radical andcompound M does yield inactive radicals, M is known as a scavenger orinhibitor of ozone decomposition (see route to 4 in Fig. 6). Many naturalsubstances such as humic substances and carbonates are known to possesssuch a role [121]. However, the case of carbonate ion is rather specialbecause it reacts with hydroxyl radicals to yield the carbonate ion radical:

CO3 HO !k3:7108 M1 s1

CO3 OH 68HCO3 HO !

k2107 M1 s1CO3 H2O 69

which is also known to react with organic compounds in water. Table 8presents a list of rate constants of these reactions. In addition, thecarbonate ion radical also reacts with hydrogen peroxide, if present inwater, to regenerate the carbonate ion and the hydroperoxide ion radicalthat eventually leads to the hydroxyl radical in the presence of ozone.Consequently, there could be a fraction of carbonates that do not inhibitthe ozone decomposition in water. Notice that in some cases when ozone isapplied, hydroxyl radical oxidation is negligible or does not developbecause direct ozonation is very fast. This is, for example, the case in theozonation of phenolic compounds under neutral or alkaline conditions,where the rate constants of the direct ozone reactions vary between 106 and

Beltran28

1010 M 1 s1. As observed, the order of magnitude is similar to or evenhigher than that of the reactions with the hydroxyl radical. An extensivestudy of this situation can be seen in a previous paper [124].

2. Chemical Kinetics

Once the mechanism of a reaction is established, kinetic studies constitutethe next step to determine the rate of pollutant degradation. Kinetic lawsmust be deduced experimentally following well-established rules [125]. Forthe degradation of a compound M through O3/H2O2/UV oxidation, the rateof M disappearance is given by Eq. (70):

rM rUV rD rH rR 70where the four terms of the right-hand side of Eq. (70) refer to thecontribution of direct reactions (photolysis, ozonation, reactions withhydrogen peroxide) and free radical oxidation, respectively. By assumingan irreversible rst-order reaction for every reactant (a global second-orderreaction), the last three rates are as follows:

rD zMkDCMCO3 71rH zHkHCMCH2O2 72rR zRkHOMCMCHO 73

where the values of zM, zH, and zR (usually 1) are the stoichiometriccoecients of M in their corresponding reactions. Notice that the negativevalue of the z coecients of M in Eqs. (71) to (73) is due to conventionalstoichiometric rules because M disappears. In the case of direct photolysis,considering the LL model, the photolysis rate is expressed by Eq. (53). Once

Table 8 Rate Constants of the Reaction Between the Carbonate Ion Radicaland Organic Compounds in Water

Compound Rate constant, M1 s1 pH

Benzene

the steady state approximation is accounted for, it is easy to show that theconcentration of hydroxyl radicals is given by Eq. (74):

CHO riXj

ksjCj74

where r1 is the rate of initiation steps and the denominator is the sum of theproducts of concentrations of substances that inhibit the decomposition ofozone and the corresponding rate constant of their reactions with thehydroxyl radical, a term usually called the scavenging factor. When a highconcentration of hydrogen peroxide is present in the reaction medium, theconcentration of hydroxyl radicals, however, can be limited by the transferof ozone into water [76]. Some investigations have also reported [76,126,127]that during ozonation, hydrogen peroxide plays a double role as initiatorand inhibitor of ozone decomposition. Thus, for concentrations usually upto 103 or 102 M, the increase of hydrogen peroxide concentration leads toan increase of the ozonation rate of M, but for concentrations above thesevalues hydrogen peroxide inhibits the ozonation rate. In these latter cases,reactions of ozone are so fast that ozone is not detected dissolved in waterand the process becomes mass transfer controlled. According to absorptiontheories [61], a complex kinetic equation can be deduced from the solutionof microscopic mass balance equations of ozone, hydrogen peroxide, andM, but a simplied equation is obtained from the macroscopic mass balanceequations as previously reported for trichloroethylene and tetrachloroethy-lene oxidation with O3/H2O2 in a semibatch system [76]. Thus, the nalequation for the concentration of hydroxyl radical is as follows [76,128]:

CHO kLaC

O3X

j

ksjCj75

In this latter case, one of the terms of the denominator in Eq. (5) containsthe concentration of hydrogen peroxide because this also plays a role as ascavenger. Thus, from Eq. (75), it follows that an increase in hydrogenperoxide concentration leads to a decrease in hydroxyl radical concentrationand hence, a decrease of rR.

The double role as scavenger and initiator, observed for hydrogenperoxide in the O3/H2O2 system, has also been reported in the UV/H2O2system. It should be noted that hydrogen peroxide does not inhibit theozone decomposition and Eq. (75) is valid only in the cases that ozone ispresent in the reaction mixture and the process is chemically controlled (lowconcentrations of hydrogen peroxide). This is because reactions of hydrogenperoxide with the hydroxyl radical release the superoxide ion radical that

Beltran30

subsequently reacts with ozone [see reaction (63)], and eventually anotherhydroxyl radical is generated (see also Fig. 6). However, in the UV/H2O2case, and in the absence of ozone, reaction (63) does not proceed, andhydrogen peroxide always acts as a terminator of propagation chainreactions through its reactions with the hydroxyl radical.

Table 9 presents the main reactions involved in the mechanism of O3/H2O2/UV radiation systems. According to this mechanism and from rateconstant data, concentrations, and the intensity applied for UV radiation,the relative importance of the initiating reactions can be deduced. Inprevious studies, comparative studies of the reaction rates of pollutantdegradation through these systems have been reported [119,120,124]. Fromthese studies, it was shown that in UV/O3 oxidation, initiation of freeradicals through the photolysis of hydrogen peroxide [reaction (58)] isnegligible if compared to reaction (61), and that reaction (59) starts to beof comparable rate to reaction (61) when the pH is higher than 10. Acomparative study between the importance of direct and free radicaloxidation pathways of ozonation based on the gas absorption theory isalso presented [61]. Thus, depending on the kinetic regime of ozoneabsorption, concentration of reactants (M, hydrogen peroxide), pH, inten-sity of UV radiation, etc., the direct or the indirect oxidation of a givencompound M will predominate.

3. Kinetic Modeling

To complete the fundamentals of these advanced oxidation systems, somediscussion should be included about the kinetic modeling studies found inthe literature. As stated previously, kinetic modeling allows predictions ofthe performance of any reacting system and represents an aid for processdesign (see Table 5). Kinetic modeling of O3/H2O2/UV systems has beenstudied extensively for the last decade with promising results especially forthe oxidation of volatile organochlorine compounds [7478]. For highermolecular weight compounds, with the exception of a few studies[79,81,129], the literature lacks information. This is probably because highmolecular weight compounds, while reacting with ozone or hydroxylradicals or being photolyzed, lead to intermediate compounds that alsoconsume oxidants and/or UV radiation, thus making the kinetics verycomplex. In a recent work [81], a kinetic model for any O3/H2O2/UVradiation system to treat aromatic compounds such as phenanthrene andnitrobenzene has been proposed. In the absence of UV radiation, theauthors found that the oxidation of phenanthrene, which only reactsdirectly with ozone, could be modeled acceptably without consideringthe presence of intermediates. A dierent situation holds for the case of the


Table 9 Reaction Mechanism for the O3/H2O2/UV System: Elementary Reactions, Reaction Rate Constant, and QuantumYield Dataa

ReactionRate constant

or quantum yield ReactionRate constant, pKa,or quantum yield

O3+OH!O2.+HO2. k1=70 HCO3+HO.!HCO3.+OH k21=2107

O3+HO2!HO.+O2.+O2 k2=2.2106 CO32+HO.!CO3.+OH k22=3.7108

O3+H2O+hm.!H2O2.+O2 /O3=0.64 CO3.+O2.!CO32+O2 k23=7.5108H2O2+hm.!2HO. /H2O2=0.5 CO3.+O3.!CO32+O3 k24=6107O3+O2.

!O3.+O2 k5=1.6109 CO3.+H2O2!HCO3+HO2 k25=8105HO3.!HO.+O2 k6=1.4105 s1 CO3.+HO2!CO32+HO2. k29=5.6107O3+HO.!HO4. k7=3.0109 H3PO4+HO.!H2O+H2PO4. k27=2.6106HO4.!HO2.+O2 k8=2.8104 s1 H2PO4+HO.!OH+H2PO4. k28=2.2107H2O2+HO.!HO2+H2O k9=2.7107 HPO42+HO.!OH+HPO4. k29=7.9105HO2.

+HO.!HO2.+OH k10=7.5109 HO2.ZH++O2. pK30=4.8HO4.+HO2.!O3+O2+H2O k11=1010 HO3.ZH++O3. pK31=8.2HO.+HO.!H2O2 k12=5109 H3PO4ZH++H2PO4 pK32=2.2HO.+O2.

!OH+O2 k13=1010 H2PO4ZH++HPO4 pK33=7.2HO.+HO3.!H2O2+O2 k14=5109 HPO4ZH++PO43 pK34=12.3HO3.+O2.

!OH+2O2 k16=1010 H2CO3ZH++HCO3 pK35=6.4HO3.+HO3.!H2O2+2O2 k16=5109 HCO3ZH++CO32 pK36=10.4HO4.+HO4.!H2O2+2O3 k17=5109 H2O2ZHO2+H+ pK37=11.7HO4.+HO3.!H2O2+O2+O3 k18=5109 M+zO3!Intermediates kD=?HO4.+HO.!H2O2+O3 k19=5109 M+HO!Intermediates kHOM=?HO4.+O2.

!OH+O2+O3 k20=1010 M+hm!Intermediates /M=?Adapted with permission from Ref. 81. Copyright 1999 American Chemical Society.a Units of rate constants and quantum yields (/) in M1 s1 and mol photon1 unless indicated. Reactions of intermediates are not given but alsotake part in the mechanism. M represents any compound present in water. Notice that the quantum yield (0.5) of hydrogen peroxide photolysis

corresponds to solutions irradiated in the presence of scavenger substances.

Beltra

n32

oxidation of nitrobenzene, which mainly reacts through hydroxyl radicals. Inthis case, the model does not lead to good results unless the mass balanceequations of intermediates are included. According to the authors [81] the keyfactor is the presence of hydrogen peroxide, proposed to be formed from theozonation of the starting aromatic compound. Thus, the system becomes anO3/H2O2 oxidation process, regardless of the initial presence of hydrogenperoxide. This observation is fundamental to matching the experimentaland predicted data. The model, however, fails when predicting the rates ofUV/O3 oxidation, probably because indirect reactions of organic peroxylradicals, reported elsewhere as a possible route of oxidation [10], were notaccounted for.

III. IN-DEPTH TREATMENT OF O3/UV/H2O2 PROCESSES

Applied treatment of these techniques requires the necessary equipment toproduce ozone and/or radiation. Hydrogen peroxide is the only reagent thatcan be purchased, transported to the plant, and used there. Ozone and UVradiation, however, require some capital investment, the importance ofwhich depends on the required ozone production rate and/or the UV lightux needed for the system. These issues constitute one of the main drawbacksof these systems: They have to be generated in situ.

In addition to the generation equipment, another important element isthe reactor or contactor chamber. Real reactors are similar in geometry or atleast in their fundamental aspects to those of laboratory or pilot plants. Inthis section, a brief summary of ozone and UV contactors is given. For moredetailed information related to this equipment, other literature sourcesshould be consulted [5,11,130].

A. Ozone-Based Processes

In actual ozonation processes, contactor chambers are similar to a bubblecolumn; the bottom chamber is equipped with many diuser plates throughwhich the oxidizing gas is fed. In other cases, however, the gas can be fedthrough injectors and intimately mixed with the water in static mixers toimprove mass transfer or to avoid problems derived from the precipitationof iron or manganese oxides.

Ozone contactors supplied with diusers are usually divided intostages (see Fig. 7) whose number depends on the major objectives ofozonation. Thus, for instantaneous or mass transfer controlled ozonations,two stages are sucient, whereas for slow ozonation reactions a highernumber is required. In order to reach plug ow behavior in the water, these


stages are separated by walls. This allows minimization of the residence timeneeded to obtain a given destruction of pollutants. In the rst stages, ozoneis fed through porous diusers located at the bottom and in the last stagesdissolved ozone circulates to provide a residual concentration of oxidant.

As was shown previously in some examples [1518] in a largeozonation plant for water treatment, residual ozone in the gas exiting theozonation stages could be sent back to the head of the water plant where it isinjected in another compartment to aid occulation, remove iron andmanganese, or reduce the trihalomethane formation potential (see Fig. 8).In these cases, it is not surprising that these plants could also have a naldisinfection ozonation step.

Ozonation contactors, however, are not always designed similar tobubble columns. In some cases where mass transfer represents an importantproblem, new contactor designs have been used, such as static mixers or thedeep-U-tube contactor.

Figure 7 Conventional waterozone contact chamber.

Figure 8 Water treatment plant with pre-ozonation and intermediate ozonationsteps. 1: Pre-ozonation chamber. 2: Equalization basin. 3: Coagulation chamber(C=coagulant). 4: Flocculation. 5: Sedimentation. 6: Ozonators. 7: Ozonation

chamber. 8. Active carbon ltration unit. 9: Final disinfection (D=disinfectant).

Beltran34

Static mixers are constituted by numerous perforated metal elements(i.e., sieve plates) housed in cylindrical tubes where mass transfer issignicantly increased. These units are especially recommended for ozona-tion processes where the rate of ozone is mass transfer controlled (i.e.,disinfection, decolorization, etc.) [131,132]. At the entrance of the tubes,water and the ozonated gas are simultaneously pumped and injected,respectively, yielding a gasliquid system with a very high specic interfacialarea for ozone transfer (1000 to 10,000 m1) and low energy dissipation (0.1to 5.0 kW m3) [130]. One possible disadvantage of static mixers is thepressure drop along the unit that limits the water ow rate applied.However, in some systems, such as those with sieve plates, the large free-area percentage, about 55%, allows a moderate pressure drop of 3000 to7000 Pa with practically 100% ozone utilization [130].

The deep-U tube is constituted by two vertical concentric tubesapproximately 20 m deep through which water and gas ow (inner tube)downwards and then upward through the outer tube. The main advantageof this system is the high driving force to dissolve ozone because of the highpressure and plug ow behavior due to the high velocity [133]. Disadvan-tages are the necessity of perforating the soil, which may involve geo-technical problems, and royalties required because it is a proprietary system.

B. UV Radiation Based Processes

For these cases, the so-called photoreactors are usually cylindrical chambersthat contain inner quartz sleeves where UV lamps are placed. These lampsare generally medium-pressure mercury vapor lamps (emitting radiationbetween 200 to 300 nm) of up to 60 kW each. The sleeve can have automaticdevices that wipe at regular intervals any precipitate formed or particlesdeposited on the sleeve walls, thus avoiding the problem of reduced lighttransmittance [134]. When combined with ozone, UV radiation photo-reactors are fed with ozone previously incorporated to the water ow[135] or with ozone fed directly into the photoreactor chamber [136]. Theformer methodology has the advantage that no ozone is destroyed in the gasphase by UV radiation; the latter process presents the advantage that theozone driving force, which favors ozone dissolution in water, is enhanced byits aqueous photolysis.

IV. DEGRADATION OF POLLUTANTS

The main objective of advanced oxidation processes involving ozone, UVlight, and hydrogen peroxide is the removal or degradation of pollutants


from water. Environmental agencies have classied some substances aspriority pollutants after a detailed study of their potential toxicity andhealth hazards [137]. Thus, phenols, aromatic hydrocarbons such as poly-cyclic aromatic hydrocarbons (PAHs), pesticides such as s-triazine com-pounds, volatile organochlorine compounds, others such as 1,4-dioxane,etc., are considered priority pollutants. As a result of industrial or agricul-tural activities, these substances have been found in many water environ-ments. In this section, brief information concerning laboratory studies onO3/UV/H2O2 oxidation is given together with a few specic examplesconcerning main pollutant families and general issues such as that of thebromate ion, detoxication, and biodegradability improvement of waste-water after a chemical oxidation step.

A. Laboratory-Scale Experimental Design

Laboratory work concerning the study of the advanced oxidation ofpollutants focuses on aspects related to the inuence of operating variableson the reaction rate (oxidant dose, UV intensity, pH, temperature, presenceof free radical scavengers, etc.), kinetics and mechanism (determination ofrate constants, quantum yields, etc.), identication of intermediates andnal products, and kinetic modeling. Most reported studies deal withlaboratory-prepared aqueous solutions, but a few deal with treatment ofnatural water and even wastewaters. By virtue of the number of studiespublished, a wide range of pollutants have been studied including dierentphenols, aromatic hydrocarbons, volatile organochlorine compounds, pes-ticides, and some other substances such as 1,4-dioxane and dissolved naturalorganic matter (such as humic substances). Table 10 provides a list of someof these studies published in the last decade (19902000). Also included inthe table are the oxidation and reactor system type, the compounds treated,and the main objectives of the work.

B. Examples of Laboratory Studies

In this section, examples of laboratory studies concerning O3/UV/H2O2advanced oxidations of some water pollutants are discussed. They have beenchosen because of the high interest that their oxidation treatment hasattracted among researchers in the eld. Three dierent types of pollutantshave been chosen: phenols of dierent nature, s-triazine herbicides, andsome volatile compounds, mainly chlorinated organics of low molecularweight (VOCs). Information is also given on the treatment of 1,4-dioxane,another important priority pollutant. These studies represent the scope andobjectives to be reached in this type of laboratory research.

Beltran36

Table 10 Studies on AOP Involving Ozone, Hydrogen Peroxide, and/or UV Radiation of Organic Compounds in Water

Compound Reacting system Main features Reference no.

Pyrene O3 BR, C0=10200 Ag L1. Intermediate identication 138Chlorophenols O3 SBBC, C=600 mg L

1. Eect on biodegradabilityand COD

139

Phenanthrene O3, O3/H2O2,O3/UV

SBBPR, LVP lamp 2.41014 photon s1. C=0.5 mg L1 140

Phenols O3, O3/H2O2,O3/UV,O3/TiO2UV/H2O2,

CBPR, LVP lamp 8 W, C=10170 mg L1.O3/H2O2 better system

141

p-Chlorophenol O3/UV SBPR, MVP lamp, C0=2.510 mg L1, TOC changes.

TOC kinetics142

Chlorophenols UV/H2O2 BPR, LVP lamp, 3,5 W. C0=3.067104 M.Absorbance eect on kinetics

143

Pyrene O3 SBBT, C0=15 mM, toxicity test. Intermediates 144

Chloro- andnitrophenols

O3/UV SBPR, Io, pH inuence, C=3104 M,LVP lamps 5 W maximum. Empirical kinetics

145

Trichlorobenzene O3, O3/H2O2,O3/UV,

UV/H2O2

CBPR, H=5 min. Kinetics, C0=1.21.6 mg L1,

LVP lamp 3.15 W146

Toluene O3, O3/H2O2 SFC, C0=0.751.5103 M, CH=2104 to 0.02 M,pH=311. Kinetics, reaction mechanism

147

Chlorophenols O3, O3/UV,UV/H2O2

BPR, SBBT, LVP lamp, 0.8 W L1,C0=10

4 to 4104 M. Reaction products,kinetics, toxicity

148

Phenol UV/H2O2 BPR, LVP lamp, 6 W, C0=103 to 102 M. COD

conversion, kinetics149

O3 /U

V/H

2 O2Oxidatio

nTechnologies

37

Compound Reacting system Main features Reference no.

Metol UV/H2O2 BPR, C0=5104 M, pH and H2O2 inuence, Kinetics 150VOCs O3 BR, C a few milligrams per liter. Removal of

chloroethylenes151

VOCs O3/UV CFPR, C0=100600 Ag L1, LVP lamp, 60 W,Io=240 W m

2. Kinetics152

Herbicides O3 BR. Kinetics. C in AM. Rate constants O3herbicidereaction

153

Atrazine O3 CFBC, C0=10 Ag L1. Intermediates, reaction mechanism 154VOCs O3/UV CFPR, LVP lamp, 60 W, C0=100600 Ag L1.

Kinetics modeling77

Chloroethanes UV/H2O2 BPR and CFPR, LVP lamps, Io=3.064.44106einstein s1, C0=0.41 Ag L1, CH=0.1105 M.Kinetic modeling

155

Atrazine O3, O3/H2O2 CSTR simulation. Kinetic modeling. C=3108 M 126Triazine herbicides O3/UV/H2O2/Fe

2+ SBBPR, LVP lamp 15 W, C0=1.9106 M 156Dyes O3, O3/UV,

O3/H2O2

SBBT, C0=260500 mg L1. Color and TOC removal,

metal release

157

Pesticides O3, O3/H2O2 BR, C0=112 Ag L1. Percentage removal atdierent conditions

158

Pesticides O3, O3/H2O2 SBBC, C0=2106 M, O3/H2O2 ratio eect 159Atrazine UV/H2O2 BPR, LVP lamp, 6.15.4106 einstein s1, C0 in AM.

Kinetic modeling79

Nonionic surfactants O3 SBR, C0=5.6104 M, pH 4 and 9.5. Reactionmechanism

160

RDX O3, O3/H2O2,UV/H2O2,

O/UV

BPR, SBPR, LVP lamp, C0=6170 Ag L1.Kinetics, intermediate identication, reaction mechanism

162

Table 10 continued

Beltra

n38

Atrazine O3, O3/H2O2,O3/UV

CBPR, pH inuence, mechanism, CATZ=6.95105 M,four LVP lamps of 15 W

163

Humic acids O3/UV SBBC, kinetics, TOC rates. Intermediates 164NOM O3 BR, changes on DOC 165Humic substances O3 SBBC, CHS=50500 mg L

1. Identication of compounds 166Fulvic acid O3, O3/H2O2,

O3/catalyst

BR, DOCo=2.84 mg L1, BDOCo=0.23 mg L

1.Changes of DOC and BDOC.

167

NOM O3 SBBT, DOCo=36 mg L1. Changes of DOC,

BDOC, THMFP

Chemical Degradation Methods for Wastes and Pollutants - M.a. Tarr

Documents

Transcript of Chemical Degradation Methods for Wastes and Pollutants - M.a. Tarr