Heterogeneous photocatalytic degradation of gallic ...

This article is published as part of a themed issue of Photochemical & Photobiological Sciences in honour of Esther Oliveros Guest edited by Marie-Thérèse Maurette and Guillermo Orellana Published in issue 7, 2009

Other articles in this issue include: Sensitized formation of oxidatively generated damage to cellular DNA by UVA radiation J. Cadet, T. Douki, J-L. Ravanat and P. Di Mascio, Photochem. Photobiol. Sci., 2009, 8, 903

Physical and chemical quenching rates and their influence on stereoselective photooxygenation of oxazolidinone-functionalized enecarbamates M. R. Solomon, J. Sivaguru, S. Jockusch, W. Adam and N. J. Turro, Photochem. Photobiol. Sci., 2009, 8, 912

Water disinfection with Ru(II) photosensitisers supported on ionic porous silicones F. Manjón, D. García-Fresnadillo and G. Orellana, Photochem. Photobiol. Sci., 2009, 8, 926

Photophysics and photochemistry of rose bengal bound to human serum albumin E. Alarcón, A. M. Edwards, A. Aspée, C. D. Borsarelli and E. A. Lissi, Photochem. Photobiol. Sci., 2009, 8, 933

Photolysis of an asymmetrically substituted diazene in solution and in the crystalline state P. A. Hoijemberg, S. D. Karlen, C. N. Sanramé, P. F. Aramendía and M. A. García-Garibay, Photochem. Photobiol. Sci., 2009, 8, 961

Heterogeneous photocatalytic degradation of gallic acid under different experimental conditions N. Quici and M. I. Litter, Photochem. Photobiol. Sci., 2009, 8, 975

Photolysis of ferric ions in the presence of sulfate or chloride ions: implications for the photo-Fenton process A. Machulek Jr., J. Ermírio F. Moraes, L. T. Okano, C. A. Silvério and F. H. Quina, Photochem. Photobiol. Sci., 2009, 8, 985

Light and dark biocidal activity of cationic poly(arylene ethynylene) conjugated polyelectrolytes T. S. Corbitt, L. Ding, E. Ji, L. K. Ista, K. Ogawa, G. P. Lopez, K. S. Schanze and D. G. Whitten, Photochem. Photobiol. Sci., 2009, 8, 998

Optimisation of the chemical generation of singlet oxygen (1O2, 1∆g) from the hydrogen peroxide–lanthanum(III) catalytic system using an improved NIR spectrometer C. Pierlot, J. Barbillat, V. Nardello-Rataj, D. Mathieu, M. Sergent, J. Marko and J-M. Aubry, Photochem. Photobiol. Sci., 2009, 8, 1024

Photocatalytic efficiencies of self-cleaning glasses. Influence of physical factors L. Peruchon, E. Puzenat, J. M. Herrmann and C. Guillard, Photochem. Photobiol. Sci., 2009, 8, 1040

PAPER

www.rsc.org/pps | Photochemical & Photobiological Sciences

Heterogeneous photocatalytic degradation of gallic acid under different experimental conditions† Natalia Quicia,b and Marta I. Litter*a,b,c Received 30th January 2009, Accepted 23rd March 2009 First published as an Advance Article on the web 29th April 2009 DOI: 10.1039/b901904a UV/TiO2 -heterogeneous photocatalysis was tested as a process to degrade gallic acid (Gal) in oxygenated solutions at pH 3. In the absence of oxidants other than oxygen, decay followed a zero order rate at different concentrations and was slow at concentrations higher than 0.5 mM. Addition of Fe3+ , H2 O2 and the combination Fe3+ /H2 O2 improved Gal degradation. In the absence of H2 O2 , an optimal Fe : Gal molar ratio of 0.33 : 1 was found for the photocatalytic decay, beyond which addition of Fe3+ was detrimental and even worse in comparison with the system in the absence of Fe3+ . TiO2 addition was beneficial compared with the same system in the absence of the photocatalyst if Fe3+ was added at low concentration (0.33 : 1 Fe : Gal molar ratio), while at high concentration (1 : 1 Fe : Gal molar ratio) TiO2 did not exert any significant effect. H2 O2 addition (1 : 0.33 Gal : H2 O2 molar ratio, absence of Fe(III)) also enhanced the heterogeneous photocatalytic reaction. Simultaneous addition of Fe3+ and H2 O2 was more effective than the addition of the separate oxidants. This system was compared with Fenton and photo-Fenton systems. At low H2 O2 concentration (0.33 : 1 : 0.2 Fe : Gal : H2 O2 molar ratio), the presence of TiO2 also enhanced the reaction. The influence of the thermal charge transfer reaction between Gal and Fe(III), which leads to an important Gal depletion in the dark with formation of quinones, was analysed. The mechanisms taking place in these complex systems are proposed, paying particular attention to the important charge transfer reaction of the Fe(III)–Gal complex operative in dark conditions.

Introduction Gallic acid (3,4,5-trihydroxybenzoic acid, Gal, MW = 170) may be considered as one of the simplest models for natural organic matter, i.e. of humic and fulvic acids.1 It is a natural product of hydrolysis of tannins and one of the main constituents of herbal roots and tea leaves.2,3 Gal is considered to be a good antioxidant4 and an effective antimicrobial compound.5 For these reasons, novel antioxidative and antimicrobial food additives are being developed using gallic acid as a starting compound. Gal is present in wastewaters from olive oil factories, distilleries or wineries and boiling cork.6,7 It is also known to prevent corrosion of mild steel and Zn pigments in aqueous solutions,4 and it was proposed to be used in combination with EDTA or NTA to dissolve magnetite or hematite in cleaning and decontaminant processes of cooling circuits of nuclear reactors.8,9 The maximum contaminant level for gallic acid in drinking water established by environmental protection organizations of Canada, USA and EEC is 2 mg L-1 .10 For all these reasons, methods for destroying Gal present in effluents are of interest. a Gerencia Qu´ımica, Centro Atómico Constituyentes, Comisión Nacional de Energ´ıa Atómica (CNEA), Av. Gral. Paz 1499, CP 1650, San Mart´ın, Prov. de Buenos Aires, Argentina b Consejo Nacional de Investigaciones Cient´ıficas y Técnicas (CONICET), Av. Rivadavia 1917, CP 1033, Ciudad de Buenos Aires, Argentina c Instituto de Investigación e Ingenier´ıa Ambiental, Universidad Nacional de San Mart´ın, Peatonal Belgrano 3563, 1◦ piso, CP 1650, San Mart´ın, Prov. de Buenos Aires, Argentina † This article was published as part of the themed issue in honour of Esther Oliveros.

Biological treatments may be not convenient for this purpose because gallic acid, as a phenolic compound, can be biorefractory. Therefore, earlier transformation to more biodegradable compounds may be a solution. For this, the use of Advanced Oxidation Processes (AOPs) as a pretreatment might be suitable. AOPs involve the generation of very active oxidising species, from which the main and most reactive one is the hydroxyl radical (HO∑ ).11–15 Among AOPs, TiO2 -heterogeneous photocatalysis (HP) is one of the most studied, together with Fenton (F) and photo-Fenton (PF) processes. In previous works, we studied HP systems for the degradation of various environmentally important aliphatic carboxylic acids, namely EDTA, NTA, oxalic acid and citric acid. The effect of addition of H2 O2 and Fe salts to the TiO2 -photocatalytic systems was also presented, in some cases together with a comparison with other AOPs such as photo-Fenton or ferrioxalate.16–25 Several papers describe Gal degradation by AOPs. Ben´ıtez et al.6 studied Gal degradation using 254 nm-photolysis and its combination with H2 O2 , Fenton and photo-Fenton processes. Gernjak et al.26 reported the rapid mineralization of Gal by photo-Fenton under artificial UV-A and solar light. Pariente et al.27 used silica supported hematite particles in a photo-Fenton process to degrade gallic acid before combination with a biological treatment. Treatment with the 2,4,6-triphenylpyrylium salt under solar irradiation was also proposed,28 while ozone and chlorine dioxide were tested as oxidants as well.10,29,30 Removal of gallic acid by electroprecipitation with a sacrificial iron anode, by electroand photoelectro-Fenton processes, and by peroxicoagulation was also described.31 Lucas et al.7 reported the use of UV,

This journal is © The Royal Society of Chemistry and Owner Societies 2009

Photochem. Photobiol. Sci., 2009, 8, 975–984 | 975

UV/H2 O2 , photo-Fenton, ferrioxalate and HP. Other studies using HP were informed by Gimeno et al.32 and by Gumy et al.33 Investigations on the degradation of gallic acid by VUVphotolysis in aqueous solution were performed previously by our group.34 Gal is rather stable at acidic pH (< 6) in the presence of oxygen, but it is readily oxidised to an o-quinone under alkaline conditions.5 As several polyphenols, it forms complexes with Fe(III) and Fe(II). In the case of Fe(III), the structure of the complex depends on pH and on the Fe : Gal molar ratio.35–38 In excess of ligand, 1 : 1, 2 : 1 and 3 : 1 complexes are formed, depending on pH;37 the formation of a 1 : 2 Gal : Fe complex in a large excess of the metal has been informed.38 When Fe(III) is added to a Gal solution (pH < 3), a dark blue colour appears, which rapidly fades to a pale yellow-green.35–38 The blue colour is attributed to the 1 : 1 Fe(III)–Gal complex (log K = 19.238 ), formed with the iron coordinated to two adjacent –OH groups,37,38 and absorbing around 690 nm.38 This complex is the primary compound in the writing “gall” inks.5 The fade of the colour is attributed to a redox process, leading to Gal degradation and Fe(II) formation, according to the following simplified equation:35,38 Fe3+ + 12 Gal Fe2+ + 12 o-quinone + H+

(1)

In this work, studies on the TiO2 -heterogeneous photocatalysis of gallic acid in oxygenated suspensions were performed, analysing the effect of addition of Fe3+ , H2 O2 and both oxidants simultaneously. A comparison with the homogenous system (absence of TiO2 ) either in the dark or under irradiation is also presented. Foreseeing a further combination with biological treatments to save costs, a low amount of H2 O2 was tested in the reported experiments to avoid a deleterious effect on the microorganisms or an expensive step for elimination of the reagent.

Experimental

Analytical techniques

Chemicals Gallic acid monohydrate (C7 H6 O5 ·H2 O) was from J.T. Baker (99.9%). TiO2 was Degussa P-25, supplied by Degussa AG, Germany. Hydrogen peroxide (Merck, 30% (w/w)) was used. Iron was added as ferric perchlorate (Aldrich, analytical grade). All other reagents were analytical grade and used without further treatment. For pH adjustments, 0.2 M HClO4 and NaOH were used. Water was purified with a Millipore Milli Q equipment (resistivity = 18 MX cm). Degradation experiments Irradiation experiments were performed in a recycling system (1.25 L min-1 flow rate) consisting of an annular reactor (415 mm length, 35 mm external diameter, 28 mm internal diameter, 85 mL total volume), a peristaltic pump and a thermostatted (298 K) 350 mL cylindrical reservoir. A black-light tubular UV lamp (Philips TLD/08, 15 W, 350 < l/nm < 410, 100% maximum emission at 366 nm) was installed inside the annular reactor as the source of illumination. Actinometric measurements were performed by the ferrioxalate method.39 An incident photon flux 976 | Photochem. Photobiol. Sci., 2009, 8, 975–984

per unit volume (qp /V , where qp is the incident photon flux and V is the irradiated volume) of 14.6 meinstein s-1 L-1 was calculated. In all cases, 300 mL of a fresh Gal solution (0.1–2 mM) were adjusted to pH 3 with drops of 0.2 M HClO4 and poured into the reservoir, where air (1.5 L min-1 ) was constantly bubbled. Solutions or suspensions in the reservoir were magnetically stirred all throughout the reaction time. In HP experiments, Degussa P-25 (1.0 g L-1 ) was suspended in the solution containing Gal (and Fe in the corresponding cases). The suspension was ultrasonicated for 2 min to homogenise the system, and then stirred into the reservoir for 30 min in the dark. Addition of Fe3+ was made from a 4 mM Fe(ClO4 )3 solution (prepared at pH 2 with HClO4 to avoid ferric ion precipitation). The solution or suspension was then adjusted to pH 3 with drops of 0.2 M NaOH. According to Pignatello,40 degradation by Fenton reactions is sensitive to the iron counterion, ClO4 - presenting the highest rate in comparison with other anions. In experiments with H2 O2 , the corresponding volume for the desired molar ratio was added immediately before starting the lamp. This was taken as the initial time for reactions under light. To compensate H2 O2 consumption, volumes equal to the initial one were added every 30 min during the course of the reactions, except in the indicated cases. In all runs, pH was measured at the beginning and at the end of the experiment. Variations higher than 0.2 units were never found. Experiments in the dark were performed similarly. Samples (1 mL) were periodically extracted for HPLC or TOC analysis; those containing TiO2 were filtered first through 0.45 mm cellulose membranes. In all experiments with H2 O2 (except indicated), methanol (1 mL) was added before analysis as a quenching reagent to interrupt the reaction. All runs were performed at least in duplicate and the results averaged. The experimental error was never higher than 10%.

The temporal evolution of Gal during the experiments was followed by HPLC using a Konik-500-A chromatograph with a UV/VIS Thermo Separation Products UV 100 detector and a Konikrom Chromatography Data System V.5.2 data acquisition system. The following conditions were used: 4.6 mm RP-C18 Prevail (Alltech) column; 5 mM KH2 PO4 at pH 2.5 (H3 PO4 ) as mobile phase; flux: 1 mL min-1 ; UV detection at 271 nm. To eliminate Fe(III) and avoid precipitation with phosphate inside the column, samples were diluted with a phosphate solution 4-fold more concentrated than that in the eluent. The precipitate formed between iron and phosphate was separated through a 0.2 mm Millipore membrane before injection. The mineralization degree was followed by total organic carbon (TOC) analysis, using a Shimadzu 5000 A TOC analyser in the non-purgeable organic carbon (NPOC) mode. Hydrogen peroxide consumption was followed spectrophotometrically using ammonium metavanadate and measurement at 454 nm.23 Spectrophotometric measurements were performed with a Hewlett–Packard diode array UV-visible recording spectrophotometer, model HP8453A.


Results Dark reactions When an aqueous solution of Gal was put in contact with TiO2 , a strong orange colour was immediately developed. After a 30 min stirring period in the dark and filtration by a 0.45 mm membrane, the resulting filtered solution was colourless but the solid retained the colour, indicative of adsorption of Gal onto the TiO2 surface and formation of a surface complex, as reported previously.41 In spite of this, changes in Gal concentration before and after the 30 min period in the dark were almost inappreciable, especially at the highest concentrations. No Gal degradation was detected in a blank in the dark performed in the presence of H2 O2 (1 : 1 Gal : H2 O2 molar ratio), as found by other authors even working with higher H2 O2 concentrations.7 As said before, Fe(III) and Gal, when mixed, formed immediately a deeply coloured complex; at pH 3, the spectrum of the mixture showed a broad band centred at 690 nm. The stability of the complex in the dark was tested with an equimolar mixture of Fe(III) and Gal (both 1 mM) at pH 3. UV-Vis spectra were taken every 10 s to follow the changes. A decrease of the 690 nm band and an increase of a new one centred at 390 nm, together with the change of colour of the solution to pale green were observed. After 60 s, spectral changes became negligible. In separate experiments, Gal concentration was measured by HPLC after mixing 1 mM Gal with Fe(III) at various concentrations (pH 3). After 90 s stirring in the dark, Gal depletions ranged 15–96% for Fe : Gal molar ratios between 0.6 : 1 and 6 : 1. No more than 96% of conversion could be attained even at higher Fe : Gal ratios. The Gal decay was however, negligible for the lowest Fe : Gal molar ratios (0.05 : 1 to 0.33 : 1). Similar decays were obtained adding 1 g L-1 TiO2 to selected samples, with the same changes in colour as in the homogeneous systems. Results are shown in Table 1. Prolonged experiments at the 1 : 1 and 0.33 : 1 Fe : Gal molar ratios did not show any further changes on Gal concentration up to 120 min stirring in the dark. These results indicate that Gal decay in systems with high iron content is important, it is minor at low iron contents and it stops Table 1 Decay of Gal in the dark measured 90 s after adding Fe(III). Experimental conditions: [Gal]0 = 1 mM, pH 3, open to air, T = 298 K Fe : Gal molar ratio

% Gal decay

0.05 : 1 0.1 : 1 0.33 : 1b 0.6 : 1 0.8 : 1 1 : 1c 2:1 2.5 : 1 3.3 : 1 6:1 8.25 : 1 10 : 1 15 : 1 25 : 1

NDa ND ND 15 21 28 59 65 82 95 96 95 95 96

ND: not detected. b Same result in the presence of 1 g L-1 TiO2 . c 30% in the presence of 1 g L-1 TiO2 .

a

in less than two minutes in all conditions. In the chromatograms of the reacting mixtures, several peaks were observed, which would correspond to intermediates of Gal degradation absorbing at 271 nm, but they were not identified. TiO2 -photocatalytic reactions The decay of Gal at different initial concentrations (0.1–2 mM, pH 3) under UV irradiation in the presence of 1 g L-1 TiO2 and oxygen is shown in Fig. 1. The working pH was 3, a value usually taken in previous HP studies of our group and suitable to avoid precipitation in systems containing iron(III). As said before, an orange colour was developed on the TiO2 surface immediately after contacting Gal with the photocatalyst, while the filtered solution after removing TiO2 was colourless. Alongside the photocatalytic runs, a yellow colour appeared on the filtrate of samples, indicative of the formation of quinones (at the 1 mM concentration the colour appeared around 30–40 min of irradiation, i.e. when ca. 10% of Gal degradation was attained). As expected, no degradation took place in the absence of TiO2 , because Gal does not absorb at the wavelengths emitted by the lamp.1,35

Fig. 1 Decay of Gal concentration under UV irradiation over TiO2 . Experimental conditions: pH 3, [TiO2 ] = 1 g L-1 , air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . Dotted lines are linear fits.

The decays were linear (zero order kinetics) with rather good correlation coefficients (R2 > 0.84). Important conversions were obtained for the lowest concentrations tested, reaching 100% for 0.1 mM in 60 min. In the chromatograms of the reacting mixtures, the same intermediates appearing in dark reactions were observed, together with other new ones. Again, no attempts to identify these products were made. A 1 mM Gal concentration was chosen in all further experiments to follow reasonably the temporal changes during the 120 min runs. Effect of Fe(III) addition to the TiO2 -photocatalytic reactions The temporal evolution of normalised Gal concentration ([Gal]/[Gal]0 , [Gal]0 = 1 mM) under UV irradiation in the presence of TiO2 , oxygen and different amounts of iron is shown in Fig. 2. For comparison, the plot in the absence of Fe(III) is also included.



Table 2 Percentages of Gal degradation after 120 min irradiation and rate constants (k) taken from Fig. 2 and 3

Fig. 2 Temporal profiles of normalised Gal concentration ([Gal]/[Gal]0 ) under UV irradiation over TiO2 in the absence and presence of Fe(III). Experimental conditions: [Gal]0 = 1 mM, pH 3, [TiO2 ] = 1 g L-1 , air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . The Fe : Gal molar ratio is indicated. Dotted lines are exponential fits with the exception of the plot in the absence of Fe(III), which is linear. For the 0.6 : 1 to 2 : 1 conditions, the normalised concentration at t = 0 is the value after the decay in the dark.

When the dark thermal reaction was important, i.e. from 0.6 : 1 to 2 : 1 Fe : Gal molar ratios, the dark decay before starting the lamp was discounted; the corresponding normalised Gal concentration a t = 0 is indicated in the plot. In the absence of iron, degradation was slow, reaching 37% after 120 min of irradiation. With iron, from 0.05 : 1 to 0.33 : 1 molar ratios, i.e. when the reaction in the dark was negligible (see Table 1), the degradation under irradiation was important (ranging 39 to 55% at 120 min), with an optimal value at the 0.33 : 1 Fe : Gal ratio. In contrast, at higher Fe(III) concentrations (0.6 : 1 to 2 : 1 molar ratios), the contribution of the reaction under light was very poor, with Gal decays lower than in the absence of Fe(III). It is important to remember that at these Fe : Gal molar ratios, after the initial decay registered in the first 60 s, no additional reaction was observed in the dark. Obviously, at the 2 : 1 condition, due to the importance of the dark reaction, global Gal depletion—including the period before starting the lamp—was the highest. In all cases, the final filtered solution presented the typical yellow colour of quinones. The same peaks as observed in the chromatograms of the photocatalytic reactions in the absence of iron were observed. In contrast with the results in the absence of iron, a different kinetic regime was obtained, rather close to an apparent first order, with good correlation coefficients (R2 = 0.94–0.99); a deceleration after 90 min could be seen in some of the systems. This change in the kinetics after addition of iron to a photocatalytic system has been also observed in most of our previously studied cases.16–19,24 The rate constants (k), calculated from the exponential plots (with the exception of the reaction in the absence of Fe, which was calculated from the linear plot), are shown in Table 2, and they reflect the same trend in conversions at 120 min. Mineralization at 120 min (measured by TOC decrease) attained 10% in the absence of Fe(III) and never surpassed 20% in the 978 | Photochem. Photobiol. Sci., 2009, 8, 975–984

Fe : Gal molar ratio

% Gal degradation

k ¥ 103 /min-1

0:1 0.05 : 1 0.1 : 1 0.33 : 1 0.33 : 1 (no TiO2 ) 0.6 : 1 0.8 : 1 1:1 1 : 1 (no TiO2 ) 2:1

37 39 49 55 9 39 36 24 21 15 (at 90 min)

3.2 4.4 5.9 7.5 0.9 3.6 3.9 2.6 2.2 1.6

presence of Fe(III), value obtained at the optimal 0.33 : 1 Fe : Gal molar ratio. Fig. 3 compares selected experiments in the presence and absence of TiO2 . It can be observed that the contribution of the photolytic reaction to the decay is very poor: 9% after 120 min of irradiation for the 0.33 : 1 molar ratio and somewhat higher (21%) for the 1 : 1 condition. Comparison with the results in the presence of TiO2 indicates that the photocatalyst significantly accelerates the decay for the 0.33 : 1 condition, reaching 55% after 120 min of irradiation, while it has almost no effect for the 1 : 1 molar ratio. Decays at 120 min and initial first order rate constants for the reactions in the absence of TiO2 are presented also in Table 2.

Fig. 3 Selected temporal profiles of normalised Gal concentration ([Gal]/[Gal]0 ) under UV irradiation in the presence of Fe(III) with and without TiO2 . Experimental conditions: [Gal]0 = 1 mM, pH 3, [TiO2 ] = 1 g L-1 , air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . The Fe : Gal molar ratio is indicated. Dotted lines are exponential fits. For both 1 : 1 conditions, the normalised concentration at t = 0 is the value after the decay in the dark.

Effect of addition of Fe(III) plus H2 O2 to the TiO2 -photocatalytic reactions Fig. 4 shows the profiles of normalised Gal decay vs. time comparing the pure TiO2 -photocatalytic system (0 : 1 : 0 Fe : Gal : H2 O2 molar ratio) with those in which H2 O2 and H2 O2 plus Fe3+ were added. It is important to note that the stoichiometric H2 O2 demand needed for the total mineralization of 1 mM Gal (in the absence of other oxidants) is 12 mM, calculated according to:


Combination of both oxidants was much more effective, and the decay increased with the increasing H2 O2 concentration (cf. 0.33 : 1 : 0.2 and 0.33 : 1 : 1), Gal being totally consumed in almost 1 h at the highest amount of H2 O2 concentration. Comparing the 0.33 : 1 : 0.2 and the 0 : 1 : 0.33 conditions, it can be observed that Fe(III) addition improved the conversion slightly, more markedly after 60 min. TOC could not be measured in these experiments due to the addition of methanol as quenching agent. Fenton and photo-Fenton reactions of Gal degradation. Effect of TiO2 addition

Fig. 4 Temporal profiles of normalised Gal concentration ([Gal]/[Gal]0 ) under UV irradiation over TiO2 in the presence of Fe(III) and H2 O2 . Experimental conditions: [Gal]0 = 1 mM, pH 3, [TiO2 ] = 1 g L-1 , air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . The initial Fe : Gal : H2 O2 molar ratios are indicated. Arrows indicate the points of H2 O2 addition in the corresponding cases.

C7 O5 H6 + 12H2 O2 → 7CO2 + 15H2 O

(2)

However, H2 O2 was added in all the experiments at initial concentrations lower than this stoichiometric amount, according to the requirements stated in the Introduction section for a further biological treatment. Fe3+ was added at the optimal condition found in the previous photocatalytic experiments (0.33 : 1). This election is justified by the negligible dark reaction between Gal and Fe3+ , which allows the extraction of conclusions about only the effect of the light. In addition, this low amount of Fe(III) is suitable to decrease the amount of sludge that could be formed and should require an additional elimination step. In the experiments in the presence of H2 O2 , an arrest of the Gal decay was observed due to depletion of the oxidant, especially at the highest H2 O2 concentration; therefore, supplementary hydrogen peroxide additions were made at 30 and 60 min, with a recovery of the decay after each addition. H2 O2 in the absence of Fe3+ was also tested; an initial 0.33 mM concentration was used, but, due to the successive additions, the total amount of H2 O2 added in this experiment was 1 mM. In the other two experiments, the amounts of H2 O2 added were 0.6 and 3 mM, respectively, but the total addition was never higher than the stoichiometric amount. After the reaction, the final filtered solution presented also the yellow colour typical of quinones; when 100% Gal degradation was obtained (0.33 : 1 : 1 HP), the yellow colour disappeared, leaving a transparent solution. The same peaks as observed in the chromatograms of the previous photocatalytic reactions were observed. Conditions in which only Fe(III) or H2 O2 was added (i.e. 0.33 : 1 : 0 and 0 : 1 : 0.33) presented rather similar initial rates, but the extent of degradation at 120 min was higher for the system containing H2 O2 (81 vs. 55%), in which supplementary H2 O2 was added.

It is important to highlight that the objective of the work was not the study of Fenton processes but only its comparison with TiO2 -photocatalytic systems combined with Fe(III) and H2 O2 . The optimal conditions found before were used, i.e. 0.33 mM Fe(III) and H2 O2 in much lower amounts than the stoichiometric amount for mineralization. Our experimental conditions are then different from those of previous studies.5–7,26 Fig. 5 shows the temporal profiles of normalised Gal concentrations for the system in the absence of TiO2 in the dark (F) and under irradiation (PF) with different initial H2 O2 additions. Analogous to the experiments of Fig. 4, after the initial H2 O2 addition, an arrest of the reaction was observed due to depletion of the oxidant, and then supplementary additions were made. Likewise as in the TiO2 experiments, an initial rapid Gal decay was observed, the rates increasing with the increasing initial H2 O2 concentration. Fig. 5 compares also selected HP experiments with those in the absence of TiO2 . There were no differences in the peaks observed in the chromatograms with and without TiO2 , indicating that same intermediates are formed in HP, F and PF reactions (at least those absorbing at 271 nm). It can be observed that TiO2 increases final conversions at 120 min when low H2 O2 concentrations (0.33 : 1 : 0.2) are used,

Fig. 5 Temporal profiles of normalised Gal concentration ([Gal]/[Gal]0 ) in the presence of Fe(III) with increasing H2 O2 amounts (Fenton and photo-Fenton processes). Selected HP experiments from Fig. 4 are included. Experimental conditions: [Gal]0 = 1 mM, pH 3, [TiO2 ] = 1 g L-1 , air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . The initial Fe : Gal : H2 O2 molar ratio is indicated. Arrows indicate the points of H2 O2 addition. The inset shows the linear increase of the initial rate constants with the H2 O2 concentration.



Table 3 Percentages of Gal degradation after 120 min and initial rate constants (kin ) of F, PF and HP reactions shown in Fig. 4 and 5 Fe : Gal : H2 O2 molar ratio

% Gal degradation

kin ¥ 103 /min-1

0 : 1 : 0 HP 0.33 : 1 : 0 HP 0 : 1 : 0.33 HP 0.33 : 1 : 0.1 F 0.33 : 1 : 0.1 PF 0.33 : 1 : 0.2 F 0.33 : 1 : 0.2 PF 0.33 : 1 : 0.2 HP 0.33 : 1 : 0.33 F 0.33 : 1 : 0.33 PF 0.33 : 1 : 1 F 0.33 : 1 : 1 PF 0.33 : 1 : 1 HP

37 55 81 33 59 53 66 88 75 79 100 100 100

3.2a 7.5a 12 24 27 41 43 29 59 49 136 122 159

a

Value taken from Table 2. Fig. 6 Decay profiles of Gal (full symbols) and TOC (empty symbols) for the F (䊉/䊊), PF (/) and HP (䉱/) systems. Experimental conditions: [Gal]0 = 1 mM, pH 3, air bubbling (1.5 L min-1 ), T = 298 K, 350 < l/nm < 410, qp /V = 14.6 meinstein s-1 L-1 . H2 O2 (0.98 M) was added as indicated in the text.

although, at the first stages, the decay is somewhat lower for the heterogeneous system. At the 0.33 : 1 : 1 molar ratio, the presence of the photocatalyst increases the initial Gal degradation, but the total Gal removal is achieved at the same time (90 min). Noticeably, in the 0.33 : 1 : 0.2 condition, the plateau observed in the homogenous system is almost absent in the presence of TiO2 , suggesting that H2 O2 is not completely consumed in the interval and that Gal oxidation proceeds by a more rapid pathway in the heterogeneous system. Because a kinetic analysis would have been very complicated, initial rate constants (kin ) were calculated in these cases taking the two first points (0 and 5 min) of the corresponding curves of Fig. 4 and 5 and are shown in Table 3 together with the Gal decay at 120 min. As H2 O2 is under the stoichiometric demand, it is rapidly consumed, and the calculated rates can be considered only lower bounds. In this table, values of previous experiments without H2 O2 (0 : 1 : 0 and 0.33 : 1 : 0) are also presented, although it is important to remark that only a very rough comparison can be made because of the intrinsic kinetic differences of the systems. It can be observed that reactions in the presence of H2 O2 are faster than in its absence, with initial rate constants around one order of magnitude higher. Initial rates of F and PF reactions are similar, but conversions are higher for PF systems at longer irradiation times, especially at low H2 O2 concentrations (0.33 : 1 : 0.1 and 0.33 : 1 : 0.2). In the other two cases, the differences between dark and irradiated systems are minor and under the limits of error, the complete Gal conversion being reached at the same time (90 min). A plot of kin vs. [H2 O2 ]in was linear for the three AOP systems and is shown in the inset of Fig. 5.

strength the oxidizing power of the system and speed up the conversion. Thus, 0.1 mmol were added at 15 and 20 min, 0.3 mmol at 25 and 30 min, 0.9 mmol at 35 min, 1.80 mmol at 60 min and 3.6 mmol at 120 min. Periodical spectrophotometric analysis of the H2 O2 content indicated that in the Fenton systems, H2 O2 became in excess after 60 min, while in the photo-Fenton and TiO2 systems, the reagent was always totally consumed before the next peroxide addition. In these experiments, to allow TOC measurements, methanol was not added to quench the reaction. Although results are not completely accurate because the systems continued reacting rather rapidly between the sampling and the analytical measurements, important conclusions can be extracted. Obviously, Gal decay shown in Fig. 6 was much faster than in experiments of Fig. 4 and 5, and complete conversions were obtained after 15 min of reaction with the three AOPs, with an almost identical shape of the three decay profiles. Gal depletion in the first 5 min was too fast, and nothing can be said about the kinetic behaviour in these short reaction times. The yellow colour of quinones was observed in the first times and it disappeared at around 15 min in the three conditions. Mineralization was almost total after 30 min for the HP and PF systems, while it reached only 45% in the Fenton case, with no further changes up to 180 min (not shown). This is in accordance with the fact that Fenton processes have a lower ability for mineralization in comparison with photo-Fenton ones.15

Fenton, photo-Fenton and HP experiments at higher H2 O2 concentrations

Discussion

In separate experiments, Fenton, photo-Fenton and PF-HP systems were compared using higher H2 O2 concentrations; results are presented in Fig. 6. H2 O2 (0.98 M) was added gradually, at fixed intervals, until a final volume corresponding to 24 mM, i.e. twice the stoichiometric amount for mineralization, was reached. Addition was made as follows. Three 0.033 mmol initial additions were made at 0, 5 and 10 min, to follow carefully the evolution of Gal concentration and mineralization degree in the first stages. Then, additions were made in accumulative concentrations to 980 | Photochem. Photobiol. Sci., 2009, 8, 975–984

A simplified sequence of reactions describes TiO2 -heterogeneous photocatalytic degradation of organic compounds such as gallic acid: TiO2 + hn → ecb - + hvb +

(3)

hvb + + HO- (H2 O) → HO∑ (+ H+ )

(4)

ecb - + O2 (+ H+ ) → O2 ∑ - (HO2 ∑ )

(5)


2HO2 ∑ → H2 O2 + O2

(6)

HO2 ∑ + H2 O2 → O2 + H2 O + HO∑

(7)

hvb + /HO∑ + Gal → oxidation products

(8) ∑

According to eqn (8), Gal can be oxidised either by HO or by hvb + , although the identity of the oxidant species in photocatalytic reactions is still controversial. The interaction of Gal with the TiO2 surface, supported by evidences of the formation of a stable surface complex (log K = 4.7041 ), is assisting largely the mechanism by holes. In addition, Gal reacts very fast with HO∑ in homogeneous systems, k ~ = 1 ¥ 1010 M-1 s-1 ,4,6 a rate that should be not much lower on or near the TiO2 surface. Thermodynamically, Gal can be easily attacked by HO∑ and holes: the reported standard one-electron reduction potential of the semiquinone/Gal couple is 0.98 V‡42 and those of HO∑ and of the value of the P-25 valence band edge are around 2.843 and 2.9 V,44 respectively. At pH < 3, Gal is mostly in the totally protonated form, with minor amounts of the carboxylate anion (pK a1 = 4.16, pK a2 = 8.55, pK a3 = 11.40 and pK a4 = 12.8035 ). Attack of HO∑ /hbv + to both Gal forms in oxygenated solution will generate species by either electron transfer or electrophilic HO∑ addition to the aromatic system, which could end in quinones or hydroxylated derivatives;34 a more prolonged reaction would lead to ring opening, formation of aliphatic carboxylic acids and mineralization. The lack of experimental evidence does not allow us to define the precise conversion routes and this should be the object of a further study. However, the yellow colour observed during the experiments, indicative of formation of quinones, is a preliminary evidence of a possible mechanism. At 1 mM, a 10% TOC decrease was observed together with 37% Gal decay in 120 min (Fig. 1 and 2). This TOC decrease does not correspond to a mere decarboxylation of Gal, as occurred under VUV photolysis,34 but to a more advanced degradation degree. Addition of H2 O2 and Fe(III) (alone or combined) enhances the TiO2 -photocatalytic oxidation of organic compounds, as we found in other systems.16–19,24,25 The chemical systems turn very complex in these conditions, dependent on the organic compound and its oxidation intermediates, with several processes for which a single overall kinetic scheme and a complete mechanism cannot be proposed. However, some general features can be given. Addition of H2 O2 to a TiO2 -photocatalytic system enhances the photodegradation, acting as an electron acceptor by inhibition of recombination and generation of additional HO∑ : H2 O2 + ecb - → HO∑ + HO-

(9)

H2 O2 + O2 ∑ - → HO∑ + HO- + O2

(10)

Reactions of H2 O2 with HO∑ or hvb + are also possible and they will produce HO2 ∑ : H2 O2 + hvb + /HO∑ → HO2 ∑ + H+ /H2 O

(11)

∑

Gal reacts very rapidly not only with HO but also with superoxide radicals.45 Eqn (9)–(10) and (11) justify then the ‡ All standard reduction potentials (E ◦ ) cited in this work are vs. NHE. Electrode potentials in aqueous solution are cited; on the TiO2 surface, they may be somewhat different.

enhancement of the HP process, even if H2 O2 is added at low amounts (cf. 0 : 1 : 0 and 0 : 1 : 0.33 in Fig. 4). Addition of Fe(III) enhances the photocatalytic reaction by inhibition of recombination, acting also as electron acceptor: Fe(III) + ecb - → Fe(II)

(12)

In the presence of oxygen, Fenton or photo-Fenton related processes contribute to Fe(III) reduction due to the formation of H2 O2 and HO2 ∑ (eqn (5) and (6)): Fe(III) + H2 O2 → HO2 ∑ (O2 ∑ - ) + Fe(II) + (2)H+

(13)

Fe(III) + O2 ∑ - (HO2 ∑ ) → Fe(II) + O2 (+ H+ )

(14)

Fe(II) + H2 O2 → Fe(III) + OH- + HO∑

(15)

Thus, a continuous Fe(III)/Fe(II) cycle takes place, with formation of reactive oxygen species (ROS: HO∑ , O2 ∑ - , etc.) or other oxidants like Fe(IV) or Fe(V) species.40,46–49 Depending on the ligand L, iron(III) complexes can behave differently. Some of them (L = EDTA, citric acid, oxalic acid, NTA, etc.) are very stable in the dark. In contrast, most catechols, Gal among them, are unstable in the presence of Fe(III) due to the occurrence of thermal (dark) LMCT reactions (16) (i.e. reaction (1) for Gal), which lead to Fe2+ and radicals (semiquinones):50 Fe(III)–L → Fe2+ + L∑

(16)

Under H2 O2 addition, Fenton reaction (15) will follow, with the promotion of an Fe(III)/Fe(II) cycle and continuous HO∑ production. In fact, due to the ability of Fe(III)–L to keep Fe(III) in solution, Gal and other similar L have been used in place of aqueous Fe(III) to catalyze the oxidation of organics by H2 O2 at neutral pH.46,51,52 In the case of stable Fe(III)–L species, reaction (16) can be photoinduced (UV and visible light), with rather high quantum yields (e.g. oxalate).§ This process was applied to the degradation of the ligand itself or to enhance that of other pollutants, sometimes even without H2 O2 addition.16–19,24,25,53,54 Gallic acid is a particular case. As said before, the Fe(III)–Gal complex undergoes the homogeneous thermal charge transfer reaction (1). Reported work using Fenton or photo-Fenton reactions as AOPs for Gal oxidation5–7,26,31 did not take into account or even mention the possible effect caused by this reaction, which, at high Fe : Gal molar ratios, induces an important Gal depletion in the dark, either in the presence or in the absence of TiO2 (Table 1). Thermal reaction (1) takes place through a twostep oxidation process, producing first a semiquinone and then an o-quinone, as shown in ref. 38. In strongly oxygenated conditions, reaction (1) will be arrested in seconds.35 Because of the closeness of the standard reduction potential for the o-quinone/Gal couple to that of the E ◦ Fe(III)/Fe(II) pair (0.79 vs. 0.77 V, respectively35 ), the overall redox reaction (1) probably does not go to completion. Moreover, the reduction potential of the Fe(III)/Fe(II) pair is more negative in the Fe(III)–Gal complex (0.20 V at pH 4.637 ). Therefore, reaction (1) will be less feasible at low Fe : Gal molar ratios. In addition, in the presence of O2 , Fe2+ (which does not form a complex with Gal at pH 337 ) is slowly reoxidised:35,42 § Although acuo- or hydroxy Fe(III) complexes are able to produce HO∑ , quantum yields are low at wavelengths above 350 nm.49



Fe2+ + 14 O2 + H+ → Fe3+ + 12 H2 O

(17)

This reaction is accelerated in excess of Gal because of the stabilization of Fe3+ and the lowering of the reduction potential by complexation. Therefore, the charge transfer is slower or stops in times as short as 60 s. Then, we have two scenarios: at Fe : Gal molar ratios lower than 0.6 : 1, Gal decay produced by the thermal LMCT is minor; at higher ratios, degradation is extensive but it stops very rapidly (Table 1). As no reactive radical species (e.g. ROS) are formed in process (1), Gal is not oxidised beyond the quinone. The effect of UV irradiation on the Fe(III)–Gal system is different from that on other Fe(III)–L like oxalate or citrate. The results of Fig. 3 demonstrate a poor contribution of the photolytic pathway at two different Fe : Gal molar ratios (absence of TiO2 ), in coincidence with results of Sun and Pignatello who reported that the Fe(III)–Gal chelate was inactive to degrade 2,4-D in the absence of H2 O2 under UV irradiation.51 By TiO2 addition, a strong Gal degradation was induced with low amounts of Fe(III) (cf. 0.33 : 1 with and without TiO2 in Fig. 3) but not at a higher Fe : Gal ratio (1 : 1), as we will see next. As shown in Fig. 2, addition of Fe(III) up to the optimal 0.33 : 1 Fe : Gal ratio improves Gal degradation in HP systems, while higher iron concentrations (0.6 : 1–2 : 1) are detrimental. In our experiments at low Fe : Gal ratios (0.05 : 1 to 0.33 : 1), Gal will be oxidised by HO∑ ; however, when Gal concentration decreases to a much lower amount in respect to Fe or when an initial high Fe : Gal ratio is used, Gal will behave as a HO∑ generator,5 promoting the degradation of their own intermediates, e.g. quinones. Fe3+ reduction by intermediate semiquinones is also very rapid, with rate constants around 102 –103 M-1 s-1 , higher than the rate of Fe3+ with H2 O2 (0.01–0.02 M-1 s-1 , reaction (13)); this will contribute even more to sustain the rate and to iron recycling. The low Gal degradation with high iron concentrations (or when Gal concentration decreases) could be also explained by a preferential reoxidation of Fe(II) by holes or HO∑ over the organic compound oxidation. In fact, a deviation of the first order is observed in almost all curves of Fig. 2 after 90 min. A filter effect due to large concentration of non-photochemically active Fe(III) species should be also considered. The decrease of the rate (on the basis of a first order kinetics) in the systems containing high Fe : Gal ratios could be also due to the decrease of the initial Gal concentration at values far from the original 1 mM concentration at the moment the lamp is started, compared with the cases where the thermal reaction was negligible. Fig. 4 shows that the combined action of Fe(III) and H2 O2 externally added to the medium produces a dramatic effect in the TiO2 -photocatalytic Gal degradation (cf. 0 : 1 : 0 HP with 0.33 : 1 : 0.2 HP and 0.33 : 1 : 1 HP). All the processes (in the dark as well as under irradiation), induce continuous formation of ROS or other oxidants, which then greatly contribute to the oxidation of organics. Comparing now homogeneous with heterogeneous reactions (Fig. 3 and 5), it can be concluded that, without or at low H2 O2 concentrations, HP is a better technique than PF (cf. 0.33 : 1 : 0 and 0.33 : 1 : 0.2 with and without TiO2 ), in contrast with other systems previously studied by us where the photocatalyst acted more as a screen, being irrelevant or even detrimental to the degradation.16,24,25 For Gal, the strong interaction with the TiO2 982 | Photochem. Photobiol. Sci., 2009, 8, 975–984

surface, with the formation of the surface complex, together with the inactivity of the Fe(III)–Gal complex under UV light, are probably the main reasons for the enhancement. When higher H2 O2 concentrations are used (Fig. 6) reaction is much more improved in the three AOP systems (F, PF and HP), leading to a faster Gal depletion together with an important mineralization degree (even in dark Fenton conditions). In these conditions, after some irradiation time, Fe(III)–L complexes that are more photochemically active (e.g. oxalate) are probably formed in the system by opening of the Gal aromatic ring, thus enhancing oxidation, as observed by Boye et al. in photoelectro-Fenton systems.31 The results are in agreement with those reported by other authors;7,32 however, as said, for coupling with a biological system, high amounts of the oxidant should be avoided.

Conclusions Degradation of gallic acid by heterogeneous photocatalysis presents some inherent important differences compared with other compounds (oligocarboxylic acids) studied before by our group. HP in the absence of additives other than oxygen is effective, although rather slow for concentrations higher than 0.5 mM. Due to its polyphenolic structure, a strong complexation of Gal with Fe(III) takes place; the Fe(III) gallate undergoes an important thermal degradation depending on the Fe : Gal ratio initially present. However, Gal depletion by this way does not go beyond the quinones as no reactive oxidising species are formed. On the other hand, the Fe(III)–Gal complex does not photolyse under UV-A light, making negligible a contribution of this type to the global degradation. Addition of TiO2 improves rather efficiently the degradation in comparison with homogeneous photocatalytic systems under similar conditions, due to the strong interaction between Gal and the TiO2 surface. For combination with a biological treatment, no biorecalcitrant or biotoxic compounds should remain in the system after the AOP treatment, which implies (a) the use of low H2 O2 amounts and (b) a careful control of the composition of the system after the advanced process.

Acknowledgements This work was performed as part of Agencia Nacional de ´ Cient´ıfica y Tecnologica ´ Promocion (ANPCyT) PICT projects 2003-13-13261 and 00512. N.Q. thanks CNEA-CONICET for a doctoral fellowship. M.I.L. is a member of CONICET.

References ´ 1 D. Sawinska, K. Polewski and J. Sławinski, Synthesis and properties of model humic substances derived from gallic acid, Int. Agrophys., 2007, 21, 199–208. 2 Y.-X. Sheng, L. Li, Q. Wang, H.-Z. Guo and D.-A. Guo, Simultaneous determination of gallic acid, albiflorin, paeoniflorin, ferulic acid and benzoic acid in Si–Wu decoction by high-performance liquid chromatography DAD method, J. Pharm. Biomed. Anal., 2005, 37, 805–810. 3 A. E. Fazary, M. Taha and Y.-H. Ju, Iron complexation studies of gallic acid, J. Chem. Eng. Data, 2009, 54, 35–42. 4 P. Dwibedy, G. R. Dey, D. B. Naik, K. Kishore and P. N. Moorthy, Pulse radiolysis studies on redox reactions of gallic acid: one electron


5 6

7 8

9

10 11 12 13 14

15

16 17

18 19

20

21 22 23 24

25

26

oxidation of gallic acid by gallic acid-OH adduct, Phys. Chem. Chem. Phys., 1999, 1, 1915–1918. M. Strliˇc, T. Radoviˇc, J. Kolar and B. Pihlar, Anti- and prooxidative properties of gallic acid in Fenton-type systems, J. Agric. Food Chem., 2002, 50, 6313–6317. F. J. Ben´ıtez, F. J. Real, J. L. Acero, A. I. Leal and C. Garcia, Gallic acid degradation in aqueous solutions by UV/H2 O2 treatment, Fenton’s reagent and the photo-Fenton system, J. Hazard. Mater., 2005, B126, 31–39. M. S. Lucas, A. A. Dias, R. M. Bezerra and J. A. Peres, Gallic acid photochemical oxidation as a model compound of winery wastewaters, J. Environ. Sci. Health, Part A, 2008, 43(11), 1288–1295. E. Danaˇciková-Popelová, J. John, K. Rosiková, F. Sebesta and E. W. Hooper, Proceedings of the 4th International Symposium and Exhibition on Environmental Contamination in Central and Eastern Europe, Warsaw, 1998. S. J. Keny, A. G. Kumbhar, G. Venkateswaran and K. Kishore, Radiation effects on the dissolution kinetics of magnetite and hematite in EDTA- and NTA-based dilute chemical decontamination formulations, Radiat. Phys. Chem., 2005, 72, 475–482. F. J. Beltrán, J. M. Encinar and J. F. Garc´ıa-Araya, Oxidation by ozone and chlorine of two distillery wastewater contaminants: gallic acid and epicatechin, Water Res., 1993, 27, 1023–1032. M. R. Hoffman, S. T. Martin, W. Choi and D. W. Bahnemann, Environmental applications of semiconductor photocatalysis, Chem. Rev., 1995, 95, 69–96. O. Legrini, E. Oliveros and A. M. Braun, Photochemical processes for water treatment, Chem. Rev., 1993, 93, 671–698. M. I. Litter, Heterogeneous photocatalysis: Transition metal ions in photocatalytic systems, Appl. Catal., B, 1999, 23, 89–114. M. I. Litter, in The Handbook of Environmental Chemistry, Part M, Environmental Photochemistry Part II, ed. P. Boule, D. W. Bahnemann and P. K. J. Robertson, Springer-Verlag, Berlin Heidelberg, 2005, vol. 2, pp. 325–366. J. J. Pignatello, E. Oliveros and A. MacKay, Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry, Crit. Rev. Environ. Sci. Technol., 2006, 36(1), 1–84. C. A. Emilio, Doctoral Thesis, University of Buenos Aires, 2006. C. A. Emilio, W. F. Jardim, M. I. Litter and H. D. Mansilla, EDTA destruction using the solar ferrioxalate advanced oxidation technology (AOT): Comparison with solar photo-Fenton treatment, J. Photochem. Photobiol., A, 2002, 151, 121–127. P. A. Babay, C. A. Emilio, R. E. Ferreyra, E. A. Gautier, R. T. Gettar and M. I. Litter, Kinetics and mechanisms of EDTA photocatalytic degradation with TiO2 , Water Sci. Technol., 2001, 44, 179–185. P. A. Babay, C. A. Emilio, R. E. Ferreyra, E. A. Gautier, R. T. Gettar and M. I. Litter, Kinetics and mechanisms of EDTA photocatalytic degradation with TiO2 under different experimental conditions, Int. J. Photoenergy, 2001, 3, 193–199. C. A. Emilio, M. I. Litter and J. F. Magallanes, Semiempirical modeling with application of Artificial Neural Networks for the photocatalytic reaction of ethylenediaminetetraacetic acid (EDTA) over titanium oxide (TiO2 ), Helv. Chim. Acta, 2002, 85, 799–813. D. Hufschmidt, D. Bahnemann, J. J. Testa, C. A. Emilio and M. I. Litter, Enhancement of the photocatalytic activity of various TiO2 materials by platinisation, J. Photochem. Photobiol., A, 2002, 148, 223–231. C. Baeza, A. Rossner, W. F. Jardim, M. I. Litter and H. D. Mansilla, Removal of EDTA by UV-C/Hydrogen Peroxide, Environ. Technol., 2003, 24, 1277–1281. G. Ghiselli, W. F. Jardim, M. I. Litter and H. D. Mansilla, Destruction of EDTA using Fenton and photo-Fenton-like reactions under UV-A irradiation, J. Photochem. Photobiol., A, 2004, 167, 59–67. N. Quici, M. E. Morgada, G. Piperata, P. A. Babay, R. T. Gettar and M. I. Litter, Oxalic acid destruction at high concentrations by combined heterogeneous photocatalysis and photo-Fenton processes, Catal. Today, 2005, 101, 253–260. N. Quici, M. E. Morgada, R. T. Gettar, M. Bolte and M. I. Litter, Photocatalytic degradation of citric acid under different conditions: TiO2 heterogeneous photocatalysis against homogeneous photolytic processes promoted by Fe(III) and H2 O2 , Appl. Catal., B, 2007, 71, 117–124. W. Gernjak, T. Krutzler, A. Glaser, S. Malato, J. Cáceres, R. Bauer and A. R. Fernández-Alba, Photo-Fenton treatment of water containing natural phenolic pollutants, Chemosphere, 2003, 50, 71–78.

27 M. I. Pariente, F. Mart´ınez, J. A. Melero, J. A. Botas, T. Velegraki, N. P. Xekoukoulotakis and D. Mantzavinos, Heterogeneous photo-Fenton oxidation of benzoic acid in water: Effect of operating conditions, reaction by-products and coupling with biological treatment, Appl. Catal., B, 2008, 85, 24–32. 28 M. A. Miranda, F. Galindo, A. M. Amat and A. Arques, Pyrylium salt-photosensitised degradation of phenolic contaminants present in olive oil wastewaters with solar light Part II. Benzoic acid derivatives, Appl. Catal., B, 2001, 30, 437–444. 29 F. J. Beltrán, J. F. Garc´ıa-Araya and I. Giráldez, Gallic acid water ozonation using activated carbon, Appl. Catal., B, 2006, 63, 249–259. 30 M. Carbajo, F. J. Beltrán, F. Medina, O. Gimeno and F. J. Rivas, Appl. Catal., B, 2006, 67, 177–186. 31 B. Boye, E. Brillas, A. Busoc, G. Farnia, C. Flox, M. Giomoc and G. Sandon, Electrochemical removal of gallic acid from aqueous solutions, Electrochim. Acta, 2006, 52, 256–262. ´ 32 O. Gimeno, M. Carbajo, M. J. Lopez, J. A. Melero, F. Beltrán and F. J. Rivasa, Photocatalytic promoted oxidation of phenolic mixtures: An insight into the operating and mechanistic aspects, Water Res., 2007, 41, 4672–4684. 33 D. Gumy, A. G. Rincon, R. Hajdu and C. Pulgarin, Solar photocatalysis for detoxification and disinfection of water: Different types of suspended and fixed TiO2 catalysts study, Sol. Energy, 2006, 80, 1376–1381. 34 N. Quici, M. I. Litter, A. M. Braun and E. Oliveros, Vacuum-UVphotolysis of aqueous solutions of citric and gallic acids, J. Photochem. Photobiol., A, 2008, 197, 306–312. 35 H. K. J. Powell and M. C. Taylor, Interactions of Iron(II) and Iron(III) with gallic acid and its homologues: a potentiometric and spectrophotometric study, Aust. J. Chem., 1982, 35, 739–756. 36 S. Arif Kazmi, M. Saqib Qureshi and Z. Maqsood, Reactivity of an iron(III) complex of gallic acid, Inorg. Chim. Acta, 1987, 137, 151– 154. 37 Z. T. Maqsood, Ph.D. Thesis, University of Karachi, 1991. ´ Coinceanainn, The kinetics and mechanisms 38 M. J. Hynes and M. O. of the reaction of iron(III) with gallic acid, gallic acid methyl ester and catechin, J. Inorg. Biochem., 2001, 85, 131–142. 39 C. G. Hatchard and C. A. Parker, A new sensitive chemical actinometer. II. Potassium ferrioxalate as a standard chemical actinometer, Proc. R. Soc. London, Ser. A, 1956, 235, 518–536. 40 J. J. Pignatello, Dark and photoassisted Fe3+ -catalyzed degradation of chlorophenoxy herbicides by hydrogen peroxide, Environ. Sci. Technol., 1992, 26, 944–951. 41 P. Z. Araujo, P. J. Morando and M. A. Blesa, Interaction of catechol and gallic acid with titanium dioxide in aqueous suspensions. 1. Equilibrium studies, Langmuir, 2005, 21, 3470–3474. 42 J. C. Danilewicz, Review of reaction mechanisms of oxygen and proposed intermediate reduction products in wine: central role of iron and copper, Am. J. Enol. Vitic., 2003, 54(2), 73–85. 43 P. Wardman, Reduction potentials of one-electron couples involving free radicals in aqueous solution, J. Phys. Chem. Ref. Data, 1989, 18, 1637–1755. 44 S. T. Martin, H. Herrmann and M. R. Hoffmann, Time-resolved microwave conductivity. Part 2. Quantum-sized TiO2 , and the effect of adsorbates and light intensity on charge-carrier dynamics, J. Chem. Soc., Faraday Trans., 1994, 90(21), 3323–3330. 45 S. V. Jovanovic, Y. Hara, S. Steenken and M. G. Simic, Antioxidant potential of gallocatechins. A pulse radiolysis and laser photolysis study, J. Am. Chem. Soc., 1995, 117(39), 9881–9888. 46 Y. Sun and J. J. Pignatello, Chemical treatment of pesticide wastes. Evaluation of iron(III) chelates for catalytic hydrogen peroxide oxidation of 2,4-D at circumneutral pH, J. Agric. Food Chem., 1992, 40(2), 322– 327. 47 W. Feng and D. Nansheng, Photochemistry of hydrolytic iron (III) species and photoinduced degradation of organic compounds. A minireview, Chemosphere, 2000, 41, 1137–1147. 48 B. C. Faust, in Aquatic and Surface Photochemistry, ed. G. R. Helz, R. G. Zepp and D. G. Crosby, Lewis Publish., Boca Raton, 1994, ch. 1, pp. 3–37. 49 R. Chen and J. J. Pignatello, Role of quinone intermediates as electron shuttles in Fenton and photoassisted Fenton oxidations of aromatic compounds, Environ. Sci. Technol., 1997, 31, 2399–2406. 50 H. Iwahashi, H. Morishita, T. Ishii, R. Sugata and R. Kido, Enhancement by catechols of hydroxyl-radical formation in the presence of ferric ions and hydrogen peroxide, J. Biochem., 1989, 105, 429–434.



51 Y. Sun and J. J. Pignatello, Activation of hydrogen peroxide by iron(III) chelates for abiotic degradation of herbicides and insecticides in water, J. Agric. Food Chem., 1993, 41(2), 308–312. 52 K. Nam, W. Rodr´ıguez and J. J. Kukor, Enhanced degradation of polycyclic aromatic hydrocarbons by biodegradation combined with a modified Fenton reaction, Chemosphere, 2001, 45, 11– 20.

984 | Photochem. Photobiol. Sci., 2009, 8, 975–984

53 O. Abida, C. Emilio, N. Quici, R. Gettar, M. Litter, G. Mailhot and M. Bolte, Degradation of 4-chlorophenol mediated by Fe(III)-NTA in homogeneous and heterogeneous systems, Water Sci. Technol., 2004, 49(4), 123–128. 54 O. Abida, G. Mailhot, M. Litter and M. Bolte, Impact of iron-complex (Fe(III)-NTA) on photoinduced degradation of 4-chlorophenol in aqueous solution, Photochem. Photobiol. Sci., 2006, 5, 395–402.
