Electrochemical Oxidation of Hydroquinone, Resorcinol, and Catechol

Aug 12, 2005 - In this sense, the direct oxidation is favored by the large oxygen evolution ... Carboxylic acids were monitored by HPLC using a Supelc...
3 downloads 0 Views 178KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 7234-7239

Electrochemical Oxidation of Hydroquinone, Resorcinol, and Catechol on Boron-Doped Diamond Anodes BENSALAH NASR,† GADRI ABDELLATIF,† PABLO CAN ˜ I Z A R E S , ‡ C R I S T I N A S AÄ E Z , ‡ JUSTO LOBATO,‡ AND M A N U E L A . R O D R I G O * ,‡ Department de Chimie Industrielle, Institut Supe´rieur des Sciences Appliquee´s et Technologie de Gabe`s, Universite´ de Gabe`s, 6072 Zrig, Gabe`s, Tunisie, and Department of Chemical Engineering, Facultad de Ciencias Quı´micas Universidad de Castilla La Mancha, Campus Universitario s/n, 13071 Ciudad Real, Spain

The electrochemical oxidation of aqueous wastes polluted with hydroquinone, resorcinol, or catechol on borondoped diamond electrodes has been studied. The complete mineralization of the organic waste has been obtained independently of the nature of each isomer. No aromatic intermediates were found during the treatment, and solely aliphatic intermediates (carboxylic acids C4 and C2, mainly) were detected in the three cases. Although as from the bulk electrolyses study no differences in the electrochemical oxidation of dihydroxybenzenes seem to exist, different voltammetric behavior between resorcinol and the other two isomers was obtained in the voltammetric study. Catechol and hydroquinone have a reversible quinonic form, and a cathodic reduction peak appears in their voltammograms. The characterization of the first steps in the electrochemical oxidation of the three dihydroxybenzenes showed the formation of a larger number of intermediates in the oxidation of catechol, although no carbon dioxide was detected in its oxidation. Conversely, the oxidation of resorcinol and hydroquinone lead to the formation of important concentrations of carbon dioxide. The nondetection of aromatic intermediates, even if small quantities of charge are passed, confirms that the oxidation must be carried out directly on the electrode surface or by hydroxyl radicals generated by decomposition of water.

Introduction Anodic oxidation with electricity-conductive diamond anodes is a new advanced oxidation process with many advantages as compared to other known chemical and photochemical processes. Diamond anodes surfaces allow one to produce large quantities of hydroxyl radicals from water electrolysis (1-3). The diamond surface does not interact with these radicals (exhibits a nonactive behavior), and, as a consequence, these radicals can only couple to form oxygen or oxidize the organic matter present in the waste. Several works * Corresponding author phone: 34902204100; fax: 34902204130; e-mail: [email protected]. † Universite ´ de Gabe`s. ‡ Campus Universitario s/n. 7234

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 18, 2005

have been published in the literature during the recent years concerning the oxidation of different compounds such as phenol (4-6), carboxylic acids (7, 8), 4-chlorophenol (9, 10), 3-methypyridine (11), benzoic acid (12), 2-naphthol (13, 14), polyacrylates (15), 4-chlorophenoxyacetic acid (16), amaranth dyestuff (17), chlorophenols (18, 19), nitrophenols (20, 21), and polyhydroxybenzenes (22). The results of these works show that the electrochemical oxidation with diamond anodes achieves a very high current efficiency as compared to other electrochemical treatments, and very high conversions of the organic carbon into carbon dioxide. In fact, the current efficiency seems only to be limited by mass transport (9, 10, 18, 20) and the mineralization by the presence of functional groups, which lead to the formation of volatile molecules (volatile organochlorinated compounds such as chloroform in the oxidation of chlorophenols) or polymers (polyaminophenols in the oxidation of aminophenols). The importance of the hydroxyl oxidation pathway has been demonstrated in several works (1, 2, 7), although this oxidation is known to be complemented by the direct oxidation of the organics on the anode surface (23). In this sense, the direct oxidation is favored by the large oxygen evolution overpotentials that exhibit the diamond electrodes. Additionally, the electricity-conductive diamond electrodes exhibit a high mechanical and chemical stability. As a consequence, diamond-film-electrodes electro-oxidation is a very promising technology that at the present time can compete with other oxidation technologies in-use. The goal of the work described here was to increase the understanding of the mechanisms involved in the electrochemical oxidation of three isomers (catechol, resorcinol, and hydroquinone) on boron-doped diamond (BDD) electrodes, and to elucidate the influence of the hydroxyls position in the aromatic ring, in the electrochemical process. These dihydroxybenzenes can be introduced into the environment from a variety of industrial and natural sources, as they are frequently used as industrial reagents in the production of rubber, dyes, plastics, pharmaceuticals, and cosmetics and are important side products in saw and pulp mills (24-26). They are known to have different oxidazabilities by chemical oxidizers, although this is the first work in which the compared electrochemistry of the three compounds with diamond electrodes is studied.

Experimental Section Analytical Procedure. The carbon concentration was monitored using a Shimadzu TOC-5050 analyzer. Chemical oxygen demand (COD) was determined using a HACH DR200 analyzer. Carboxylic acids were monitored by HPLC using a Supelcogel H column (mobile phase, 0.15% phosphoric acid solution; flow rate, 0.15 mL min-1). The UV detector was set at 210 nm. Aromatics were also monitored by HPLC using a Nucleosil C18 column (mobile phase 60% water-40% methanol; flow rate, 0.50 mL min-1). In this case, the UV detector was set to 270 nm. Determination of the Instantaneous Current Efficiency (ICE). The chemical oxygen demand (COD) method was used for the determination of the current. In this method, the COD was measured during electrolysis and the instantaneous current efficiency was calculated using the relation

ICE )

|CODt+∆t - CODt|FV 8I∆t

(1)

where CODt and CODt+∆t are the chemical oxygen demand (in g of O2 dm-3) at times t and t +∆t (in seconds), respectively, 10.1021/es0500660 CCC: $30.25

 2005 American Chemical Society Published on Web 08/12/2005

FIGURE 1. (a) Experimental setup. (b) Detail of the electrochemical cell. I is the current intensity (A), F is the Faraday constant (96 487 C mol-1), V is the volume of the electrolyte (dm3), and 8 is a dimensional factor for unit consistence ((32 g of O2‚mol-1 of O2)/(4 mol of e-‚mol-1 of O2)). Galvanostatic Electrolysis Setup. The galvanostatic electrolyses were carried out in a single-compartment electrochemical flow cell (Figure 1). Diamond-based material was used as anode, and stainless steel (AISI 304) was used as the cathode. Both electrodes were circular (100 mm diameter) with a geometric area of 78 cm2 each and an electrode gap of 9 mm. The electrolyte was stored in a glass tank (500 mL) and circulated through the electrolytic cell by means of a centrifugal pump. A heat exchanger was used to maintain the temperature at the desired set point. The experimental setup also contained a cyclone for gas-liquid separation, and a gas absorber to collect the carbon dioxide contained in the gases evolved from the reactor into sodium hydroxide. Preparation of the Diamond Electrode. Boron-doped diamond films were provided by CSEM (Switzerland) and synthesized by the hot filament chemical vapor deposition technique (HF CVD) on single-crystal p-type Si 〈100〉 wafers (0.1 Ω cm, Siltronix). The temperature range of the filament was 2440-2560 °C, and that of the substrate was 830 °C. The reactive gas was methane in excess dihydrogen (1% CH4 in H2). The dopant gas was trimethylboron with a concentration of 3 mg dm-3. The gas mixture was supplied to the reaction chamber at a flow rate of 5 dm3 min-1, giving a growth rate of 0.24 µm h-1 for the diamond layer. The resulting diamond film thickness was about 1 µm. This HF CVD process produces columnar, random texture, and polycrystalline films with an average resistivity of 0.01 Ω cm. Prior to use in galvanostatic electrolysis assays, the electrode was polarized during 30 min with a 1 M H3PO4 solution at 50 mA cm-2 to remove any kind of impurity from its surface. Voltammetry Experiments. Electrochemical measurements were obtained using a conventional three-electrode cell in conjunction with a computer-controlled potentiostat/ galvanostat (Autolab model PGSTAT 30, Eco Chemie B.V., Utrecht, The Netherlands). Diamond was used as the working electrode, Hg/Hg2Cl2‚KCl (sat) as a reference, and platinum as a counter electrode. The BDD electrode was circular (25 mm diameter) with a geometric area of 4.91 cm2. Voltammetry experiments were performed in unstirred solutions (75 mL). Anode was anodically polarized during 5 min with a 1 M H3PO4 solution at 0.1 A prior to each experiment. Bulk Electrolysis. Galvanostatic electrolyses were carried out to determine the main intermediates formed in the process. The synthetic wastewaters used in the experiments contained 13.5 mM of hydroquinone, resorcinol, or catechol, 3333.33 mg of Na3PO4 dm-3, and H3PO4 in suitable amounts to give a pH of 2. The pH was kept constant by the continuous introduction of orthophosphoric acid (or sodium hydroxide) to the electrolyte reservoir. The cell potential was constant

FIGURE 2. COD variation with electrical charge during the complete electrolyses of wastes polluted with 13.5 mM catechol (2), resorcinol (]), or hydroquinone (×) (3333.33 mg of Na3PO4 dm-3, pH 2, j ) 30 mA cm-2, T ) 25 °C). Solid line: mathematical model proposed in the literature (9). during the electrolysis, indicating that appreciable deterioration of the electrode or passivation phenomena did not take place. The electrolyte flow rate through the cell was 2500 cm3 min-1. The linear velocity of the fluid was 4.62 cm s-1, and the space velocity was 4.16 min-1.

Results and Discussion Galvanostatic Electrolysis. Figure 2 shows the COD variation during the electrolysis of wastes polluted with the three isomers. The complete mineralization of the organics contained in the wastes is obtained, and the mineralization rates do not seem to depend on the nature of each isomer. On the contrary, the electrochemical efficiencies depend mainly on the concentration of COD. As it can be observed, an electrochemical model previously proposed in the literature (9), in which only mass transfer limitations are considered, fits well the experimental data. Figure 3 shows the main intermediates found in the experimental essays. It can be observed that carboxylic acids C4 and C2 are the main intermediates detected in the three cases and that the concentrations measured are low in comparison with the initial concentration of pollutant. The oxidation of catechol leads also to the formation of formic acid. This acid is not detected during the electrochemical oxidation of the other two isomers, and the ratios in which it is formed in the oxidation of catechol are similar to those of C4 carboxylic acids. No aromatic intermediates were formed during the treatment except for the formation of ortho- and para-benzoquinone, whose presence is detected even previously to the oxidation process, due to the acid base equilibrium between the keto and the enolic forms of hydroquinone and catechol. The small concentration of intermediates and the nondetection of aromatic intermediates suggest that, once a dihydroxibenzene molecule starts its oxidation, this process follows up to the complete mineralization of the molecule in a zone close to the anode surface. This behavior was previously reported for the oxidation of other organics on BDD anodes and supports the fact that the oxidation of the organics on these electrodes occurs directly on the anode surface or in a small volume very close to this surface, mediated by hydroxyl radicals generated by water oxidation (8, 20, 27, 28). The higher concentrations of C2 carboxylic acids indicate their lower oxidazability in comparison with the initial aromatic compounds and with the C1 and C4 carboxylic acids. Figure 4 shows the average ratios of the different C2 carboxylic acid measured with respect to the oxalic acid measured. It can be observed that the oxidation of hydroquinone leads only to the formation of oxalic acid and that the oxidation VOL. 39, NO. 18, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7235

FIGURE 3. Variation of intermediates with the specific electrical charge passed in the electrochemical oxidation of dihydroxybenzenes (C0 ) 13.5 mM, 3333.33 mg of Na3PO4 dm-3, pH 2, j ) 30 mA cm-2, T ) 25 °C). 0, Electrolyses of hydroquinone; 2, electrolyses of catechol; ], electrolyses of resorcinol. Concentrations are expressed in percentage of carbon with respect to the initial total organic carbon (dihydroxybenzenes).

FIGURE 4. Average ratios of the different C2 carboxylic acid measured with respect to the oxalic acid measured. 9, [glyoxalic acid]/[oxalic acid]; 0, [glycolic acid]/[oxalic acid]. of the other two isomers leads also to the formation of important amounts of glycolic and glyoxalic acids. 7236

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 18, 2005

FIGURE 5. Cyclic voltammograms on BDD anodes of catechol (a), resorcinol (b), and hydroquinone (c) solutions (13.5 mM) on sodium phosphate/phosphoric acid (pH 2, 3333.33 mg dm-3). (1) First cycle; (2) second cycle; (3) third scan; and (4) nonorganic matter. Auxiliary electrode: Platinum. Reference electrode: SCE. Scan rate: 100 mV s-1. A: Anodic start of the CV. These important differences in the ratios of the intermediates formed indicate that the reaction stages must be different. Nevertheless, the efficiency and the decrease of the organic load during the treatment are similar during the treatment of the three isomers. Two C4 carboxylic acids were detected: fumaric and maleic acids. The average concentration of maleic acid during the treatment was 5 times higher than that of the fumaric acid in the oxidation of the three isomers. This ratio has been previously observed in the oxidation of other phenols with BDD anodes (18-22). Voltammetric Study. Cyclic voltammograms with BDD electrodes of solutions containing 13.5 mM of catechol, resorcinol, and hydroquinone and 3333.33 mg dm-3 of Na3PO4 at pH 2 are shown in Figures 5 (potential range from -2.0 to 2.5 V vs SCE) and 6(potential range 0.0-2.5 V vs SCE). In both figures, the curve obtained under the same experimental conditions but without organic matter is also shown for the sake of comparison. As it can be observed, the voltammetric behavior is very similar for catechol and

FIGURE 6. Cyclic voltammograms on BDD anodes of catechol (a), resorcinol (b), and hydroquinone (c) solutions (13.5 mM) on sodium phosphate/phosphoric acid (pH 2, 3333.33 mg dm-3). (1) First cycle; (2) second cycle; (3) third scan; and (4) nonorganic matter. Auxiliary electrode: Platinum. Reference electrode: SCE. Scan rate: 100 mV s-1. A: Anodic start of the CV. hydroquinone. For both isomers, an anodic oxidation peak (with a similar current) is obtained at about 1.35 V vs SCE. This peak has a reverse peak at about -0.6 V for catechol and about -0.8 V for hydroquinone. The size of the cathodic peaks is smaller than that of the anodic, suggesting EC mechanisms. The anodic oxidation peak decreases in size with the number of scans. This decrease is important in the voltammograms with 0.0 V as the lower potential, and smaller in those with a lower potential value of -2.0 V vs SCE. The cathodic peak increases slightly with the number of cycles in the case of hydroquinone and is maintained in the voltammograms of catechol. The hydrogen evolution starts at higher overpotentials with the presence of both catechol and hydroquinone. The oxygen evolution does not seem to be affected by the presence of catechol. For hydroquinone, a significant shift toward the right is observed in the oxygen evolution. The voltammetric behavior of resorcinol is completely different from that shown by the other two isomers. The current densities are smaller, and no reverse peak is observed. Likewise, the anodic peak decreases strongly in both figures. In the voltammograms of Figure 6 (lower potential value 0.0 V vs SCE), it can be observed that the oxidation peak is overlapped with others in the second and in the third scans. The hydrogen evolution is shifted toward higher overpotentials more markedly for this compound, and oxygen evolution is not affected by its presence.

To justify the differences observed between the three isomers, two points have to be considered: (1) Hydroquinone and catechol have a reversible quinonic form. This justifies the presence of a cathodic reduction peak in their voltammograms. The quinonic compound generated in the first stage of the electro-oxidation of resorcinol is not thermodynamically stable, and thus no reduction peak is observed. (2) The reactivity of the aromatic ring activated with an OH group increases in the ortho- and para-positions. Thus, hydroquinone and catechol have the entire aromatic ring activated, while carbon 5 of resorcinol is not activated. This can justify its lower reactivity. The decrease in the peak size in the voltammograms of Figure 6 can be justified in terms of adsorption of the dihydroxybenzene onto the BDD surface and also in terms of polymer formation. In the voltammograms with 0.0 V vs SCE as the lower potential, the enolic form cannot be regenerated after each cycle and some active sites remain occupied by the quinonic form. Thus, the peak current obtained is smaller, and it will decrease in size with each cycle. When the lower potential is -2.0 V vs SCE (Figure 5), some of the active sites become occupied by the enolic form (after the reduction of the corresponding benzoquinone), and thus the peak size is almost maintained. On the other hand, according to the literature, the first stage in the oxidation of dihydroxybenzenes is the formation of a phenoxy-type radical (29). This radical can be further oxidized to the quinone form or can couple with other radicals or dihydroxybenzenes to form polymers. This formation of polymer may occur in the oxidation of the three isomers, and it can justify the smaller size of the cathodic peak in the oxidation of hydroquinone and catechol (EC mechanism) and also the shift in the oxygen evolution observed in the voltammograms of hydroquinone. However, the polymer formed must be easily removed by the hydroxyl radicals formed during water decomposition as no important differences between successive cycles can be observed for catechol and hydroquinone in the voltammograms of Figure 6. The shift in the hydrogen evolution process toward higher overpotentials has been reported in the literature previously for the reduction of water onto BDD in the presence of aromatics (6, 18, 20-22), but a clear explanation does not exist for this observation. Characterization of the Initial Stages. To characterize the first steps in the electrochemical oxidation of the dihydroxybenzenes, some essays were carried out in the lab-scale batch electrochemical cell used to conduct the voltammetric study. These essays were carried out for the three isomers, and they consisted of a sequence of four chronopotentiometries and cyclic voltamograms during the firsts 2.0 Ah dm-3 of the galvanostatic electrolyses. The same current density employed in the galvanostatic electrolysis essays described before was used (30 mA cm-2) in these essays. Results of these essays show no change in the cell potential during the electrolysis of the three isomers. This indicates no formation of polymer film onto the BDD surface. Figures 7 and 8 show, respectively, the variation of the initial organic pollutant and the intermediates found during the experiments, and Figure 9 shows the voltammograms carried out to characterize the processes that happen on the BDD surface during the first stages of the electrolysis process. It can be observed that during the first stages other carboxylic acids (not detected in the general galvanostatic electrolysis described before) were found. The more important was tartaric acid, which appeared in an important concentration in the oxidation of hydroquinone and catechol. The formation of glyoxal can also be observed in the three cases. The oxidation of catechol led to the formation of a VOL. 39, NO. 18, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7237

FIGURE 7. Variation of the concentration of the initial organic pollutant with the specific electrical charge passed (C0 ) 13.5 mM, 3333.33 mg of Na3PO4 dm-3, pH 2). ×, Hydroquinone; 2, catechol; 0, resorcinol.

FIGURE 9. Cyclic voltammograms on BDD anodes obtained during the first stages of the electrolysis process of catechol (a), resorcinol (b), and hydroquinone (c) solutions (13.5 mM) on sodium phosphate/ phosphoric acid (pH 2, 3333.33 mg dm-3). Electrical charge passed: (1) 0 Ah dm-3; (2) 0.5 Ah dm-3; (3) 1 Ah dm-3; (4) 1.5 Ah dm-3; (5) 2 Ah dm-3. Cyclic voltammogram on BDD anodes of 3333.33 mg of Na3PO4 dm-3 at pH 2 (6). Auxiliary electrode: Platinum. Reference electrode: SCE. Scan rate: 100 mV s-1. A: Anodic start of the CV.

FIGURE 8. Variation of intermediates with the specific electrical charge passed in the electrochemical oxidation of catechol (a), resorcinol (b), and hydroquinone (c). (C0 ) 13.5 mM, 3333.33 mg of Na3PO4 dm-3, pH 2, j ) 30 mA cm-2, T ) 25 °C). [, Glyoxal; 4, oxalic acid; +, maleic acid; 2, formic acid; *, fumaric acid; 9, tartaric acid; 0, inorganic carbon. larger number of intermediates and also to the accumulation of formic acid. No carbon dioxide was detected in its oxidation. Conversely, the oxidation of resorcinol and hydroquinone led to the formation of carbon dioxide during the first stages, and important concentrations were quantified. Despite the high concentration of dihydroxybenzenes, and the small quantities of charge passed, no aromatic intermediates were found; only benzoquinones were presented, but their concentrations were almost the same as those measured initially and caused by the pH equilibria. 7238

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 18, 2005

This confirms that the oxidation must be carried out directly on the electrode surface or by hydroxyl radicals generated by decomposition of water. To determine the electrochemical efficiency of the system, the theoretical oxygen demand was calculated from the intermediates concentration values. It was found that the instantaneous current efficiencies (ICE) were close to the maximum (100%) in the three experiments. Figure 10 shows the time variation of the ThOD (calculated according to the intermediates measured during the electrolyses of the three isomers) and the COD predicted if the current efficiency is maximum. As it can be observed, although the intermediate distribution is different for the three isomers, the current efficiency is maximum as the experimental points are wellfitted by the model. The small differences observed can be justified by the experimental error in the measurements. The successive voltammograms carried out during the electrolyses show different behaviors for resorcinol and the other two isomers. For hydroquinone and catechol, the anodic and the cathodic peaks decrease with time, and the size of both peaks is directly related to the concentration of the isomer remaining in the treatment. The small size of the cathodic peaks and the shift in the oxygen evolution confirm that the main direct reaction in the region of water stability is the formation of polymer.

FIGURE 10. Time variation of the ThOD (solid line) and the COD in the electrochemical oxidation of hydroquinone (×), catechol (2), and resorcinol (0) (C0 ) 13.5 mM, 3333.33 mg of Na3PO4 dm-3, pH 2, j ) 30 mA cm-2, T ) 25 °C). The voltammograms of resorcinol do not show a decrease in the peak current but a shift toward higher overpotential. However, the more important point is the high current values obtained in the reverse scans for high anodic potentials. The value of the current is similar to those obtained in the voltammograms of the other two isomers and could be justified in terms of the oxidation of reaction intermediates after the removal of the polymeric film by the action of hydroxyl radicals. In previous papers by our group (8), it was found that the presence of carboxylic acids shifted oxygen evolution toward lower overpotentials (due to the overlapping effect of the oxidation of carboxylic acid with that of the water). This shift is not observed in the successive voltammograms carried out in this work (despite their presence, measured by HPLC), indicating that the polymeric film inhibits this reaction. This confirms the important role of the hydroxyl radicals formed by the water decomposition, which are responsible for the destruction of the polymeric film and for the later aromatic ring cleavage to yield the different intermediates found in the bulk reaction solution.

Acknowledgments This work was supported by the MCT (Ministerio de Ciencia y Tecnologı´a, Spain) and by the EU (European Union) through the Project REN2001-0560. The authors also want to thank the Spanish and the Tunisian governments for the project 34/04/P/E.

Literature Cited (1) Foti, G.; Gandini, D.; Comninellis, Ch. Curr. Top. Electrochem. 1997, 5, 71-91. (2) Marselli, B.; Garcı´a-Go´mez, J.; Michaud, P.-A.; Rodrigo, M. A.; Comninellis, Ch. J. Electrochem. Soc. 2003, 150, D79-D83.

(3) Kraft, A.; Stadelmann, M.; Blaschke, M. J. Hazard. Mater. 2003, B103, 247-261. (4) Hagans, P. L.; Natishan, P. M.; Stoner, B. R.; O’Grady, W. E. J. Electrochem. Soc. 2001, 148, E298-E301. (5) Iniesta, J.; Michaud, P. A.; Panizza, M.; Cerisola, G.; Aldaz, A.; Comninellis, Ch. Electrochim. Acta 2001, 46, 3573-3578. (6) Can ˜ izares, P.; Lobato, J.; Garcı´a-Go´mez, J.; Rodrigo, M. A. J. Appl. Electrochem. 2004, 34, 111-117. (7) Gandini, D.; Mahe, E.; Michaud, P. A.; Haenni, W.; Perret, A.; Comninellis, Ch. J. Appl. Electrochem. 2000, 30, 1345-1350. (8) Can ˜ izares, P.; Garcı´a-Go´mez, J.; Lobato, J.; Rodrigo. Ind. Eng. Chem. Res. 2003, 42, 956-962. (9) Rodrigo, M. A.; Michaud, P. A.; Duo, I.; Panizza, M.; Cerisola, G.; Comninellis, Ch. J. Electrochem. Soc. 2001, 148, D60-D64. (10) Gherardini, L.; Michaud, P. A.; Panizza, M.; Comninellis, Ch.; Vatistas, N. J. Electrochem. Soc. 2001, 148, D78-D82. (11) Iniesta, J.; Michaud, P. A.; Panizza, M.; Comninellis, Ch. Electrochem. Commun. 2001, 3, 346-351. (12) Montilla, F.; Michaud, P. A.; Morallon, E.; Vazquez, J. L.; Comninellis, Ch. Electrochim. Acta 2002, 47, 3509-3513. (13) Panizza, M.; Cerisola, G. Electrochim. Acta 2003, 48, 3491-3497. (14) Panizza, M.; Cerisola, G. Electrochim. Acta 2004, 49, 3221-3226. (15) Bellagamba, R.; Michaud, P. A.; Comninellis, Ch.; Vatistas, N. Electrochem. Commun. 2002, 4, 171-176. (16) Boye, B.; Michaud, P. A.; Marselli, B.; Dieng, M. M.; Brillas, E.; Comninellis, Ch. New Diamond Front. Carbon Technol. 2002, 12, 63-72. (17) Hattori, S.; Doi, M.; Takahashi, E.; Kurosu, T.; Nara, M.; Nakamatsu, S.; Nishiki, Y.; Furuta, T.; Iida, I. J. Appl. Electrochem. 2003, 33, 85-91. (18) Can ˜ izares, P.; Garcı´a-Go´mez, J.; Sa´ez, C.; Rodrigo, M. A. J. Appl. Electrochem. 2003, 33, 917-927. (19) Can ˜ izares, P.; Garcı´a-Go´mez, J.; Sa´ez, C.; Rodrigo, M. A. J. Appl. Electrochem. 2004, 34, 87-94. (20) Can ˜ izares, P.; Sa´ez, C.; Lobato, J.; Rodrigo. M. A. Ind. Eng. Chem. Res. 2004, 43, 1944-1951. (21) Can ˜ izares, P.; Sa´ez, C.; Lobato, J.; Rodrigo, M. A. Electrochim. Acta 2004, 49, 4641-4650. (22) Can ˜ izares, P.; Sa´ez, C.; Lobato, J.; Rodrigo. M. A. Ind. Eng. Chem. Res. 2004, 43, 6629-6637. (23) Fujishima, A.; Rao, T. N.; Tryk, D. A. Electrochim. Acta 2000, 45, 4683-4690. (24) Milligan, P. W.; Ha¨ggblom, M. M. Environ. Toxicol. Chem. 1998, 17, 1456-1461. (25) Environmental fate and effects of bleached pulp mill effluents. So¨dergren, A., Ed.; SEPA conference, Stockholm, Sweden, 1991. (26) Vasquez, R. A.; Maestro, D. R.; Graciano, C. E. Grasas Aceites 1974, 25, 341-345. (27) Can ˜ izares, P.; Garcı´a-Go´mez, J.; Lobato, J.; Rodrigo, M. A. Ind. Eng. Chem. Res. 2004, 43, 1915-1922. (28) Can ˜ izares, P.; Garcı´a-Go´mez, J.; Lobato, J.; Rodrigo, M. A. Ind. Eng. Chem. Res. 2004, 43, 1923-1931. (29) Streitwieser, A.; Heathcock, C. H.; Kosower, E. M. Introduction to Organic Chemistry, 4th ed.; MacMillan: New York, 1992.

Received for review January 13, 2005. Revised manuscript received June 30, 2005. Accepted July 13, 2005. ES0500660

VOL. 39, NO. 18, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7239