Journal of Electroanalytical Chemistry 702 (2013) 3136

Contents lists available at SciVerse ScienceDirect

Journal of Electroanalytical Chemistry

journal homepage: www.elsevier.com/locate/jelechem

Electro-Fenton degradation of anti-inammatory drug ibuprofen in hydroorganic medium

Silvia Loaiza-Ambuludi a, Marco Panizza b, Nihal Oturan a, Ali zcan c, Mehmet A. Oturan a,
a

Universit Paris-Est Marne-la-Valle, Laboratoire Gomatriaux et Environnement, 5 boulevard Descartes, Champs-sur-Marne, 77454 Marne-la-Valle Cedex 02, France Department of Civil, Chemical and Environmental Engineering, University of Genoa, Piazzale Kennedy 1, Genoa 16129, Italy c Anadolu University, Faculty of Science, Department of Chemistry, 26470 Eskis ehir, Turkey
b

a r t i c l e

i n f o

a b s t r a c t
The oxidative degradation of ibuprofen in hydroorganic medium water/acetonitrile has been investigated by electro-Fenton processes. Experiments have been performed in a one-compartment cell with a Pt or BDD anode and a commercial 3D graphite felt cathode. The effect of operating conditions such as applied current, catalyst concentration, and supporting electrolyte nature has been studied. Ibuprofen decay kinetics and the evolution of its aromatic intermediates have been monitored during the electrolysis by HPLC and GCMS analyses. The experimental results have shown that ibuprofen has been completely removed in all experimental conditions either by homogeneous OH formed in the bulk of the solution through electrochemically generated Fentons reagent (H2O2 and Fe2+) and heterogeneous BDD(OH) at the anode surface from water oxidation. The removal rate has been more effective using a Pt anode than a BDD anode because of the greater regeneration of Fe2+ catalyst on the former. With both the anodes, the decay kinetics of ibuprofen always followed a pseudo-rst-order reaction and the oxidation rate largely depends on applied current, Fe2+ concentration and electrolyte nature. Several aromatic by-products such as 1-(1-hydroxyethyl)-4-isobutylbenzene, 4-isobutylacetophenone, 4-isobutylphenol and benzoquinone has been identied, and a plausible reaction mechanism has been proposed. 2013 Elsevier B.V. All rights reserved.

Article history: Received 5 March 2013 Received in revised form 3 May 2013 Accepted 7 May 2013 Available online 16 May 2013 Keywords: Ibuprofen oxidation Electro-Fenton Boron-doped diamond anode Hydroorganic media

1. Introduction Ibuprofen (2-(4-isobutylphenyl)propionic acid) is a nonsteroidal anti-inammatory drug (NSAID) of the propionic acid. Thanks to its analgesic and antipyretic properties, it is widely employed in the treatment of rheumatoid arthritis, osteoarthritis, but it is also used for the alleviation of mild to moderate pain, inammation and fever caused by many and diverse diseases. Because of its widespread use and persistence, concentration of ibuprofen below 10 lg dm3 has been detected in river waters [13]. It is also considered to be one of the most important pharmaceutical contaminants in sewage treatment plant (STP) inuents [4]. Since ibuprofen, like most of pharmaceutically active compounds, is poorly biodegradable and it affects signicantly the growth of several bacterial and fungal species [5,6], an efcient destruction of ibuprofen and that of its degradation by-products needs powerful oxidation methods for achieving its complete removal from wastewaters. The abatement of ibuprofen from aqueous solutions has been studied under applications of different water treatment technologies such as ozonation [7], advanced oxidation processes (AOPs) [8,9] and electrochemical oxidation [1012]. For example, Zwiener
Corresponding author. Tel.: +33 1 49 32 90 65.
E-mail address: mehmet.oturan@univ-paris-est.fr (M.A. Oturan). 1572-6657/$ - see front matter 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jelechem.2013.05.006

and Frimmel [13] reported that simple ozonation was ineffective to remove ibuprofen (i.e. 12% of degradation efciency) while the application of AOP O3/H2O2 almost quantitatively degraded the drugs (i.e. 99.4%). In a previous paper [14] we reported that using boron-doped diamond (BDD) anode, ibuprofen has been completely destroyed under all the conditions tested, following pseudo rst-order kinetics. Almost complete mineralization (i.e. TOC removal higher than 96%) of ibuprofen was also obtained in 8 h of electrolysis. Similar results were also obtained by Ciriaco et al. [10]. Recently, there is an increasing interest in the use of electrochemical Advanced Oxidation Processes (EAOPs) such as electroFenton (EF) process for wastewater remediation. In the EF process, the highly powerful oxidizing agent, hydroxyl radicals (OH) are produced in the bulk of the polluted solution using the Fenton reaction (Eq. (2)) through electrogenerated Fentons reagent (mixture of H2O2 + Fe2+) [1517]. H2O2 is supplied by in situ electrogeneration from the two-electron reduction of O2 (Eq. (1)) and Fe2+ is continually regenerated from Fe3+ reduction (Eq. (3)). The process needs thus the introduction of a catalytic amount of an iron (III) salts to the solution.

O2 2H 2e ! H2 O2 Fe2 H2 O2 ! Fe3 OH OH

1 2

32

S. Loaiza-Ambuludi et al. / Journal of Electroanalytical Chemistry 702 (2013) 3136

Fe3 e ! Fe2

These reactions occur with high yield and satisfactory rate only at gas diffusion cathodes (GDE) [1820] and three-dimensional carbon-based cathodes, such as graphite felt [2126]. This process has been efciently applied for the treatment of some phenols and polyphenols, dyes and pesticides enabling the complete mineralization of the wastewater [2732]. Using an undivided electrolytic cell for the EF process, the organic compounds can be also destroyed by heterogeneous hydroxyl radicals (M(OH)) produced at the anode surface (M) from water oxidation following the Eq. (4), thus accelerating the mineralization rate of the pollutants [3338]:

M H2 O ! M OH H e

In this eld, the degradation of ibuprofen solution has been comparatively studied by Skoumal and co-workers [12] using different methods such as electro-Fenton and solar photoelectro-Fenton using an O2-diffusion cathode and a Pt or boron-doped diamond (BDD) anode. Higher mineralization was attained using BDD instead of Pt, because the former produces greater quantity of OH enhancing the oxidation of pollutants. The mineralization rate also increased under solar irradiation by the rapid photodecomposition of complexes of Fe(III) with acidic intermediates. The most potent method was solar photoelectro-Fenton with BDD anode giving 92% mineralization. The solubility of ibuprofen in water is low (about 45 mg L1 at pH 3.0 and 25 C) [39], while it has large solubility in many organic solvents. In the present paper, we report a detailed study on the performance of the electro-Fenton process for the degradation of ibuprofen in a hydroorganic medium containing 20% acetonitrile. Hydroorganic medium was chosen for a technical reason (see experimental part). The experiments were carried out in an undivided cell using commercial graphite felt cathode and a Pt or BDD anode. The effect of some operating parameters such as applied current, catalyst dose, electrolyte nature and anode material on the removal rate was investigated. Aromatic intermediates were identied by gas chromatography-mass spectrometry (GCMS), and, based on identied reaction intermediates, a plausible reaction scheme was also proposed for the oxidative degradation of ibuprofen. 2. Experimental 2.1. Chemicals Ibuprofen, 2-(4-(2-methylpropyl)phenyl)propanoic acid (C13H18 O2) was obtained from SigmaAldrich and used without further purication. Anhydrous sodium sulfate, sodium chloride, hexahydrated ammonium iron(II) sulfate (Mohrs salt), (NH4)2Fe(SO4)26H2O, as ferrous ion (catalyst) source and pentahydrated ferric sulfate used as ferric iron ions (catalyst) source were analytical grade from Fluka. Deionized water used for the preparation of solutions and HPLC eluents was obtained from a Millipore Milli RO6 system, with resistivity >18 MX cm. Methanol (chromanorm grade) used as organic solvent for the liquid chromatography mobile phase and sulfuric acid used for pH adjustment were supplied by VWR International. 2.2. Electrochemical cell The electrolyses were performed in a small, open, cylindrical, one-compartment electrochemical cell of 6-cm diameter and 230 mL capacity, stirred by a magnetic PTFE follower during the treatment to enhance the mass transport towards electrodes. Either a cylindrical Pt mesh or a 25 cm2 boron-doped diamond (BDD) electrode (BDD thin-lm deposited on a niobium substrate

from CONDIAS, Germany) was used as the anode, and a large surface area graphite felt (14 cm 5 cm each side, 0.5 cm in width, from Carbone-Lorraine, France) as the cathode. In all cases, the anode was centred in the electrochemical cell and surrounded by the cathode, which covered the inner wall of the cell. Experiments were carried out in hydroorganic medium containing 20% acetonitrile with 0.05 M Na2SO4 or 0.100 M NaCl as supporting electrolyte and were investigated at room temperature (23 2 C) by applying a constant current in the range 50500 mA. The solution pH was xed to 3, since much higher pH values hamper the development of Fenton-based systems due to the Fe(OH)3 precipitation, which lead to both the decrease of dissolved iron ion [16]. Continuous saturation of the solution by O2 gas at atmospheric pressure was ensured by bubbling compressed air having passed through a frit at about 1 L min1 in the solution to be treated, starting 10 min before electrolysis to reach a stationary O2 concentration. The electrolyses were performed with a Hameg HM8040 triple power supply at constant current. This instrument displayed the cell voltage along the treatments as well. The solution pH was measured with a CyberScan pH 1500 pH-meter from Eutech Instruments. 2.3. Analytical procedures Ibuprofen solutions were prepared in wateracetonitrile (80:20 (v/v)) medium because the experiments conducted in aqueous solution did not yield reproducible results in electro-Fenton degradation over a relatively long time. We observed that the bubbling of compressed air or O2 through the solution leads to a decrease of the concentration of ibuprofen even if it was completely disappear over a long treatment time. Preliminary experiments, with and without Fe(II), with and without compressed air bubbling, bubbling with O2 or N2, were carried out for monitoring the concentration of the ibuprofen by HPLC. When the solution was bubbled with pure O2 and/or compressed air, the decrease in the concentration of ibuprofen was accompanied by the appearance of white crystals on the inner walls of the cell. The nature of this substance is not investigated but it was probably formed by oxidation of the acid group of ibuprofen, leading to the formation of an oxidation product which was less soluble than ibuprofen. Samples withdrawn from the treated solution at different electrolysis times were microltered onto a hydrophilic membrane (Millex-GV Millipore, pore size 0.22 lm) before analysis. The time course of the concentration of ibuprofen was followed by reversed-phase HPLC using a Merck Lachrom liquid chromatograph, equipped with a L-7100 pump, tted with a Purospher RP18 5 lm, 25 cm 4.6 mm (i.d.) column at 40 C, and coupled with a L-7455 photodiode array detector selected at optimum wavelengths of 228 nm for ibuprofen. The analyses of the ibuprofen decay were carried out isocratically with a methanol/water (with 1% phosphoric acid) 68:32 (v/v) mixture as the mobile phase. A ow rate of 0.5 mL min1 was always used. The corresponding retention time (tR) for ibuprofen was 11.6 min. In all cases, the identication of intermediates was made by comparison of tR and UV spectra with those of pure standards. GCMS analysis was performed by using a Thermo Finnigan PolarisQ GCMS analyzer, equipped with a TRB-5-MS column. The column was held at 40 C for 2 min, and then a gradient temperature program at 10 C min1 was applied between 40 and 280 C. The temperature was held at 280 C for 2 min. The temperature of the injection part and the MS transfer line was 250 C. Helium was used a carrier gas at a ow rate of 1.0 mL min1. 3. Results and discussion A set of electrolysis was carried out with 0.2 mM ibuprofen in hydroorganic solutions to determine the inuence of the main

S. Loaiza-Ambuludi et al. / Journal of Electroanalytical Chemistry 702 (2013) 3136

33

operating parameters on the destruction rate. The kinetics for the reaction between ibuprofen and hydroxyl radicals (OH) was followed by reversed-phase HPLC chromatography. Ibuprofen exhibits a well-dened peak with a retention time (tr) of 11.6 min under operating conditions given in 2.3. In the electro-Fenton process the applied current is an important parameter for the operational cost and process efciency, because both the production rate of H2O2 and the regeneration rate of Fe2+ is affected by this parameter. The effect of the applied current on the decay kinetics of ibuprofen was investigated by conducting constant current electrolysis at different current values, i.e., 50, 100, 200, 300 and 500 mA in the presence of 0.2 mM of Fe3+ as a catalyst and using 0.05 M Na2SO4 or 0.1 M NaCl as supporting electrolyte. As revealed in Fig. 1a, when Na2SO4 was used as supporting electrolyte, for all the applied current values, ibuprofen was completely removed, and the degradation rate increased with applied current. This faster oxidation rate at higher applied current values could be ascribed to acceleration of H2O2 and Fe2+ formation rates according to reactions (1) and (3) respectively, leading to the generation of higher amount of homogeneous OH hydroxyl radicals from Fentons reaction (Eq. (2)) and also from the electrogeneration of M(OH) at the anode surface (Eq. (4)). The exponential decrease of ibuprofen concentration indicates a pseudo rst order reaction kinetics for oxidation of ibuprofen by (homogeneous and heterogeneous) hydroxyl radicals. Working in galvanostatic condition, the concentration of hydroxyl radicals can be approximated in a steady state and therefore, the oxidation rate expression can be written as follows:

0.25 0.2

(a)
Ln [C/C 0 ]

8 6 4 2 0 0 10 20 30 40

[Ibu] /mM

0.15 0.1 0.05 0 0 10

Time / min

20

30

40

50

60

Time / min
0.25 0.2

(b)
[Ln [C/C 0 ]

5 4 3 2 1

[Ibu] / mM

0.15 0.1 0.05 0 0 10

0 0 10 20 30 40

Time / min

20

30

40

50

60

Time / min
Fig. 1. Inuence of the applied current on the decay kinetics of ibuprofen (IBU) (C0 = 0.2 mM) solutions containing 0.2 mM Fe3+ as catalyst, and (a) 0.05 M Na2SO4 and (b) 0.1 M NaCl as supporting electrolyte, by electro-Fenton process with Pt anode. I (mA): 50 (h), 100 (s), 200 (D), 300 (e) and 500 (). The insert represent the kinetic analysis following pseudo-rst order reaction between IBU and hydroxyl radicals.

dIBU k OHIBU kapp IBU dt

with apparent rate constant, kapp = k[OH], k being absolute (or second order) rate constant. Apparent rate constants for oxidative degradation of ibuprofen, obtained from the slop of straight lines of the inset of Fig. 1 were reported in Table 1. The rst conclusion from these apparent rate constant values was that they were about 10 times smaller than previously published results with similar aromatic compounds [17,24,27,31]. Oxidation kinetics is slowed at a large extent because of the competition reaction in the presence of acetonitrile:

Table 1 Apparent rate constants calculated from graph Ln(C0/Ct) = f(t) following pseudo rst order reaction kinetics for oxidation of ibuprofen by OH. Run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Anode Pt I (mA) 50 100 200 300 500 50 100 200 300 500 50 50 50 50 200 500 Electrolyte Na2SO4 Na2SO4 Na2SO4 Na2SO4 Na2SO4 NaCl NaCl NaCl NaCl NaCl Na2SO4 Na2SO4 Na2SO4 Na2SO4 Na2SO4 Na2SO4 [Fe3+] (mmol L1) 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.05 0.1 0.5 0.2 0.2 0.2 kapp (s1 103) 1.40 1.57 1.83 2.97 7.57 1.75 2,08 2.18 2.30 2.48 1.11 1.15 1.17 0.30 0.66 1.23

CH3 CN OH ! CH2 CN H2 O k 2:2 107 M1 s1

Since OH are consumed by acetonitrile, the oxidation of ibuprofen and the value of the kinetic constants related to this reaction was decreased. Comparing with our previous results [14] obtained in aqueous medium by anodic oxidation using BDD anode, we obtain better rate constants even in the presence of the scavenging effect of acetonitrile. For example, the values of 1.4 103 and 3.58 104 s1 were obtained for apparent rate constant of ibuprofen under electro-Fenton and anodic oxidation conditions respectively, at 50 mA when Na2SO4 was used as a supporting electrolyte. The total disappearance of ibuprofen was reached more quickly in electro-Fenton (about 20 min) at hydro-organic media then that of anodic oxidation (about 200 min) in water. These results show that the electro-Fenton process is more efcient in the oxidation kinetics than anodic oxidation, because the reaction takes place in the bulk, and it is also not limited by the mass transfer of ibuprofen to the BDD surface as in the case of anodic process. An almost complete mineralization (96%) of the ibuprofen solution was reached by an anodic oxidation using BDD anode, since oxidation and mineralization occured simultaneously near the anode surface. Unfortunately, the presences of acetonitrile do not permit the determination of TOC removal in the present study for comparison. It must be also noted that the rate constants do not rise proportionally to the current increase, due to the progressive

BDD

enhancement of the parasitic reactions of water formation from oxygen reduction (Eq. (7)) or hydrogen evolution (Eq. (8)) at the cathode, or oxygen evolution (Eq. (9)) at the anode. An additional wasting reaction contributing to the lower efciency of the process is the consumption of OH by acetonitrile (Eq. (2)). Even the reactivity of acetonitrile toward OH is relatively low (k = 2.2 107 M1 s1) its relatively high concentration makes the reaction (2) one of the most important wasting reactions.

O2 4H 4e ! 2H2 O 2H 2e ! H2

7 8

34

S. Loaiza-Ambuludi et al. / Journal of Electroanalytical Chemistry 702 (2013) 3136

0.25 0.2

6 5

0.25 0.2

5 4

Ln [C/C0]

Ln [C/C0]

4 3 1 2 0 0 20 40 60 80

3 2 1 0 0 50 100 150 200

[IBU] / mM

[IBU] / mM

0.15 0.1 0.05 0 0 10

0.15 0.1 0.05

Time / min

Time / min 20 30 40 50 60 70 0 0 100 200 300

Time / min
Fig. 2. Inuence of the Fe3+ (catalyst) concentration on the decay kinetics of ibuprofen, IBU (C0 = 0.2 mM) solutions containing 0.05 M Na2SO4 as supporting electrolyte at I = 50 mA by electro-Fenton with Pt anode. [Fe3+] (mM): 0.05 (h), 0.1 (s), 0.2 (D), 0.5 (e). The insert represent the kinetic analysis following pseudo-rst order reaction between IBU and OH.

Time / min
Fig. 3. Inuence of the applied current on the decay kinetics of ibuprofen, IBU (C0 = 0.2 mM) solutions containing 0.05 M Na2SO4 as supporting electrolyte, [Fe3+] = 0.2 mM by electro-Fenton with BDD anode. I (mA): 50 (h), 200 (s) and 500 (e). The insert represent the kinetic analysis following pseudo-rst order reaction between IBU and OH.

H2 O ! 1=2O2 2H 2e

This indicated that the process is under mass transfer limitation and the oxidation is affected by the diffusion of oxygen or Fe3+ to the cathode, which are the key step for the production of H2O2 and catalyst regeneration, or the diffusion of the ibuprofen to the anode. When NaCl was used as supporting electrolyte (Fig. 1b), the ibuprofen removal follows also a pseudo-rst order kinetics, however the oxidation rate was only slightly affected by the applied current. Contrary to what it was expected and to results obtained in a previous study about anodic oxidation [14], the presence of NaCl did not accelerated the removal of ibuprofen. The comparative behavior between ibuprofen abatement in Na2SO4 and NaCl medium is more clearly seen in Table 1 that reports the kinetic analysis of the plots shown in Fig. 1. The kapp values are close to each other at low currents, whereas at 500 mA the value of kapp is signicantly higher in Na2SO4 medium. This behavior can be explained by the fact that, using a Pt anode, Na2SO4 is a stable electrolyte, while chloride ions can be oxidized at the anode surface to active chlorine according to the following equation:

2Cl ! Cl2 2e

10

The Fe3+ concentration is another important parameter in the electro-Fenton process. In an attempt to optimize reaction conditions on the EF system, degradation of 0.2 mM ibuprofen was investigated in the presence of different Fe3+ concentrations at pH 3 applying a constant current of 50 mA. As reported in Fig. 2, the removal rate increases with increasing Fe3+ concentration from 0.05 to 0.2 mM. This effect can be related to an increasing quantity of Fe2+ regenerated from reaction (Eq. (3)) that enhances the production of strong oxidant OH by Fentons reaction (Eq. (2)) and hence, its reaction with ibuprofen. For Fe3+ concentration higher than 0.2 mM the removal rate undergoes to a rapid decrease, needing longer time for complete removal of the drug. This negative effect can be associated with a progressive fall of OH content in solution, mainly due to the participation of non-oxidizing reaction (13), which is strongly accelerated when much larger amounts of Fe2+ are formed from reaction (3) as more Fe3+ is added to the starting solution. That means that the maximum concentration of OH in the Pt/graphite felt cell at 50 mA is attained using 0.2 mM Fe3+, that is, when this radical formed from Fentons reaction (2), it is wasted in smaller extent with regenerated Fe2+.

The electrogenerated active chlorine can acts as oxidation mediator in the bulk of the solution, accelerating the reaction rate during anodic oxidation. However, in EF process, the electrogenerated chlorine can also oxidize Fe2+ (Eq. (11)) or decompose hydrogen peroxide (Eq. (12)), thus reducing the production rate of strong oxidant OH by Fentons reaction (Eq. (2)) and hence, its reaction with ibuprofen:

Fe2 OH ! Fe3 OH

13

2Fe2 Cl2 ! 2Fe3 2Cl

11 12

2H2 O2 Cl2 ! 2HCl H2 O O2

The inset panel of the Fig. 3 presents the excellent correlation decay considering a pseudo-rst-order reaction rate. The values of the apparent rate constants obtained from the kinetic analyses of the plots in Fig. 3 are reported in Table 1. Several experiments were also performed replacing the Pt anode with an high O2 evolution overpotential anode such as BDD. The effect of applied current on the destruction of ibuprofen by EF with 0.2 mM Fe3+ was studied in the range 50500 mA and the results are reported in Fig. 3. As in the case of Pt anode, the oxi-

Table 2 Aromatic intermediates detected during the EF degradation of 0.2 mM ibuprofen at 50 mA with Pt anode in the presence of 0.2 mM Fe3+. No 1 2 3 4 5 6 7 Compound p-Benzoquinone 4-Isobutylphenol 1-(1-Hydroxyethyl)-4isobutylbenzene 4-Isobutylacetophenone Ibuprofen Isomer molecule 1 Isomer molecule 2 Retention time 5.66 11.79 13.01 13.41 15.96 20.31 21.08 Mass fragmentation 107.98 (M+), 82.03, 80.04, 61.10, 54.10 150.03 (M+), 133.00, 109.10, 108.14, 107.09, 79.13, 73.12, 50.98 178.16 (M+), 163.22, 134.16, 117.15, 91.13, 79.19, 65.00, 57.10 175.97 (M+), 161.13, 133.21, 119.00, 105.24, 91.25, 90.19, 77.65 206.09 (M+), 163.07, 161.17, 145.20, 119.19, 107.21, 91.27, 77.21 249.08 (M+), 234.18, 208.11, 207.13, 192.13, 179, 166.18, 164.12, 150.13, 136.15, 122.20, 108, 91.19 249.12 (M+), 234.18, 208.11, 207.13, 192.13, 179, 166.18, 164.12, 150.13, 136, 122.11, 108, 91.11, 77.16

S. Loaiza-Ambuludi et al. / Journal of Electroanalytical Chemistry 702 (2013) 3136

35

Fig. 4. A plausible oxidative degradation pathway for oxidation of ibuprofen by electro-Fenton process in hydroorganic media containing acetonitrile.

dation of ibuprofen was accelerated by increasing the applied current because of progressively large production of OH. However, contrary to what can be expected, the removal rate of ibuprofen was signicantly reduced with the BDD anode in comparison to that found for electrolyses with Pt anode in the same conditions. For example, at 50 mA constant current, the time for complete drug removal was 50 min for Pt anode and 240 for BDD anode. This behavior, that is in agreement with the results obtained by Sires et al. [18], can be explained by the fact that using BDD anode, that has high O2 evolution overpotential, sulfate ions can be oxidized to persulfates (Eq. (14)) and these ions can further react with Fe2+ to give Fe3+ (Eq. (15)):
2 2SO2 4 ! S2 O8 2e

4. Conclusions In this paper, an hydroorganic solution of ibuprofen has been treated by electro-Fenton process using a graphite felt cathode and a Pt or BDD anode. Complete removal of ibuprofen has been obtained during galvanostatic electrolyses, due to the formation of hydroxyl radicals in the bulk from Fenton reaction via in situ electrogenerated Fentons reagent or at the anode surface from water oxidation. With both the anodes, the decay kinetics of ibuprofen follows pseudo-rst-order reaction; however, the use of a Pt anode instead of BDD yields a faster oxidation rate because of the generation of secondary oxidant (persulfate ion) that oxidize ferrous iron to ferric iron ion, thus harming the concentration of Fe2+ catalyst. With Pt anode, the oxidation rate increase with applied current, and it present a maximum value for catalyst (Fe3+) concentration of 0.2 mM. The removal of ibuprofen decreased also when NaCl was used as supporting electrolyte instead of Na2SO4, because the electrogeneration of active chlorine species from chloride ions reduce the production of OH radicals from Fentons reaction.

14

2 3 S 2 O2 ! 2SO2 8 2Fe 4 2Fe

2+

15

The waste of Fe catalyst thus reduces the production of OH radicals by Fentons reaction causing a lower removal rate of ibuprofen in the BDD-graphite felt cell. In order to identify the main aromatic oxidation intermediates, an hydroorganic solution of 0.2 mM ibuprofen was electrolyzed at 50 mA with 0.2 mM Fe3+ using a Pt anode. The identication of the reaction intermediates was performed by using GCMS analyses, comparing the obtained spectra (molecular mass and fragmentation values) with those of standard compounds found in the library (NIST) and with fragmentation available in literature. Table 2 reports the identied aromatic intermediates obtained during the EF process and the mass spectra characteristics. On the basis of the detected compounds, a plausible reaction sequence for the initial degradation of ibuprofen by electro-Fenton oxidation in hydroorganic media has been proposed in Fig. 4. The isomer structures 6 and 7 are not integrated in Fig. 4 since these structures are different to the expected one corresponding to addition of CH2 CN radical on aromatic moiety of ibuprofen, and their formation could not be explained. The oxidation process started with the hydroxylation on the C(2) position of the propionic acid group followed by direct decarboxylation to give 1-(1-hydroxyethyl)-4-isobutylbenzene. Further attack of the OH radicals onto the 1-(1-hydroxyethyl)-4-isobutylbenzene gave 4-isobutylacetophenone which is then decarboxylated to 4-isobutylphenol. This compound can be hydroxylated and decarboxilated in the isobutyl substituent to give nally hydroquinone (non-detected) which is oxidized quickly to benzoquinone. A similar reaction mechanism was also proposed by Skuomal et al. [12] for the oxidation of ibuprofen in aqueous media.

References
[1] N.M. Vieno, T. Tuhkanen, L. Kronberg, Environ. Sci. Technol. 39 (2005) 8220 8226. [2] N. Nakada, T. Tanishima, H. Shinohara, K. Kiri, H. Takada, Water Res. 40 (2006) 32973303. [3] G.L. Brun, M. Bernier, R. Losier, K. Doe, P. Jackman, H.-B. Lee, Environ. Toxicol. Chem. 25 (2006) 21632176. [4] M. Winkler, J.R. Lawrence, T.R. Neu, Water Res. 35 (2001) 31973205. [5] A.K. Sanyal, D. Roy, B. Chowdhury, A.B. Banerjee, Lett. Appl. Microbiol. 17 (1993) 109111. [6] B. Chowdhury, D. Roy, U. Chavan, S. Mukhopadhyay, Med. Sci. Res. 24 (1996) 801802. [7] E.C. Wert, F.L. Rosario-Ortiz, S.A. Snyder, Water Res. 43 (2009) 10051014. [8] N. Nakada, H. Shinohara, A. Murata, K. Kiri, S. Managaki, N. Sato, H. Takada, Water Res. 41 (2007) 43734382. [9] M.M. Huber, S. Canonica, G.-Y. Park, U. Von Gunten, Environ. Sci. Technol. 37 (2003) 10161024. [10] L. Ciriaco, C. Anjo, J. Correia, M.J. Pacheco, A. Lopes, J. Correra, Electrochim. Acta 54 (2009) 14641472. [11] L. Ciriaco, D. Santos, M.J. Pacheco, A. Lopes, J. Appl. Electrochem. 41 (2011) 577587. [12] M. Skoumal, R.M. Rodrguez, P.-L. Cabot, F. Centellas, J.A. Garrido, C. Arias, E. Brillas, Electrochim. Acta 54 (2009) 20772085. [13] C. Zwiener, F.H. Frimmel, Water Res. 34 (2000) 18811885. [14] S. Loaiza-Ambuludi, N. Oturan, A. zcan, M.A. Oturan, Environ. Sci. Pollut. Res. 20 (4) (2013) 23812389. [15] M.A. Oturan, J. Appl. Electrochem. 30 (2000) 477482. [16] E. Brillas, I. Sirs, M.A. Oturan, Chem. Rev. 109 (2009) 65706631. [17] A. Dirany, I. Sirs, N. Oturan, A. zcan, M.A. Oturan, Environ. Sci. Technol. 46 (2012) 40744082. [18] I. Sirs, N. Oturan, M.A. Oturan, R.M. Rodrguez, J.A. Garrido, E. Brillas, Electrochim. Acta 52 (2007) 54935503.

36

S. Loaiza-Ambuludi et al. / Journal of Electroanalytical Chemistry 702 (2013) 3136 [29] M.H. Zhou, Q.Q. Tan, Q. Wang, Y.L. Jiao, N. Oturan, M.A. Oturan, J. Hazard. Mater. 215216 (2012) 287293. [30] S.-H. Yuan, X.-H. Lu, J. Hazard. Mater. 118 (2005) 8592. [31] A. Kesraoui-Abdessalem, N. Oturan, N. Bellakhal, M. Dachraoui, M.A. Oturan, Appl. Catal. B: Environ. 78 (2008) 334341. [32] E. Brillas, B. Boye, I. Sires, J.A. Garrido, R.M. Rodrguez, C. Arias, P.-L. Cabot, C. Comninellis, Electrochim. Acta 49 (2004) 44874496. [33] N. Oturan, E. Brillas, M.A. Oturan, Environ. Chem. Lett. 10 (2012) 165170. [34] P. Caizares, J. Lobato, R. Paz, M.A. Rodrigo, C. Sez, Water Res. 39 (2005) 26872705. [35] D. Montanaro, E. Petrucci, Chem. Eng. J. 153 (2009) 138144. [36] E. Guinea, E. Brillas, F. Centellas, P. Canizares, M.A. Rodrigo, C. Saez, Water Res. 43 (2009) 21312138. [37] E. Petrucci, D. Montanaro, Chem. Eng. J. 174 (2011) 612618. [38] X.J. Ma, J.F. Ding, L.X. Bian, M.H. Zhou, Curr. Org. Chem. 16 (2009) 19721977. [39] S.D. Bhattamishra, R.K. Padhy, Indian J. of, Chem. Technol. 16 (2009) 426430.

[19] I. Sirs, C. Arias, P.-L. Cabot, F. Centellas, J.A. Garrido, R.M. Rodrguez, E. Brillas, Chemosphere 66 (2007) 16601669. [20] G.R. Agladze, G.S. Tsurtsumia, B.-I. Jung, J.-S. Kim, G. Gorelishvili, J. Appl. Electrochem. 37 (2007) 375383. [21] A. Da Pozzo, L. Di Palma, C. Merli, E. Petrucci, J. Appl. Electrochem. 35 (2005) 413419. [22] A. zcan, Y. S ahin, A.S. Koparal, M.A. Oturan, Water Res. 42 (2008) 28892898. [23] M. Panizza, M.A. Oturan, Electrochim. Acta 56 (2011) 70847087. [24] I. Sirs, N. Oturan, M.A. Oturan, Water Res. 44 (2010) 31093120. [25] S.C. Elaoud, M. Panizza, G. Cerisola, T. Mhiri, J. Electroanal. Chem. 667 (2012) 1923. [26] A. zcan, Y. S ahin, A.S. Koparal, M.A. Oturan, Electroanal. Chem. 616 (2008) 7178. [27] N. Oturan, M. Panizza, M.A. Oturan, J. Phys. Chem. A 113 (2009) 1098810993. [28] M.A. Oturan, I. Sirs, N. Oturan, S. Procheau, J.-K. Laborde, S. Trvin, J. Electroanal. Chem. 624 (2008) 329332.