A PHP Error was encountered

Severity: Notice

Message: Undefined index: fileUrl

Filename: controllers/content.php

Line Number: 95

Direct and indirect electrochemical oxidation of organic pollutants | Follow Science
A place to share and follow researchSubmit an Article or Content File →
Home  » 

Direct and indirect electrochemical oxidation of organic pollutants

University Department of Ferrara of Chemistry

Ph.D. in Chemical Sciences Direct and indirect electrochemical oxidation of organic pollutants

Supervisor: Prof. A. De Battisti Ph.D. thesis work by: Dr. Carlos Alberto Martínez Huitle Advisor: Dr. S. Ferro

Ph.D. in Chemical Sciences, Seventeenth Cycle Coordinator: Prof. G. Gilli

?Recuerda que la historia esta repleta de logros increíbles alcanzados por aquellos que fueron bastante locos para creer en sí mismos?

Acknowledgements The present Ph. D thesis was supported by National Council of Science and Technology of México Under the program ?Ph. D international scholarships?.

Ph. D. student on Chemistry Department, Laboratory of Electrochemistry Ferrara University, Italy.

I am grateful to Professor Achille De Battisti for giving me the opportunity to work in his research group, for his aid, motivation, comments, advices and friendship. Thanks a lot for many things, that he made for me, but especially, for the opportunity to participate in different international congress meetings, in particularly, in the Brazil and Greece ISE meetings.

I would like to thanks the CONACYT - MEXICO for the financial support during the period of my study and scientific research.

Thanks to Professor Chistos Comninellis for his collaboration and for establishing the contact with Professor Achille De Battisti. It was the first step for starting my Ph. D.

in Ferrara, Italy. Also, I am grateful to Professor Comninellis for accepting me in his research group two times in Switzerland, during my Bachelor (2000) and Ph. D.

studies (2003). These stages gave me a very important professional and personal experiences.

Thanks to all the members of the jury and thanks to Professor Christos Comninellis, Professor Achille De Battisti, Dr. Sergio Ferro for ?reading? my thesis and giving me helpful comments.

Thanks to Beatrice, Ilaria Duo, Gabrielle, Fabio, Flavio, Ricardo, Bahaa, Ivan and Lassinè for the great moments during the time, that we spent together in the EPFL, Switzerland.

I am grateful with my colleges and friends in Department of Chemistry, Ferrara University, during all of these years. Thanks to all of them: Ghiro, Luca, Stefano, Maria, Anna, Letizia, Loretta, Catia, Simone, Gennaro, Nicola, Nikla, Martino, Violetta, Nicola Marcheti, Simona.

Thanks a lot to students and scientists I worked with in the laboratory of the electrochemistry: Tiziano ?trop bel?, Sara ?esquiava?, Christian ?cioccolatino bianco?, Massimo ?Max?, Roberto, Chiara ?la bella bionda?, Federico ?accidenti- accidenti?, Elisa, Costantino ?tino?, Morozov ?aglio?, Dinara ?Dov'é..?, Davide ?tarocco dati?, Martina ?tina? and Paula ?addormentata?. In particularly, to Matteo Maria Borghato for his help with Italian language during my first days in Italy and for his friendship.

Rosa (El Salvador), Claudio (México), Lily (México), Bo (Korea), Rafael (México) and Argentinos, thank you all. You have been my first friends in Italy.

I would like to thank specially to Ilaria Boari and Lulu for great moments in the lab, Spanish, English and Italian class, good conversations and different motivation phrases to me. Non dimenticherò mai tutti i belli momenti fuori e dentro l'università con voi. Grazie mille per la vostra amicizia e incomparabile compagnia.

Grazie Lulu per dare ancora più vita alla comunità messicana nel lab. Y por cada una de las grandes pláticas y discusiones que tuvimos, asimismo los grandes consejos en los momentos dificiles y las innumerables risas.

Grazie Ilaria per la compagnia nel laboratorio; per le infinite conversazioni e tutte le lezioni d'italiano; per i test fati sul computer e per il bello state del 2003. Ed specialmente per la tua simpatia e incomparabile amicizia.

Thanks to Sergio Ferro my great Italian friend: his English corrections, motivation phrases, technical support, precious advice and helpful discussion. Grazie mille per le sue parole: ?che pazienza che ho?, ?Non ho capito niente?, ?Cosa stai facendo?, ?sei un porco? e ?Finto?. Ti ringrazio per tutte le belle giornate a casa tua, per l'aiuto durante la scrittura degli articoli e le buone conversazioni che abbiamo avuto. Non dimenticherò mai i piccoli regali e le correzioni del mio italiano.

Grazie mille. A todos esos grandes amigos que, desde el otro lado del mar siempre me echaron muchas porras. En especial a Arancha, Edith Villascan y Lucy.

Thank you Elaine. Grazie Mille per i più belli momenti della mia vita. Per il sostegno e il tuo amore. Por los grandes y maravillosos viajes, momentos de espera, risas y la gran comprensión. Você é muito especial em meu coração e em minha vida.

Agradezco infinitamente la ayuda proporcionada cada vez que hablábamos y los grandes consejos que me dieron día con día. Por el ?coraje? y ?tranquilidad? que me trasmitieron para poder seguir adelante.

Uriel Y Paola, gracias por darme su confianza y ayuda en los momentos difíciles. Por no dejarme solo y darme siempre palabras de aliento.

Electrochemical incineration of Tartaric acid 129 Introduction 129 Experimental 131 Cell and electrolyses conditions 131 Techniques and instrumentation 131 Discussion and results 132 Anodic oxidation of Tartaric acid 132 Electrochemical oxidation pathway 142 Conclusions 145 References 147 Chapter 5. Reactivity and Engineering parameters 149 Introduction 149 Experimental part 151 Cell and electrodes 151 Techniques and instrumentation 154 Results and discussion 156 Mass transport coefficient 156 Anodic oxidation of oxalic acid Oxalic acid oxidation and kinetics 160 Conclusions 165 References 167 Chapter 6. Indirect or mediated electrochemical oxidation 171 Anodic oxidation of tartaric and oxalic acids in the presence of halides 171 Introduction 171 Experimental section 175 Cell and electrodes 175 Techniques and instrumentation 175 Discussion and results 176 Electrochemical measurements 176 Anodic oxidation 179 Influence of media 179 Influence of halides 182 Influence of applied current density 190 Variation of pH 190 Representation data as ?Volcano Plot? 193 Conclusions 195 References 196

Results and discussion 202 Conclusions 235 References 238 Chapter 8. Hydroxylation 241 Electrochemical and chemical hydroxylation of salicylic acid 241 Introduction 241 Experimental 243 Results and Discussion 245 Chemical hydroxylation 245 Electrochemical hydroxylation 249 Comparison 252 Hydroxylation selectivity 253 Conclusions 256 References 257

Abstract In recent years, the perspectives of application of electrochemistry for environmental pollution abatement have been thoroughly investigated. The feasibility of electro-chemical incineration of organic substrates in wastewater, in particular, has drawn much attention since pioneering papers by Dabrowski in the 70's, Kirk, Stucki, Kotz, Chettiar and Watkinson in the 80's, and Comninellis in early 90's to ours days.

In these works, the influence of the nature of the electrode material on the faradaic efficiency of the anodic mineralization of organics has been considered in detail, showing that optimal conditions for the process in question are met at high-oxygen- overpotential anodes. Basing on these evidences, Comninellis has proposed a general model, which satisfactorily accounts for the different results described in the literature.

More recent results, obtained at conductive diamond electrodes, characterized by very high oxygen overpotential, also fit in the model predictions quite well. Many model organic substrates, like glucose, aldonic and aldaric acids among others, show a complex reactivity toward the anodic mineralization. In fact, in these cases, besides the central role of adsorbed hydroxyl radicals, also the mode of adsorption of the organic species has to be taken into account.

Simpler molecules, like oxalic acid, also support the view that co-electrosorption of hydroxyl radicals and organic species decides of the rate of the anodic mineralization, volcano-plot-approaches of the type applied long since for ethylene electrochemical oxidation being possibly a good interpretative tool. These considerations afford further evidence to the importance of the nature of the electrode material in electrochemical incineration.

Taking into consideration the role of mediators, like Cl-, of particular interest for its common presence in many different types of wastewater, interpretations may be attempted, bringing again the role of the electrode material in a prominent position. Chloride-mediated incineration can be properly accounted for by assumptions on co-adsorbed hydroxy- and chloro-radicals at the electrode surface and on the effect of this on the rate of oxygen evolution reaction. Converting the results of the indirect oxidation, as second alternative in the elimination of the organic pollutants from water.

The aim of this thesis work is emphasizing the role of the electrode material in direct and indirect electrochemical oxidation for wastewater treatment.

List of Symbols List of symbols A electrode surface area (m2) C bulk species concentration (mol m-3) D diffusion coefficient (m2 s-1) de duct equivalent diameter (m), de ( ) = 2ws w + s F faradaic constant (96 487 C mol-1) h a fourth of the disk circumference IL electrolysis limiting current (A) K mass transfer coefficient (m s-1) L electrode length (m) Re Reynolds number s inter-electrode distance (m) Sc Schmidt number Sh Sherwood number, based on cell or channel equivalent diameter w section of cell or channel (m2) z electrons exchanged in electrode reaction v mean fluid velocity in cell or channel (m s-1) Q volumetric flow rate (m3 s-1)

Chapter 1. Introduction

Introduction The intensification of industrial activities, since the second half of XIX century and during all the XX century, has inevitably caused severe environmental pollution with dramatic consequences in atmosphere, waters, and soils. The consequent restrictions imposed by new legislations require effective initiatives for pollution abatement not only in gaseous emissions and industrial aqueous effluents, but also adequate decontamination in soils. Typically, in the latter case, different classes of pollutants may have accumulated during long periods of uncontrolled waste disposal and reclamation may represent a serious technological problem. In all these cases, universal strategies of reclamation can hardly be found, due to the very different features of pollution phenomena.

Generally, the wastewater treatment is effectuated using primary, secondary or tertiary methods, according to type of pollutants. As far as organic pollutants in wastewaters are concerned, biological abatement may be sometimes impossible, due to the bio-refractory character of substrates. For this reason, physical-chemical methods are applied, but an oxidation with e.g. ozone or chlorine dioxide is not always effective and also transportation and storage of reactants may be a significant inconvenience for safe processing.

An alternative can be the application of electrochemical technologies for wastewater treatment, taking profit of advantages like versatility, environmental compatibility and potential costs effectiveness among others. Accordingly, both the direct and the mediated electrochemical oxidation/incineration can be considered, anyhow representing an interesting subject for different research groups and industries, which are looking for new technologies for the wastewater treatment (Chapter 2).

Basically, the electrochemical incineration proceeds under the action of strong oxidants, similarly to the chemical incineration, but the in-situ electro-generation

Chapter 1. Introduction

allows better efficiency of the abatement of the organic substrates and avoids the need for transportation and storage.

The direct electrochemical incineration has been investigated, with particular efforts, since the end of the eighties, testing different anodic materials for the oxidation of diverse organic pollutants dissolved in water. A mechanism considering the different stabilization exerted by the electrode material on electrosorbed hydroxyl radicals has been proposed; accordingly, electrodes have been classified as active and non- active, on the basis of their electrocalytic properties. In this context, an important result of this PhD work can be found in the confirmation of the influence of the electrode material on the efficiency of the process.

On the other hand, the indirect oxidation (also called mediated electrochemical oxidation) is based on the activity of strongly oxidant species, like e.g. Cl-, S2O8= and Ce(IV), and may also represent an interesting alternative to the above- mentioned wastewater treatments. The Cl--mediated mineralization has been shown to give good results at low-oxygen-overvoltage electrodes, like Pt. The addition of chloride ions in the electrolyte allows an increase of the removal efficiency, and a degradation of pollutants can be obtained due to the participation of active chlorine.

Similarly to chlorides, also bromides (e.g., NaBr) can be effectively used for the anodic oxidation of organic pollutants, but this anion has been scarcely investigated and only a limited number of examples are reported in the literature.

For both the direct and indirect oxidations, many model organic substrates have been considered, as well as different experimental conditions and anode materials;

nevertheless, in many cases, the electrochemical process leads to the formation of stable carboxylic acids such as maleic, formic, acetic, malonic and oxalic acid. Also, these molecules may represent the polluting content of industrial wastewaters, as in the case of e.g. the olive oil manufacturing.

Chapter 1. Introduction

defining optimal process yields, at least in terms of electrode material. Experimental evidences should be of help to understand the interaction between the organic substrate and the electrode surface, thus allowing an optimization of the process conditions and possibly increasing the elimination of water pollutants (Chapter 4).

On the basis of the results obtained investigating the above organic substrate, an alternative compound, tartaric acid, has been selected to extend the investigation to more complex molecular frames, possibly involving the formation of intermediates blocking, at different extents, the electroactive sites (again in Chapter 4). In fact, the latter carboxylic acid has a higher complexity, in terms of molecule structure, with respect to oxalic acid. Similarly to what has been done with oxalic acid, tartaric acid has been studied at different electrode materials (Ti/PbO2, BDD and Pt), trying to understand the interactions of this organic substrate and hydroxyl radicals with the electrode surface.

In consideration of the fact that also the cell design and the way electrodes are prepared (characteristic and methodology of synthesis) can play a role in the electrochemical oxidation process, among other variables, investigations have been carried out at different anode materials (Pb/PbO2, BDD, Ti/Pt and Ti/IrO2-Ta2O5), making use of oxalic acid as a model organic pollutant, with the aim of testing a disk-shaped, parallel-plate electrochemical cell based on a new design, with respect to those reported in the literature. The cell has only one central inlet and several peripheral exit sections, thus allowing the obtainment of a rather complex hydro- dynamics. Measurements of mass transfer coefficient were made using the limiting diffusion current technique based on ferricyanide ion reduction, and overall mass transfer coefficients were correlated to Reynolds numbers ranging from 100 to 800.

Thanks to the above features, and comparing the obtained data with those available in the literature, for similar devices, higher mass transfer coefficients could be obtained with the new cell design, thus allowing an improvement for the whole electrochemical process (Chapter 5).

Chapter 1. Introduction

While a thorough optimization of the process can suggest the use of a particular anode material, other motivations can drive the choice, like e.g. the anode material availability and cost. For these reasons, in Chapter 6, the ?mediated electrochemical oxidation? (MEO) was also studied, accomplishing the anodic oxidation of both carboxylic acids in the presence of halides (NaCl, NaBr and NaF); both the influence of the nature of the anion, and its concentration were analyzed. In this context, the chloride mediation is of particular interest because of the ubiquitous character of Cl- species in wastewaters, and because of their quite effective action;

as a result, different papers have already appeared in the literature, suggesting possible ?direct? and ?indirect? roles for the anion in the electrochemical reaction.

On the contrary, the effects of both NaBr and NaF were not previously investigated and reported to the scientific community, which makes these aspects particularly interesting. In addition, the study can help in explaining the specific role of halides at the electrode surface.

Since oxalic acid (OA) represents one of the proposed metabolites of the anodic oxidation of more complex organic molecules, in spite of its simple structure, its mineralization is of particular interest. Experimentally, it appears to be strongly dependent on the nature of the electrode material at which the process is carried out.

Thus, it is important to understand the kinetic mechanism of the electrooxidation, on the basis of results obtained at different electrodes. In the literature, the investigation was carried out at Rh, Pd, Os, Ir, Pt, Au, a DSA-type anode like RuO2-TiO2, and glassy carbon; while repeating and partly confirming the above results, in Chapter 7, the analysis of the dependence of the OA electro-oxidation on the nature of the electrode material has been further extended to highly conductive, boron-doped diamond (BDD) electrodes, with either oxygen or fluorine at their surface, as well as to other DSA-type electrodes.

Chapter 1. Introduction

the spin-trap characteristics of salicylic acid and analysing the obtained metabolites through HPLC measurements. Since hydroxyl radicals have been suggested to play a significant role, especially in the direct electrochemical oxidation, many efforts have been made to confirm their generation, interaction and chemical reactivity at ?non- active electrodes? like BDD. This last part of the research was realized at EPFL (Switzerland), in the frame of the scientific collaboration existing between this research group and that of Professor Comninellis.

Finally, in Chapter 9, a general discussion and future perspectives about the application of the electro-chemical oxidation as an alternative to common wastewater treatments are proposed.

Chapter 2. Bibliography

Bibliography A thorough analysis of wastewater treatments, as they can be found in the literature, is presented. A distinction was made between these treatments and the so-called ?advanced oxidation processes? (AOP). Particular attention has been dedicated to physical- chemical processes, as they are the most significant methods for the treatment of organics dissolved in water effluents, and considering that oxidation is anyway involved In this context, an examination of several AOP is presented: in fact, different methods have been investigated by several research groups for the elimination of organic Concerning the development of new technologies, a mention is deserved by the application of electrochemistry for the protection of the environment, since it offers promising approaches for the prevention of pollution problems in the process industry. The inherent advantage of electrochemical methods has to be found in their environmental compatibility. The fundamental concepts of the method and the different investigations carried out at several electrode materials will be presented, emphasizing On the other hand, a general description of electrode properties will be also given, moving from the nature of electrode reactions (inner and outer sphere) to considerations of electrocatalytic activity for their classification (active or non active).

Chapter 2. Bibliography

Wastewater Treatment [1] The treatment of effluents represents a serious problem, especially for the chemical industry. Over the last twenty-five years, big efforts have been made to limit at the source this type of pollution, by improving processes, recycling products and controlling the treatment of wastes at the production stage. However, considering the large amount of industrial effluents to be treated, for example to retrieve certain solvents, there are inevitably residues requiring a final transformation, which is often delicate. Traditional incineration methods, for their part, pose problems of corrosion and, more seriously, of From the point of view of the industry, the problem must be examined as a whole, since there are no universal and simple methods in this area. The great variety of industrial discharges means that a diversification of techniques must be sought, adapting the treatment to each situation, as much as possible. In spite of the efforts made to develop ever-clean processes, the increasingly severe environmental laws should encourage the research for better-performing treatments, making it possible to obtain environmentally Actually, the processes for the treatment of wastewater may be divided into three main categories of primary, secondary and tertiary treatments, each of which will be discussed separately.

Chapter 2. Bibliography devices shred and grind solids in the sewage. Particle size may be reduced to the extent that the particles can be returned to sewage flow.

Secondary treatment The most obvious harmful effect of biodegradable organic matter in wastewater is BOD, consisting of a Biological Oxygen Demand for dissolved oxygen by microorganisms- mediated degradation of organic matter. Secondary wastewater treatment is designed to remove BOD, usually by taking advantage of the same kind of biological processes that would otherwise consume oxygen in water receiving the wastewater. Secondary treatment by biological processes takes many forms but consists basically of the following: microorganisms provided with added oxygen are allowed to degrade organic material in solution or in suspension, until the BOD of the waste has been reduced to acceptable levels. The waste is biologically oxidized under condition controlled for optimum bacterial growth and at a site where this growth does not influence the environment.

Chapter 2. Bibliography addition, potentially hazardous toxic metals may be found among the dissolved inorganics. The major problem is due to the dissolved organics, because these are the Complete physical-chemical wastewater treatment systems offer both advantages and disadvantages relative to biological treatment. The capital costs of these facilities can be less than those of biological treatment facilities, and they usually require less land. Basically, a physical-chemical treatment process involves [1, 3]: 9 Removal of scum and solid objects 9 Clarification, generally with addition of coagulants, and frequently with the addition of other chemicals 9 Filtration, to remove filterable solids 9 Sedimentation 9 Elimination of heavy metals by precipitation with Ca(OH)2, H2S and/or ionic exchange resins 9 Activated carbon adsorption 9 Inorganic dissolved compounds removal by electrodialysis, ionic exchange and/or inverse osmosis on semi-permeable membrane 9 Disinfection with Cl2, ClO and/or O3 In addition, the wastewater can be treated by a variety of chemical processes, including Therefore, in recent years, combined treatment methods are finding an increasing use, particularly physical-chemical treatment with subsequent biological treatment, and physical-chemical processes employed as tertiary treatment after a biological treatment. Flow sheets of treatment facilities using physical-chemical methods alone are also being developed [4].

Chapter 2. Bibliography

Chapter 2. Bibliography special attention given to fundamental chemical aspects. Advanced oxidation processes, although making use of different reacting systems, are all characterized by the same chemical feature: the production of OH radicals. Hydroxyl radicals are extraordinarily reactive species: they attack the most part of organic molecules with rate constants usually in the order of 106?109M?1 s?1 [12, 13]. They are also characterized by a little selectivity of attack, which is a useful attribute for an oxidant used in wastewater treatment and for solving pollution problems. The versatility of AOP is also enhanced by the fact that they offer different possible ways for OH radicals production, thus allowing a better compliance with the specific treatment requirements. A suitable application of AOP to wastewater treatments must consider that they make use of expensive reactants as H2O2 and/or O3 and therefore it is obvious that their application should not replace, whenever possible, more economic treatments as the biological degradation. The potentialities offered by AOP can be exploited to integrate biological treatments by an oxidative degradation of toxic or refractory substances entering or leaving the Another aspect, concerning the opportunity of AOP application, is that referring to the polluting load of wastes normally expressed as COD. Only wastes with relatively small COD contents can be suitably treated by means of these techniques, since higher COD contents would require the consumption of too large amounts of expensive reactants. Wastes with more massive pollutants contents can be more conveniently treated by means of wet oxidation or incineration [14]. Wet oxidation makes use of oxygen or air to achieve pollutant oxidation at high temperatures (130-300 °C) and pressure (0.5-20 MPa). For these wastes, the cost evaluation of fuel consumption will give the selection criteria for the application of AOP or wet oxidation treatment.

Chapter 2. Bibliography

Fenton processes The renewed interest of researchers for this classic, old reactive system, discovered by Fenton the last century [15], is today underlined by a significant number of investigations devoted to its application in wastewater treatments. It has been demonstrated that the Fenton's reagent is able to destroy toxic compounds in waste- The production of OH radicals by the Fenton reagent [16] occurs by means of addition of H2O2 to Fe2+ salts: Fe2+ + H O ? Fe3+ + OH- + OH? (1) 22 This is a very simple way of producing OH radicals, neither special reactants nor special apparatus being required. This reactant is an attractive oxidative system for wastewater treatment due to the fact that iron is a very abundant and non toxic element and hydrogen peroxide is easy to handle and environmentally safe. It must be stressed that the behavior of the system cannot be completely explained on the basis of the sole In fact, as it has been pointed out in many recent studies [17] the adoption of a proper value of pH (2.7-2.8) can result into the reduction of Fe3+ to Fe2+ (Fenton-like) Fe3+ + H O H+ + FeOOH2+ (2) 22

FeOOH2+ ? HO ? + Fe2+ (3) 2

proceeding at an appreciable rate. In these conditions, iron can be considered as a real catalyst.

Chapter 2. Bibliography process, which takes advantage from UV-VIS light irradiation at wavelength values higher than 300 nm. In these conditions, the photolysis of Fe3+ complexes allows Fe2+ regeneration and the occurrence of Fenton reactions due to the presence of H2O2. Despite the great deal of work devoted by researchers to these processes, scanty indications have been found about their industrial applications. This is not surprising, since Fenton processes application requires strict pH control and sludges can be formed with related disposal problems.

3+ UV/Fe -Oxalate/H2O2 An improvement of photo-assisted Fenton processes is the UV-VIS /Ferrioxalate/ H2O2 system, which has been very recently demonstrated [20] to be more efficient than photo- Fenton for the abatement of organic pollutants: ferrioxalate is the oldest and best-known photo-active example of Fe3+-polycarboxylate complex [21]. Irradiation of ferrioxalate in acidic solution generates carbon dioxide and ferrous ions Fe2+ free or complexed with oxalate, which, in combination with H2O2, provides a continuous source of Fenton's reagent.

Photocatalysis Photocatalytic processes make use of a semiconductor metal oxide as catalyst and of oxygen as oxidizing agent [22]. Many catalysts have been so far tested, although only TiO2 in the anatase form seems to have the most interesting attributes, such as a high stability and good performance [23, 24]. The initial event in the photocatalytic process is the absorption of the radiation with the formation of electron-hole pairs. The considerable reducing power of formed electrons allows them to reduce some metals and

Chapter 2. Bibliography

dissolved oxygen with the formation of the superoxide radical ion O2??, whereas remaining holes are capable of oxidizing adsorbed H2O or HO? to reactive HO radicals: TiO(h+)+HO?TiO+HO·+H+ (4) 2 2 ad 2 ad

TiO2(h+)+HOad-?TiO2+HOad· (5) These reactions are of great importance in oxidative degradation processes, due to the high concentration of H2O and HO? adsorbed on the particle surface. Unfortunately, a significant part of electron-hole pairs recombine, thus reducing the quantum yield. In the most part of the works devoted to photocatalysis, the possible exploitation of the wavelengths of the solar spectrum is stressed. However, this is only partially true since the overlapping between the absorption spectrum of TiO2 and that of the sun at ground is rather poor [24]. Intensive researches are carried out worldwide to obtain modified (doped) TiO2 with broader absorption spectrum and characterized by a higher quantum yield. Despite the great efforts devoted to the study of photocatalytic processes, no indications have been found in the literature of their application on industrial scale.

Chapter 2. Bibliography enhance the O3 decomposition with formation of OH radicals. The influence of pH is also evident, since in the ozone decomposition mechanism the active species is the conjugate base HO2?, whose concentration is strictly dependent upon pH. The increase of pH and the addition of H2O2 to the aqueous O3 solution will thus result into higher rates of OH radicals production and the attainment of higher steady-state concentrations It must be remarked that the adoption of the H2O2/O3 process does not involve significant changes to the apparatus adopted when only O3 is used, since it is only necessary to add an H2O2 dosing system.

2+ Mn /oxalic acid/ozone The system Mn2+/oxalic acid can be conveniently used to enhance ozone decomposition to produce HO radicals. Mn2+ catalyzed ozonation of oxalic acid has been shown to develop according to a radical mechanism at pH > 4.0 at which Mn(III)-dioxalate and Mn(III)-trioxalate are formed. In these conditions, the oxidation process proceeds presumably through the formation of OH radicals as a result of a reaction between manganese complexes and ozone [27]. The system has been demonstrated to be effective for the abatement of refractory pollutants such as pyrazine and pyridine [28].

H2O2 photolysis This process is realized by irradiating the pollutant solution containing H2O2 with UV light having wavelengths smaller than 280 nm. This causes the homolytic cleavage of H2O2 [29]: H O ???2OH· (6) hv 22

Chapter 2. Bibliography Since H2O2 itself is attacked by OH radicals, the overall quantum yield becomes one. The rate of photolysis of aqueous H2O2 has been found to be pH dependent and to increase when more alkaline conditions are used [1]. This may be primarily due to the higher molar absorption coefficient of the peroxide anion HO2?, which is 240M?1 cm?1, at 254 nm.

O3/UV O3/UV process is an advanced water treatment method for the effective oxidation and destruction of toxic and refractory organics in water. Basically, aqueous systems saturated with ozone are irradiated with UV light of 254 nm in a reactor convenient for such heterogeneous media. The extinction coefficient of O3 at 254 nm is 3600M?1 cm?1, much higher than that of H2O2. The O3/UV oxidation process is more complex than the other ones, since OH radicals are produced through different reaction pathways; under these conditions, the system has the chemical behavior of both O3/H2O2 and H2O2/UV systems [5, 10]. From the photochemical point of view, the absorption spectrum of ozone provides a much higher absorption cross section than H2O2, and inner filter effects by, e.g., aromatics are less problematic [5].

Electrochemistry for the Environment An increase world population with growing industrial demands has lead to a situation where the protection of the environment has become a major issue and crucial factor for the future development of industrial process, which will have to meet the requirements of sustainable development.

Chapter 2. Bibliography Electrochemistry offers promising approaches for the prevention of pollution problems in the process industry. The inherent advantage is its environmental compatibility, due to the fact that the main reagent, the electron, is a clean reagent. The strategies include both the treatment of effluents and waste and the development of new processes or products with less harmful effects, often denoted as process-integrated environmental protection The application of the electrochemistry for the protection of the environmental has been the topic of several books and reviews [31-38]. Besides the process-oriented benefits, electrochemistry is also playing a key role in the sensor technology. Electroanalytical techniques for monitoring and trace level detection of pollutants in the air, water and soil as well as of microorganisms are needed for process automation. Sensors for environmental applications have been already reviewed [32, 34], while an interesting view on the role of electrocatalysis for electrochemistry and environment has recently The electrochemical processes offer several promising approaches for the prevention and remediation of pollution problems. Among the main characteristics that are attractive, the following can be chosen [34]: Versatility: direct/indirect oxidations and reductions, phase separations, biocide functions, concentrations or dilutions; can deal with many pollutants: gases, liquids Energy efficiency: these processes generally have lower temperature requirements than those of equivalent non-electrochemical counterparts (e. g., thermal incineration); the potentials can be controlled and electrodes and cells can be designed to minimize power losses due to poor current distribution, voltage drops and side reactions.

Chapter 2. Bibliography Amenability to automation: the electrical variables used in the electrochemical processes (I, E) are particularly suited for facilitating data acquisition, process Environmental compatibility: the electron is a clean reagent, and the high selectivity of many of these processes can be used to prevent the production of unwanted side- Cost effectiveness: the required equipment and operations are generally simple and, if properly designed, are also inexpensive.

Therefore, intensive research continues with the goal of discovering more efficient techniques, processes, materials, technologies and applications of the electrochemistry for the remediation and/or prevention of pollution problems.

Electrochemical technologies for wastewater treatment Electrochemical technologies have recuperated their importance in the world during the past two decades. There are different companies supplying facilities for metal recoveries, treating drinking water or process water, treating various wastewaters resulting from tannery, electroplating, dairy, textile processing, oil and oil-in-water emulsion, etc. At the present time, electrochemical technologies have reached such a state that they are not only comparable with other technologies in terms of cost, but also more efficient and compact. The development, design and application of electrochemical technologies in water and wastewater treatment have been focused particularly to some technologies such as electrodeposition, electrocoagulation, electrofloculation and electrooxidation. These techniques will be explained in brief, being the electrochemical oxidation ours central point.

Chapter 2. Bibliography

Electrodeposition The electrochemical recovery of metals has been used in form of electrometallurgy since long time ago [40]. The first recorded example of electrometallurgy was in mid-17th century in Europe [41]. It involved the recovery of copper from cupriferous mine water electrochemically. During the past two and half centuries, electrochemical technologies have grown into areas as energy storage, chemical synthesis, metal production, surface The electrochemical recovery of metals can be used in the metal surface finishing industry. It has to bear in mind that it is unable to provide a complete solution to the industry's waste management problems because it cannot treat all the metals either technically or economically. The electrolytic recovery of metals involves two steps: collection of heavy metals and stripping of the collected metals. The first step involves plating and the stripping can be accomplished chemically or electrochemically. Actually, metals powders can be formed on the surface of carbon electrodes and a physical Another application is in the printed circuit board manufacturing industry: for treating dilute effluent, an ion-exchange can be used, while high concentration streams can be treated directly, using a recovery system as in metal surface finishing industry.

Chapter 2. Bibliography plate form or packed form of scraps. The advantages of electrocoagulation include high particle removal efficiency, compact treatment facility, relatively low cost and This electrochemical technology is efficient in removing suspended solids as well as oil and greases. It has been proven to be effective in water treatment such as drinking water supply for small or medium sized community. Electrocoagulation is effective in removing the colloidals found in natural water so that both the turbidity and color are reduced. Also, it is used in removal or destruction of algaes or microorganisms; it can be also used to remove irons, silicates, humus, dissolved oxygen, etc. [43]. Electrocoagulation was particularly employed in wastewater treatment [44]. It has been employed in treating wastewaters from textile [45-48], catering [49, 50], petroleum [51], carpet wastewater [52], municipal sewage [53], chemical fiber wastewater [54], oil- water emulsions [55, 56], oily wastewater [57], clay suspension [58], nitrite [59] and dyestuff [60] from wastewater. Also, copper reduction, coagulation and separation were effective [61].

Chapter 2. Bibliography [75], wastewater from coke-production [76], mining wastewater [77], groundwater [78], food processing wastewater [79], fat-containing solutions [80], restaurant wastewater [50], or food industry effluents [81], dairy wastewater [82], urban sewage [83], pit waters [84], colloidal particles [85], heavy metals containing effluents [86-89], gold and silver recover from cyanide solution [90] and many other water and wastewaters [71, 91, 92].

Electrochemical Oxidation: An alternative in wastewater treatment Studies on the electrochemical oxidation for wastewater treatment go back to the 19th century, when the electrochemical decomposition of cyanide was investigated [93]. Extensive investigation of this technology commenced since late of 1970s. During the last two decades, research works have been focused on the efficiency in oxidizing various pollutants on different electrodes, improvement of the electrocatalytic activity and electrochemical stability of the electrode materials, investigation of factors affecting the process performance and the exploration of the mechanisms and kinetics of the pollutant degradation. Experimental investigations, focused mostly on the behavior of anodic materials, have been realized by different research groups. Attempts for an electrochemical oxidation/incineration treatment for waste or wastewater can be subdivided in two important categories [30]: 9 Direct oxidation at the anode 9 Indirect oxidation using appropriate, anodically-formed oxidants

Chapter 2. Bibliography

Direct and Indirect Electrochemical Oxidation Direct anodic oxidation Electrochemical oxidation of pollutants can occur directly at anodes through the generation of physically adsorbed ?active oxygen? (adsorbed hydroxyl radicals, ?OH) or chemisorbed ?active oxygen? (oxygen in the oxide lattice, MOx+1) [94]. This process is usually called ?anodic oxidation? or ?direct oxidation? and the course for the anodic oxidation has been described by Comninellis [94]; the complete combustion of the organic substrate or its selective conversion into oxidation products is schematically represented in Figure 1. The electrochemical conversion transforms only the toxic non- biocompatible pollutants into biocompatible organics, so that a biological treatment is still required after the electrochemical oxidation. In contrast, the electrochemical combustion yields water and CO2 and no further purification is necessary. Nevertheless, the feasibility of this process depends on three parameters: (1) the generation of chemically or physically adsorbed hydroxyl radicals, (2) the nature of the anodic material and (3) the process competition with the oxygen evolution reaction. The anodic oxidation does not require the addition of large amounts of chemicals to wastewater or to feed O2 to cathodes; moreover, there is no tendency of producing secondary pollution and fewer accessories are required. These advantages make the anodic oxidation more attractive than the other oxidation processes. As previously commented, the most important parameter in this process is obviously the anode material. Among the investigated anode materials, the following can be mentioned: glassy carbon [95], Ti/RuO2, Ti/Pt-Ir [96, 97], carbon fibers [98], MnO2 [99, 100], Pt- carbon black [101, 102], porous carbon felt [103], stainless steel [55] and reticulated vitreous carbon [104, 105]. Unfortunately, none of them has sufficient activity and, at

Chapter 2. Bibliography the same time, satisfactory stability. Pt, PbO2, IrO2, SnO2, and conductive diamond films Table 1 presents a comparison among the different anodes for the degradation of some important pollutants, under different conditions. Also, it is worth mentioning that this table contains the more relevant researches in the frame of the direct electrochemical oxidation from the beginning of the application of this method to nowadays. Different parameters have been resumed; nevertheless, the current density and the current efficiency (CE) [106] are of particular interest, as well as the possible intermediates. The current efficiency [106] is a measure of the process effectiveness; it can be alternatively expressed by means of the Electrochemical Oxidation Index (EOI), the Apparent Current Phenol and derivates are among the mostly investigated examples in electrochemical studies. Dabrowski et al. in 1975 [224] have also tried the use of electrochemical oxidation for the destruction of phenolic waste on a pilot-scale plant; Chettiar and Watkinson used synthetic wastewater solutions in 1981 [225] and 1983 [226].

Chapter 2. Bibliography Anode Pollutant Current Efficiency Removal Comments Metabolites Ref. density, (ICE, EOI, efficiency intensity ACE, %) or potential 3D carbon felt nitroaromatic 500- - 10-50% COD H2SO4 from 50- 2-methyl [143] waste 1000A/m2 removal 96%, T=25- quinoxaline, 2,3- 40°C dimethyl quinoxaline, quinoxaline Au benzoquinone 10mA/cm2 46% (after 48 Vol=50 ml Hydroquinone, [127] hrs) [BQ]= 50ml resorcinol, p- soln. stock, benzoquinone, [sol. pyrocatechol, stock]=100mg/ formic, maleic, L succinic, malonic, fumaric, acetic acids Au Pentachlorophenol Polymerization [144]

Chapter 2. Bibliography Anode Pollutant Current Efficiency Removal Comments Metabolites Ref. density, (ICE, EOI, efficiency intensity ACE, %) or potential Packed bed reactor Phenol I=3.0V 16% (after 1.5hrs.) Phenol removal [Phenol]=1.4x10- Phenol, [151] of PbO2 pellets 72-100%, 2M, 1M H2SO4, benzoquinone, depending First Electro- maleic acid and CO2 [phenol]; ?J-? % chemical oxidation removal, ? studies, Innovative [phenol]- ? % reactor, Study at removal, ? different [phenol], [H2SO4]- ? % applied current removal, ?T - ? density, [H2SO4], [benzoquinone] T=25-50°C, conversion and ? dissolved oxygen [CO2] conversion after 1. 5h and considerable [CO2] achieved PbO2 Indoles I=0.5-500mA 32% 0.05 Na2SO4 Polymerization [113]

PbO2 1-Methyl indol I=0.5-500mA 99% 0.05 Na2SO4 Polymerization [113] PbO2 2-Methyl-indol I=0.5-500mA 58% 0.05 Na2SO4 Polymerization [113]

PbO2 3-Methyl-indol I=0.5-500mA 27% 90% (60mA) and 0.05 Na2SO4 Polymerization [113] 99% (100mA) PbO2 Tryptophan I=0.5-500mA 37% 0.05 Na2SO4 Polymerization [113]

Chapter 2. Bibliography Anode Pollutant Current Efficiency Removal Comments Metabolites Ref. density, (ICE, EOI, efficiency intensity ACE, %) or potential PbO2 glucose 100-900A/m2 EOI=30-60% 100% 1M H2SO4, T=25- 9 derivates of [116] 57°C, Specific glucose interaction between Pb (IV) sites and carboxylic groups PbO2 Chloranilic acid 6.3-50mA/cm2 GCE=0.7-0.2 90% H2SO4, T=25-60°C Hydroquinone, [122] ketomalonic and oxalic acid PbO2 p-Benzoquinone 10mA/cm2 CE(%)=0.6 [p-BQ]=5x10-4 M muconic acid, maleic [124] maleic acid assuming one- and oxalic acids electron reaction PbO2 Urine-waste E=1.8 V vs. 95% H2SO4, T=80°C NO2, NO, NO-2 3, [130] biomass mixtures NHE, around NH4+ (0.2-0.4 mA/cm2)

Chapter 2. Bibliography Anode Pollutant Current Efficiency Removal Comments Metabolites Ref. density, (ICE, EOI, efficiency intensity ACE, %) or potential Pt or Ti/Pt Phenol 300A/m2 30%, TOC pH=12, Stop oxidation with [131] [phenol]=1000mg/ maleic and oxalic L, Na2SO4, acids competition with oxygen evolution reaction Pt/WOx Formic and oxalic 10mA/cm2 100%, CE aprox. 95% [HCOOH]=8x10- CO2 [133] acids 4M, [(COOH)2]=1.0x10- 3M, 0.5M H2SO4

Chapter 2. Bibliography Anode Pollutant Current Efficiency Removal Comments Metabolites Ref. density, (ICE, EOI, efficiency intensity ACE, %) or potential Ti/Pt/Ir

Chapter 2. Bibliography 50 mg/l at Si/BDD, and to about 300, 650 and 950 mg/l at Ti/SnO2, Ta/PbO2 and Pt, respectively. Another aspect to be taken into consideration is the production of powerful oxidants, like the peroxodisulphate [179, 180]; these species can participate in the As can be observed in Table 1, other anode materials have been employed for the direct oxidation of pollutants: pure metals [127, 144], DSA® electrodes, carbonaceous materials (e.g.: glassy carbon, carbon felt, granular carbon, graphite) [95, 103, 127, 138, Also the column of intermediates is a key-point for the evaluation of data in Table 1: in fact, several metabolites are generally produced during the oxidation of the original organic substrate. Starting from an aromatic compound, hydroxylated derivatives are found as initial intermediates but, in the final stages of the oxidation process, several carboxylic acids are produced, the last being usually oxalic acid. The formation of these acids increases the time process and highlights possible mass transport limitations; interestingly, some anode materials are more efficient than others for their elimination.

Chapter 2. Bibliography application of this technique [96, 184]. As another drawback, if the chloride content in the raw wastewater is low, a large amount of salt must be added to increase the process efficiency [184]. Pollutants can also be degraded by the electrochemically generated hydrogen peroxide [185-190]. The electrically generated ozone is also reported for wastewater treatment [191, 192]. Farmer et al. [193] proposed another kind of mediated electrooxidation for the treatment of mixed and hazardous wastes; in this case, metal ions, usually called mediators, are oxidized at an anode from a stable, low-valence state to a reactive, high-valence state, which can directly attack organic pollutants. The reaction may also produce free hydroxyl radicals, which are useful for the destruction of the organic pollutants. Subsequently, the mediators are regenerated at the anode, thus forming a closed cycle. Typical mediators are Ag2+, Co3+, Fe3+, Ce4+ and Ni2+ [193-198]. The mediated electrooxidation usually needs to operate in highly acidic media; unfortunately, the resultant pollution from the added heavy metals limits its application. Table 2 summarizes different examples in which a mediated electrochemical oxidation has been adopted for the elimination of organic pollutants.

Chapter 2. Bibliography Anode Pollutant Current Efficiency Efficiency Removal Removal Mediator Notes Metabolites Ref. density, (ICE, EOI, (ICE, EOI, efficiency efficiency intensity %) %) (mediator (mediator or mediator mediator absence) presence) potential absence presence BDD Chloro 30mA/cm2 95% (Before 100%, Hypochlori Na2SO4 Oxalic, [214] phenols 15 Ah/L) COD te or 5000mg/L, maleic and peroxo- T=30- fumaric disulphate 60°C acids, hydro- quinone, benzo- quinone, trichloro- acetic acid DSA landfill 7.5 A/dm2 29%, Chlorine/ [200] leachate COD; hypochlori 36.1% te Ammoniu m.

Graphit landfill 7.5 A/dm2 21%, Chlorine/ [200] e leachate COD; hypochlori 11% te Ammoniu m.

Chapter 2. Bibliography Anode Pollutant Current Efficiency Efficiency Removal Removal Mediator Notes Metabolites Ref. density, (ICE, EOI, (ICE, EOI, efficiency efficiency intensity %) %) (mediator (mediator or mediator mediator absence) presence) potential absence presence Stainless steel Nitrite 2.00 AM 30% Hypochlorite pH [205] dependence Ti/ Sn-Pd- Ru-oxide landfill 7.5 A/dm2 30.3%, COD; 67.6%, COD; Chlorine/ at 15A/dm2 chloramines [200] leachate 37.6% 92%, hypochlorite and 7500mg/l Ammonium. Ammonium NaCl Ti//Pt/Bi- PbO2 phenol 100mA/cm2 24% (after 24 26% (after 79%, COD 98%, COD Active pH = 12 CHCl3 [211] Ah/L) 26Ah/L) (After 24 Ah/L)(After 24 Ah/L)Chlorine [NaCl]=0.1M, [NaCl]=0.1M, 30% (after 99%, COD 20Ah/L) (After 20 Ah/L) [NaCl]=0.5M [NaCl]=0.5M Ti/IrO2 phenol 0.1 A/cm2 EOI=0.65 EOI=0.65 25%, [phenol] 98%, [phenol] Electro- Selective organochlorinate [199] ; generated oxidation, d compounds ClO- Na2SO4 + and after to NaCl volatile chlorinated compounds Ti/PbO2 landfill 7.5 A/dm2 27.4%, COD; Chlorine/ [200] leachate 33.1% hypochlorite Ammonium.

Chapter 2. Bibliography

Electrode Materials In every new technology, materials play an important role. In this section, some significant parameters, relating to the above subject and pertaining to the application of direct and indirect oxidations, will be briefly commented.

Overpotential for the oxygen evolution [205] For an anode material, the catalytic activity toward organic oxidation depends on the value of oxygen evolution overpotential. Table 3 gives a comparison of most extensively investigated anode materials [205].

Anode Value Conditions Pt 1.3 0.5M H2SO4 Pt 1.6 0.5M H2SO4 IrO2 1.6 0.5M H2SO4 Graphite 1.7 0.5M H2SO4 PbO2 1.9 1M HClO4 SnO2 1.9 0.5M H2SO4 Pb?Sn (93:7) 2.5 0.5M H2SO4 Ebonex® (titanium oxides) 2.2 1M H2SO4 Si/BDD 2.3 0.5M H2SO4 Ti/BDD 2.7 0.5M H2SO4 DiaChem 2.8 0.5M H2SO4

Chapter 2. Bibliography

density or in the presence of a high concentration of chlorides or metallic mediators. When the current density is high, significant decrease of the current efficiency is expected, due to the production of oxygen.

Oxidants Formation potential H2O/·OH (hydroxyl radical) 2.8 O2/O3 (ozone) 2.07 SO42?/S2O82? (peroxodisulfate) 2.01 MnO2/MnO42? (permanganate ion) 1.77 H2O/H2O2 (hydrogen peroxide) 1.77 Cl?/ClO2? (chlorine dioxide) 1.57 Ag+/Ag2+ (silver(II) ion) 1.5 Cl?/Cl2 (chlorine) 1.36 Cr3+/Cr2O72? (dichromate) 1.23 H2O/O2 (oxygen) 1.23 Table 4. Formation potential for some chemical reactants [218].

The boron-doped diamond (BDD) films deposited on titanium substrate [155] or other valve metals, as in DiaChem electrodes [218], give the highest value of oxygen evolution overpotential (Table 3). As a result, the anodic oxidation can take place, at this electrode surface, at significantly high current density with minimal contributions from the oxygen evolution side reaction. An effective process is thus expected, and indeed this material is the most active anode for the oxidation of various pollutants; more details will be given in the following sections.

Chapter 2. Bibliography

homogeneous or heterogeneous chemical reactions occur at the electrode surface, in the region associated to surface phenomena. The electronic transfer at the electrode surface follows these reactions. Finally, the mass transfer from the electrode surface to the bulk The electron transfer reaction is influenced by the nature and the structure of the reacting species, the potential, the solvent, the electrode material and the adsorbed layers on the electrode. In order to understand these influences (interactions between reactant and electrode surface), microscopic theories have been developed, on the basis of two main concepts, which are known as inner-sphere and outer-sphere electron transfer reactions.

Chapter 2. Bibliography

Outer-sphere electron transfer reaction The term outer-sphere is used to describe a reaction, in which the activated complex maintains the coordination sphere originally presented in the reactant species (Figure 2A).

Chapter 2. Bibliography

Inner-sphere electron transfer reaction A reaction is described in terms of inner-sphere when the reactants share a ligand in the activated complex (Figure 2B). Therefore, both the reactant and the products species, as well as the activated complex, are involved in very strong interactions with the electrode surface (specific adsorption). This kind of reaction implies multistep electron transfer reactions [219]. The benzoquinone/hydroquinone redox equilibrium is a typical inner- sphere reaction that involves complex electron and proton transfer mechanisms.

Chapter 2. Bibliography

Frequently, the electrochemical oxidation of some organics in aqueous media takes place, without any loss in electrode activity, only at high potentials and with concomitant evolution of oxygen [106, 129, 199, 221]. Furthermore, it has been found that the nature of the electrode material influences strongly both the selectivity and the efficiency of the process [94, 221]. To interpret these observations, a comprehensive model for the anodic oxidation of organics in acidic medium, including the competition with the oxygen evolution reaction, has been proposed by Comninellis [94]. The model (Figure 3) permits to illustrate the differences between two limiting cases, i.e. the so-called ?active? and ?non-active? anodes.

Figure 3. Scheme of the electrochemical oxidation of organic compounds on active (reactions a, b, c, d) and non active anodes (reactions a, e, f). M designates an active site at the anode.

Chapter 2. Bibliography

In both cases, the first reaction (Equation `a' in the figure) is the oxidation of water molecules leading to the formation of adsorbed hydroxyl radicals: + ? ( ? ) + + + ? (a) M H2O M HO H e Both the electrochemical and chemical reactivity of adsorbed hydroxyl radicals depends At active electrodes, there is a strong interaction between the electrode (M) and the hydroxyl radical (OH?). In this case, adsorbed hydroxyl radicals may interact with the anode, forming a so-called higher oxide MO (Equation `b'). This may be the case when higher oxidation states are available, for the electrode material, above the thermo- dynamic potential for the oxygen evolution (1.23 V vs. SHE).

( ? ) ? + + + ? (b) M HO MO H e We can speculate that, at active electrodes, the redox couple MO/M acts as a mediator in the oxidation of organics (Equation `c'). This reaction is in competition with the side reaction of oxygen evolution, which is due to the chemical decomposition of the higher oxide (Equation `d'): MO + R ? M + RO (c) 1 MO ? M + O2 (d) 2 The oxidative reaction via the surface redox couple MO/M (Equation `c') may be much more selective than the reaction involving hydroxyl radicals (Equation `e'). A typical example of an active electrode is the case of IrO2.

Chapter 2. Bibliography

Chapter 2. Bibliography

This model assumes that the electrochemical oxidation is mediated by hydroxyl radicals, either adsorbed at the surface (in the case of active electrodes) or free, in the case of the non-active ones.

Chapter 2. Bibliography

References [1] S. E. Manahan, Environmental Chemistry, Boca Raton: Lewis Publishers, USA; [4] V. E. Cenkin, A. N. Belevtsev, Effluent and Water Treatment Journal, (1985) 243. [9] W.H. Glaze, J.W. Kang, E.M. Aieta, Ozone?Hydrogen peroxide systems for control of organics in municipal water supplies, in: Proc. 2nd Int. Conf. on the Role of Ozone in Water and Wastewater Treatment. Tek Tran Intern., Ltd., [10] W.H. Glaze, J.W. Kang, D.H. Chapin, Ozone Sci. Eng. 9 (1987) 335. [11] E.M. Aieta, K.M. Regan, J.S. Lang, L. McReynolds, J.W. Kang, W.H. Glaze, J. [12] Farhataziz, A.B Ross, Natl. Stand. Ref. Data Ser., (USA Natl. Bur. Stand.), 1977, [14] V.S. Mishra, V.V. Mahajani, J.B. Joshi, Ind. Eng Chem. Res. 34 (1995) 2?48. [17] J.J. Pignatello, Environ. Sci. Technol. 26 (1992) 944.

Chapter 2. Bibliography

[18] J. Kiwi, C. Pulgarin, P. Peringer, M. Gratzel, Appl. Catal. B: Environ. 3 (1993) 85. [20] A. Safarzadehet-Amiri, J.R. Bolton, S.R. Cater, Solar Energy 56 (5) (1996) 439. [21] C.G. Hatchard, C.H. Parker, Proc. R. Soc. London, A235 (1956) 518?536. [25] J. Hoigné, The Handbook of Environmental Chemistry, vol. 5, part C, Quality and Treatment of Drinking Water, Part II, Springer, Berlin Heidelberg, 1998. [27] R. Andreozzi, V. Caprio, A. Insola, M.G. D'Amore, Water Res. 26 (7) (1992) 917. [28] R. Andreozzi, V. Caprio, M.G. D'Amore, A. Insola, Environ. Technol. 16 (1995) [30] K. Juttner, U. Galla, H. Schmieder, Electrochim. Acta, 45 (2000) 2575. [31] J. O'M. Bockris (Ed.), Electrochemistry of cleaner environments, Plenum, New [32] D. Pletcher, F. Walsh (Eds.), Industrial Electrochemistry, Chapman and Hall, [33] D. Pletcher, N.L. Weinberg, Chem. Eng. (London) (1992) Aug. 98-103, Nov. 132- [34] K. Rajeshwar, J.G. Ibáñez, G.M. Swain, J. Appl. Electrochem. 24 (1994) 1077. [35] K. Rajeshwar, J.G. Ibáñez, Fundamentals and Application in Pollution Abatement, Academic Press, San Diego, CA, 1997.

Chapter 2. Bibliography

[36] C.A.C. Sequeira (Ed.), Environmentally oriented electrochemistry, Elsevier, [37] P. Tatapudi, J.M. Fenton, in: H. Gerischer, C.W. Tobias (Eds.), Advances in electrochemical sciences and engineering, Vol. 4, VCH-Veragsgesellschaft, [40] G. Dubpernel, Selected topics in the history of electrochemistry. The [41] B. Fleet, Evolution of electrochemical reactor system for metal recovery and pollution control, in: J.T. Stock, M.V. Orna (Eds.), Electrochemistry, Past and [42] J.J. Leddy, Industrial Electrochemistry, in: J.T. Stock, M.V. Orna (Eds.), Electrochemistry, Past and Present, American Chemical Society, Washington, DC, [43] L.A. Kul'skii, P.P. Strokach, V.A. Slipchenko, E.I. Saigak, Water Purification by [44] V.A. Matveevich, Elektronnaya Obrabotka Materialov, 5 (2000) 103. [48] L.J. Gao, Y.F. Cheng, Environ. Pollut. Control. 14 (5) (1992) 10. [51] R.R. Renk, Ener. Prog. 8 (1998) 205.

Chapter 2. Bibliography

[55] G.V. Sleptsov, A.I. Gladikii, E.Y. Sokol, S.P. Novikova, Elektronnaya Obrabotka [56] U.B. Ogutveren, S. Koparal, J. Environ. Sci. Health A. 32 (9-10) (1997) 2507. [57] T. Ya. Pazenko, T.I. Khalturina, A.F. Kolova, I.S. Rubailo, J. Appl. USSR. 58 [58] J. Szynkarczuk, J. Kan, T.A.T. Hassan, J.C. Donini, Clay Miner. 42 (1994) 667. [59] N.S. Abuzaid, Z. Al-Hamouz, A.A. Bukhari, M.H. Essa, Water Air Soil Pollut. [62] G.B. Raju, Khangaonkar, Trans. Indian Inst. Met. 37 (1) (1984) 59. [63] V.E. Nenno, V.I. Zelentsov, E.V. Mel'nichuk, A.M. Romanov, T. Ya. Datsko, [67] N.T. Manjunath, I. Mehrotra, R.P. Mathur, Water Res. 34 (2000) 1930. [68] R.L. Vaughan, B.E. Reed, G.W. Roark, D.A. Masciola, Environ. Eng. Sci. 17 [71] V.I. Il'in, O.N. Sedashova, Chem. Petrol. Eng. 35 (7-8) (1999) 480.

Chapter 2. Bibliography

[72] M.Y. Ibrahim, S.R. Mostafa, M.F.M. Fahmy, A.I. Hafez, Sep. Sci. Technol. 36 [74] V.I. Il'in, Khimicheskoe i Neftyanoe Mashinostroenie. 5 (2002) 41. [75] M.F. Prokop'eva, V.N. Tkacheva, E.Yu. Kirshina, Khimiya i Tekhnologia Vody, [76] I.V. Aleksandrov, O.I. Rodyushkin, K.S. Ibraev, Koks I Khimiya 7 (1992) 41. [77] L. Alexandrova, T. Nedialkova, I. Nishkov, Int. J. Miner. Process. 41 (1994) 285. [80] O.R. Shendrik, E.E. Andreeva, M.I. Ponomareva, I.B. Ivanenko, Khimiya i [81] T.D. Kubritskaya, I.V. Drako, V.N. Sorokina, R.V. Drondina, Surf. Eng. Appl. [82] M.N. Rabilizirov, A.M. Gol'man, Khimiya i Tekhnologia Vody, 8 (4) (1986) 87. [83] V. Il'in, V.A. Kolesnikov, Yu. I. Parshina, Glass Ceram. 59 (7-8) (2002) 242. [84] I.A. Zolotukhin, V.A. Vasev, A.L. lukin, Khimiya i Tekhnologia Vody, 5 (3) [86] V.A. Kolesnikov, V.I. Il'in, S.O. Varaksin, V.T. Shaturov, Russ. J. Heavy Mach. 1 [87] V. Srinivasan, M. Subbaiyan, Sep. Sci. Technol. 24 (1-2) (1989) 145. [88] G. Ramadorai, J.P. Hanten, Minerals and Minerallurgical Processing (1986) 149. [89] V.I. Zelentsov, K.A. Kiselev, Elektronnaya Obrabotka Materialov, 4 (1986) 50.

Chapter 2. Bibliography

[90] V.E. Nenno, V.I. Zelentsov, T.Ya. Datsko, E.E. Dvornikova, T.M. Radzilevich, [91] C. Llerena, J.C.K. Ho, D.L. Piron, Chem. Eng. Commun. 155 (1996) 217. [92] C. Camilleri, Industrie Mineral, Les Techniques 67 (1) (1985) 25. [96] J. Naumczyk, L. Szpyrkowicz, F.Z. Grandi, Water Sci. Technol. 62 (1995) 111. [97] O.J. Murphy, G.D. Hitchens, L. Kaba, C. E. Verostko, Water Res. 26 (1992) 443. [98] L. Szpyrkowicz, J. Naumczyk, F.Z. Grandi, Toxicol. Environ. Chem., 44 (1994) [100] N.N. Rao, K.M. Somasekhar, S.N. Kaul, L. Szpyrkowicz, J. Chem. Technol. [101] J.L. Boudenne, O. Cerclier, J. Galea, E.V. Vlist, Appl. Catal. A: Gen. 143 (1996) [104] J. Manriquez, J.L. Bravo, S. Gutierrez-Granados, S.S. Succar, C. Bied-Charreton, [107] D. Gandini, P.-A. Michaud, I. Duo, E. Mahé, W. Haenni, A. Perret, Ch. [108] Ch. Comninellis, C. Pulgarin, J. Appl. Electrochem., 21 (1991) 703.

Chapter 2. Bibliography

[110] E. Brillas, B. Boye, I. Sires, J. A. Garrido, R.M. Rodriguez, C. Arias, P-L. Cabot, [111] P. Cañizares, J. Garcia-Gomez, J. Lobato, M.A. Rodrigo, Ind. Eng. Chem. Res., 42 [112] P. Cañizares, M. Diaz, J.A. Dominguez, J. Garcia-Gomez, M.A. Rodrigo, Ind. [115] J. Iniesta, P.-A. Michaud, M. Panizza, G. Cerisola, A. Aldaz, Ch. Comninellis, [116] F. Bonfatti, S. Ferro, F. Lavezzo, M. Malacarne, G. Lodi, A. De Battisti, J. [117] D. Gandini, E. Mahè, P.-A. Michaud, W Haenni, A. Perret, Ch. Comninellis, J. [118] Ch. Comninellis, C. Pulgarin, J. Appl. Electrochem., 23 (1993) 108. [119] R. Bellagamba, P.-A. Michaud, Ch. Comninellis, N. Vatistas, Electrochem. [120] M. Panizza, P.-A. Michaud, G. Cerisola, Ch. Comninellis, J. Electroanal. Chem. [121] A.M. Polcaro, A. Vacca, S. Palmas, M. Mascia, J. Appl. Electrochem., 33 (2003) [122] C.A. Martinez-Huitle, M.A. Quiroz, Ch. Comninellis, S. Ferro, A. De Battisti, Electrochim. Acta, 50 (2004) 949.

Chapter 2. Bibliography

[123] S.K. Johnson, L.L. Houk, J. Feng, R.S. Houk, D.C. Johnson, Environ. Sci. [125] A. Morão, A. Lopes, M.T. Pessoa de Amorim, I. C. Gonçalves, Electrochim. Acta, [126] C. Seignez, C. Pulgarin, P. Péringer, Ch. Comninellis, Swiss Chem., 14 (1992) 25. [127] L.L. Houk, S.K. Johnson, J. Feng, R.S. Houk, D.C. Johnson, J. Appl. Electrochem. [128] E. Brillas, M.A. Baños , J. A. Garrido, Electrochim. Acta, 48 (2003) 1697. [130] L. Kaba, G.D. Hitchens, J. O'M Bockris, J. Electrochem. Soc., 137 (1990) 1341. [131] R. Kötz, S. Stucki, B. Carcer, J. Appl. Electrochem. 21 (1991) 14. [132] S. Stucki, R. Kötz, B. Carcer, W. Suter, J. Appl. Electrochem. 21 (1991) 99. [134] C. Bock, A. Smith, B. MacDougall, Electrochim. Acta, 48 (2002) 57. [135] J.D. Rodgers, W. Jedral, N.J. Bunce, Environ. Sci. Technol. 33 (1999) 1453?1457. [136] A.M. Polcaro, S. Palmas, F. Renoldi, M. Mascia, J. Appl. Electrochem. 29 (1999) [137] G. Chen, E.A. Betterton, R.G. Arnold, J. Appl. Electrochem. 29 (1999) 961. [138] K.-W. Kim, M. Kuppuswamy, R.F. Savinell, J. Appl. Electrochem. 30 (2000) 543. [139] S. Tanaka, Y. Nakata, T. Rimura, Yustiawati, M. Kawasaki. H. Kuramitz, J. Appl. [140] R.H. de Lima Leite, P. Cognet, A.-M. Wilhelm, H. Delmas, J. Appl. Electrochem. [141] L. Szpyrkowicz, J. Naumczyk, F. Zillio-Grandi, Wat. Res. 29 (1995) 517.

Chapter 2. Bibliography

[142] Carmem L.P.S. Zanta, P.-A. Michaud, Ch. Comninellis, A.R. De Andrade, J.F.C. [143] N.E. Jiménez Jado, C. Fernández Sánchez, J.R. Ochoa Gómez, J. Appl. [144] M. Grattrell, B. MacDougall, J. Electrochem. Soc., 146 (1999) 3335. [146] L. Szpyrkowicz, C. Juzzolino, S.N. Kaul, S. Daniele, M.D. De Faveri, Ind. Eng. [147] Y. Xiong, P. J. Strunk, H. Xia, X. Zhu, H. T. Karlsson, Wat. Res. 35 (2001) 4226. [148] L. Codognoto, S.A.S. Machado, L.A. Avaca, J. Appl. Electrochem., 33 (2003) [154] C. Pulgarin, N. Alder, P. Péringer, Ch. Comninellis, Water Res., 28 (1994) 887. [156] W. Kirk, H. Sharifian, F.R. Foulkes, J. Appl. Electrochem., 15 (1985) 285. [157] A.M. Polcaro, S. Palmas, F. Renoldi, M. Mascia, J. Appl. Electrochem. 29 (1999) [158] F. Montilla, P.-A. Michaud, J.L. Vazquez, Ch. Comninellis, Electrochim. Acta., [159] M. R.V. Lanza, R. Bertazzoli, Ind. Eng. Chem. Res., 41 (2002) 22. [160] D. Rajkumar, K. Palanivelu, Ind. Eng. Chem. Res., 42 (2003) 1833.

Chapter 2. Bibliography

[161] P. Cañizares, C. Sáez, J. Lobato, M.A. Rodrigo, Ind. Eng. Chem. Res., 43 (2004) [162] A. Buso, L. Balbo, M. Giomo, G. Farnia, S. Sandonà, Ind. Eng. Chem. Res., 39 [164] J.J. Carey, J.C.S. Christ, S.N. Lowery, US Patent 5,399,247 (1995). [165] M. Fryda, D. Hermann, L. Schafer, C.P. Klages A. Perret, W. Haenni, Ch. Comninellis, D. Gandini, New Diamond Front. C. Technol. 9 (1999) 229?240. [166] A. Perret, W. Haenni, N. Skinner, T.M. Tang, D. Gandini, Ch. Comninellis, B. [167] Y.M. Awad, N.S. Abuzaid, J. Environ. Sci. Health A 32 (1997) 1393. [168] Y.M. Awad, N.S Abuzaid, Separation and Purification Technology (2000), 18 [169] N. Kannan, S.N. Sivadurai, J.L. Berchmans, R.Vijayavalli, J. Environ. Sci. Health, [171] R. Cossu, A.M. Polcaro, M.C. Lavagnolo, M. Mascia, S. Palmas, F. Renoldi, [172] J. Grimm, D. Bessarabov, W. Maier, S. Storck, R.D. Sanderson, Desalination 115 [173] X. Chen, G. Chen, F. Gao, P. L. Yue, Environ. Sci. Technol. 37 (2003) 5021. [174] C.A. Vincent, D.G.C. Weston, J. Electrochem. Soc. 119 (1979) 518. [176] J.M. Kelsselman, O. Weres, N.S. Lewis, M.R. Hoffmann, J. Phys. Chem., B 101, (1997) 2637.

Chapter 2. Bibliography

[179] P.-A. Michaud, E. Mahe, W. Haenni, A. Perret, Ch. Comninellis, Electrochem. [180] P.-A. Michaud, Ch. Comninellis, W. Haenni, A. Perret, M. Fryda, International [181] J. Naumczyk, L. Szpyrkowicz, M.D.D. Faveri, F. Zilio-Grandi, Trans. IChemE B [182] A.G. Vlyssides, C.J. Israilides, Environ. Pollut. 97 (1?2) (1997) 147. [183] A.G. Vlyssides, C.J. Israilides, M. Loizidou, G. Karvouni, V. Mourafeti, Water [185] T. Matsue, M. Fujihira, T. Osal, J. Electrochem. Soc. 128 (1981) 2565. [186] E. Brillias, R.M. Bastida, E. Llosa, J. Electrochem. Soc. 142 (1995) 1733. [187] E. Brillias, E. Mur, J. Casado, J. Electrochem. Soc. 143 (1996) L49. [188] E. Brillias, R. Sauleda, J. Casado, J. Electrochem. Soc. 144 (1997) 2374. [189] E. Brillias, R. Sauleda, J. Casado, J. Electrochem. Soc. 145 (1998) 759. [190] E. Brillias, E. Mur, R. Sauleda, L. Sanchez, F. Peral, X. Domenech, J. Casado, [191] S. Stucki, H. Baumann, H.J. Christen, R. Kotz, J. Appl. Electrochem. 17 (4) [192] W. El-Shal, H. Khordagui, O. El-Sebaie, F. El-Sharkawi, G.H. Sedahmed, [193] J.C. Farmer, F.T. Wang, R.A. Hawley-Fedder, P.R. Lewis, L.J. Summers, L. Foiles, J. Electrochem. Soc. 139 (1992) 654.

Chapter 2. Bibliography

[195] R.G. Hickman, J.C. Farmer, F.T. Wang, American Chemical Society, 1993, pp. [196] F. Bringmann, K. Ebert, U. Galla, H. Schmieder, J. Appl. Electrochem. 25 (1995) [197] V. Cocheci, C. Radovan, G.A. Ciorba, I. Vlaiciu, Revue Roumaine de Chimie 40 [198] A. Paire, D. Espinoux, M. Masson, M. Lecomte, Radiochim. Acta 78 (1997) 137. [201] T. Inokuchi, S. Matsumoto, S. Torii, J. Org. Chem., 56 (1991) 2416. [202] J. Bringmann, K. Ebert, U. Galla, H. Schmieder, J. Appl. Electrochem., 27 (1997) [203] G.A. Bogdanovskii, T.V. Savel'eva, T.S. Saburova, Russian J. Electrochem., 37 [204] A.S. Vaze, S.B. Sawant, V.G. Pangarkar, J. Appl. Electrochem., 29 (1999) 7. [205] N.S Abuzaid, Z. Al-Hamouz, A.A. Bukhari, M.H. Essa, Water, Air, and Soil [207] U. Galla, P. Kritzer, J. Bringmann, H. Schmieder, Chem Eng. Technol., 23 (2000) [208] F. Bonfatti, A. De Battisti, S. Ferro, G. Lodi, S. Osti, Electrochim. Acta, 46 (2000) [209] Y.H. Chung, S.-M. Park, J. Appl. Electrochem., 30 (2000) 685.

Chapter 2. Bibliography

[210] F. Bonfatti, S. Ferro, F. Lavezzo, M. Malacarne, G. Lodi, A. De Battisti, J. [211] J. Iniesta, J. Gonzalez-Garcia, E. Exposito, V. Montiel, A. Aldaz, Wat. Res. 35 [213] A.G. Vlyssides, P.K. Karlis, G. Mahnken, J. Appl. Electrochem. 33 (2003) 155. [214] P. Cañizares, J. Garcia-Gomez, C. Saez, M.A. Rodrigo, J. Appl. Electrochem. 34 [218] I. Tröster, M. Fryda, D. Herrmann, L. Schafer, W. Hanni, A. Perret, M. Blaschke, [224] A. Dabrowski, J. Mieluch, A. Sadkaoski, J. Wild, P. Zoltowski, Prezm. Chem., 54 [225] V. Smith de Sucre, A. P. Watkinson, Can. J. Chem. Eng., 59 (1981) 52. [226] M. Chettiar, A. P. Watkinson, Can. J. Chem. Eng., 61 (1983) 568.

Chapter 3. Global Parameters

Global parameters Total organic carbon (TOC) The total organic carbon (TOC) is amount of organic carbon in the sample, expressed in concentration of carbon (mol C/L or mg/L). The persulphate combined with irradiation UV (is the technique of analyzer TOC Dohrmann DC-80). The sample reacts with ion, associated at the same time to irradiation UV. The oxidation rate is remarkably increased respect to the method with single UV or only persulphate, due to the simultaneous ionization of dissolved organics; production of free sulphate radicals and hydroxyls highly reactive. After the CO2 produced is transported with a air or nitrogen, oxygen flow until the detector.

Chemical Oxygen Demand (COD) The normality of the dichromate is that specified by these authors. For samples in the range of 20-900 mg/L (Standard Range), the COD is determined using a spectrophotometer at 600 nm by measuring the concentration of the produced Cr (III) ion. For samples in the range of 5-150 mg/L, the COD can be determined using a spectrophotometer at 440 nm by measuring the decrease in concentration of the Cr (VI) ion. The path length of the tubes is 1.40 cm.

Chapter 3. Global Parameters

Instantaneous (ICE), Total (TCE), General Current Efficiency (GCE) and Electrochemical Oxidation Index (EOI) The instantaneous current efficiency for the anodic oxidation has been calculated, using the following relation [2]:

? [(COD) ? (COD) +? ] ? ICE = FV ? ? (1) t tt

? 8I ?t ? where (COD)t and (COD)t+?t are the chemical oxygen demands (g dm-3) at times t = 0 (initial) and t, respectively; I is the current (A), F the Faraday constant (96487 C mol-1), V the volume of electrolyte (dm3) and 8 is the oxygen equivalent mass (g eq- 1).

The general current efficiency for the anodic oxidation has been calculated from values of COD, using the following relation:

? [(COD) ? (COD) +? ] ? GCE = FV ? ? (2) 0 tt

? 8I ?t ? where (COD)0 and (COD)t+?t are the chemical oxygen demands (g dm-3) and the other variables are described above. This equation is similar to that proposed in [2], for the determination of instantaneous current efficiency (ICE), although the expression used for GCE represents an average value between the initial time t and t+?t.

Chapter 3. Global Parameters

from the ICE - time curves the EOI value has been calculated. The EOI for a determined organic compound supplies information about the reactivity of the considered species in the comparison of the oxidative degradation. The appraisal of this parameter, during a period of sufficiently short time, and in particular in the beginning of the electrolysis, can be considered specific for the substrate.

Electrodes Ti/PbO2, Ti/Pt and Ti/IrO2-Ta2O5 anode materials were supplied by De Nora S.p.A. (Milano, Italy); Pt and Au were obtained as pure metals. HBDD electrodes were kindly supplied by CSEM (Neuchatel, Switzerland). BDD films were grown on conductive p-Si substrate (0.1 ?·cm, Siltronix) via a hot filament, chemical vapor deposition technique (HF-CVD) [3, 4]. This procedure gave a columnar, randomly textured, polycrystalline diamond film, with a thickness of about 1µm and a resistivity of 15 m?·cm (±30%) onto the conductive p-Si substrate.

References [4] A. Perret, N. Skinner, Ch. Comninellis, D. Gandini, Electrochem. Soc. Proc., 32 (1997) 275.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Electrochemical Incineration of Oxalic Acid Introduction The reactivity of different organic substrates toward electrochemical oxidative degradation and destruction has been widely investigated in recent years. Oxidative electrochemical processes, promising versatility, environmental compatibility and cost effectiveness, have a continuously growing importance in degradation of organics pollutants [1]. In these processes, the aim is the complete oxidation of organics to CO2 [2,3] or the conversion of the toxic organics to biocompatible compounds [4].

As mentioned in a previous chapter, several anodic materials have been tested for the electrochemical oxidation of a number of model compounds; for example, the anodic oxidation of phenol has been investigated at Ti/IrO2 and Ti/SnO2-Sb2O5 anodes [5]. Possible mechanisms of the electrochemical incineration are essentially based on oxygen-transfer stages, in which electrosorbed hydroxyl radicals should play a decisive role [6]. Their reactivity, in turn, is largely dependent on the nature of the electrode material, high-oxygen-overvoltage electrodes exhibiting a much better performance, compared with the classical anode materials of industrial electrochemistry, like DSA®'s (dimensionally stable anodes), which are characterized by low oxygen overvoltage.

In this context, electrochemical incineration at Pt [7,8,9], PbO2- [7,10] and SnO2- based electrodes [5,10] and, more recently, conductive diamond [1,11-14], have been thoroughly studied, aiming at optimization of process efficiency and costs for the electrochemical oxidation of organic pollutants in wastewaters. As far as the nature of the organic substrate is concerned, it has been shown that carboxylic acids, common intermediates of the oxidative degradation of several organic substrates, are

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

rather stable and are mineralized at longer times [1]. This seems to hold, in particular, for oxalic acid (OA), last intermediate in wastewater treatment processes.

Oxalic acid is a metabolite of the catalytic oxidation of phenol (a water pollutant present in many industrial wastewaters) and of other aromatic substrates like, for instance, coumaric acid (a by-product of olive oil manufacturing) [15]. As previously described, the oxidation of OA has been studied at different electrode materials and the influence of a number of process variables, such as concentration, pH, electrode potential and solvent, has been described [15-17]. Most of the investigations have been carried out at the Pt electrode [8,9,15-24] but interest was addressed also to other metals, like Pd [25,26] and Au [26,27], as well as to TiO2 [28], WOX [29] and RuO2 [28]. The wealth of experimental evidence has allowed to formulation of different oxidation mechanisms, although no complete agreement has been attained on the nature of adsorbed intermediates and details of the reaction mechanism are not completely elucidated so far.

In the present part of the Chapter, the anodic incineration of oxalic acid has been studied, with the scope of verifying optimal process yields in terms of electrode material, current density and solution temperature. The research has been performed at Ti/IrO2-Ta2O5, Ti/PbO2, highly boron-doped diamond (BDD), Au and Pt electrodes; with the only exception of gold, all of them are available for industrial production.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Experimental Cell and Electrodes Bulk electrolyses have been carried out in a single-body, thermostated pyrex glass cell of 250 ml; the anode was a plate or a cylinder, with a geometrical area of 15 cm2, and the test solution volume was 230 mL. Experiments were performed at 25 °C for studying the role of applied current density; furthermore, the effect of the temperature has been investigated carrying out experiments in the range from 25 to 80 °C. Solutions were vigorously stirred, by means of a magnetic stirrer. The applied current density range was established from 100 to 600 A m-2. The anode was symmetrically positioned between two zirconium plate cathodes in the case of the Ti/PbO2, Ti/IrO2-Ta2O5 and Au electrodes; a cylindrical platinum grid was used as the counter when the cylindrical Pt anode was investigated; finally, only one Zr cathode was used in the case of BDD.

BDD films were grown on conductive p-Si substrate (0.1 ?·cm, Siltronix) via a hot filament, chemical vapor deposition technique (HF-CVD) [30,31].

Techniques and Instrumentation During the electro-oxidation test, the OA concentration was determined by means of a conventional KMnO4 titration method, using a 0.1 N KMnO4 solution. Electrochemical analyses were carried out by an Autolab PGSTAT20 (EcoChemie, The Netherlands). Cyclic voltammetries (CV) were performed in solutions at different concentration of OA, degassing by bubbling nitrogen, and applying scan rates of 50, 100 and 250 mV s-1. The chosen range potential was cycled using a step potential of 2.5 mV and repeating the measurements at least five times, or until reproducible signals were obtained; in every case, the last cycle was recorded.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Results and Discussion Cyclic voltammetry Preliminary experiments have been carried out by cyclic voltammetry, to obtain information on the electroactivity of OA at electrodes, like IrO2-Ta2O5, PbO2 and BDD, prior to anodic oxygen evolution. At Pt and Au electrodes, whose CV features are already well known in the literature, analogous measurements have been made, for the need of internal and external comparison. Voltammograms were obtained at different carboxylic acid concentrations, in acidic media. In all cases, the CV curves were recorded below the decomposition potential of water and/or supporting electrolyte. Figure 1 shows CV data recorded at 50, 100 and 250 mV s-1 at the Ti/IrO2-Ta2O5, in the absence and in presence of OA (100 mM).

Figure 1. CV curves for the Ti/IrO2-Ta2O5 electrode, in the presence of the pure supporting electrolyte (0.5 M H2SO4) and with OA in solution (100 mM); data obtained at different scan rates, and at room temperature.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

The addition of the organic substrate does not seem to have a significant effect on the shape of CV curves, with the exception of a very small increase of currents in the oxygen evolution potential range. CV experiments at BDD (Figure 2) clearly show that OA is electroactive at this electrode, its oxidation-taking place about 0.1 V before the oxygen evolution.

Figure 2. CV curves for the BDD electrode, in the presence of the pure data obtained at 50 mVs-1 scan rate. Inset: magnification of CV curves obtained at different scan rates.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

observed at the Ti/PbO2 electrode, where, as shown in Figure 3, a significant current shift could be recorded when OA was added to the solution, compared with the curves recorded in 0.5 M H2SO4, at the same J value.

Figure 3. CV curves for the Ti/PbO2 electrode, in the presence of the pure supporting electrolyte (0.5 M H2SO4) and with OA in solution (100 mM); data obtained at different scan rates, and at room temperature.

In Figure 4, the behavior of OA at a polycrystalline Au electrode is reported; again, cyclic voltammograms have been recorded at scan rates of 50, 100 and 250 mV s-1.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

prior to the potential region where the electrochemical reduction of the oxide may take place. With this assumption, the effect of the potential scan rate on the voltammogram shape may be explained, considering that the higher the former, the higher the amount of unreacted auric oxide when the sweep reaches the electrochemical oxide reduction range. The formation of a fresh metal surface through a chemical/electrochemical reaction allows an increase of the rate of anodic OA oxidation (positive J values in the cathodic branch of CV, just after the oxide reduction range) down to the potential limit allowed by its electrochemical reactivity.

Figure 4. CV curves for the Au electrode in the presence of the pure data obtained at different scan rates, and at room temperature.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 5. CV curves for the Pt electrode, in the presence of the pure supporting electrolyte (0.5 M H2SO4) and of different amount of OA in solution; data obtained at 50 mVs-1 scan rate, and at room temperature.

Inset: dependence of the voltammetric peak current on the oxalic acid concentration.

In agreement with the literature [15,16,32-34], CV curves at the Pt electrode in the presence of OA show that the organic molecule is adsorbed on the electrode surface, as witnessed by the hydrogen adsorption and desorption voltammetric peaks, which are shifted towards more negative potentials when OA is added to the solution (Figure 5). This behavior is characteristic of a weak adsorption, similar to that generally observed in the case of anions. The OA oxidation peak appears as a single signal between 0.8 and 0.9 V and, in the explored range from 1mM to 100mM, its height depends on OA concentration, as shown in Figure 5.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Galvanostatic oxidation experiments At all the electrode materials taken into consideration in this work, OA anodic oxidation experiments were performed under galvanostatic conditions, allowing a more complete description of the role of different factors, which, like the oxidative modification of the electrode surfaces, can become important and sometimes decisive only in the J and cell potential range applicable in electrochemical wastewater treatments. Tests at different current density were carried out (J = 100, 200, 600 A m-2) and also the influence of temperature on the OA oxidation rate was studied. During each electrolysis, the carboxylic acid concentration was followed by a titration method with KMnO4.

At the Ti/IrO2-Ta2O5 electrode, the OA electro-oxidation takes place quite slowly and its efficiency is higher at the lower current densities (Figure 6). The process is anyway lengthy, the gain in efficiency being overwhelmed by the lower current values applied. This result may be not surprising on the basis of the previously discussed CV overview, which indicated a too weak interaction between OA species and the mixed-oxide surface across the whole potential sweep range. This includes the initial part of the oxygen evolution region, where also no changes are induced by the presence of the organic molecule. Any further increase of the electrode potential and anodic current density, to the levels attained in Fig. 6, necessarily leads to higher hydroxyl radical coverages, both assuming electrochemical or chemical oxide formation paths for o.e.r. at IrO2-based electrodes. This favorable kinetic feature results however in higher o.e. rates, rather than in a more effective OA oxidation.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Likewise, the poor adsorption of OA species can but decrease the rate of the direct oxidation, based on electron transfer between electrode surface and adsorbed substrate(s), which has been hypothesized by several authors at different electrode materials (e.g.: [8, 9, 24]).

Figure 6. Decrease of OA content in solution, at the Ti/IrO2-Ta2O5 electrode, as a function of charge, at different current densities (25 °C).

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

justify the assumption that the hydroxyl radicals are not directly involved in the OA electrochemical incineration.

Experiments have been carried out also at another bulk-oxide electrode material: PbO2, under the form of electrodeposited film onto Ti plate [37]. This is one of the extreme examples of classical high-oxygen-overpotential material and therefore expected to perform quite well in electrochemical mineralization of organics.

Recently, some role of short-range interaction between Pb(IV) sites and carboxyl groups have been assumed to play an important role [7]. In the present study, experiments have been performed at J = 100, 200 and 600 A m-2, attaining the mineralization of 91-98% of the initial OA amount in 8, 5 and 3 hours, respectively (see Figure 7). The faradaic yield of the process is practically 100%, with the exception of the final stages for J = 100 and 200 A m-2. For J = 600 A m-2, the loss in efficiency starts at an earlier stage of the electrolysis and the whole set of data might be understood essentially in terms of a dependence on temperature of the Pt- oxide formation mechanism.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 7. Decrease of OA content in solution, at the Ti/PbO2 electrode, as a function of charge, at different current densities. Data obtained at room temperature (25 °C).

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

compared with the 14 Ah dm-3 requested at Ti/PbO2 electrodes. This experimental evidence supports the idea that the interaction between the organic substrate and the electrode surface is likely to be a sufficient condition for a fast incineration to take place. A weaker interaction, of the type possibly taking place between OA and BDD, is apparently not sufficient to promote a fast oxidation, although favoring in principle a possible hydroxyl radical electrosorption.

Figure 8. Decrease of OA content in solution, at the BDD electrode, as a function of charge, at different current densities. Data obtained at room temperature (25 °C).

OA mineralization rate data at Au surfaces are substantially similar to those obtained at BDD, in spite of its higher catalytic activity toward oxygen evolution. As shown in Figure 9, the similarity remains the same at different J and T values. In fact, CV results indicate a higher reactivity of OA at Au, than at BDD. However, the reactivity of the substrate is clearly conditioned by the state of the electrode surface.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

In the potential range explored by cyclic voltammetry, the electro-oxidation takes place at Au metal surfaces or anyway at surfaces at their first stages of oxidation.

Within the range of current densities applied for electrochemical incineration, electrode potentials are raised well inside the oxygen evolution region and, under these polarization conditions, the Au electrode is covered by auric sesquioxide film, whose thickness increases with polarization time [19]. This change is possibly also responsible for the strong decrease in incineration efficiency observed by increasing J from 200 to 600 A m-2. The considerable improvement in incineration efficiency, observed with increasing the solution temperature, is quite significant and is at variance with the results at BDD electrodes.

Figure 9. Decrease of OA content in solution, at the Au electrode, as a function of charge, at different current densities (25 °C). Inset: effect of the temperature, on electrolyses carried out at 600 Am-2.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

As far as the Pt electrode is concerned (Figure 10), the mineralization of OA is fast, but its rate decreases to almost zero below an OA concentration limit that is higher, the higher the anodic J value. At 100 A m-2, for instance, the concentration limit at which deactivation takes place is quite low, but it becomes already rather important at J = 300 and 600 A m-2. In concomitance with the observed drop of efficiency, an increase of about 0.7 V in the Pt anode potential has been observed, from about 1.28 to 1.96 V (vs. S.C.E.), which is not the case at the lowest J limit of 100 A m-2, where the potential increases rather smoothly from about 1.01 V, to about 1.59 V (vs.

S.C.E.). The jump in the anode potential at the higher J's can be tentatively related with nature and thickness of the Pt oxide formed during the anodic polarization, when the limiting current of OA oxidation is approached. Apparently, neither OA residual concentration, nor oxygen evolution rate alone can determine the potential jump and concomitant drop in faradaic efficiency for OA mineralization. It is rather an optimal combination of the two parameters, and a very critical one, which causes the change.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 10. Decrease of OA content in solution, at the Pt electrode, as a function of charge, at different current densities (25 °C). Inset: effect of the temperature, on electrolyses carried out at 600 Am-2.

In fact, in a widely cited paper by Giner [38], a deactivation of Pt towards OA oxidation has been reported to occur at potentials at which a complete monomolecular layer of oxygen is formed, and attempts of interpretation have been made, based on adsorption of oxygen species [38,39]. The work by Fioshin and collaborators has supplied further support [40]. On the other hand, the complex electrochemistry of Pt oxidation at high anodic potentials has been thoroughly studied by Volodin et al. [41], also in the presence of organic solutes. Different types of adsorbed oxygen, up to the level of more stable species, have been hypothesized. At potentials higher than 2.0 V, a three-dimensional oxide phase is formed, with improved catalytic activity towards the parasite o.e.r., which could

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

somewhat explain the drop in efficiency of the OA oxidation above given polarization limits.

The effect of temperature (inset in Fig. 10) may be understood along the same lines, considering the improved mass transport related with lowering of the solution viscosity, while increasing the solution temperature.

The comparison of results on the electrochemical incineration of OA under galvanostatic conditions, with J values from moderate to high, and the consequent involvement of the parasite o.e.r., is certainly a complex task. The wealth of literature on the subject of OA oxidation under much less drastic polarization conditions suggests a number of possible mechanisms, which could be of help also in the analysis of our results.

For the OA anodic oxidation at Pt electrodes, Bockris and co-workers proposed the following mechanism [18]: H2C2O4 ? HC2O4(ads) + H + e (1) +

HC2O4(ads) ? 2CO2 + H + e (2) + or: H2C2O4 ? HC2O4(ads) + H + e (3) +

HC2O4(ads) ? HCO2(ads) + CO2 (4) HCO2(ads) ? CO2 + H + e (5) +

Accordingly, an important role is played by the interaction between the radicals formed after the first electron transfer step and the electrode surface. J.W. Johnson et al. have extended the same mechanism to the reaction at gold electrodes [19], with step (4) determining the process rate. In these cases, the role played by the interaction between the radicals formed after the first electron transfer step and the electrode surface becomes primary.

On the basis of a series of studies, again at Pt electrodes, Sargysian and Vasil'ev [8] have proposed a mechanism of the type: HC2O4 ? HC2O4(ads) (6) ??

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

HC2O4(ads) ? HCO2(ads) + CO2 + e (7) ? HCO2(ads)? ? CO2 + H+ + e (8) where the anion adsorption step (defined as weak chemisorption by the authors) and the second decarboxilation are fast, compared with the first decarboxilation, which is the rate-determining one.

In a successive paper, the same authors have extended the above mechanism [9] for the OA anodic oxidation at other electrode materials, including Au, Pd, Rh, Ir, a RuO2-TiO2 oxide mixture and glassy carbon. Also in their analysis, the deactivation of Pt electrodes at higher potentials is mentioned, with distinction of more reactive Pt oxide forms. Thorough adsorption studies by Horanyi and co-workers have also proved the occurrence of significant adsorption of oxalic acid at Pt metal surfaces [42].

According to Inzelt and Szetey [21], one more mechanism should be considered at Pt electrodes: H2O ? (OH)ads + H + e (9) +

H2C2O4 ? (H2C2O4)ads (10) (H2C2O4)ads + (OH)ads ? (HCO2)ads + CO2 + H2O (11) (HCO2)ads ? CO2 + H + e (12) +

where semi-reaction (11), a decarboxilation assisted by adsorbed hydroxyl radicals, would be rate-determining.

Lamy-Pitara and co-workers have assumed the same path for catalytic and field- assisted oxidation of OA at Pt, without and with applied field, pointing out the weak character of the adsorption of hydroxyls at the Pt surface, prior to the first stages of its electro-oxidation [15].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

and hydroxyl radicals. The latter would dominate at low, the former at high electrode potentials.

Referring to BDD electrodes, Comninellis et al. [1,43] have proposed a mechanism of oxidation of OA based on the assumption of a small reaction volume (?reaction cage?) confined nearby the electrode surface. According to this hypothesis, hydroxyl radicals could react with the organic substrate in the solution phase: ?OH ? ? H2C2O4 + HC2O4 + H2O (13) HC O ? ?COOH + CO (14) ? 24 2

? ?OH ? COOH + CO2 + H2O (15) The modest adsorptive ability assigned to BDD surfaces supports this hypothesis, particularly in the J?E range where the electrochemical incineration takes place.

On the basis of the mechanisms cited above, the reactivity of OA toward anodic oxidation at a number of electrode materials, generally high, is conditioned by adsorption of the reactant (as undissociated or partially dissociated molecule), as well as radicalic intermediates undergoing the first decarboxilation stage. In fact, assuming for the H2C2O4 molecule and for its mono-anion HC2O4?, a planar structure [44], the strongly delocalized ? electron system can give origin to weak chemisorption at metal surfaces, eventually leading to de-electronation.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

electrochemical incineration proposed by Comninellis [6]. The co-existence of two reaction mechanisms at Pt electrodes, assumed by Petrii et al. [24], and the concomitant sensitivity of the reaction to electrode surface conditioning, also reflects this aspect.

When the study of OA anodic oxidation is extended to higher J values, sustained also by o.e.r., the situation becomes more complex, because of the significant changes undergone by the electrode surface. At the IrO2-based electrode, the stabilization of hydroxyl radicals, by the interaction with the oxide lattice and the oxidation state changes of the metal cations, keeps a dominant inhibiting effect under all the explored conditions. On the contrary, the lack of higher oxidation states in the case of the PbO2 electrode, together with a carboxyl?Pb(IV) interaction at the electrode surface, favor the OA anodic oxidation. At the Pt electrode, when the depolarizing action of OA loses its effectiveness, the oxidation of Pt starts, leading to different oxide species, as a function of the anode potential attained and therefore of the J value imposed in the galvanostatic experiment. Also, in consideration of studies in ref. [42], more reactive forms of Pt oxide precursors are probably formed up to around 100 A m-2; for higher J values, with the earlier mentioned sharp increase of the Pt anode potential, phase oxide is formed, with good catalytic activity toward o.e.r.. In the case of Au, at variance with Pt, no deactivation is observed, irrespective of the applied J and of the stage of the OA incineration.

However, this is in agreement with the mechanism of growth of anodic Au oxide and its chemical and mechanical instability.

Concerning the BDD electrode, operating at higher J values does not substantially modify the state of its surface, and changes in the efficiency of the OA oxidation process are essentially related with the competition with the o.e.r. at oxygen- terminated BDD surfaces [46].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Conclusions According to the model proposed by Comninellis [6], on a general basis the electrochemical incineration of organics at a given electrode can take place at satisfactory rates only in the potential region where oxygen evolution also takes place, electrosorption of hydroxyl radicals and their reactivity being the decisive factor. The many papers published on the subject of electrochemical abatement of organic pollutants in aquatic media have largely confirmed this approach, oxalic acid being, at some extent, an exception because of its relatively low reactivity towards hydroxyl radicals [47]. The results discussed in this chapter substantially confirm the deviation of the behavior of OA, with respect to the above hypothesis, in substantial agreement with most of the literature, where its electrosorption is in fact assumed as a necessary prerequisite for a fast electro-oxidation to take place [8, 9, 18, 19, 24, 29].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Pt electrodes, apparently in an intermediate position between PbO2 and IrO2 electrode materials, the final result is decided by the potential attained during the electrolysis, and therefore by the value of J and residual concentration of OA. The good performance at Au electrodes may have different reasons under different polarization conditions. In the potential range where the metal is stable, the anodic oxidation takes place with practically 100% faradaic yield. The onset of Au surface oxidation and oxide film growth, taking place when a large percent of the organic substrate has been oxidized, do not strongly inhibit the elimination of the residual part, because of the chemical/mechanical instability of the oxide film [19].

In a wider sense, the reactivity of oxalic acid toward anodic oxidation at different electrodes could be thus seen as one more example of ?volcano plot?, of the type proposed for ethylene oxidation.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

References [1] D. Gandini, E. Mahè, P.A. Michaud, W. Haenni, A. Perret, Ch. Comninellis, J. Appl. Electrochem., 30 (2000) 1345.

[4] C. Pulgarin, N. Alder, P. Péringer, Ch. Comninellis, Water Res., 28 (1994) 887.

[10] B. Correa, Ch. Comninellis, A. De Battisti, J. Appl. Electrochem., 27 (1997) 970.

Comninellis, D. Gandini, New Diamond Frontier Carbon Technol., 9 (1999) 229.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

[18] J.W. Johnson, H. Wroblowa, J.O'M. Bockris, Electrochim. Acta, 9 (1964) 639.

[19] J.W. Johnson, S.C. Mueller, W.J. James, Trans. Faraday Soc., 67 (1971) 2167.

[22] G. Horanyi, D. Hegedüs, E. M. Rizmayer, J. Electroanal. Chem., 40 (1972) 393.

[26] N.B. Morozova, G.E. Shcheblykina, A.V. Vvedenskii, Russ. J. Electrochem., 35 (1999) 310.

[28] Z. Alaune, R. Mazeikiene, Liet. TSR Mokslu Akad. Darb. Ser. B., 2 (1987) 11.

[31] A. Perret, N. Skinner, Ch. Comninellis, D. Gandini, Electrochem. Soc. Proc., 32 (1997) 275.

[32] J.M. Orts, J.M. Feliu, A. Aldaz, J. Clavilier, A. Rodes, J. Electroanal. Chem., 281 (1990) 199.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

[39] S.E.S. El Wakkad, S.E. Khalafalla, A.M. Shams El Din, Egypt. J. Chem., 1 (1958) 23.

[41] Yu.V. Battalova, L.A. Smirnova, G.F. Volodin, Yu.M. Tyurin, Elektrokhimiya, 11 (1975) 1276.

[44] J. Higgins, X. Zhou, R. Liu, T.T.-S. Huang, J. Phys. Chem. A, 101 (1997) 2702.

[45] J.O'M. Bockris, A.K.N. Reddy, Modern Electrochemistry, Vol. 2, Plenum Press, New York, 1970, Chapter 10.

[46] S. Ferro, M. Dal Colle, A. De Battisti, ?Chemical surface characterization of electrochemically and thermally oxidized boron-doped diamond film electrodes?, Carbon., accepted for publication.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Electrochemical Incineration of Tartaric Acid Introduction Several biological, chemical and physicochemical methods can be used to detoxify industrial wastewaters containing organic pollutants. Recent research has demonstrated that electrochemistry is an attractive alternative to traditional methods for wastewater treatment [1]. In particular, during the last decade, electrochemical incineration has drawn much attention, often affording interesting alternatives to already existent processes.

The electrochemical abatement of organic pollutants in a given wastewater is carried out directly or indirectly, by electrochemical oxidation. The main goal in this process is the complete oxidation of the organic substrate to CO2, or its conversion to biocompatible compounds. For the optimisation of the electrochemical oxidation method, the role of the anodic material is clearly important; traditional electrode materials, such as Pt, PbO2, IrO2, Pt-SnO2 [2] and Ti/SnO2-Sb2O5 [3], have been taken into consideration. More recently, highly boron-doped diamond (HBDD) electrodes have also been suggested representing a potentially important alternative, because of their very high oxygen overpotential [4].

According to information in the Chapter 2, several anodes have so far been tested in the electrochemical oxidation of model organic compounds, aiming at the optimisation of process efficiency and cost of the electrochemical oxidation of organic pollutants in wastewaters. It had already been found that electrochemical oxidation of most organics in aqueous media only occurs at high potentials with the concomitant evolution of molecular oxygen and without any loss in electrode activity [3, 5]. It had also been demonstrated that the nature of the electrode material strongly influences both the selectivity and the efficiency of the process [2, 3, 6, 7].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

As far as the nature of the organic substrate is concerned, it has been shown that carboxylic acids, common intermediates of the oxidative degradation of several organic substrates, are rather stable and are mineralized at longer times [8]. For this reason the TA was studied, being it considered as a last intermediate of the degradation of some organic pollutants in the wastewater treatment processes.

In a previous experimental work [6], the electrochemical incineration of oxalic acid in acidic media at Ti/PbO2, highly boron-doped diamond (HBDD), Pt, Au and Ti/IrO2-Ta2O5 electrodes was studied, obtaining the best incineration rates at the Pt electrode. The mechanism and products of several anodic reactions are known to depend on the anode material. As confirmed by the oxidative behaviour of OA at the Pt electrode, complex interactions may exist between the organic substrate and the electrode surface.

On the basis of the results obtained investigating the above organic substrate, an alternative compound, TA, has been selected to extend the investigation to more complex molecular frames, possibly involving the formation of intermediates blocking, at different extents, the electroactive sites.

In fact, the latter carboxylic acid has a higher complexity, in terms of molecule structure, with respect to oxalic acid [6]. Similarly to what has been done with oxalic acid, tartaric acid has been studied at different electrode materials (Ti/PbO2, BDD and Pt), trying to understand the interactions of this organic substrate and hydroxyl radicals with the electrode surface.

On the other hand, TA, along with other organic compounds, are part of the wastewater generated by wineries, distilleries and other grape processing industries [9]. This originates mainly from various washing operations during the crushing and pressing of grapes, as well as rinsing of fermentation tanks, barrels and other equipment or surfaces [10]. This organic substrate has been studied according with different methods of degradation such as KMnO4 [11], H2O2 [12], Ce(SO4)2 [13], vanadium(V) in acidic media [14], and electrochemical oxidation [15-17].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

In the present part of this Chapter, the anodic incineration of TA has been studied, with the scope of verifying optimal process yields in terms of electrode material and current density. The research has been performed at Ti/PbO2, highly boron-doped diamond (HBDD) and Pt electrodes; being all of them available for industrial production.

Experimental Cell and electrolyses conditions Bulk electrolyses have been carried out in a single-body, thermostated pyrex glass cell of 250 ml; the anode was a plate or a cylinder, with a geometrical area of 15 cm2, and the test solution volume was 200 mL. Experiments were performed at a constant temperature of 25 °C for studying the role of applied current density.

Solutions were vigorously stirred, by means of a magnetic stirrer. The applied current density range was established from 300 to 1200 Am-2, depending on the anodic material and experiment. The anode was symmetrically positioned between two zirconium plate cathodes in the case of the Ti/PbO2; a cylindrical platinum grid was used as the counter when the cylindrical Pt anode was investigated; and finally, only one Zr cathode was used in the case of BDD and Ti/Pt.

BDD films were grown on conductive p-Si substrate (0.1 ?·cm, Siltronix) via a hot filament, chemical vapour deposition technique (HF-CVD) [18, 19].

Instantaneous Current Efficiency (ICE) and Total Current Efficiency (TCE) The ICE and TCE values for the anodic oxidation has been calculated, using the relation described in Chapter 3[20].

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Techniques and instrumentation Measurements of total organic carbon (TOC) were carried out with a Dohrmann Model DC-80 analyser, follow the carbon concentration of carboxylic acids. The identification of intermediates was realized: The electrolysis sample was extracted with diethyl acetate and then derivatized with diazomethane, prior to the gas chromatography (GC) and thin-layer chromatography (TLC) monitoring. The first screening of metabolites was performed by thin-layer chromatography (TLC) on silica gel Kiesegel 60F254. Eluent used was Alcohol/water/ammonia in the ratio 100:12:16. After evaporation of the solvent the chromatographic plate was treated with a iodine solution; obtaining the Rf values 0.08 and 0.68 for TA and principal product, respectively. Subsequently, a gas- chromatographic analysis was carried out with a Carlo Erba HRG 50160 and a SE (25 m x 0:32 mm) fused silica column. The carrier gas was helium (0.55 atm). A temperature program was applied (first step: 50±200°C with a 5°C min-1 heating rate; second step: 200°C for 2 min. A flame ionization detector was used.

Discussion and results Anodic oxidation of TA TA electrochemical oxidation experiments were performed under galvanostatic conditions at different values of current density (J= 300, 600 and some cases, 1200 A m-2). During each electrolysis, the carboxylic acid elimination was followed by TOC measurements. At the Ti/PbO2 electrode, under the form of electrodeposited film onto Ti plate [21], the oxidation process of TA in 0.5M H2SO4 solutions at 25 °C was carried out at J values of 300 and 600 Am-2. The course of the reaction was followed as described in the experimental section.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 1. Electrochemical oxidation of TA at PbO2 electrode. TOC Trend line hypothesized the 100% elimination efficiency.

The variations in the elimination of TA concentration as a function of Q (specific electrical charge passed), at the different J values, are shown in Figure 1.

Recently, the short-range interaction between Pb(IV) sites and carboxyl groups has been assumed to play an important role [2]. As can be observed, 90% of TA elimination at two values of applied current density (300 and 600 A·m-2) were obtained in the beginning of the oxidation process where 20 Ah dm-3 were consumed. The oxidation curves reveal a rapid removal of TA at the lowest current density; while at 600 A·m-2 the elimination process depend on mass transport limitations, at the least in the final stages. As can be seen in the Figure 2, high values

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

of ICE were obtained, at both applied current densities. TCE values for this anodic material were also calculated, giving figures of 87 and 63% at 300 and 600 A·m-2, respectively.

Figure 2. ICE curves at different applied current densities for the elimination of TA at PbO2 electrode.

Analogous experiments were carried out at HBDD electrodes; under comparable conditions, the trend of the elimination is very similar to that obtained at the Ti/PbO2 electrode, despite the difference in the reaction mechanism. The increase in the efficiency at the latter material may be due to interactions between Pb(IV) sites and carboxyl groups, while for the HBDD electrode the mechanism is through the participation of hydroxyl radicals. In fact, as proposed by Comninellis et al. [7, 8, 22], the reaction at HBDD requires the participation of hydroxyl radicals, with the

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

formation close to the electrode surface of a reaction cage, in which organic molecules and hydroxyl radicals react through a homogeneous reaction.

In addition, the incineration could be increased by the formation of other oxidants on the diamond surface (peroxodisulphuric acid, O3) which could also participate in the oxidation of the carboxylic acid and intermediates near the electrode surface and/or in the bulk of the electrolyte [8]. The change of TOC concentration, as a function of specific electrical charge passed (Q), is depicted in the Figure 3, for the anodic oxidation at the HBDD electrode: similar behaviors were attained at the two applied current densities.

Figure 3. TOC concentration values vs Q during the elimination of TA, for BDD electrode (at 300 and 600 Am-2, and at 25 °C).

This suggests that the incineration of TA depends on the participation of the hydroxyl radicals. In addition, the TCE values for the TA incineration were obtained

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

according to the equation described in the experimental section. At 600 Am-2 a TCE value of 51% was obtained, which increased to 64% at 300 Am-2. On the other hand, the ICE curves are represented in Figure 4, where a near 100% instantaneous current efficiency was observed at both applied current densities during the initial 500 min of electrolysis. A comparison of results of Figures 1 and 2 clearly confirms that a fast incineration of TA is obtained at the beginning, at both PbO2 and HBDD electrodes, while slower elimination is observed at HBDD in the final stages, as a result of a mass transport limitation, which increases the oxidation times [6, 23].

Figure 4. ICE curves vs Q, for the elimination of TA at HBDD electrode, 300 and 600 Am-2 and at 25 °C.

TA mineralization rate data at Ti/Pt surfaces were extremely slower. This behavior was not understood because the Pt has shown a higher catalytic activity towards organic substrates, as observed with the oxalic acid (OA) studied by the authors

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

previously [6]. In fact, the results indicate a higher reactivity of OA at Pt than at other anodic materials. However, the reactivity of the substrate is clearly conditioned by the state of the electrode surface.

As shown in Figure 5, the curves remain similar at different J values at Ti/Pt. In this figure, it can be observed that the TA incineration is independent of the applied current density, because 1200 A m-2 of applied current density was also used at this anodic material, confronting this value with the others. On the other hand, experiments at Pt (pure metal) under similar conditions (300 and 600 Am-2) were performed, where a similar behavior than the one at Ti/Pt was observed (Figure 6).

The last anodic material was taken into consideration because a different composition and preparation path of the Pt electrode could give a considerable improvement in the faradaic yield. Nevertheless, the results were practically the same. The ICE and TCE values indicate that TA is hardly oxidized by Pt and Ti/Pt, ICE values for this latter material are shown in Figure 7: as expected, they are practically negligible. Similar TCE values that were as low as 4.0, 3.7 and 1.5% were obtained at 300, 600 and 1200 Am-2, respectively. Meanwhile, at the pure Pt electrode, the TCE values at 600 Am-2 was 9.4%.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 5. Effect of current density on the electrochemical incineration of TA. TOC data vs specific electrical charge passed (Q); Ti/Pt at 25 °C.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 6. Comparison between Ti/Pt and Pt electrodes at different values of current density.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Figure 7. ICE curves vs Q, for the elimination of TA Finally, an experiment was carried also at Pt, but in alkaline media: the oxidation reaction was a little bit faster, but no major improvements could be achieved (Figure 8). As the literature suggests, the results obtained at Ti/Pt and Pt electrodes could be considered as a poor elimination, possibly due to particular interactions of the TA molecule toward the electrode surface. Yan et al. have demonstrated that adsorption interactions are suffered according to the metal and/or OH group of the tartaric acid molecule positions (TA enantiomers). The adsorption of TA on Cu(111) has been studied by electrochemical scanning tunneling microscopy (STM) in aqueous solution. It has been showed that (R, R)-TA and (S, S)-TA can form a well-ordered

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

adlayer on the Cu(111) surface with a (4 x 4) symmetry and a dimeric structure in the temporary model from STM observation was proposed [24]. Also, other investigations were carried out on Ni(110) in an ultrahigh vacuum (UHV) by experimental and theoretical methods. The enantioselectivity of catalysts was attributed to the chiral reconstruction of the substrate induced by the adsorption of (R, R)-TA molecules [25]. Recently, Jones and Baddeley found that (R, R)-TA produced three distinct ordered structures on the Ni(111) surface at different temperatures in an UHV [26].

In this case, the studies reveal the presence of terraces, steps and kinks on the Pt surface where adsorption of (R, R)-TA caused a negative shift of the hydrogen potential under potential features and the {1 1 1} anion feature of the cyclic voltammograms, showing that tartrate ions were more strongly adsorbed than the sulphate and bisulphate ions. Hence a subsequent slower electro-oxidation was obtained.

According to the literature, it is possible to hypothesize a particular interaction of TA towards the Pt surface; assuming that TA structure and geometry can favor strong adsorption phenomena on the anode due to the generation of active sites at the PtOx surface, which are no longer suitable for TA adsorption and/or oxidation intermediates, impeding the anodic oxidation of this organic substrates.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

Comparison between acidic and alkaline media at Pt electrodes at 600 Am-2 of current density and 25 °C. Initial [TA] = 0.1 M.

Electrochemical oxidation pathway In order to achieve the information required to elucidate the oxidative mechanism of TA, an analysis of the solution composition obtained at a Pb/PbO2 electrode, at 300 A m-2, was performed by TLC and GC. Peaks related with the different metabolites, among which tartaric acid (TA), oxalic acid (OA) and glyoxal (GXL) were identified. Other intermediates were not identified during the oxidation process.

From the thin-layer chromatography (TLC) was possible to obtain Rf values, 0.08 and 0.68 for TA and principal product, respectively.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

products, thus confirming the TOC results reported in Figure 1. The GC data confirm the active character of the PbO2 electrode toward the mineralization of TA, which takes place through an almost instantaneous decarboxylation. The detection of GXL and OXA allows to suggest the chemical reaction sequence of Figure 9 as a possible TA oxidation mechanism. This mechanism is in agreement with the results obtained by [28] where the glyoxal and other intermediates were detected.

The first step is represented by decarboxylation and the conversion of TA to GXL, the latter compound being responsible for the absorptive peak observed at a retention time of about 3 min in the GC results (Figure 10). In addition, the production of hydroxyl radicals from acidic media permit the generation of OA, which are further oxidized a complete mineralization, at a rate that strongly depends on the anode material [6, 23].

OH O HO CO2 OH O OH O TARTARIC ACID

GLYOXAL CO2 O · OH OH HO H O HO CO2 OO TARTRONALDEHYDIC ACIDOH O OXALIC ACID

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

OA GXL TA Figure 10. Typical GC analysis for a 0.1M TA solution in 0.5M H2SO4, electrolyzed at 300 A m-2 (Fig.1) at a Ti/PbO2 anode, indicating the intermediates identified after 12 Ah dm-3 of specific charge passed.

? It has been suggested [8] that, at BDD electrodes, the OH radicals formed by water oxidation can be either electrochemically oxidized to dioxygen or contribute to the complete oxidation of the organic compounds, especially carboxylic acids Other oxidants formed at the diamond surface (H2S2O8, O3) can participate in the oxidation of the carboxylic acids, in the proximity of the electrode surface and/or in the bulk of the electrolyte.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

As a final remark about PbO2-based electrodes, recently, some role of short-range interaction between Pb(IV) sites and carboxyl groups have been assumed to play an important role [2]; favoring the decarboxylation step. Also, the formation of tartronaldehic acid could be probable at these electrodes (BDD and PbO2), but it converts to GXL rapidly.

Conclusions In this research, the electrochemical oxidation of TA at Ti/PbO2, Ti/Pt, Pt and HBDD anodes has been studied under different experimental conditions, i.e.

different values of applied current density. The analysis of results of bulk TA electrolyses leads to some final considerations: A complete elimination of the organic reagent has been achieved at PbO2 and HBDD electrodes, while only a minor attack takes place at the Pt anodes. The influence of the anodic material on the elimination of TA seems to be very important, Ti/PbO2 being the electrode at which the best incineration efficiency has been obtained.

Ti/PbO2 anodes exhibit good electrocatalytic properties towards the oxidation of TA due to the interactions between Pb(IV) sites and carboxyl groups. Also, this material offers reasonable service life and resistance to deterioration by the characteristic feature in its preparation.

In acidic media, the mechanism at HBDD electrode is through hydroxyl radicals participation, where

H O ? ( ? OH) + e (2) 2 ads

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

? ?1 + ( OH) O + H + e (3) ads 2 2

Carboxylic acids ??? ??? CO2 + H2O (4) ( OH) ads

the organic molecules and hydroxyl radicals react through a homogeneous reaction[7, 8, 22].

In acidic and alkaline media, the percentage of TA elimination was minimum at Pt electrodes. In this case, changes seems to generate active sites at the PtOx surface, which are no longer suitable for TA adsorption and/or oxidation intermediates, impeding the anodic oxidation of the organic substrates.

During the TA incineration at the same electrolysis conditions applied at Pt, the Ti/Pt electrode obtained a TCE from 4 to 1.5% approximately.

The ICE and TCE values for the electrochemical elimination of TA from aqueous solutions follows the sequence: PbO2 > HBDD > Pt anodes, and depend on the applied current density.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

References Comninellis, D. Gandini, New Diamond and Frontier Carbon Technol., 9 (1999) 229.

[6] C. A. Martinez-Huitle, S. Ferro, A. De Battisti, Electrochim. Acta, 49 (2004) 4027.

[8] D. Gandini, E. Mahè, P.A. Michaud, W. Haenni, A. Perret, Ch. Comninellis, J. Appl. Electrochem., 30 (2000) 1345.

[9] L. Malandra, G. Wolfaardt, A. Zietsman, M. Viljoen-Bloom, Water Res., 37 (2003) 4125.

[11] V. M. Bhale, P. G Sant, S. L. Bafna, I. Holkar Coll., J. Sci. & Ind. Res., 15 (1956) 45.

[12] A. Ya. Sychev, G. G. Duka, Biologicheskie i Khimicheskie Nauki, 6 (1981) 68.

[14] R. A. Ando, C. Raminelli, J. W. Barreto, K. Takashima, Monatshefte fuer Chemie, 134 (2003) 1321.

Chapter 4. Direct electrochemical oxidation. Role of the electrode material

[19] A. Perret, N. Skinner, Ch. Comninellis, D. Gandini, Electrochem. Soc. Proc., 32 (1997) 275.

[23] C. A. Martínez-Huitle, M. A. Quiroz, Ch. Comninellis, S. Ferro and A. De Battisti, Electrochim. Acta, 50 (2004) 949.

[24] H.-J. Yan, D. Wang, M.-J. Han, L.-J. Wan and C.-L. Bai, Langmuir 20 (2004) 7360.

[25] V. Humblot, S. Haq, C. Muryn, W. A. Hofer, R. Raval, J. Am. Chem. Soc., 124 (2002) 503.

[27] O. A. Hazzazi, G. A. Attard, P. B. Wells, J. Mol. Catalysis A: Chemical, 216 (2004) 247.

Chapter 5. Reactivity and engineering parameters

Electrochemical Incineration of Oxalic Acid Reactivity and engineering parameters

Introduction Many electrochemical processes are carried out under limiting or near limiting current conditions, in order to maximize the space-time yield of the electrolyser. As a result, mass transport determines the rate of conversion of reactants to products and it is common to use inert turbulence promoters, like e.g. baffles, and/or high fluid velocity to enhance the mass transport toward the electrode surface and, hence, the cell current density [1]. Cells consisting of planar electrodes in a parallel plate configuration are among the most frequently used type of electrochemical reactors for industrial applications. Such reactors combine simplicity of design, manufacture and versatility, allowing their use in several processes. However, an increase in the mass transfer coefficient is often the result of a complex combination of geometry and hydrodynamics, especially in the inlet and outlet sections of the cell. The relationship between cell flow conditions and the mass transfer performance, in parallel plate cell geometry, has been recently described for some specific cases [2].

Chapter 5. Reactivity and engineering parameters

have been realized [8-15], concerning the experimental conditions, phenomena, interactions, inert turbulence promoters, design of cells, as well as the mathematical modelling of processes, allowing important hydrodynamic improvements for the electrochemical processes under investigation. Besides, different types of cells have been proposed throughout recent years; an efficient cell design has been obtained through the modification of specific electrode area and/or mass transport coefficient optimisation [16].

In the field of wastewater treatments, in spite of the more traditional applications in the abatement of heavy metals [16], electrochemistry may also find interesting applications in the incineration of organic compounds [15], offering both process versatility and simplicity of reactors, in terms of construction and management. In addition, electrochemical oxidation processes promise environmental compatibility and potential cost effectiveness for the degradation of different organic pollutants [17]. The electrochemical oxidation method has been proposed, in particular, for the destruction or/and the conversion of mixed wastes, containing refractory organic pollutants or toxic substances. The main purpose of the wastewater treatment is the complete oxidation of organics to CO2 [18-21] or the conversion of toxics to biocompatible compounds [22, 23]. For the elimination of organic pollutants, several anodic materials have been investigated, including metal-oxide anodes such as IrO2 [24, 25], PbO2 [26-29], SnO2 [21, 30] and SnO2-Sb2O5 [31]; more recently, the list of possible candidates has been further enriched with the involvement of Ti/Diamond (Ti/BDD) [32] and Si/Diamond (Si/BDD) [33].

The aim of this studio was carried out mass transfer studies in an electrochemical cell with a new design, and to test this cell for the electro-oxidation of oxalic acid.

The latter investigation has been performed using Ti/Pt, Si/BDD, Pb/PbO2 and Ti/IrO2-Ta2O5 electrodes. As evident in the information in the Chapter 2, carboxylic acids are common intermediates in the oxidative degradation of several organic substrates; also, they are rather stable and their mineralization is quite elaborate [34].

Chapter 5. Reactivity and engineering parameters

This seems to hold, in particular, for oxalic acid (OA), which is the last intermediate in a number of wastewater treatment processes as is indicated in the Table 1 (Chapter 2); for this reason, it has been selected as a model organic substrate in this research.

Experimental Cell and Electrodes Figure 1 shows schematic diagram of the electrolytic flow cell with parallel plate electrodes, whose schematic diagram is reported in Figure 1. The cell has been assembled with two parallel disc plate electrodes: an anode made of nickel and a Ti/Pt cathode, both 0.5 mm thick with an active area of 63.5 cm2 and the inter- electrode gap of 10 mm. The cell has only one inlet and several outlet sections, all positioned along its central element. Electrolyte circulation through the parallel disc cell was achieved with a high-flow peristaltic pump. The cell compartments were made of PVC and constructed so as to accommodate the disc plate electrodes, as shown in Figures 2 and 3. The cell inlet was designed in such a way as to sprinkle the solution at the center of the anodic disk electrode, thus obtaining a rather homogeneous circulation of the fluid inside the cell, as described in Figure 1c. The electrolyte was stored in a glass tank and circulated through the electrolytic cell (Figure 1b), at a constant flow rate.

Chapter 5. Reactivity and engineering parameters

1a 1b 1c

Figure 1. Design of the electrochemical cell. 1a) Different compartments of the cell: A) anodic compartment, B) electrolysis compartment and C) Cathodic compartment. 1b) Electrolysis compartment: One inlet and several outlet sections. 1c) Electrolyte distribution in the central compartment and toward electrode surfaces.

Chapter 5. Reactivity and engineering parameters

Chapter 5. Reactivity and engineering parameters

Techniques and Instrumentation Characterization of the flow cell, was achieved in an aqueous solution of potassium ferro/ferri-cyanide (considering range from 10 to 30 mM) in 0.5 M NaOH.

Equimolar ferro/ferri-cyanide concentrations have been used, to ensure a constant composition for the electrolyte during the measurements; the physical properties of the test solution have been already described in the literature [4], and are presented in Table 1.

Chapter 5. Reactivity and engineering parameters

Density, ? 1020.5 Kg m-3 Dynamic viscosity, µ 1.105× 10-3 Kg m-1 s-1 Kinematic viscosity, ? 1.083× 10-6 m2 s-1 Diffusity of ferri-cyanide ion 6.631× 10-10 m2 s-1 Schmidt number, Sh 1633 ---

The reactions describing the process: 3? + ? 4? Fe(CN) e Fe(CN) at the cathode 66

4? ? 3? + Fe(CN) Fe(CN) e at the anode 66 The mass transfer coefficient was determined through the well-known diffusion limiting current technique [5, 35], i.e., by using the limiting current plateau regions of current-voltage curves obtained at different electrolytic flows. The mass transfer coefficient value was then obtained using the following relation:

Chapter 5. Reactivity and engineering parameters

grown on conductive p-Si substrate (0.1 ?·cm, Siltronix) via a hot filament, chemical vapor deposition technique (HF-CVD) [36, 37]. This procedure gave a columnar, randomly textured, polycrystalline diamond film, with a thickness of about 1µm and a resistivity of 15 m?·cm (±30%) onto the conductive p-Si substrate.

The Ti/Pt and Ti/IrO2-Ta2O5 electrodes were provided by De Nora (Milan, Italy). Constant current experiments were performed with an Amel model 553 galvanostat, controlling the cell flow rate by using a peristaltic pump. During the various electro- oxidation tests, the OA concentration was determined by means of a conventional KMnO4 titration method, using a 0.1 N KMnO4 solution. Fresh 0.1 M solutions of OA (Fluka, dihydrate salt) were prepared, in 0.5 M H2SO4 (Riedel-de-Haën) using distilled water; the 0.1 N KMnO4 solution was prepared with Fluka reagent, and standardized by titration of a known amount of anhydrous and pure oxalic acid.

Results and Discussion Mass transfer coefficient Polarization curves for the cell, in the particular configuration adopted in this work, are shown in Figure 4; the reported data were obtained at a constant flow rate (Q = 1.83× 10-5 m3/s) but at different ferro/ferri-cyanide concentrations (from 10 to 30 mM, in 0.5 M NaOH). Figure 5 shows the expected linear dependence of the limiting current from the ferro/ferri-cyanide concentration. Applying results to equation 1, gave a value of 2.93× 10-5 m s-1 for the mass transfer coefficient K. This represents a considerable improvement with respect to the value of 1.77× 10-5 m s-1, obtained with a former design of the electrochemical cell (one peripheral inlet and only one outlet, diametrically opposed) [41]. In addition to mass transfer coefficient increase, the new cell design allows a better flow control, increasing the effective distribution of the electrolyte onto the electrode surface.

Chapter 5. Reactivity and engineering parameters

Figure 4. Polarization curves for the engineering characterization of the electrochemical cell, using the ferro/ferri-cyanide redox couple ( from 10 to 30 mM, in 0.5M NaOH). Inset: Determination of the limiting current at different flow rates (% speed) and 30 mM, in 0.5M NaOH.

Chapter 5. Reactivity and engineering parameters

Figure 5. Dependence of the limiting current on the ferro/ferri-cyanide concentration.

Mass transfer coefficient was determined theoretically, based on the Sherwood number (which relates Sh with K [38]), and taking profit applied to proposed by Leveque (equation 3 below), for a laminar flow in a rectangular channel [39, 40], and the definitions of Reynolds (Re) and Schmidt (Sc) parameters [38]:

Chapter 5. Reactivity and engineering parameters

where: v?d Re = e µ (4)

S = v (5) c D

To use the Leveque relationship, the present electrochemical cell, with planar disk electrodes having a diameter of 9.0 cm, has been approximated to a rectangular channel having a width equal to a fourth of the disk circumference, and a height of 10 mm (inter-electrode gap); hence: = × = × ?4 2 ? × ?2 w h s 7.07 10 m d 1.75 10 m e

Once the characteristic length parameter (de) is known, the theoretical mass transfer coefficient for this electrochemical cell can be estimated: for an electrolyte flow of 1.83× 10 m s , Re and Sh numbers of 420 and 94.6 are respectively obtained -5 3 -1

(equations 3 and 4); the K value, calculated using equation 2, results therefore to be 3.5× 10-6 m s-1. Interestingly, the experimentally determined value of K is more than eight times higher than the theoretical value just obtained. This discrepancy has to be ascribed to an inhomogeneity of flow in the cell: certain zones of the electrode could have a higher coefficient of local transfer (inlet and outlet areas), whereas other parts might suffer from mass transfer limitations.

Chapter 5. Reactivity and engineering parameters

Speed Pump (%) 20 30 40 50 60 70 80 90 Flow (m3/s × 10-6) 4.16 7.66 11.2 14.7 18.3 21.8 25.0 28.0 Re 95.5 175.8 257.1 337.4 420.0 500.3 573.8 642.6 Sh 57.7 70.7 80.3 87.9 94.6 100.2 104.9 109.0 K (m s-1 × 10-6) 2.1 2.6 2.9 3.2 3.5 3.7 3.8 4.0

On the other hand, different values for the deciding engineering parameters can be obtained, for the electrochemical cell, depending on the electrolyte flow and, in turn, on the performance of the peristaltic pump; data in Table 2 have been obtained according to the above-discussed approximation for the cell geometry and thus they represent under-estimations of the possible experimental results. With the results obtained in the previous experiments, we decided to probe the new design cell's performance for the electrochemical oxidation of organics, and of oxalic acid in particular.

Anodic oxidation of oxalic acid Oxalic acid oxidation and kinetics The anodic oxidation of OA was carried out at constant temperature (25 °C) and at a current density J = 100 Am-2. The efficiency of the process was studied at highly- boron-doped diamond, PbO2 Ti/ IrO2-Ta2O5 and Ti/Pt electrodes. During electrolysis, the carboxylic acid concentration was followed by titration with 0.1 N Results obtained at a BDD electrode, at different flow rates, are shown in Figure 6.

The fastest degradation of the organic substrate was achieved at an electrolytic flow rate of 1.83× 10-5 m3/s, equivalent to a mass transfer coefficient of 2.93× 10-5 m s-1.

Chapter 5. Reactivity and engineering parameters

Faster pumping rates induce a decrease in the substrate oxidation rate, which may possibly be due to cavitation phenomena. Figure 6 shows that OA oxidation can be achieved with a charge consumption of 7 Ah dm-3 corresponding to a total faradaic yield of 90.0%. Although a straight comparison with results obtained in a batch-type cell [42] is not possible (because of the different composition and preparation path of the diamond electrode used) the considerable improvement in the faradaic yield achieved in the current flow-cell has is ascribed to a more effective mass transport.

Figure 6. Electro-oxidation of OA at the BDD electrode, at 100 Am-2 and at different flow rates, depending on % speed pump.

Chapter 5. Reactivity and engineering parameters

molecules and hydroxyl radicals react through a homogeneous reaction, could be in agreement with the observed significant difference between results obtained in the batch- and in the flow-cell.

Anodic oxidation of OA was also carried out at a Ti/Pt electrode (under the same conditions applied for the BDD electrode). According to figure 7, the main features of the oxidation process do not differ significantly from those ascertained at BDD.

As already observed in [42], the anodic oxidation process initially exhibits a faradaic yield close to 100%, which decreased to zero close to the final stages. According to from previous research [42] and their interpretation, the improved transport conditions do not eliminate blocking of the incineration process at lower OA concentration (it rather better defines its position). This highlights further support for the hypothesis that when during the electrolytic treatment, low concentrations of organic depolarizer are attained, the role of the concomitant oxygen evolution becomes decisive; this increases the efficiency of the parasitic reaction (experiments are carried out at constant J) causes changes at the Pt electrode surface, that decreases its activity toward the OA oxidation.

Again in good agreement with the measurements in batch reactor, the faradic efficiency of the OA oxidation at the Pb/PbO2 electrode is quite high, practically 100% down to very low residual substrate concentrations (Figure 7). Also these results, as well as those at BDD and Ti/Pt, have been obtained at electrode materials not identical with those used in [42], differing in terms of preparation, in the detail of their surface texture and, at a certain extent, also in their composition and microstructure (case of Pb/PbO2). On the other hand, the anodic oxidation/incineration of OA seems to take place with efficiency close to 100% at electrode materials with quite different properties.

Chapter 5. Reactivity and engineering parameters

Figure 7. Comparison of elimination of OA at Pt, PbO2, BDD and IrO2- Normalized OA concentration vs specific charge passed.

Chapter 5. Reactivity and engineering parameters

experiment. Apparently, in the case of thermally (IrO2) or anodically-prepared (Pt) group VIII noble-metal oxides, the nature and reactivity of active sites favor the interaction with water molecules and hydroxyl radicals that, displacing the OA species, makes its oxidation extremely slow.

Moreover, a linear trend was found for the logarithm of the OA concentration (normalized to the initial concentration) as a function of time, indicating a first order kinetics:

? [OA]? ? ? A ? ln ? t ? = ? ? ? kt (9) ?[OA]t=0 ? ?V ? where [OA]?t is the OA concentration at a given time, and [OA]t=0 is the initial OA concentration (Figure 8).

Chapter 5. Reactivity and engineering parameters

Flow Cell i / A m-2 k × 106 / m s-1 R2 PbO2 100 7.13 0.990 Pt 100 7.10 0.989 BDD 100 6.53 0.992 IrO2-Ta2O5 100 1.26 0.991

Table 3 collects the kinetic constant values (k) obtained from a linear regression of experimental data, elaborated according to equation 9, for the flow cell and the investigated anodic materials. It's worth mentioning that correlation coefficients were always ? 0.98, thus confirming the first order kinetics of the OA incineration process. Obviously, the rate of disappearance of the organic substrate is influenced by an increase of K; however, other factors can be considered to explain the results obtained with the flow-cell.

Conclusions The proposed flow-cell scheme guarantees an effective and reproducible mass transport. The inter-electrode space is properly exploited and the cell geometry and flow direction, from the center toward the periphery, guarantees a progressive lowering of the solution flow rate while the reactant is consumed. Some problems encountered at higher flow-rates denounce however the need for further improvements in the cell design.

Chapter 5. Reactivity and engineering parameters

stage. It is confirmed, in particular, that the anomalous decrease of the process rate at Pt electrodes, when the substrate concentration decreases below certain critical values, can be related with changes of the electrode surface at the higher anodic potentials attained as a consequence of the lack of organic depolarizers [42]. These changes seem to generate active sites at the PtOx surface, which are no longer suitable for OA adsorption, thus resembling the case of IrO2-Ta2O5 electrode, typically a very good catalyst for the oxygen evolution reaction.

Chapter 5. Reactivity and engineering parameters

References [2] W.N. Taama, R.E. Plimley, K. Scott, ?Mass transfer rates in a DEM electrochemical cell?, Proceedings of the 4th European Symposium on Electrochemical Engineering (CHISA); Institute of Chemical Technology, Prague (1996) 289-295.

[3] F. Goodridge, G.M. Mamoor, R.E. Plimley. IChemE Symp. Ser. 98 (1985) 61.

[12] A. Djati, M. Brahimi, J. Legrand, B. Saidani, J. Appl. Electrochem. 31 (2001) 833.

[14] A.A. Wragg, A. Leontaritis, in Electrochemical Cell Design and Optimization, Dechema Monographs 123 (1991) 345.

[15] A.M. Polcaro, A. Vacca, S. Palmas, M. Mascia, J. Appl. Electrochem. 33 (2003) 885.

[17] D. Gandini, E. Mahè, P.A. Michaud, W Haenni, A. Perret, Ch. Comninellis, J. Appl. Electrochem. 30 (2000) 1345.

Chapter 5. Reactivity and engineering parameters

[22] Ch. Comninellis, E. Plattner, C. Seignez, C. Pulgarin, P. Péringer, Swiss Chem, 14 (1992) 25.

[23] C. Pulgarin, N. Alder, P. Péringer, Ch. Comninellis, Water Res., 28 (1994) 887.

[24] A.M. Polcaro, M. Mascia, S. Palmas, A. Vacca, Ind. Eng. Chem Res. 41 (2002) 2874.

[26] A.M. Polcaro, S. Palmas, F. Renoldi, M. Mascia, J. Appl. Electrochem. 29 (1999) 147.

[30] G. Foty, D. Gandini, Ch. Comninellis, Current Topics in Electrochemistry, 5 (1997) 71.

Comninellis, D. Gandini, New Diamond and Frontier Carbon Technol., 9 (1999) 229.

[34] D. Gandini, E. Mahè, P.A. Michaud, W. Haenni, A. Perret, Ch. Comninellis, J. Appl. Electrochem., 30 (2000) 1345.

Chapter 5. Reactivity and engineering parameters

[37] A. Perret, N. Skinner, Ch. Comninellis, D. Gandini, Electrochem. Soc. Proc., 32 (1997) 275.

[38] D. Pletcher, Industrial Electrochemistry. University Press, Cambridge (1982) Chapter 2.

[42] C.A. Martinez-Huitle, S. Ferro, A. De Battisti, Electrochim. Acta. 49 (2004) 4027.

Chapter 6. Indirect Electrochemical Oxidation

Anodic Oxidation of Tartaric and Oxalic acids in the presence of Halides

Introduction The environmental application of electrochemistry has demonstrated that many organic compounds can be treated by electrochemical oxidation instead of traditional methods [1-4]. Electrochemical incineration of organic pollutants can be attempted by direct or indirect oxidation. In the latter case, anodically formed oxidants (Cl2, hypochlorite, peroxide, ozone, Fenton's reagent, peroxodisulphate) react with organic substrates [5], eventually leading to their complete conversion to CO2, H2O and other inorganic components. In parallel to the study of the optimization of the electrochemical incineration of organic substances, the DSA anodes, in the last years has provoked interest also by the possibility to improve the efficiency of the electrochemical oxidation, using ion of transition metals in their more elevated state oxidation. In practical, the process previews the generation in the reaction media a strong oxidant, through the anodic oxidation of the precursor salt: the generated oxidant can reacts with the organic molecule at the electrode surface and/or in the volume of the solution, reducing itself. It will be able but of new oxidizing itself to the anode, becoming still available, so as to form a reaction cycle, as was schematized by Galla [6] (Figure 1).

However, the greater part of the mediators is presented from redox couples shown in Table 1 [7]. The ?indirect electrochemical oxidation? (IEO) has a wide range of applications, considering quality characteristics mediators as was described in the Chapter 2.

Chapter 6. Indirect Electrochemical Oxidation

Figure 1. Indirect oxidation Mediators with a high standard potential (e. g. Ag2+, Co3+) are mainly used for the oxidation total in the waste treatment as was indicated in the Chapter 2. In presence of ion chlorine or chlorinated compounds, the more opportune mediator could be considered the Co3+/Co2+ redox couple, because doesn't exist the precipitation phenomena like in the case of the Ag2+/Ag+ couple. Relatively at the Co3+/Co2+ redox couple a problem is represented from the presence of ion Co3+, in solution at the end of the process, when a part or all the organic species has been eliminated. In fact, it is not excluded that the presence of these ions in wastewaters, and their recovery for precipitation, does not give problems. From this point of view, exist an optimal redox couple for IEO, it is the couple Ag2+/Ag+, than it is obtained for oxidation of the AgNO3 mediator in HNO3. Nitric acid (E° = 0,79 V) impedes the reduction of Ag+ to Ag0, reducing itself to nitrous acid.

Mediators with minor standard potentials (Ce, Mn) are mainly use in the selective organic synthesis [8]. Finally, the redox couples Ti3+/Ti2+ have a potential standard negative, and these can be used in the organic synthesis for reduction. The transition metals are not the only possible electrochemical mediators; in fact is possible to generate species strongly oxidants at the anode from other chemical species.

Chapter 6. Indirect Electrochemical Oxidation Mediator Standard potential (V) Ag2+/Ag+ 1,98 Co3+/Co2+ 1,81 Ce4+/Ce3+ 1,61 H2SO4 media MnO4- / MnO2 1,69 Mn3+/Mn2+ 1,51 Cr3+ / Cr2O 2- 7 1,38 Cl2/Cl- 1,36 Br2/Br- 1,09 Fe3+/Fe2+ 0,77 Cr3+/Cr2+ ?0,41 Ti3+/Ti2+ ?0,37 Table 1. Different mediators and standard potential

Chapter 6. Indirect Electrochemical Oxidation

pollutants in presence of NaCl in alkaline media, considered the anodic hypochlorite formation and the successive indirect oxidation of the organic substrate in the volume solution or proximity of the electrode. At the same time existing the generation of parallel reactions that decreases the process efficiency, as the chlorate formation and the chlorine evolution. While in the case of the glucose studied by De Battisti and coworkers [12], the presence of species strongly oxidants in solution like ClO- (from solutions of sodium hypochlorite) was observed.

On the basis of a series of studies, both chloro and oxy-chloro radicals, co-generated at the electrode surface, have to be considered in the mechanism of the electrochemical incineration, thus representing an extension of the model initially proposed by Comninellis [4], for the direct electrochemical oxidation, and subsequently extended by De Battisti and coworkers, to the case of the active chlorine mediation, where oxygen transfer can be attained through absorbed oxy- chloro species [12-14], where those take play a decisive role in the electrochemical oxidation, as a whole naturally with the hydroxyl radicals.

Alternatively, the application of NaBr for the anodic oxidation of organics has been previously studied for the case of, e.g., isosorbide [15], methyl orange [16], hydrazodicarbonamine [17], lactic acid [18], also the mediator effect showed by NaBr was employed for indirect oxidation of alcohols by a double mediator system [19]. In all cases the reaction conditions were optimized and different yields were obtained due to the presence of NaBr and the kinetic was elucidated for same process.

Looking at the literature, in the indirect electrochemical oxidation no attempts to degrade or to eliminate organic pollutants from aqueous solutions used of NaF like an alternative method has been published yet.

Chapter 6. Indirect Electrochemical Oxidation

particular, the influence of some parameters such as current density, halogen salt concentration and media.

Experimental section Cell and electrodes Bulk electrolyses have been carried out in a single-body, thermostated pyrex glass cell of 250 ml; the anode was a plate or a cylinder, with a geometrical area of 15 cm2, and the test solution volume was 200 mL. Experiments were performed at 25 °C for studying the role of applied current density (J) and mediator. Solutions were vigorously stirred, by means of a magnetic stirrer. The J range was established from 300 to 600 A m-2, it depended the anodic material and experiments. A cylindrical platinum grid was used as the counter when the cylindrical Pt anode was investigated; finally, only one Zr cathode was used in the case of Ti/Pt. During the polarization curves measurements a Pt wire was used, with a real area of 7.06 mm-2.

Instantaneous (ICE), Total (TCE), General current efficiency (GCE) and Electrochemical Oxidation Index (EOI) The ICE, TCE and GCE for the anodic oxidation, were calculated using the relations described in Chapter 3.

Techniques and instrumentation Measurements of total organic carbon (TOC) were carried out with a Dohrmann Model DC-80 analyzer, follow the carbon concentration of carboxylic acids.

Electrochemical analyses were carried out by an Autolab PGSTAT20 (EcoChemie, The Netherlands). Polarization curves were performed in solutions at different

Chapter 6. Indirect Electrochemical Oxidation

concentration of NaCl, NaBr and NaF on alkaline media. Potential values were referred to a saturated calomel electrode (SCE).

Solutions 0.1 M solutions of TA (Fluka) and OA (Fluka, dihydrate salt) were prepared in alkaline media composed: 0.25 M NaOH (Riedel-de-Haën) and 0.5 M Na2SO4 (Baker analyzed), using distilled water. The NaCl, NaBr and NaBr (Fluka, salt).

Discussion and Results Electrochemical measurements Polarization curves were obtained in solutions of different sodium halides (NaCl, NaBr, NaF) at different salt concentration, in alkaline media and at room temperature (~25°C), using a Pt wire as anode.

Figure 2, NaCl Solutions: In the presence of Cl- (NaCl concentration range was from 0.005 to 2 M), the onset of the oxygen evolution reaction (o. e. r.) at 1,4 V vs SCE was attained, it is shifted towards more positive values for NaCl concentrations between 0.005 and 0.5 M.

Above the latter value, the opposite effect is observed. In the J range within which we carried out the electrochemical incineration of organic substrates, the anode potential increases up to about 2.0 V (vs. SCE). The linear polarization curves of Pt electrode have been obtained in 0.25 M NaOH and 0.5 M Na2SO4 at scan rate of 0.0005 V s-1. These results evidenced the favoring of the electrochemical oxidation in presence of NaCl on the alkaline media than oxygen evolution reaction.

Chapter 6. Indirect Electrochemical Oxidation

Figure 2. Polarization curves at Pt electrode; scan rate of 0.5 mV s-1, electrode area: 4.66 mm2; [NaOH]=0.25 M; [Na2SO4]=0.5 M. Effect divergent from [NaCl]=0.6 to 2M.

Figure 3, NaBr Solutions: The experiments were carried out under similar conditions that NaCl effect, but the range of NaBr concentrations was from 0.0001 to 0.4 M. These concentrations of halogen salts are approximately from 41.2 to 1.03 g/L. At very small NaBr concentrations (< 0.01 M), a shift of the J/E curves in the positive direction is still observed, although the effect is much less evident than in Fig. 2. Above 0.01 M NaBr, the anode potential buffering by the halide electro- activity prevails. This observation may indicate the formation of Br- species, but it also involves the mechanisms formation of other species generated from Br- and media.

Chapter 6. Indirect Electrochemical Oxidation

Figure 3. Polarization curves at Pt electrode; scan rate of 0.5 mV s-1, electrode area: 4.66 mm2; [NaOH]=0.25 M; [Na2SO4]=0.5 M.

Figure 4, NaF solutions: The effect of NaF is quite interesting, considering that F- is not electroactive in aqueous solutions. The NaF concentrations were from 0.005 to 0.5 M in the order of 0.21 to 21 g/L. Quite small NaF additions (e.g.: 0.005 M) cause an abrupt positive shift of the o. e. r., which is further enhanced by increasing the halide concentration.

The role of the anion nature in the anodic Pt-oxide formation and o. e. r. is here evident, also in agreement with literature data. The NaF effect observed on Figure 4, it was understood like an inhibition of the o. e. r. In this case, the behavior is the same to NaCl, nevertheless the results during the anodic oxidation confirmed other expectative.

Chapter 6. Indirect Electrochemical Oxidation

Figure 4. Polarization curves at Pt electrode; scan rate of 0.5 mV s-1, Similar effect that NaCl with low and higher concentrations of NaF.

Anodic oxidation Influence of Media In previous works [20, 21], the electrochemical incineration of carboxylic acids was studied. At the Pt electrode in acidic media is concerned (Figure 5), the mineralization of OA is fast, but its rate decreases to almost zero below an OA concentration limit that is higher, the higher the anodic J value. At 100 A m-2, for instance, the concentration limit at which deactivation takes place is quite low, but it becomes already important at J = 300 and 600 A m-2.

Chapter 6. Indirect Electrochemical Oxidation

Figure 5. Decrease of OA content in solution, at the Pt electrode, as a function of charge, at different current densities (25 °C). Inset: effect of the temperature, on electrolyses carried out at 600 Am-2.

This behavior was eliminated changing the media to alkaline (Figure 6), where the complete elimination of OA is attained at higher J values. Nevertheless, this last modification in the media does not offered good results with TA (Figure 7); that at this anodic material in acidic and alkaline media was achieved a poor elimination due to possible adsorption of TA molecules on the active sites of Pt [21]. For the behavior showed at Pt electrode with OA in solution, the alkaline media was used in the experiments with absence and presence of halogen salts.

Chapter 6. Indirect Electrochemical Oxidation

Chapter 6. Indirect Electrochemical Oxidation

Comparison between acidic and alkaline media at Pt electrodes at 600 Am- 2 of current density and 25 °C. Initial [TA] = 0.1 M.

Influence of halides OA and TA anodic oxidation experiments were performed under galvanostatic conditions at different current density were carried out (J= 300, 600 Am-2). During electrolysis, the carboxylic acid elimination was follow by TOC measurements.

At the Ti/Pt electrode, the OA electro-oxidation in presence of halides is faster and its efficiency is higher at NaBr (90% approximately). In the Figure 8 is observed the effect generated for each halogen salt in the electrochemical process. In the presence of NaCl (5 gr dm-3), the incineration of OA requires 20 Ah dm-3 respects to 70 Ah dm-3 in absence of halides (alkaline media). The behavior obtained was comparable to that of previous investigation using Pt and SnO2 during oxidation of glucose [12]

Chapter 6. Indirect Electrochemical Oxidation

and Ti-Ru-Sn during the 2-napthol oxidation [13] electrodes. NaBr is even more effective, when it is present in solution, allowing a decrease to 12 Ah dm-3 of the electric-charge consumption.

Figure 8. Decrease of [OA] vs. Q at Ti/Pi electrode in absence and presence of mediator.

The Figure 9 is a zoom of behavior of the Figure 8 excluding the alkaline media curve, where we can see good results obtained at 600 and 300 A m-2 with NaBr present in solution. The better GCE values were obtained in presence of NaBr, as can be seen in Figure 10. While at lower values of J, higher ICE values were observed at the initial stages of process Table 2.

Chapter 6. Indirect Electrochemical Oxidation

Oxalic acid incineration efficiency Alkaline 300 A m-2, 600 A m-2, 300 A m-2, 600 A m-2, Ti/ Pt electrode media NaCl NaCl NaBr NaBr %EOI (i) 24.31 67.13 49.9 79.2 55.56 %ICE (i) 22 80 75 94 85 %TCE 10.0 39.3 34.4 37.2 34.4 Tartaric acid incineration efficiency Alkaline 300 A m-2, 600 A m-2, 300 A m-2, 600 A m-2, Ti/ Pt electrode media NaCl NaCl NaBr NaBr %EOI (i) 4.2 32.9 39.51 62.72 49.69 %ICE (i) 7.2 30 41 52 50 %TCE 4.7 25 28,1 36,2 62,9

Chapter 6. Indirect Electrochemical Oxidation

Finally, under similar conditions were carried out different experiments with cylindrical Pt electrode, because with this electrode were performed the experiments in presence of NaF. The last anodic material was taken into consideration because the support (Ti) could have interaction in presence of NaF in solution. The comparison of experiments at J=300 and 600 Am-2 shows that the faster oxidation of OA is anyway achieved in the presence of NaF (Figure 11).

Chapter 6. Indirect Electrochemical Oxidation

Figure 11. Decrease of OA content in solution, at the Pt electrode, as a function of charge, at different current densities (25 °C).

Chapter 6. Indirect Electrochemical Oxidation

Although an increase in the faradaic efficiency of organic substrate incineration is the common feature in the cases, which we have reported here, the mechanism behind this effect may be different. In the case of F- the anion may have the only role to change the stoichiometry and micorstructure of the oxide film, inhibiting o.e.r. and consequently favoring the electrochemical incineration (Figure 4 and 11).

In the case of Br-, (see also Figs. 3, 11 and 12), the main factor seems to be the electrogeneration of strong oxidants (Br-based oxy-anions), mainly through volume reactions. The Cl- effect seems to be a hybrid of the two former cases, with possible electrogeneration of strong oxidants [14], but also with changes in the electrode film properties, which make it less active toward o.e.r.

Figure 12. Curves for the Ti/Pt electrode, in the presence of the supporting electrolyte (H2SO4 or NaOH/Na2SO4 in halides absence) and with TA in solution (100 mM); data obtained at 300 and 600 Am-2 in presence and absence of NaCl and NaBr in solution.

Chapter 6. Indirect Electrochemical Oxidation

Other experiments were carried out at BDD electrode, in absence and presence of mediators (NaCl and NaBr). The particular characteristics of BDD, its chemical inertness is well recognized; for this reason, this material is considered as an ideal ?non-active? electrode, on which organics oxidation and oxygen evolution take place through the formation of weakly adsorbed and very reactive hydroxyl radicals;

Chapter 6. Indirect Electrochemical Oxidation

easy production of hydroxyl radicals that participate in the oxidation of the OA. On the other hand, the most interesting effect is observed when NaCl was added in solution because the behavior is the same to absence it. Nevertheless, NaBr increases the incineration rate of OA; this effect confirms the role of Cl- explained through its inhibiting effect on the oxygen evolution reaction, at the electrode surface. While the NaBr role is through the generation of species strongly oxidant that react in the volume of reaction.

Figure 14. Decrease of OA at BDD electrode in presence of NaCl, NaBr in solution. Experimental conditions: J = 600 A m-2, 25°C, [OA]=0.12 M, alkaline media (0.25 M NaOH and 0.5 M Na2SO4).

Chapter 6. Indirect Electrochemical Oxidation

Influence of applied current density According at the results, the process is independent of applied current density, while it depends only of specific charge passed and the mediator. For this reason, the effects of halides concentration ([NaX] 5 g dm-3) and current density studied at J = 300 and 600 Am-2 are shown in Fig. 10 and 11, respectively, in terms of the [carboxylic acid] dependence from electrolysis time indicate that the behavior is the same using different applied current density. This observation is in agreement with data obtained the oxidation of 2-naphthol [13] and glucose [12] previously reported.

Variation of pH As shown in Fig. 15, the solution pH (originally around 12, for the experiments carried out in 0.25 M NaOH and 0.5 M Na2SO4) decreases rather sharply during the first two hours of electrolysis to values of 9-10. The original pH value is restored in the last phase of the electrochemical incineration, when the OA was oxidizing in alkaline media, as can be seen in Figure 15. This observation is similar to that of the incineration of glucose [12] as previously reported.

However, in the 0.25 M NaOH solution in presence of NaCl or NaBr, the pH decreases to about 9 during incineration of OA (Figure 16); the larger quantity of sodium hydroxide allows to carry out the electrochemical process in better buffered media.

Chapter 6. Indirect Electrochemical Oxidation

Figure 15. Evolution of pH as a function of specific charge passed (Q) during the incineration of OA. Experimental conditions: J = 600 A m-2, alkaline media, Pt electrode.

This can also justify the smaller amount of chloride requested. In fact the loss of chloride as evolving chlorine (encountered when the pH reaches acidic values) is drastically reduced, as well as the chlorate formation (produced at weakly acidic pH, by reaction between hypochlorite and hypochlorous acid in the bulk of the solution).

Chapter 6. Indirect Electrochemical Oxidation

Figure 16. Evolution of pH as a function of specific charge passed (Q) during the incineration of OA in presence of NaCl and NaBr. Experimental conditions: J = 300 and 600 A m-2, alkaline media, Pt electrode.

Chapter 6. Indirect Electrochemical Oxidation

Figure 17. Evolution of pH as a function of specific charge passed (Q) during the incineration of TA in presence of NaCl and NaBr. Experimental conditions: J = 300 and 600 A m-2, alkaline media, Pt electrode.

In the case of TA, the values of pH decrease to about 8, when NaBr was added in solution and after 40 Ah dm-3 of specific charge passed the pH increases to about 9.

Chapter 6. Indirect Electrochemical Oxidation

Figure 18 shown the dependency of potential electrode at a current density of 350 Am-2. In this graphic are reassumed the polarization curves in presence of halides previously described, emphasizing the potential values achieved under these conditions. Increasing the concentration of NaCl, the o.e.r. is displaced gradually to potentials more positives respect to the potential in absence of mediator. For concentrations up to 0.5 mol dm-3 is observed a change of direction (potential decrease).

Chapter 6. Indirect Electrochemical Oxidation

The curve in presence of NaBr differs totally from that one observed in presence of NaCl: increasing the concentration of the mediator, the curves are displaced to potentials less positives and these always decreasing gradually to values relative to the o.e.r. In this case, as previously was commented, the species production containing bromide is hypothesized. Br- primarily acts in the volume of the solution, with the formation of strong oxidants.

Conclusions Mediated Electrochemical Oxidation at the Pt electrode: According to the results, the efficiency of the incineration process does not strongly depend on applied current density, while it's strongly affected by halide nature and concentration; Br- primarily acts in the volume of the solution, with the formation of strong oxidants; Cl- action could be explained through its inhibiting effect on the oxygen evolution reaction, at the electrode surface, although electrosynthesis of strong oxidants has to be taken into consideration.

The role of F- is played only through modifications of anodic oxide at the Pt surface and of its electrocatalytic properties: the main consequence is an inhibiting effect toward the o.e.r.

Chapter 6. Indirect Electrochemical Oxidation

References [1] K. Rajeshwar, J. G. Ibañez, G. M. Swain, J. Appl. Electrochem. 24 (1994) 1077.

[2] J. D. Rodgers, W. Jedral, N. J. Bunce, Environ. Sci. Technol. 33 (1999) 1453.

[5] K. Jüttner, U. Galla, H. Schmieder, Electrochim. Acta, 45 (2000) 2575 [6] U. Galla, P. Kritzer, J. Bringmann, H. Schmieder, Chem. Eng. Technolo., 23 (2000) 3.

[7] D. Dobos, Electrochemical Data, Elsevier Scientific Publishing Company - Amsterdam-Oxford-New York, (1975).

[9] Boscolo Boscoletto, A. De Battisti et al. J. Appl. Electrochem., 24 (1994) 1852.

[10] J. Mieluch, A. Sadkowski, J. Wild, P. Zoltowski, Prezm. Chem., 59 (1975) 513.

[14] F. Bonfatti, A. De Battisti, S. Ferro, G. Lodi, S. Osti, Electrochim. Acta, 46 (2000) 305.

[15] F. Jacquet, A. Gaset, J. Simonet, G. Lacoste, Electrochim. Acta., 30 (1985) 477.

[16] M-L. Tsai, W-L. Lee, T-C. Chou, J. Chinese Inst. Chem. Eng., 32 (2001) 517.

Chapter 6. Indirect Electrochemical Oxidation

[20] C. A. Martínez-Huitle, S. Ferro, A. De Battisti, Electrochim. Acta. 49 (2004) 4027.

Chapter 7. Kinetic Mechanism of the Electroxidation

Kinetic Mechanism of the Electroxidation of Oxalic Acid at different Electrode Materials

Introduction Oxalic acid (OA) represents one of the proposed metabolites of the anodic oxidation of more complex organic molecules; in spite of its simple structure, its mineralization is strongly dependent on the nature of the electrode material at which the process is carried out. Sargysian and Vasil'ev [1] pointed out such dependence, investigating the kinetic behavior of OA at different metal electrodes (Rh, Pd, Os, Ir, Pt and Au), at a Dimensionally Stable Anodes (DSA) (RuO2-TiO2) and at glassy carbon (GC). Their conclusions, also in agreement with our recent results on the topic [2], highlighted the important role played by the anion adsorption step, claiming that OA is oxidized with increasing difficulty at electrode materials having higher oxygen affinity.

The oxidation of OA on the platinum metals is frequently considered as a relatively simple modeling electrocatalytic process involving transfer a two electrons and yielding the only one product, CO2. Studying this process is also of interest due to the search for regularities of certain stages of an applicationally important process of the ethylene glycol oxidation, and in the last years in the electrochemical oxidation process to be considered a complicate organic substrate in the wastewater treatment.

Chapter 7. Kinetic Mechanism of the Electroxidation

Chollier-Brym, F. Epron, E. Lamy-Pitara, J. Barbier, (1999) [10], N, V. Smirnova, G. A. Tsirlina, S. N. Pron'kin, O. A. Petrii (1999) [11], Ch. Comninellis, (2000) [12]. Most of the investigations have been carried out at the Pt electrode [1, 13,5, 6, 8-11, 14, 16, 17] but interest was addressed also to other metals, like Pd [18, 19] and Au [19, 20], as well as to TiO2 [21], WOX [22] and RuO2 [21]. The wealth of experimental evidence has allowed to formulation of different oxidation mechanisms, although no complete agreement has been attained on the nature of adsorbed intermediates and details of the reaction mechanism are not completely elucidated so far.

The data on the composition of the OA adsorption products [8, 9, 16, 23, 24] largely refer to potential ranges where no oxidation ocurrs at a noticeable rate under steady- state conditions. In most works two adsorbate types were fixed. One is a strongly bound species with the stoichiometry of the type CHO which completely desorbs at E=0.4-0.6 V where the potential with respect to the irreversible hydrogen electrode in the same solution. In what follows, we call it adsorbate I. The other is weakly bound anion like species (adsorbate II).

According to these adsorbates, two groups of mechanisms of the OA oxidation can be considered [11]: One involves the slow electron transfer to a particle of an organic adsorbate. Depending on the nature of this adsorbate, one can segregate at least two versions of the mechanism: with the participation of a strongly bound organic adsorbate [5] or reversible adsorbed oxalate particle of anion type [25]: H2C2O4 HC2O4(ads) + H+ + e (1) -

Chapter 7. Kinetic Mechanism of the Electroxidation

(H2 2O4)ads + 2(OH)ads 2 2O (5) C 2CO + 2H A further detailing (for example, a consideration of the issue concerning the slowness of transfer of the first or second electron [9, 5] in the two electrons reactions written above) seems to be impossible without a preliminary selection of a total scheme for the process.

In this Chapter, the analysis of the dependence of the OA electroxidation on the nature of the electrode material has been extended to highly conductive, boron- doped diamond (BDD) electrodes, with both oxygen and fluorine at their surface.

Also, these kinetic mechanism studies were carried out at DSA electrodes, Pt and glassy carbon.

Experimental Cell, Electrodes, Techniques and Instrumentation The electrolysis cell was made of Teflon® and glass, with the electrode placed at the bottom of the cell and with only the diamond thin film surface making contact with the electrolyte solution. The nominal area of the diamond electrode was 0.78 cm2 (a disk with a diameter of 1 cm). A cylindrical platinum grid was used as counter electrode, and potentials are reported vs. a double-walled saturated calomel electrode (SCE), with an intermediate saturated NaNO3 solution. The latter modification was necessary in order to avoid precipitation of KClO4 in the porous glass joints in the case when the supporting electrolyte was perchloric acid and/or sodium perchlorate.

Measurements were performed at room temperature using 1 M HClO4 (Fluka) as a background electrolyte; NaClO4 monohydrate (Fluka) was used in order to maintain a constant ionic strength. Chemicals were of analytical grade, and all solutions were prepared with triply distilled water. Electrochemical analyses were carried out by an Autolab PGSTAT20. Cyclic voltammetries (CV) were performed at room

Chapter 7. Kinetic Mechanism of the Electroxidation

temperature in solutions stirred by bubbling nitrogen and at a scan rate of 100 mV s- 1; the chosen range potential was cycled using a step potential of 2 mV and repeating the measurement at least five times or until these were reproducible signals; in every case, the last cycle was recorded. Quasi steady-state polarization curves were carried out at a scan rate of 0.5 mV s-1 and with a step potential of 0.45 mV. Curves were recorded starting from the higher value and conditioning the electrode at the initial potential for 15 s. The stability of the electrode material was tested recording a CV curve (as previously discussed) before and after every polarization. The electrodes used were glassy carbon (GC), Ti/Pi, Pt, Ti/IrO2-Ta2O5, Ti/IrO2-2SnO2, Ti/Ir0.67Ru0.33Sn2 O6, Ti/RuSn2O6, Fluorinated-BDD (F-BDD), strongly oxidized BDD, mildly oxidized BDD.

Commercially available BDD electrodes, synthesized at CSEM by the hot filament chemical vapor deposition technique (HFCVD) on p-type, low-resistivity (1-3 m?cm), {100} silicon wafers (Siltronix; diameter, 10 cm; thickness, 1.0 mm), have been used. The plasma-fluorination reaction was performed at CSEM using CF4 as the reactive gas, in a microwave plasma reactor at 13.56 MHz to obtain F-BDD electrode by the postdeposition treatment of the BDD film. The pristine BDD film had a thickness of 1 µm (±10%) and a resistivity of 15 m?cm (±30%), consistent with a boron concentration between 3500 and 5000 ppm.

On the other hand, the BDD electrode electrochemically oxidized was obtained by a pretreatment at +3V vs. SCE, for 20 min (mildly oxidized BDD); while the strongly oxidized BDD electrode, at 400°C for 30 min under oxygen.

Chapter 7. Kinetic Mechanism of the Electroxidation

concentrations of OA, specifically from 100 to 750 mM in solutions containing HClO4 as supporting electrolyte and NaClO4 monohydrate in order to maintain a constant ionic strength.

Ti/Pt, Pt and Glassy Carbon (GC) In agreement with the literature [10, 14, 28-30], CV curves at the Pt electrode in the presence of OA show that the organic molecule is adsorbed on the electrode surface, as witnessed by the hydrogen adsorption and desorption voltammetric peaks, which are shifted towards more negative potentials when OA is added to the solution. This behavior is characteristic of a weak adsorption, similar to that generally observed in the case of anions. The OA oxidation peak appears as a single signal between 0.8 and 0.9 V and, in the explored range from 1mM to 100mM, its height depends on OA concentration, as shown in Figure 5 in the Chapter 4. These experiments were carried out at acidic media (H2SO4); nevertheless the catalytic activity of platinum in the oxidation of OA has studied by other authors [10, 14] where this process depended of the pH and electrolyte support. In addiction, M. J. Chollier et al. have determined of this reaction is quite inhibited by the presence of chloride anions and more favored in the presence of ClO4- and HSO4-, the platinum activity being higher in the case of perchloric acid than in sulfuric acid, at a pH = 0.5.

Chapter 7. Kinetic Mechanism of the Electroxidation

hysteresis. This result is at variance with the inhibition of the process caused by the preliminary maintaining of the potential in the oxygen range of potentials described by authors of [16] on basis of steady-state curves measured in solutions containing HClO4 as supporting electrolyte. It should be assumed that the nature of the anion of the supporting electrolyte that undergoes adsorption simultaneously with the oxygen and oxalic acid may essentially affect the ratio between components in a mixed adsorption layer. Experiments performed by the authors of [23] in sulfuric acid solutions of different concentrations have proved that sulfate anions can force out products of adsorption of OA. Under these conditions where the surface is blocked by anions of sulfuric acid, additional effects of the inhibition by atoms adsorbed oxygen may be manifested.

Also, the values to current density were determined to obtain the values occupied during anodic oxidation described in precedent Chapters.

In the Figure 1, we can see the polarizations curves recorded to different OA concentrations. Nevertheless, it present a very interesting effect where is observed a intersection between the HClO4 polarization curve and the curves at different OA concentrations.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 1. Polarization curves for the electrooxidation of OA on oxidized electrodes of Pt. Inset: Magnification of the bottom part from 1.6 to 1.8 V vs. SCE and 10 mA cm-2.

This intersection is attained to about 1.8 V vs SCE and applied current density of 10 mA cm-2; for this reason is possible that the anodic oxidation of OA occurs with good faradic efficiencies at lower values of current density; while oxygen evolution reaction (o.e.r.) is favored to higher current densities [2].

Although a straight comparison with results obtained in a Pt given considerable improvements in the faradaic yield may be due to differences in the composition and preparation path of the Pt electrode used. In this case, the intersection between polarization curves was not attained (supporting electrolyte and different OA

Chapter 7. Kinetic Mechanism of the Electroxidation

concentration), as shown Figure 2. At Ti/Pt the oxidation of OA achieved higher efficiencies at different applied current densities.

Figure 2. Polarization curves for the electrooxidation of OA on oxidized electrodes of Ti/Pt.

Since of polarization curves were determined the variation of potential as a function of Log J (Figs. 3 and 4), where the Tafel portion with a slope of 165-175 mV/Dec and 183-190 mV/Dec for Pt and Ti/Pt, respectively were studied. These results are in agreement with results obtained by Sargisyan and Vasil'ev [9], and Bockris et al.

Chapter 7. Kinetic Mechanism of the Electroxidation

reaction's slow stage of an organic absorbate on a heterogeneous surface at sufficiently high surface coverages [11].

Figure 3. Current-potential curves for electrooxidation of OA at Pt electrode: effect of OA concentration.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 4. Current-potential curves for electrooxidation of OA at Ti/Pt electrode: effect of OA concentration.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 5. Determination of the reaction order for Pt electrode

Chapter 7. Kinetic Mechanism of the Electroxidation

At lower reagent concentrations, the authors of [8] obtained values of slope approaching 0.6. Taking into consideration that the nature of the slow stage may depend on the concentration, of interest are the values of reaction order determined in narrow concentration intervals.

Allowing for these Tafel plots can be considered the two groups of mechanism of the OA oxidation presented above in the introduction section. Of great importance for choosing between path A (reaction 1 and 2) and path B (reactions 3, 4 and 5) is the magnitude of the reaction order by hydroxyl ions, whose determination is essentially complicated by the fact that the balance between three types of species in solution (no dissociated molecules, univalent anions and divalent anions of OA) depends on solution pH. An exhaustive analysis of this problem was given by Inzelt and Szetey [8].

Figure 7. CV curves for the GC electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and with OA in solution.

Chapter 7. Kinetic Mechanism of the Electroxidation

In Figure 7, the behavior of OA at a glassy carbon (GC) electrode is reported; again, cyclic voltammograms have been recorded at scan rate of 250 mV s-1 for different OA concentrations in solution. As we can be seen in Figure 8 the CV experiments at GC clearly show that OA is electroactive at this electrode, its oxidation-taking place about 0.1 V before the oxygen evolution. Nevertheless, an interaction about 0.9 V vs SCE was observed to higher concentrations of OA (500 and 750 mM).

In agreement with the evidence available in the literature, the oxidation of the organic substrate takes place across a wide potential range at the surface (for example, see refs. [1]). It anyway extends into the polarization where it exhibited a good electroactivity toward OA molecules. This effect was achieved increasing the current density and displacing the curves to values less positive of potential;

according the increment in the OA concentration in solution (Figure 9). In addition,

Chapter 7. Kinetic Mechanism of the Electroxidation

the reaction order was obtained at several values of potential to about 0.7; being it calculated from the Tafel plots obtained for this anodic material (Figure 10).

Figure 9. Polarization curves for the GC electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and of different amount of OA in solution; data obtained at 0.5 mV s-1 scan rate.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 10. Current-potential curves for electrooxidation of OA: effect of OA concentration. The slopes are from 110-121 mV/Dec.

However, the results obtained by the authors in [1] that have obtained a slope value in the Tafel plots to about 220mV/Dec respect to the slope value reported for GC in Figure 10. This discrepancy may be due to the change of the electrode surface state for the increase in the anodic potential. With the results obtained is possible realize a comparison with the new anodic materials and study the mechanisms of electrochemical oxidation in them.

BDD electrodes The same measurements (polarization curves) were also carried out for BDD electrodes: F-BDD, strongly oxidized BDD and mildly oxidized BDD.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 11 represents the polarization curves obtained at F-BDD electrode. From these measurements, it is possible to study the dependence between the current- potential and the OA concentrations. The current density increases at higher concentrations of OA; nevertheless these increments are considered insignificants because it presents a lower electroactivity toward the OA species during the oxidation respect to other BDD electrodes. This lower electroactivity observed could be due to scarce interaction of the F-terminated surface with hydrogen and hydroxyl radicals implies, in turn, a scarce interaction with water molecules: a hydrophobicity of the electrode surface otherwise expected on the basis of the properties of C-F bond. A primary effect of this would then be an easier organization of the solvent molecules around the reacting species and consequently a slower electron transfer;

Chapter 7. Kinetic Mechanism of the Electroxidation

In this context, the slope of Tafel plots was obtained for this anodic material (Figure 12) and forms these curves; the electrochemical reaction order was calculated as shown Figure 13. The values of the slope were to about 160 mV/Dec.

Figure 12. Current-potential curves for electrooxidation of OA at F-BDD electrode: effect of OA concentration. The slopes are from 160-170 mV/Dec.

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 13. Determination of the reaction order for F-BDD electrode

Electrooxidation of the OA under similar conditions were realized at strongly oxidized BDD. As can see from Figure 14, the character of polarization curves at this electrode is similar to latter BDD electrode. Polarization curves showed an increase on the current density values when was increased OA concentration respect to data obtained at F-BDD electrode. Also, the o.e.r is attained at similar potentials in both materials (Strongly Oxidized-BDD and F-BDD) to about 2.2 V vs SCE;

Chapter 7. Kinetic Mechanism of the Electroxidation

Figure 14. Polarization curves recorded at different OA concentrations at strongly oxidized BDD electrode. Scan rate 0.5 mV s-1;

From data obtained in the Tafel plots were possible to obtain the electrooxidation reaction order as shown Figure 15 and 16, respectively. In fact, these results reveal a similar surface activity toward the species of OA oxidation. Also a moderate increase in electroactivity by strongly oxidized BDD electrode respect to F-BDD was attained.

Chapter 7. Kinetic Mechanism of the Electroxidation

Zoom 1 Zoom 2 Figure 15. Current-potential curves for electrooxidation of OA at strongly oxidized BDD electrode: effect of OA concentration. Zoom 1) from 2.3 to 2.5 V and 2) from 2.0 to 2.3 V. The slopes are from 200-245 mV/Dec.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Figure 16. Determination of the reaction order for strongly oxidized BDD electrode at two potential values according to Tafel plots areas.

Chapter 7. Kinetic Mechanism of the Electrooxidation

evolution and these results are in agreement with those obtained with BDD anodes [12].

Figure 17. Polarization curves recorded at different OA concentrations at mildly oxidized BDD electrode. Scan rate 0.5 mV s-1; electrolyte 1 M HClO4. Inset: Magnification of range from 2.0 to 2.4 V vs. SCE.

Among the factors that can influence the electrochemical behavior of boron-doped diamond electrodes, the crystallographic structure [33, 34], the surface functional groups [35-37], the boron doping level [38, 39], and the presence of non-diamond amorphous carbon species (sp2) [33, 40, 41] are probably the most important [42].

The preparation conditions and the surface treatment have a strong effect on the electrochemical behavior of diamond electrodes [36]. Many kinds of surface

Chapter 7. Kinetic Mechanism of the Electrooxidation

treatment have been used so far, and the resulting changes in electrode properties have been investigated.

Figure 18. Current-potential curves for electrooxidation of OA at mildly oxidized BDD electrode. The slopes are from 180-210 mV/Dec.

Chapter 7. Kinetic Mechanism of the Electrooxidation

species that can be more or less numerous on the diamond surface, depending on the preparation conditions. The importance of non-diamond species for electrode activity is not completely understood as yet. Nevertheless Duo et al. [44] have clarified the influence of surface heterogeneity on the electronic properties of the diamond material, and particularly the importance of the sp2 species in electron- transfer reactions.

Figure 19. Determination of the reaction order for mildly oxidized BDD electrode at two potential values.

The studios realized by authors in [44] have revealed that the electrochemical response of the electrodes varied as a function of surface treatment. Untreated electrodes (BDD as-grown or ?fresh electrode?) exhibited the highest surface activity. The voltammogram revealed in [44] many surface redox transitions represented by overlapping current waves in the voltammetric curves. The electrode response decreased significantly after mildly anodic polarization. In fact, the

Chapter 7. Kinetic Mechanism of the Electrooxidation

voltammetric response of mildly oxidized BDD electrodes revealed a decrease in surface activity, and only one well-defined irreversible pair of peaks was still visible probably attributable to surface redox functions. Also, the peak current decreased and the irreversibility of the surface redox couple increased with increasing severity of the treatment, where the BDD electrodes were polarized under severe conditions were obtained by 576 h of anodic polarization at 1 A cm-2 of as-grown electrodes in 1M HClO4 at 40°C; being denominated BDD severe in this work. The latter BDD anode can be considered different to strongly oxidized BDD electrode used during this Chapter, but similar conditions in the oxidation surface permit a relative comparison.

According to results obtained with F-BDD, Strongly and Mildly Oxidized BDD, the decrease in surface activity is observed at different pre-treatment BDD electrodes in agreement with results obtained by [44], where the chemical and electrochemical properties of diamond electrodes were found to be strongly influenced by the surface treatment. A relatively mild polarization process was sufficient to transform the surface from hydrophobic (BDD as-grown) to hydrophilic (BDD mildly oxidized) without changes in the crystal shape and size. The electrochemical properties were also strongly modified. The voltammetric charge decreased probably due to the decrease in the concentration of active surface sites. After stronger oxidation processes the electrodes had experienced morphological changes involving both crystal size and crystal shape (BDD strongly oxidized). The surface became smoother, and the electrochemical activity measured in terms of voltammetric charge decreased strongly. For this reason, better results were obtained at mildly oxidized BDD during the oxidation of OA and good electrochemical activity is achieved.

Chapter 7. Kinetic Mechanism of the Electrooxidation

DSA electrodes In the present part DSA formulations has been studied, which could be an important result for comparison with different anodic materials above described.

Cyclic voltammetry The electrodes used were: Ti/IrO2-Ta2O5, Ti/IrO2-2SnO2, Ti/Ir0.67Ru0.33Sn2O6 and Ti/RuSn2O6. A thermo-decomposition method was employed to prepare oxide coatings. Prior to use, Ti plates were treated to support the oxide coating. The oxide coating was synthesized at variable compositions. The electrochemical surface characterization of these mixed oxide electrodes was carried out by linear sweep voltammetry between 0.15-1.15 V (vs SCE as reference). In consideration of the importance of the OA anodic oxidation a study of this reaction was also carried out making use of quasi-steady polarization data.

Figure 20. CV curves for the Ti/IrO2-Ta2O5 electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and with OA in solution.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Cyclic voltammograms for Ti/IrO2-Ta2O5 anode were recorded at a potential rate of 100 mV s -1 are shown in Fig.20. The addition of the organic substrate does not seem to have a significant effect on the shape of CV curves, with the exception of a very small increase of currents in the oxygen evolution potential range. The shape of the voltammograms is substantially the same, with increasing OA concentration.

CV experiments at Ti/IrO2-2SnO2 (Figure 21) clearly show that OA is electroactive at this electrode, its oxidation-taking place about 0.9 V vs. SCE before the oxygen evolution.

Figure 21. CV curves for the Ti/IrO2-2SnO2 electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and with OA in solution; data obtained at 50 mVs-1 scan rate.

Chapter 7. Kinetic Mechanism of the Electrooxidation

In Figure 22, the behavior of OA at a Ti/Ir0.67Ru0.33Sn2O6 electrode is reported; again, cyclic voltammograms have been recorded at scan rate of 100 mV s-1. A very interesting behavior was observed, where an important current shift was be evidenced when OA was added to the solution, compared with the curves recorded at Ti/IrO2-2SnO2 electrode. As clearly show the Figure 22, that OA is more electroactive at this electrode respect to other materials. In basis these results, is possible confirm the hypothesis realized in Chapter 4 where the electrooxidation of OA is strongly dependent of anodic material.

Figure 22. CV curves for the Ti/Ir0.67Ru0.33Sn2O6 electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and with different OA concentrations in solution.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Ti/Ir0.67Ru0.33Sn2O6 and Ti/IrO2-Ta2O5. In latter material does not observed a change with increasing OA concentration and it explain the poor elimination of OA respect to other materials researched in Chapter 4.

In this context, determination of the influence of the main components of the mixed- oxide films was realized. It was using an electrode that excludes Ta and Ir components.

Figure 23. CV curves for the Ti/RuSn2O6 electrode, in the presence of the pure supporting electrolyte (1 M HClO4) and with different OA concentrations in solution.

Chapter 7. Kinetic Mechanism of the Electrooxidation

can be used as a parameter to evaluate the electrocatalytic activity for the anodic oxidation of OA. So far, the inclusion of iridium as part of composition is very important, as well the studio of the role of Ta in the electrode composition during oxidation process. In addition, is essential to remark the importance of nominal composition of binary and/or ternary oxide mixtures because the improvements in the oxidation are dependent of this parameter.

Quasi-steady state The kinetics of the electrooxidation of OA was studied analyzing quasi-steady polarization curves. As shown in Fig. 24 the general features of the Potential vs. J curves are not significantly affected by the OA.

Figure 24. Polarization curves recorded at different OA concentrations at Ti/IrO2-Ta2O5 electrode. Scan rate 0.5 mV s-1; electrolyte 1 M HClO4.

Chapter 7. Kinetic Mechanism of the Electrooxidation

behavior was recorded (Figure 25). This result may be not surprising on the basis of the previously discussed CV overview, which indicated a too weak interaction between OA species and the mixed-oxide surface across the whole potential sweep range. This includes the initial part of the oxygen evolution region, where also no changes are induced by the presence of the organic molecule. The reaction order by OA, estimated on the basis of data in Figure 25, amounted to approximately 0.5. But it did not given important information about the oxidation mechanism due to anomalous behavior.

Figure 25. Current-potential curves for electrooxidation of OA at Ti/IrO2-Ta2O5 electrode. The slope at 750mM is 238 mV/Dec.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Experiments were also carried out at Ti/IrO2-2SnO2 electrode. Actually, as shown in Figure 27, the electrochemical reactivity of OA towards incineration is higher than at Ti/IrO2-Ta2O5; in addition, as indicated by CV in Fig. 21, the onset of OA oxidation is located at lower potentials, compared with that of oxygen evolution, which points out that the occurrence of the oxidation of the organic substrate may take place without participation of adsorbed hydroxyl radicals, and might even shift their formation towards more positive potentials.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Figure 27. Polarization curves recorded at different OA concentrations at Inset: Magnification of OA interaction before o.e.r.

From data obtained in the Tafel plots (Figure 28) were possible to obtain the electrooxidation reaction order to about 0.75. In fact, these results reveal a strong surface activity toward the species of OA oxidation. The slopes of the Tafel Plots at different OA concentrations were 100, 110, 118 and 157 mV/Dec for 100, 250, 500 and 750 mM, respectively.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Figure 28. Current-potential curves for electrooxidation of OA at Ti/IrO2-2SnO2 electrode.

On the other hand, the electrooxidation experiments of OA at Ti/Ir0.67Ru0.33Sn2O6 anodes were also carried out, exploring the behavior at this material. As shown the Figure 29, the more strong interaction was observed before the oxygen evolution reaction respect to behavior achieved at Ti/IrO2-2SnO2. In fact, CV results indicate a higher reactivity of OA at Ti/Ir0.67Ru0.33Sn2O6, than all DSA-type electrodes. However, the reactivity of the substrate is clearly conditioned by the state of the electrode surface Figure 22. It is probably connected with the kind of mechanism described in the literature.

Chapter 7. Kinetic Mechanism of the Electrooxidation

The Tafel plots where the slopes are to about 90-95mV/Dec for different OA concentrations (Figure 30). Under these conditions, the mechanism proposed by Johnson et al. [5] and Orts et al. [25] (Equations 1 and 2) should be expected.

Recently, Petrii and co-workers have proposed that the OA electro-oxidation at Pt electrodes could take place through two parallel paths [11]. One of them involves a slow electron-transfer step between adsorbed OA and electrode surface, while the other would be kinetically controlled by the surface reaction between adsorbed OA and hydroxyl radicals. The latter would dominate at low, the former at high electrode potentials. Thus, is possible that the OA electrooxidation at DSA-type anodes could take place through two parallel paths as proposed by Petrii for the case of Pt; excluding the Ti/IrO2-Ta2O5 electrode due to lower catalytic activity observed.

Figure 29. Polarization curves recorded at different OA concentrations at Ti/Ir0.67Ru0.33Sn2O6 electrode. Scan rate 0.5 mV s-1; electrolyte 1 M HClO4.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Figure 30. Current-potential curves for electrooxidation of OA at Ti/Ir0.67Ru0.33Sn2O6 electrode. The slope to around 90-97 mV/Dec.

Chapter 7. Kinetic Mechanism of the Electrooxidation

Conclusions According to results described in this Chapter, the model proposed by Comninellis [12, 45], on a general basis the electrochemical incineration of organics at a given electrode can take place at satisfactory rates only in the potential region where oxygen evolution also takes place, electrosorption of hydroxyl radicals and their reactivity being the decisive factor. This assumption is valid for the Ti/Pt, Pt and GC electrodes; where oxalic acid is relatively low reactivity towards hydroxyl radicals [47].

The case of BDD electrodes is also quite interesting, in that the very poor adsorptive ability of their surface and their great stability toward oxidation allow the reaction to take place with reactant and intermediates in a non-adsorbed state. The hydroxyl radicals may react with OA but also recombine, eventually producing molecular oxygen, justifies the yield results. Alternately, the decrease in the surface activity when the electrode received a pre-treatment and/or surface modification (in the case of F-BDD) plays an important role. These results are in agreement with results obtained by Duo et al. [44], where the chemical and electrochemical properties of diamond electrodes are strongly influenced by the surface treatment. In the case, F- BDD, its lower electroactivity observed could be due to scarce interaction of the F- terminated surface: hydrophobicity.

In contrast, the results discussed for the DSA-type electrodes substantially confirm the hypothesis, in substantial agreement with most of the literature, where its electrosorption is in fact assumed as a necessary prerequisite for a fast electro- oxidation to take place [1, 5, 6, 11, 13, 22]. Nevertheless, the electrocatalytic activity showed at Ti/IrO2-2SnO2, Ti/Ir0.67Ru0.33Sn2O6 and Ti/RuSn2O6 electrodes suggest the assumption about two parallel mechanisms of oxidation of OA because the increase in the surface activity was observed. The opposite occurrence is observed at the IrO2-based electrode, where the lack of any interaction between OA and

Chapter 7. Kinetic Mechanism of the Electrooxidation

electrode surface, and the strongly favored o.e.r,; strongly inhibit the reaction of interest.

This feature inevitably makes the rate of OA incineration strongly dependent on the nature of the electrode material, as observed, for instance, by Sargisyan and Vasil'ev [9]; they reported the results for different anodic materials as shown the Figure 31.

Figure 31. Polarization curves of electrooxidation of OA in H2SO4 on different electrodes: 1-Pd, 2-Au, 3-Pt, 4-DSA, 5-Rh, 6-Ir and 7-Glassy carbon (CY-12). [OA] = 0.1M.

As can see from this Figure, in H2SO4 the character of polarization curves of the electrooxidation of OA on the Pt, Ir, Rh, Pd and DSA are completely identical.

Chapter 7. Kinetic Mechanism of the Electrooxidation

On the other hand, the investigations realized in this chapter are resumed in the Figure 32, where the electrocatalytic activity is higher to DSA electrodes than BDD electrodes. In addition, Pt and GC electrodes, apparently in an intermediate position between DSA and BDD electrodes. Finally these results have remarked the importance of nature of the electrode material for the pollutants oxidation.

Figure 32. Tafel plots of electrooxidation of OA in HClO4 on different electrodes. [OA] = 750mM.

Chapter 7. Kinetic Mechanism of the Electrooxidation

References [2] C. A. Martínez-Huitle, S. Ferro, and A. De Battisti, Electrochim. Acta, 49 (2004) 4027.

[4] S.E.S. El Wakkad, S.E. Khalafalla, A.M. Shams El Din, Egypt. J. Chem., 1 (1958) 23.

[5] J.W. Johnson, H. Wroblowa, J.O'M. Bockris, Electrochim. Acta, 9 (1964) 639.

[6] J.W. Johnson, S.C. Mueller, W.J. James, Trans. Faraday Soc., 67 (1971) 2167.

[10] M.J. Chollier-Brym, F. Epron, E. Lamy-Pitara, J. Barbier, Catal. Today, 48 (1999) 291.

[12] D. Gandini, E. Mahè, P.A. Michaud, W. Haenni, A. Perret, Ch. Comninellis, J. Appl. Electrochem., 30 (2000) 1345.

[16] G. Horanyi, D. Hegedüs, E. M. Rizmayer, J. Electroanal. Chem., 40 (1972) 393.

Chapter 7. Kinetic Mechanism of the Electrooxidation

[20] R. Albalat, E. Gomez, M. Sarret, E. Valles, Monatsh. Chem., 120 (1989) 651.

[21] Z. Alaune, R. Mazeikiene, Liet. TSR Mokslu Akad. Darb. Ser. B., 2 (1987) 11.

[24] N.V. Smirnova, O.A. Petrii and A. Grzejdziak, J. Electroanal. Chem., 251 (1988) 73.

[25] J. M. Orts, J. M. Feliu, A. Aldaz, et al., J. Electroanal. Chem., 281 (1990) 73.

[32] G. Sine, L. Ouattara, M. Panizza and Ch. Comninellis, Electrochem.Solid- State Lett. 6 (9) (2003) D9.

Chapter 7. Kinetic Mechanism of the Electrooxidation

[35] K. Hayashi, S. Yamanaka, H. Okushi and K. Kajimura, Appl. Phys. Lett., 68 (1996) 376.

[38] R .J. Zhang, S. T. Lee and Y. W. Lam, Diamond Relat. Mater., 5 (1996) 1288.

[39] C. Levy-Clement, F. Zenia, N. A. Ndao and A. Deneuville, New Diamond Front. Carbon Technol., 9 (1999) 189.

Chapter 8. Hydroxylation

Comparison between Chemical and Electrochemical hydroxylation processes

Introduction Synthetic BDD films represent a new electrode material, and have recently received great attention [1-6]. The wide potential window and the high anodic stability of BDD permit its applications in electroanalysis [7-9] and in the preparation of powerful oxidants [10, 11]. Also, several studies have been realized, with the goal of developing applications in the electrochemical oxidation of organics for wastewater treatment [12-14] and for electrosynthesis [15]. Comninellis et al. [16-18] have found that the electrochemical oxidation of most organics in aqueous media occurs only at high potentials, with a concomitant evolution of O2; in addition, the nature of the anodic material strongly influences the process selectivity and efficiency. Hence, a model for the anodic oxidation of organics in acidic media, including competition with oxygen evolution, has been proposed [16-18]. In this model, the electrochemical and chemical reactivity of the adsorbed hydroxyl radicals are explained, underlining the important role of the nature of the anode material.

Depending on the interactions between hydroxyl radicals and the electrode surface, anodes can be distinguished as ?actives? or ?non-actives? materials [17, 19]. Among the particular characteristics of BDD, its chemical inertness is well recognized; for this reason, this material is considered as an ideal ?non-active? electrode, on which organics oxidation and oxygen evolution take place through the formation of weakly adsorbed and very reactive hydroxyl radicals.

Chapter 8. Hydroxylation

approach is not the unique form to produce them with the purpose of eliminating organic pollutants: alternative methods, where hydroxyl radicals are generated with the participation of chemical reagents or promoters, have been used for the wastewater treatment, comprising i.e. the Fenton reaction and H2O2/UV process.

Oxidation using liquid hydrogen peroxide (H2O2) in the presence of native or supplemental ferrous iron (Fe2+) produces the so-called Fenton reagent, which yields free hydroxyl radicals (·OH) [20]. This strong and nonspecific oxidant can rapidly degrade a variety of organic compounds. Fenton reagent oxidation is most effective at very acidic pH (e.g., at pHs from 2 to 4), becoming ineffective under moderate or strongly alkaline conditions. In general, reactions are extremely rapid and follow a second-order kinetics. The photolysis of hydrogen peroxide with ultraviolet radiations (H2O2/UV) represents another way to produce hydroxyl radicals [21], and this approach has been effectively used by several researchers, since the sixties, to oxidize various organic substances in water. Recently, commercial units employing this process have been developed for on-site oxidation of organic pollutants in groundwater (Chapter 2).

In this Chapter, we present an investigation on the electrochemical hydroxylation of salicylic acid, which has been chosen as a model organic compound, by hydroxyl radicals produced during the oxidation of water at BDD electrodes. This electrochemical synthesis has been compared to the chemical hydroxylation that can be realized by both the Fenton reaction and the H2O2/UV process.

This studio was realized in scientific collaboration between University of Ferrara and Institute of Chemical Engineering of the Swiss Federal Institute of Technology - Lausanne, Switzerland supervised by Professor Comninellis.

Chapter 8. Hydroxylation

Experimental Cell and electrodes BDD films were synthesized at CSEM (Neuchâtel, Switzerland) by the hot filament chemical vapor deposition technique (HF-CVD) on conductive p-Si substrate (Siltronix). The temperature of the filament was adjusted from 2440 to 2560 °C and that of the substrate was monitored at 830 °C. A 1% mixture of methane in hydrogen, containing also 1-3 ppm of trimethylboron, was used as the reactive gas, and supplied to the reaction chamber at a flow rate of 5 l min-1, to give a growth rate of 0.24 µm h-1 for the diamond layer. The obtained diamond film had a thickness of about 1 µm and a resistivity between 10 and 30 m? cm. The HF-CVD process optimized at CSEM produces columnar, pinhole-free, randomly textured and polycrystalline diamond films [6, 14].

Electro-oxidation experiments were carried out in an undivided electrochemical flow cell with parallel plate electrodes (Figure 1A); circular electrodes were used, having a diameter of 9 cm and exposing to the aqueous solution a geometrical area of 63.5 cm2; the interelectrode gap was set to 10mm. A zirconium disc was used as the cathode, and the electrochemical cell was provided with inlet and outlet for electrolyte circulation, which was obtained by using a peristaltic pump. The electrolyte was stored in a thermoregulated glass tank and circulated through the electrolytic cell (Figure 1B), at a flow rate of 165 dm3 h-1; under these conditions, the mass transfer coefficient (determined using the ferri/ferrocyanide couple) was 2.0 × 10-5 m s-1.

Chapter 8. Hydroxylation

Figure 1. Undivided electrochemical cell for the oxidation of organics at a BDD electrode. (A) Set-up used: (1) thermoregulated reservoir; (2) electrochemical cell; (3) power supply; (4) pump. (B) Electrochemical cell: (1) Inlet; (2) BDD anode; (3) Zr cathode; (4) electrolysis compartment; (5) and (6) electrical contacts; (7) outlet.

Chemical oxidation processes Fenton experiments were realized on a batch reactor with 250 ml of SA solution, and performing the process with different values for the [H2O2]/[SA] ratio. Subsequently, a specific ratio was selected for allowing the comparison with the other oxidation methods. In the case of the H2O2/UV process, the reaction was carried out in a photoreactor provided with a low-pressure mercury lamp, which was positioned in the center of the reactor.

Chapter 8. Hydroxylation

column and using a mixture of a 1% phosphoric acid solution and acetonitrile (80:20) mobile phase, at a flow rate of 1 ml min-1.

Electrochemical oxidation Galvanostatic assays were performed at 0.1 A and at 25 °C. During the electrolysis experiments, the oxidation products were analyzed by using the same HPLC apparatus previously described.

Solutions Different solutions of SA (7.25 mM, Fluka reagent) were prepared, using distilled water: the electrochemical investigation was carried out both in 1M HClO4 and in NaClO4, adjusting the pH to ~4.3 with a CH3COONa / CH3COOH pH buffer. Chemical oxidations, on the other hand, were performed in buffered solutions as well as in 0.1M NaOH. A 3% hydrogen peroxide was used as received, and Fe2+ additions were made starting from a 0.6M solution of (NH4)2Fe(SO4)2·6H2O.

Results and discussion Hydroxylation Process Fenton Reaction According to the Fenton reaction [22], the generation of hydroxyl radicals takes place by adding H2O2 to a Fe+2 solution (Equation 1): 2+ + ? 3+ + ? + ? Fe H O Fe OH OH (1) 22

While this reactant is an attractive oxidative system for wastewater treatments, the Fenton process requires a strict pH control because better yields are obtained at pH values ranging between 3.0 and 5.0.

Chapter 8. Hydroxylation

Figure 2. Hydroxylation of SA through the Fenton reaction (pH adjusted to ?4.0 with NaOH).

Figure 2 shows results of a reaction between SA and the Fenton reagent, in aqueous solution adjusted at a pH of 4 with NaOH; the production of 2,3-DHBA was relatively larger than that of 2,5-DHBA, and a complete conversion of SA into intermediates was attained.

Adjusting the pH at a value of ~4 with the acetate-based buffer solution, the yield of 2,5-DHBA was moderately increased, in comparison with the previous conditions, as shown in Figure 3; however, both experimental conditions favored the synthesis of 2,3-DHBA over 2,5-DHBA.

Chapter 8. Hydroxylation

Figure 3. Replication of the experimental investigation of Figure 2, but controlling the pH through the CH3COOH/CH3COONa buffer solution (pH ?4.3).

Different values for the [H2O2]/[SA] ratio were studied (the following ratio were considered: 2, 4, 6, 7, 8, 10, 12 and 14) and the best results were attained at a ratio equal to 7; accordingly, this specific condition was chosen to compare the results obtained with the other processes.

H2O2/UV process This process is obtained by adding H2O2 to the pollutant solution and irradiating with UV light having wavelengths smaller than 280 nm; this causes the homolytic cleavage of H2O2, as schematized in Equation 2 [23]:

Chapter 8. Hydroxylation

H O ??? 2OH ? (2) hv 22

The reaction was carried out at 25 °C, with an initial concentration of SA of 7.25 mM; as for the H2O2 addition, the amount was chosen as to obtain a [H2O2]/[SA] ratio equal to 7, as done in the case of the Fenton reaction. In the hydroxylation of SA, using NaOH for adjusting the pH, the main product was again the 2,3-DHBA, while catechol and 2,5-DHBA were produced in smaller amounts. In addition, the complete elimination of SA was not attained (Figure 4). Adjusting the pH with the buffer solution, the production of 2,3-DHBA was moderately increased, as was the formation of 2,5-DHBA, which reached a concentration of about 150 ppm (Figure 5). Interestingly, the complete oxidation of SA was finally obtained, but the requested time was not comparable with that necessary under the previously discussed experimental conditions.

Figure 4. H2O2/UV process for the hydroxylation of SA: solution pH adjusted to ?4.0 with NaOH.

Chapter 8. Hydroxylation

Figure 5. Replication of the experimental investigation of Figure 4, but controlling the pH through the CH3COOH/CH3COONa buffer solution (pH ?4.3).

Electrochemical hydroxylation Salicylic acid was chosen as the test molecule because of its ability to act as a spin- trap for free hydroxyl radicals in biological systems. Marselli et al. [19] have used this compound to realize the spin trapping of hydroxyl radicals formed in close proximity of a BDD surface electrode, obtaining good results. SA reacts with the ·OH radicals to form dihydroxylated aromatic compounds, the most important products of this reaction being the 2,3- and 2,5-dihydroxybenzoic acids (DHBA);

the formation of catechol, due to a decarboxylation process, is also possible, as schematized in Figure 6. To investigate the electrochemical hydroxylation at the BDD anode, several experiments were realized, at a current intensity of 0.1 A.

Chapter 8. Hydroxylation

Figure 7 shows the results obtained electrolyzing a SA solution prepared in 1M HClO4; the main product was the 2,5-DHBA, while 2,3-DHBA and catechol were produced in minor amounts.

Chapter 8. Hydroxylation

practically the same amount of 2,5-DHBA but, as shown in Figure 8, the concentration of 2,3-DHBA was significantly increased, with respect to the previous conditions.

Figure 7. Electrochemical hydroxylation of SA, carried out at a BDD electrode in 1M HClO4.

Chapter 8. Hydroxylation

Figure 8. Replication of the experimental investigation of Figure 7, but controlling the pH through the CH3COOH/CH3COONa buffer solution (pH ?4.3, supporting electrolyte: NaClO4).

Comparison between the electrochemical and chemical hydroxylation In the chemical hydroxylation through the Fenton and H2O2/UV processes, the time requested to attain a complete elimination of SA was substantially different, a shorter time being requested by the Fenton reaction. On the other hand, the chemical attack exerted by H2O2/UV proved to be unable to convert all the SA into DHBA and pertaining metabolites, whereas the oxidation of SA and products is complete under the conditions of the Fenton reaction. Concerning the electro-chemical hydroxylation, 11-12 hours of electrolysis were requested, under the two

Chapter 8. Hydroxylation

investigated pH conditions, only to convert the SA into the different metabolites (a charge consumption between 1.4 and 1.6 Ah dm-3 is needed).

While the chemical oxidation of SA led to the production of 2,3- and 2,5-DHBA, at both investigated pH, the electrochemical approach gave analogous results only when carried out under buffered conditions; in acidic media, on the contrary, only the 2,5-DHBA could be measured, the 2,3-geometric isomer being formed only at the level of traces.

Chapter 8. Hydroxylation

works [28, 33, 41-43] dealing with the Fenton reaction, both the products of the hydroxylation of SA were obtained: the preferential formation of 2,3-DHBA was justified suggesting an interaction between Fe2+ and the OH group of the molecule (or the formation of ferryl ions, which interact with the hydroxylated substrate, often giving complexes). Mason [44] postulated the existence of a ferryl specie (FepO++) as the active intermediate, in which iron is complexed by a chelating structure occupying up to five of the six available coordination positions. On the other hand, Groves et al. [45, 46] suggested a mechanism to describe the hydroxylation of cyclohexanol, on account of the formation of stereoselective hydroxylation products.

Similar interactions were observed by Lunak et al. [47] and Lang et al. [48] during the photochemical hydroxylation of SA, where the formation of 2,3-DHBA resulted to be favored. According to the above literature, the attack of ·OH radicals to the SA molecule may take place preferentially in the 3-position, during the Fenton reaction, because of the orienting properties of the ring substituents. On the other hand, the possible formation of the ferryl ion, which subsequently interacts with the OH group of the SA molecule according to Groves [45, 46] and Mason [44], may provide a method of specifically transferring an O atom, favoring, in this way, the hydroxylation at the ortho position.

Chapter 8. Hydroxylation

photodissociation of hydrogen peroxide into two hydroxyl radicals, which subsequently attack the ring. Their results on substrate sensitized reactions [50] showed that, in the hydroxylation of SA by photolysis of H2O2 (254 nm) and/or SA excitation (313 nm), the isomer ratio Y= [2,3-DHBA]/[2,5-DHBA] ranged from 1.12 to 2.46. In addition, the roles of pH, H2O2 concentration and irradiation time were studied, and similar yields were obtained, the favored product being always the ortho-hydroxylated product.

Finally, referring to the electrochemical hydroxylation, some experiments have been realized [28, 33, 41] and both the 2,3- and 2,5-dihydroxybenzoic acids were obtained. Oturan et al. [33, 41] used a Pt anode and carried out the investigations in the presence of Fe2+ in solution: the production of 2,5-DHBA was less important than that of 2,3-DHBA, possibly because of the iron effects (as previously discussed commenting the Fenton reaction). Alternatively, an interaction of the organic substrate with the Pt electrode surface can play the same role of the iron ions, thus favoring the insertion of the OH group in the ortho position, with respect to the pre- existing one. On the contrary, the hydroxylation of SA at a BDD electrode surface allows the obtainment of 2,5-DHBA, which could be explained assuming two events: 1) the characteristic of ?non-active? electrode of BDD, and 2) the moderate steric hindrance of SA.

Chapter 8. Hydroxylation

BDD has an inert character and shows very weak adsorptive properties. These qualities make it an ideal ?non-active? electrode [19], at which both the oxidation of organics and the oxygen evolution reaction take place via the formation of hydroxyl radicals, which form a sort of ?reaction cage? in the very close proximity of the electrode surface. Since the formation of 2,5-DHBA, due to the hydroxylation of SA, cannot be due to an adsorption of SA onto the electrode surface, the preferred attack of hydroxyl radicals may be possibly justified by the moderate steric hindrance generated by the presence of both the COOH and OH groups in the molecule [52].

Conclusions We showed in this Chapter that, during all investigated processes; hydroxyl radicals react with SA leading to the formation of dihydroxylated products (2,3- and 2,5- DHBA). The Fenton reaction allows a complete conversion of the test organic molecule, and a detailed investigation is possible only when a proper [H2O2]/[SA] ratio is chosen. On the contrary, under analogous conditions ([H2O2]/[SA] =7), the The hydroxylation selectivity depends on the experimental approach: 2,3-DHBA is the main product when hydroxyl radicals are chemically produced, while 2,5-DHBA is preferentially formed when electrochemistry at BDD electrodes is investigated, supporting the idea that ·OH radicals are important intermediates for the organic oxidation at this electrode material.

Finally, the electrochemical hydroxylation method for the production of 2,5-DHBA (also called gentisic acid) may represent a promising synthetic pathway: in fact, this molecule has a broad spectrum of biological activity such as anti-inflammatory, anti- rheumatic and anti-oxidant properties, which are also independent of the action of SA [53].

Chapter 8. Hydroxylation

References [2] N. Vinokur, B. Miller, Y. Avyigal, R. Kalish, J. Electrochem. Soc. 146 (1999) 125.

[5] Y. Yano, D.A. Tryk, K. Hashimoto, A. Fujishima, J. Electrochem. Soc. 145 (1998) 1870.

[7] F. Bouamrane, A. Tadjeddine, R. Tenne, J.E. Butler, R. Kalish, C. Levy- Clement, J. Phys. Chem. B. 102 (1998) 134.

[9] M. Panizza, I. Duo, P.A. Michaud, G. Cerisola, Ch. Comninellis, Electrochem. Solid-State Lett. 3 (2000) 429.

[10] P.A. Michaud, E. Mahè, W. Haenni, A. Perret, Ch. Comninellis, Electrochem. Solid-State Lett. 3 (2000) 77.

[11] M. Panizza, I. Duo, P.A. Michaud, G. Cerisola, Ch. Comninellis, Electrochem. Solid-State Lett. 3 (2000) 550.

[12] J.J. Carey, W. Henrietta, C.S. Christ Jr., S.N. Lowery, US Patent 5 399 247 (1995).

[14] A. Perret, N. Skinner, Ch. Comninellis, D. Gandini, Electrochem. Soc. Proc., 32 (1997) 275.

[15] J. Iniesta, P.A. Michaud, M. Panizza, Ch. Comninellis, Electrochem Commun. 3 (2001) 346-351.

Chapter 8. Hydroxylation

[18] G. Foti, D. Gandini, Ch. Comninellis, Current Topics in Electrochemistry, 5 (1997) 71.

[23] R. Andreozzi, V. Caprio, A. Insola, R. Marotta, Catalysis Today 53 (1999) 51.

[27] G. Zhi-Hu, F. Yamaguchi, A. Itoh, A. Kitani, K. Sasaky, Electrochim. Acta 37 (1992) 345.

[30] R. Richmond, B. Halliwell, J. Chauhan, A. Darbre, Anal. Biochem. 118 (1981) 328.

Chapter 8. Hydroxylation

[47] K. Lang, D.M. Wagnerova, J. Brodilova, Collect. Czech. Chem. Commun. 59 (1994) 2447.

[48] S. Lunak, J. Muzart, J. Brodilova, Collect. Czech. Chem. Commun. 59 (1994) 905.

[49] Y. Ogata, K. Tomizawa, Y. Yamashita, J. Chem. Soc. Perkin Trans. 2 (1980) 616.

[50] K. Lang, J. Brodilova, S. Lunak, Collect. Czech. Chem. Commun. 61 (1996) 1729.

[53] J.F. Cunha, F.D. Campestrini, J.B. Calixto, A. Scremin, N. Paulino, Brazilian J. Med. and Bio. Res. 34 (2001) 381.

Chapter 9. General Discussion

General Discussion In this chapter, we will summarize the principal studies of this work. The first part concerns at direct electrochemical oxidation of carboxylic acids, which are studied by different electrode materials.

Then we will discuss the advantages of the flow cell, a new electrochemical cell with well establishing hydrodynamic conditions.

On the other hand, the role of the halides in the indirect electrochemical oxidation As the mechanism of the OA electroxidation depends strongly on the nature of the electrode material; which was studied at highly conductive, boron-doped diamond (BDD) electrodes, with both oxygen and fluorine at their surface. Also, these kinetic mechanism studies were carried out at DSA electrodes, Pt and glassy carbon.

The reactivity of hydroxyl radicals was underlined by both chemical (Fenton and H2O2/UV) and electrochemical (at BDD electrode) hydroxylation of salicylic acid and with the study of the intermediates distribution.

Finally, some perspectives will be proposed about the electrochemical oxidation process.

Chapter 9. General Discussion

Electrochemical Incineration of Oxalic Acid The electrochemical oxidation of Oxalic Acid (OA) was investigated under galvanostatic conditions at different electrode materials (Chapter 4), as a function of applied current density and temperature. A complete elimination of the organic reagent has been achieved at Pt, Au, PbO2 and BDD electrodes, while only a minor attack takes place at the IrO2-Ta2O5 anode. The influence of the anode material on the elimination of OA seems to be very important, PbO2 being the electrode at which the best incineration efficiency has been attained in initial stages (Figure 1).

Figure 1. Decrease of OA content in solution, at different electrodes, as a function of charge, at 600 A m-2 of applied current density.

Chapter 9. General Discussion

this material is considered as an ?active electrode? for oxygen evolution, the use of smaller values of current density could be suggested [1].

Figure 2. Decrease of OA content in solution, at different electrodes, as a function of charge, at at 600 A m-2 of applied current density and 60°C of temperature.

Nevertheless, Pt electrode increases in the elimination efficiency was achieved when the temperature was increased (Figure 2). In conclusion, for the electrochemical elimination of OA from aqueous solutions follows the sequence: PbO2 > Pt > BDD ? Au > IrO2, depending on the current density and temperature.

Chapter 9. General Discussion

Electrochemical Incineration of Tartaric Acid In the Chapter 4, the electrochemical oxidation of Tartaric Acid (TA) at Ti/PbO2, Ti/Pt, Pt and HBDD anodes has been studied. A complete elimination of the organic reagent has been achieved at PbO2 and HBDD electrodes, while only a minor attack takes place at the Pt anodes. The influence of the anodic material on the elimination of TA seems to be very important, Ti/PbO2 being the electrode at which the best incineration efficiency has been obtained. Ti/PbO2 anodes exhibit good electrocatalytic properties towards the oxidation of TA due to the interactions between Pb(IV) sites and carboxyl groups. In acidic and alkaline media, the percentage of TA elimination was minimum at Pt electrodes. In this case, changes seems to generate active sites at the PtOx surface, which are no longer suitable for TA adsorption and/or oxidation intermediates, impeding the anodic oxidation of the organic substrates.

The electrochemical elimination of TA from aqueous solutions follows the sequence: PbO2 > HBDD > Pt anodes, and depend on the applied current density.

Electrochemical Incineration of Oxalic Acid. Reactivity and engineering parameters In Chapter 5, a new type of electrochemical cell, which presents some advantages, was developed. From the characterization experiments, these have possible to establish the hydrodynamic parameters; and predict the mass transfer coefficient.

We obtained a good correlation with the Leveque equation. As explained in the conclusions part, the cell presents local turbulences. This electrochemical cell proved to be an efficient for bulk electrolysis, obtaining improvements in the OA electrooxidation.

Chapter 9. General Discussion

Anodic Oxidation of Tartaric and Oxalic acids in the presence of Halides From direct electrooxidation of organic compounds like Oxalic and Tartaric Acids (Chapter 4) at different anodic materials permitted to explore the mediated electrochemical oxidation at Pt electrode through addition of halides in solution (Chapter 6). According to the results, the efficiency of the incineration process does not strongly depend on applied current density, while it's strongly affected by halide nature and concentration (Figure 3).

Figure 3. Decrease of OA content in solution, at the Pt electrode, as a function of charge, at different current densities (25 °C).

Chapter 9. General Discussion

incineration is the common feature in the cases, which we have reported in Chapter 6, the mechanism behind this effect may be different. In the case of F- the anion may have the only role to change the stoichiometry and microstructure of the oxide film, inhibiting o.e.r. and consequently favoring the electrochemical incineration. In the case of Br-, the main factor seems to be the electrogeneration of strong oxidants (Br- based oxy-anions), mainly through volume reactions. The Cl- effect seems to be a hybrid of the two former cases, with possible electrogeneration of strong oxidants [2], but also with changes in the electrode film properties, which make it less active toward o.e.r.

Kinetic Mechanism of the Electroxidation of Oxalic Acid at different Electrode Materials From the results obtained in Chapter 7, the potential values and a current density were considered for the graphic representation of surface activity electrode toward the species of OA, are shown in Figure 4. Interestingly, from data by Sargisyan and Vasil'ev [3], the anodic oxidation of OA is completely inhibited at Os electrodes. Its rate decreases as follows: Pd ? Au ? Pt > Ir > Rh, exhibiting a minimum at glassy carbon (GC). The two extreme cases of Os and GC, with the optimal oxidation rate at Au, Pd and Pt, support the idea that, when adsorption of OA species is hindered, either by adsorbed hydroxyl radicals (Os surfaces) or by a scarce adsorptive ability of the electrode surface (the case of GC in ref. [3] or BDD, in ref. [4] and this Chapter), the rate of anodic oxidation of the substrate reaches minimum values.

Using these results and including other electrode materials was possible to extend this classification in basis to rate decrease as follows: Ti/IrO2-2SnO2 ? Ti/Ir0.67Ru0.33Sn2O6 > Ti/RuSn2O6 > GC > Ti/Pt ? Pt > mildly oxidized BDD > F-BDD ? strongly oxidized BDD.

Chapter 9. General Discussion

Figure 4. Graphic representation of the potential to electrode surface activity toward OA molecule at different electrodes.

Finally, the experimental data suggest that two OA oxidation processes occur in parallel (with and without participation of absorbed oxygen-containing species);

with the exception of BDD electrodes where mechanism occurs through the hydroxyl radicals may react with OA but also recombine, eventually producing molecular oxygen; due to that the BDD behave as a ?non-active? electrode. And it depends of the pre-treatment received and/or surface modification (in the case of F- BDD).

Chapter 9. General Discussion

Comparison between Chemical and Electrochemical hydroxylation processes The reaction between hydroxyl radicals and salicylic acid (SA) yields 2,3- and 2,5- dihydroxylated benzoic acid (2,3-DHBA and 2,5-DHBA, respectively), and when decarboxylation occurs, catechol is produced. The hydroxylation was performed at BDD electrode and the reaction was compared with chemical hydroxylations of SA The hydroxylation selectivity depends on the experimental approach: 2,3-DHBA is the main product when hydroxyl radicals are chemically produced, while 2,5-DHBA is preferentially formed when electrochemistry at BDD electrodes is investigated, supporting the idea that ·OH radicals are important intermediates for the organic oxidation at this electrode material [5].

Perspectives Indirect oxidation is still a viable technology for treating toxic o biorefractory pollutants although there are concerns about the formation of chlorinated intermediates in the case of using chlorine ions; being considerable the use of others (Br or F halides).

Direct anodic oxidation represents one of the simplest technologies in the pollutant mineralization provided the anode materials stable and have high overpotential of oxygen evolution. The investigation of various materials have showed that depending of pollutant is possible select the correct electrode as candidate for industrial application.

Chapter 9. General Discussion

References [2] F. Bonfatti, A. De Battisti, S. Ferro, G. Lodi, S. Osti, Electrochim. Acta, 46 (2000) 305.