Mineralization of Metoprolol by Electro-Fenton and Photoelectro

ARTICLE pubs.acs.org/JPCA

Mineralization of Metoprolol by Electro-Fenton and Photoelectro-Fenton Processes Eloy Isarain-Chavez, Jose Antonio Garrido, Rosa María Rodríguez, Francesc Centellas, Conchita Arias, Pere Lluís Cabot, and Enric Brillas* Laboratori d’Electroquímica dels Materials i del Medi Ambient, Departament de Química Física, Universitat de Barcelona, Martí i Franques 1-11, 08028 Barcelona, Spain ABSTRACT: Solutions of about 0.25 mM of the β-blocker metoprolol tartrate (100 mg L-1 total organic carbon) with 0.5 mM Fe2þ in the presence and absence of 0.1 mM Cu2þ of pH 3.0 have been comparatively degraded under electroFenton (EF) and photoelectro-Fenton (PEF) conditions. The electrolyses were carried out with two systems: (i) a single cell with a boron-doped diamond (BDD) anode and an air-diffusion cathode (ADE) for H2O2 electrogeneration and (ii) a combined cell with a BDD/ADE pair coupled with a Pt/carbon felt (CF) cell. Overall mineralization was reached in all PEF treatments using both systems due to the efficient production of hydroxyl radical (•OH) from Fenton’s reaction induced by UVA light and the quick photolysis of Fe(III) carboxylate complexes formed. In EF, the combined cell was much more potent than the single one by the larger •OH generation from the continuous Fe2þ regeneration at the CF cathode, accelerating the oxidation of organics. However, almost total mineralization in EF was feasible using the combined cell in the presence of 0.1 mM Cu2þ, because of the parallel quick oxidation of Cu(II) carboxylate complexes by •OH. Metoprolol decay always followed a pseudo-first-order reaction. Aromatic products related to consecutive hydroxylation/oxidation reactions of metoprolol were detected by gas chromatography-mass spectrometry. The evolution of the aromatic 4-(2-methoxyethyl)phenol and generated carboxylic acids was followed by HPLC. The degradation rate and mineralization degree of metoprolol tartrate were limited by the removal of Fe(III) and Cu(II) complexes of ultimate carboxylic acids such as formic, oxalic, and oxamic. NH4þ ion and to a lesser extent NO3ion were released in all treatments, being quantified by ionic chromatography.

1. INTRODUCTION Over the past decade, an increasing number of pharmaceutical drugs, coming from their widespread consumption in human and veterinary medicine, have been detected in surface, ground, and drinking waters at low contents of micrograms per liter.1-7 This pollution is caused by emission from production sites, direct disposal of surplus drugs in households, excretion after drug administration to humans and animals, and treatments throughout the water in fish and other animal farms. Accumulation of these compounds is due to their inefficient destruction in sewage treatment plants (SWTs).5,7-9 Some works have reported that low contents of drugs in the environment can interact with living beings, as in the case of β-blockers that can affect the endocrine systems of fish and can exert toxic effects on algae and invertebrates.3,10-14 Research efforts are then required to develop potent oxidation methods to remove drugs and their metabolites from wastewaters to avoid their effects on the health of living beings. Electrochemical advanced oxidation processes (EAOPs) based on Fenton’s reaction chemistry are eco-friendly methods that have recently received much attention for water remediation.15 The most popular EAOP is the electro-Fenton (EF) process, in which H2O2 is continuously supplied to an acidic polluted solution from the two-electron reduction of injected O2 gas by reaction 1, whereas Fe2þ ion is added as catalyst to react with generated H2O2 originating homogeneous hydroxyl radical r 2011 American Chemical Society

(•OH) and Fe3þ ion from Fenton’s reaction (2):16 O2ðgÞ þ 2Hþ þ 2e- f H2 O2

ð1Þ

Fe2þ þ H2 O2 f Fe3þ þ • OH þ OH-

ð2Þ

17-22

Cathodes such as carbon felt (CF) and gas (air or O2) diffusion23-26 electrodes are typically employed in EF. While the former favors the regeneration of Fe2þ from the cathodic reduction of Fe3þ by reaction 3, thus promoting Fenton’s reaction (2), the latter allows a higher generation of H2O2 via reaction 1. Fe3þ þ e- f Fe2þ

ð3Þ

•

OH thus formed in the bulk is a strong oxidant with so high a standard reduction potential (E°(•OH/H2O) = 2.80 V/SHE) that it can react nonselectively with organics via dehydrogenation and/ or hydroxylation until mineralization (conversion into CO2, water, and inorganic ions). Using an undivided cell, the oxidation power of EF is enhanced by the parallel destruction of organic pollutants with heterogeneous hydroxyl radical produced at a high O2-overpotential anode from water oxidation.24,27 Boron-doped diamond (BDD) thin-film electrodes are the most potent anodes known because they possess an inert Received: November 10, 2010 Revised: January 10, 2011 Published: February 2, 2011 1234

dx.doi.org/10.1021/jp110753r | J. Phys. Chem. A 2011, 115, 1234–1242

The Journal of Physical Chemistry A

ARTICLE

surface with low adsorption, remarkable corrosion stability, and extremely wide potential windows (>3 V) in aqueous medium.28-31 These properties for BDD give a very weak electrode-•OH interaction that results in a higher O2-overpotential than conventional anodes such as Pt and the formation of higher amounts of heterogeneous BDD(•OH) from reaction 4, which reacts more rapidly with organics as found for several aromatics and carboxylic acids:32-38 BDD þ H2 O f BDDð• OHÞ þ Hþ þ e-

ð4Þ

A variant of EF is the photoelectro-Fenton (PEF) process, where the solution is degraded under EF conditions and simultaneously irradiated with UVA light of λmax = 360 nm.15,24,25 The positive action of this radiation is complex and can involve (i) the photolysis of Fe(OH)2þ, the predominant species of Fe3þ in the pH range 2.5-4.0, producing more Fe2þ and •OH from reaction 516,39 and (ii) the photodecarboxylation of Fe(III) carboxylate complexes, as shown in reaction 6 for Fe(III) oxalate complexes (Fe(C2O4)þ, Fe(C2O4)2-, and Fe(C2O4)33-):40 FeðOHÞ2þ þ hν f Fe2þ þ • OH

ð5Þ

2FeðC2 O4 Þn ð3 - 2nÞ þ hν f 2Fe2þ þ ð2n-1ÞC2 O4 2- þ 2CO2 ð6Þ Other approaches have also been proposed to increase the oxidation ability of the EF process. Thus, several works have tested the addition to the solution of cocatalysts such as Cu2þ that can yield Fenton-like reactions with electrogenerated H2O2 increasing the formation of •OH and/or more easily degradable complexes with oxidation products.41-43 Another possibility involves the use of combined cells such as a BDD/gas diffusion reactor coupled with a Pt/CF cell, which accelerates the generation of •OH from Fenton’s reaction (2) because Fe2þ is quickly regenerated by Fe3þ reduction at the CF cathode while H2O2 is largely produced at the gas-diffusion electrode.44 In this paper, we report a study on the degradation of the β-blocker metoprolol (1-(isopropylamino)-3-[4-(2-methoxyethyl)phenoxy]-2-propanol) by EF and PEF. To try to establish the more potent oxidation conditions for both EAOPs, Fe2þ and/or Cu2þ ions were added as catalysts and single BDD/air-diffusion electrode (ADE) and combined BDD/ADE-Pt/CF cells were used to remove the drug in its commercial formula of metoprolol tartrate (2:1 complex). A 0.246 mM solution of this complex (100 mg L-1 total organic carbon (TOC)) was degraded to compare the oxidation power of all procedures and to detect its oxidation products by gas chromatography-mass spectrometry (GC-MS) and chromatographic techniques. Metoprolol is used to treat hypertension and is metabolized by the liver, but about 10% unchanged drug is excreted in the urine to be accumulated in the environment. It has been found up to 3 μg L-1 in European rivers,3,4 and can persist >100 days in aquatic medium,6 producing very toxic metabolites.13 A limited number of papers reported its destruction by ozonation,45,46 UVC,47,48 and UV/H2O2.48 Mineralization of a mixture of 0.15 mM β-blockers including metoprolol by EF with catalytic Fe2þ in a single Pt/CF cell has also been described by Sires et al.22

(Madrid, Spain). 4-(2-Methoxyethyl)phenol was of reagent grade from Sigma. Maleic, tartaric, formic, oxalic, and oxamic acids were of reagent grade from Panreac. Sulfuric acid, anhydrous sodium sulfate, ferrous sulfate heptahydrate, and cupric sulfate pentahydrate were of analytical grade from Merck and Fluka. Solutions were prepared with pure water obtained from a Millipore Milli-Q system with resistivity >18 MΩ cm at 25 °C and TOC < 1 μg L-1. Organic solvents and other chemicals used were of either HPLC or analytical grade supplied by Aldrich, Merck, and Panreac. 2.2. Electrolytic Systems. All electrolyses were conducted in an open and undivided cylindrical cell containing 100 mL of solution stirred with a magnetic bar at 800 rpm to ensure its homogenization and the transport of reactants toward/from the electrodes. The cell was surrounded with a double jacket for circulation of external thermostated water to regulate the solution at 35 °C. The anodes were a Pt sheet of 99.99% purity from SEMPSA (Barcelona, Spain) and/or a BDD electrode from Adamant Technologies (La-Chaux-de-Fonds, Switzerland), which was synthesized by the hot filament chemical vapor deposition technique on single-crystal p-type Si(100) wafers (0.1 Ω cm, Siltronix). The cathodes were a carbon-PTFE ADE from E-TEK (Somerset, NJ, USA) and/or a CF from Sofacel (Sant Feliu de Llobregat, Barcelona). The geometric area of all electrodes was 3 cm2. Two monopolar cell configurations, BDD/ ADE and BDD/ADE-Pt/CF, were used.44 The ADE cathode was fed with 20 mL min-1 air with a pump to electrogenerate H2O2 from reaction 1. A constant current was applied to each couple of electrodes using an Amel 2053 potentiostat-galvanostat and/or an EG&G PAR 363 potentiostat-galvanostat. Before the BDD anode and the ADE cathode were used in the electrolytic assays, they were polarized during 60 min in a 0.05 M Na2SO4 solution at 300 mA for their activation and removal of the surface impurities. Comparative EF and PEF treatments were made in both cells using 0.246 mM metoprolol tartrate solutions in 0.05 M Na2SO4 as background electrolyte. All trials were carried out after addition of 0.5 mM Fe2þ as catalyst and at pH 3.0 by regulation with concentrated H2SO4. These conditions were chosen because they were optimal for similar treatments of other aromatics.15,22-25 A low Cu2þ concentration was added to the solution to study the effect of this cocatalyst on the degradation process. For the PEF process, the solution was irradiated with a Philips TL/6W/08 fluorescent black light blue tube of λmax = 360 nm, placed at the top of the open cell at 2 cm above the solution, with a photoionization energy input of 1.4 W m-2. 2.3. Apparatus and Product Analysis Procedures. The solution pH was determined with a Crison GLP 22 pH meter. Aliquots withdrawn from electrolyzed solutions were filtered with 0.45 μm PTFE filters from Whatman before analysis. Solution TOC was obtained with a Shimadzu VCSN total organic carbon analyzer. Reproducible TOC values with an accuracy of (1% were found by injecting 50 μL aliquots into the analyzer. From these data, the mineralization current efficiency (MCE, in %) for each treated solution was estimated by eq 7:24 MCE ¼

2. EXPERIMENTAL SECTION 2.1. Chemicals. Metoprolol tartrate of 99% purity was pur-

chased from the pharmaceutical company AstraZeneca Espa~na

nFVs ΔðTOCÞexp ð4:32 107 ÞmIt

100

ð7Þ

where F is the Faraday constant (96 487 C mol-1), Vs is the solution volume (L), Δ(TOC)exp is the solution TOC decay 1235

dx.doi.org/10.1021/jp110753r |J. Phys. Chem. A 2011, 115, 1234–1242


ARTICLE

(mg L-1), 4.32 107 is a conversion factor to homogenize units (3600 s h-1 12 000 mg mol-1), m is the number of carbon atoms of metoprolol tartrate (34 C atoms), I is the applied current (A), and t is the electrolysis time (h). The number of electrons (n) consumed per drug molecule was taken as 162 considering that it is mineralized to CO2 according to eq 8, with the release of NH4þ as main primary ion, as will be discussed below. ðC15 H25 NO3 Þ2 C4 H6 O6 þ 56H2 O f 34CO2 þ 2NH4 þ þ 160Hþ þ 162e-

ð8Þ

The evolution of aromatics was followed by reversed-phase HPLC using a Waters 600 liquid chromatograph fitted with a Spherisorb ODS 5 μm, 150 mm 4.6 mm (i.d.), column at 35 °C and coupled with a Waters 996 photodiode array detector selected to λ = 223 nm, the maximum wavelength for metoprolol and 4-(2-methoxyethyl)phenol. Carboxylic acids were detected by ion-exclusion HPLC using the above liquid chromatograph fitted with a Bio-Rad Aminex HPX 87H, 300 mm 7.8 mm (i.d.), column at 35 °C and the photodiode array detector selected at λ = 210 nm. In all HPLC measurements, 20 μL aliquots were injected into the chromatograph. A 36:36:28 (v/v/v) acetonitrile/methanol/water (with 2 g L-1 sodium dodecyl sulfate at pH 3.0) mixture at 1.5 mL min-1 and 4.0 mM H2SO4 at 0.6 mL min-1 were used as mobile phase for reversedphase and ion-exclusion HPLC, respectively. Inorganic ions released were quantified by ionic chromatography using a Shimadzu 10 Avp HPLC coupled with a Shimadzu CDD 10 Avp conductivity detector by injecting 25 μL aliquots. The NH4þ concentration of electrolyzed solution was measured with a Shodex IC YK-421, 125 mm 4.6 mm (i.d.), cation column at 40 °C and a mobile phase of 5.0 mM tartaric acid, 2.0 mM dipicolinic acid, 24.2 mM boric acid, and 15.0 mM corona ether at 1.0 mL min-1. The NO3- concentration of the same solutions was determined using a Shim-Pack IC-A1S, 100 mm 4.6 mm (i.d.), anion column at 40 °C and a mobile phase of 1.0 mM p-hydroxybenzoic acid and 1.1 mM N,Ndiethylethanolamine at 1.5 mL min-1. To identify the aromatic intermediates, the metoprolol tartrate solution was electrolyzed under EF conditions with 0.5 mM Fe2þ in a BDD/ADE cell at 20 mA for 60 min. About 15 mL of the treated solution was lyophilized, and the remaining solid was eluted in 5 mL of CH2Cl2. The resulting organic solution was filtered, concentrated to 1.5 mL, and finally analyzed by GC-MS using an HP 5890 Series II gas chromatograph coupled with an HP 5989A mass spectrophotometer operating in EI mode at 70 eV. Aromatics were separated with a nonpolar J&W DB-5MS 0.25 μm, 30 m 0.25 mm (i.d.), column using a temperature ramp of 50 °C for 3 min, 10 °C min-1 up to 300 °C, and hold time 5 min. Mass spectra were analyzed with a NIFT05 data library.

3. RESULTS AND DISCUSSION 3.1. Effect of Current on TOC Removal by EF. A first series of experiments were carried out by electrolyzing the 0.246 mM metoprolol tartrate solution with 0.5 mM Fe2þ of pH 3.0 by EF in single BDD/ADE cell at a constant current between 60 and 210 mA for 360 min. In all cases, the pH remained practically

Figure 1. Effect of applied current on (a) TOC removal and (b) mineralization current efficiency calculated from eq 7 with electrolysis time for the electro-Fenton (EF) degradation of 100 mL of a 0.246 mM metoprolol tartrate solution in 0.05 M Na2SO4 with 0.5 mM Fe2þ at pH 3.0 and 35 °C using an open and undivided cylindrical cell with a 3 cm2 boron-doped diamond (BDD) anode and a 3 cm2 air-diffusion electrode (ADE) as cathode. Current: (O) 60, (0) 90, ([) 120, (4) 150, and (3) 210 mA.

unchanged, decreasing to 2.7-2.9 at the end of treatment. Figure 1a shows a gradual TOC removal for these trials, more rapidly when current rises from 60 to 120 mA, but practically at the same rate at higher currents. After 360 min, for example, TOC is reduced by 80, 82, 86, 86, and 87% for increasing currents of 60, 90, 120, 150, and 210 mA. The fact that only partial mineralization is achieved by EF is indicative of the formation of very refractory byproducts that are difficultly oxidized by •OH formed from Fenton’s reaction (2) and BDD(•OH) produced from reaction 4. The effect of current on the degradation rate of the EF process can be better analyzed from the corresponding MCE. Figure 1b illustrates that the efficiency is maximal at about 90 min of all electrolyses, which decays as current rises from 47% at 60 mA to 14% at 210 mA. Note that increasing current causes the production of a greater amount of •OH since more H2O2 is generated at the ADE cathode by reaction 1 enhancing Fenton’s reaction (2),15,44 as well as a greater amount of BDD(•OH) since water oxidation at the BDD anode by reaction 4 is accelerated.30,31 The fall in MCE then indicates that the increase in current is mainly consumed in parasitic nonoxidizing reactions of hydroxyl radicals, because organic pollutants react very hardly with them. The most important side reaction corresponds to the direct oxidation of BDD(•OH) to O2 via reaction 9.30-34 Other waste reactions involve the dimerization of •OH to H2O2 by reaction 10 or its destruction with H2O2 giving hydroperoxyl radical (HO2•) and with Fe2þ from reactions 11 and 12, respectively.24,25 BDD(•OH) generation can also be inhibited by the parallel production of weaker oxidants such as S2O82- ion from SO42- ion 1236



ARTICLE

pH 3.0 to ensure the total regeneration of Fe2þ at the CF cathode from the reduction of Fe3þ formed at the Pt anode. Higher currents accelerate more strongly the oxidation of Fe2þ to Fe3þ at the Pt anode, causing the decay of Fe2þ content in solution. 3.2. Comparative Degradation by EF and PEF. The possible influence of Cu2þ ion as cocatalyst was first analyzed for EF in a BDD/ADE cell. To do this, several 0.246 mM metoprolol tartrate solutions of pH 3.0 containing 0.5 mM Fe2þ and 0.11.0 mM Cu2þ were treated at 120 mA for 360 min. Figure 2a evidences a slight enhancement of the degradation rate of the solution in the presence of only 0.1 mM Cu2þ, since its TOC is reduced by 88% at the end of treatment, a value slightly higher than the 86% found for 0.5 mM Fe2þ alone. However, similar TOC removals were obtained operating up to 1.0 mM Cu2þ, indicating that a Cu2þ content as low as 0.1 mM is enough to study the effect of this ion in the EF and PEF systems tested. The slight increase in oxidation power of EF with 0.5 mM Fe2þ in BDD/ADE cell by adding Cu2þ ion could be accounted for by the generation of small amounts of •OH in the bulk from the Cu2þ/Cuþ catalytic system. This involves the reduction of Cu2þ to Cuþ with HO2• (e.g., formed from reaction 11) by reaction 15 and/or with organic radicals R• by reaction 16, followed by Cu2þ regeneration from Cuþ oxidation with H2O2 by the Fenton-like reaction (17):49,50

Figure 2. TOC abatement with electrolysis time for (a) EF and (b) photoelectro-Fenton (PEF) treatments of 100 mL of 0.246 mM metoprolol tartrate solutions in 0.05 M Na2SO4 with 0.5 mM Fe2þ at pH 3.0 and 35 °C. (O, b) Single BDD/ADE cell at 120 mA and (4, 2) combined BDD/ADE-Pt/carbon felt (CF) cell at 120-12 mA. In b and 2 0.1 mM Cu2þ was added to the solution as cocatalyst. The PEF processes were carried out under 6 W UVA irradiation of λmax = 360 nm.

oxidation by reaction 13 and O3 from reaction 14:30,32 2BDDð• OHÞ f 2BDD þ O2ðgÞ þ 2Hþ þ 2e-

ð9Þ

2• OH f H2 O2

ð10Þ

H2 O2 þ • OH f HO2 • þ H2 O

ð11Þ

Fe2þ þ • OH f Fe3þ þ OH-

ð12Þ

2SO4 2- f S2 O8 2- þ 2e-

ð13Þ

3H2 O f O3ðgÞ þ 6Hþ þ 6e-

ð14Þ

The existence of the side reactions 9-14 limits the oxidation power of EF in a BDD/ADE cell to destroy metoprolol tartrate, and hence, the best current to be applied has to be chosen. From the results of Figure 1a, one can conclude that the use of 120 mA is preferable under the experimental conditions tested, because no significant greater TOC abatement is reached at higher applied current. A current of 120 mA was then supplied to the BDD/ADE pairs of single and combined cells in further experiments to compare their oxidation power under EF and PEF conditions, as well as the effect of Cu2þ ion as cocatalyst. For the Pt/CF pair of the combined BDD/ADE-Pt/CF cell, a current of 12 mA was applied. In previous work,44 we found that this is the maximum current that can be supplied to a Pt/CF cell filled with 100 mL of a 0.05 M Na2SO4 and 0.5 mM Fe2þ solution at

Cu2þ þ HO2 • f Cuþ þ Hþ þ O2

ð15Þ

Cu2þ þ R • f Cuþ þ R þ

ð16Þ

Cuþ þ H2 O2 f Cu2þ þ • OH þ OH-

ð17Þ

Figure 2a also shows that the alternative use in EF of the combined BDD/ADE-Pt/CF cell using 0.5 mM Fe2þ at 120-12 mA highly increases the degradation rate compared with that of the single BDD/ADE one, giving a final TOC removal of 92%. The greater oxidation power of the combined system can be related to the fast Fe2þ regeneration from Fe3þ reduction at the CF cathode by reaction 3, which accelerates the formation of •OH by Fenton’s reaction (2) and, hence, the destruction of organic pollutants. However, a much higher increase in TOC decay can be observed in Figure 2a after addition of 0.5 mM Fe2þ þ 0.1 mM Cu2þ, and the solution reaches almost total mineralization (96% TOC removal) at the end of electrolysis. This phenomenon can be ascribed to the competitive formation of Cu(II) complexes with some byproducts that are rapidly oxidized with the excess of •OH produced from Fe2þ regeneration at the CF cathode in the combined cell.41-43 In contrast, results of Figure 2b evidence that UVA irradiation causes the total mineralization (>98% TOC removal) of the metoprolol tartrate solution by the comparative PEF processes. Overall decontamination is achieved after about 180 and 120 min of treatment in the single BDD/ADE and combined BDD/ ADE-Pt/CF cells, respectively, practically regardless of the presence and absence of Cu2þ ion. This potent action of UVA light can be due to the generation of more •OH from Fenton’s reaction (2) induced by the photocatalytic reaction (5) and/or the additional photodecomposition of complexes of Fe(III) with byproducts such as carboxylic acids.15,23-25 The generation of a greater quantity of •OH from the acceleration of Fe2þ regeneration at the CF cathode explains the faster degradation in the combined cell than in the single cell. However, no significant 1237



ARTICLE

Figure 3. Mineralization current efficiency calculated from eq 7 vs electrolysis time for the experiments of Figure 2 with 0.5 mM Fe2þ þ 0.1 mM Cu2þ as catalyst. (O) EF in BDD/ADE at 120 mA, (4) EF in BDD/ ADE-Pt/CF cell at 120-12 mA, (b) PEF in BDD/ADE at 120 mA, and (2) PEF in BDD/ADE-Pt/CF cell at 120-12 mA.

influence of the Cu2þ/Cuþ system takes place in PEF, probably because it is much quicker the photolysis of Fe(III) complexes formed. The much greater oxidation power of the PEF than EF processes can be analyzed from the corresponding MCE values calculated from eq 7, taking currents of 120 mA for the single BDD/ADE cell and 132 mA for the combined BDD/ADE-Pt/ CF one. The results obtained for 0.5 mM Fe2þ þ 0.1 mM Cu2þ are presented in Figure 3. While an analogous maximum efficiency of 78-82% can be observed at 40 min of the PEF treatment in both cells as a result of the large influence of UVA light, the use of EF decreases this maximum value to 50% at the same time for the combined cell, which even falls to about 30% for the single one. When electrolysis time is prolonged, however, the efficiency decays dramatically in all cases due to the production of byproducts that are more difficultly destroyed and, then, similar MCE values are reached in each process, independent of the cell configuration used. Thus, Figure 3 shows that the presence of Cu2þ ion in EF using the combined cell enhances the degradation of metoprolol tartrate during the initial 180 min of treatment, whereupon efficiencies similar to those of the single cell are obtained. This confirms that the faster destruction of organics in the more potent combined cell mainly takes place at the first stages of the EF treatment. 3.3. Kinetics for Metoprolol Decay. When the (2:1) metoprolol tartrate complex is dissolved in aqueous medium, it dissociates into its components. This was corroborated by determining by ion-exclusion chromatography that the initial 0.246 mM metoprolol tartrate solutions contained 0.246 mM tartaric acid and, hence, a metoprolol concentration C0 = 0.492 mM. The role of generated •OH in the destruction of metoprolol was then analyzed from its concentration decay measured by reversed-phase HPLC, where it exhibited a welldefined absorption peak at a retention time (tr) of 3.59 min. Blank experiments performed with and without 20 mM H2O2 under UVA illumination did not show any significant removal of metoprolol, indicating that it is neither attacked by electrogenerated H2O2 nor directly photolyzed by UVA light. Figure 4 depicts that the drug is completely destroyed in all EAOPs tested by the efficient attack of •OH and BDD(•OH). As can be seen in Figure 4a for EF with 0.5 mM Fe2þ in the BDD/ ADE cell, metoprolol disappears in 32 min, whereas with the use of the BDD/ADE-Pt/CF cell it is much more rapidly removed

Figure 4. Decay of metoprolol concentration (C0 = 0.492 mM) with electrolysis time during the treatment of 0.246 mM metoprolol tartrate solutions by the (a) EF and (b) PEF processes under the same conditions as given in Figure 2. Each inset panel illustrates the corresponding kinetic analysis assuming that metoprolol follows a pseudo-first-order reaction.

in only 21 min. Slightly shorter removal times can also be observed for each cell when 0.1 mM Cu2þ is added, indicating that larger •OH production takes place by the acceleration of Fenton’s reaction (2) from Fe2þ regeneration at the CF cathode than by reaction 17 from the Cu2þ/Cuþ system, in agreement with TOC results of Figure 2a. The concentration decays for the above assays were well fitted to a pseudo-first-order kinetic equation, as shown in the inset panel of Figure 4a, and a pseudo-first-order rate constant (k1) of 1.7 10-3 s-1 (square of the regression coefficient (R2) = 0.999) for 0.5 mM Fe2þ in BDD/ADE cell, 2.0 10-3 s-1 (R2 = 0.987) for 0.5 mM Fe2þ þ 0.1 mM Cu2þ in BDD/ADE cell, 3.2 10-3 s-1 (R2 = 0.998) for 0.5 mM Fe2þ in BDD/ADE-Pt/CF cell, and 3.7 10-3 s-1 (R2 = 0.986) for 0.5 mM Fe2þ þ 0.1 mM Cu2þ in BDD/ADE-Pt/ CF cell are obtained. This behavior is indicative of a constant generation of •OH in the medium from Fenton’s reaction (2) and reaction 17, and of BDD(•OH) at the anode surface from reaction 4. Figure 4b evidences a large •OH promotion from reaction 5 for the PEF treatments, where metoprolol also follows a pseudofirst-order decay (see inset panel). For PEF with 0.5 mM Fe2þ in the BDD/ADE cell, the drug undergoes the slower decay with k1 = 2.9 10-3 s-1 (R2 = 0.993) and is completely removed in 21 min, i.e., much quicker than the comparative EF process, but slightly slower than using the BDD/ADE-Pt/CF cell. The Fe2þ regeneration at the CF cathode in the combined cell is then more effective to produce a greater quantity of •OH than UVA irradiation, although the latter is much superior to that of the Cu2þ/Cuþ system. However, Figure 4a also shows a quite similar 1238



ARTICLE

Table 1. Aromatic Intermediates Detected by GC-MS for the Electro-Fenton Degradation of Metoprolol Tartrate

and slightly faster removal of the drug for the other PEF processes with 0.1 mM Cu2þ and/or the combined cell. The k1 values under these conditions vary between 3.3 10-3 and 3.5 10-3 s-1, being practically equal to that of the more potent EF process with 0.5 mM Fe2þ þ 0.1 mM Cu2þ in the BDD/ ADE-Pt/CF cell (see Figure 4a). That means that these EAOPs have already reached their maximum oxidation ability, probably by the limitation of the mass transfer of the drug toward the anode and the diffusion of reactants in the bulk. 3.4. Identification of Aromatic Products. Table 1 summarizes the primary aromatic intermediates, along with their main characteristics, detected by GC-MS after 60 min of electrolysis of the 0.246 mM metoprolol tartrate solution by the less potent EF method with 0.5 mM Fe2þ in the BDD/ADE cell at a current as low as 20 mA, where the drug persists more than 90 min. Taking into account these products, a plausible path for the initial degradation of metoprolol under the action of BDD(•OH) and •OH as the main oxidizing species is proposed in Figure 5. The process is initiated by the hydroxylation of metoprolol with breaking of its C(1)-O bond giving 4-(2methoxyethyl)phenol and the release of (2-hydroxy-3-isopropylamine)propoxyl radical. Further oxidation of 4-(2-methoxyethyl)phenol leads to methyl 4-hydroxyphenylacetate, followed by its hydroxylation to 2-hydroxy-2-(4-hydroxyphenyl)acetic acid with loss of methanol. The oxidation of this product causes the release of formic acid yielding 4-hydroxybenzaldehyde, which is probably degraded by cleavage of its benzene moiety to produce shorter byproducts such as linear carboxylic acids. To clarify the degradative behavior of the above aromatics in the EAOPs tested, the commercially available 4-(2-methoxyethyl)phenol was taken as an example and its evolution was followed by reversed-phase HPLC. The chromatograms for electrolyzed solutions exhibited a well-defined peak related to this compound at tr = 1.45 min, confirmed by comparing their retention times and UV-vis spectra, measured on the diode array detector, with those of the pure compound. Figure 6a illustrates that 4-(2-methoxyethyl)phenol is quickly formed and destroyed for all EF processes, attaining maximum contents of

Mineralization of Metoprolol by Electro-Fenton and Photoelectro

Recommend Documents