oxidative treatment with superior

Applied Catalysis B: Environmental 59 (2005) 249–257 www.elsevier.com/locate/apcatb

Reductive/oxidative treatment with superior performance relative to oxidative treatment during the degradation of 4-chlorophenol P. Raja a, A. Bozzi a, W.F. Jardim b, G. Mascolo c, R. Renganathan d, J. Kiwi a,* a

Laboratory of Photonics and Interfaces, Institute of Chemical Sciences and Engineering, Swiss Federal Institute of Technology, Lausanne 1015, Switzerland b Institute of Chemistry, UNICAMP, State University of Campinas, Campinas, Sau Paulo, Brazil c Water Research Institute, CNR-IRSA, Via F de Blasio 5, Bari 70123, Italy d Department of Chemistry, Bharathidasan University, Tiruchirappalli, Tamilnadu, India Received 20 September 2004; received in revised form 17 January 2005; accepted 12 February 2005 Available online 14 March 2005

Abstract This study obtained information on the effectiveness of the photo-assisted Fenton oxidation of 4-chlorophenol (4-CP) combined with zero-valent pretreatment. The kinetic rate parameters of the process as well as the operating conditions were determined. Homogeneous photo-assisted Fenton enhanced processes lead to 33% mineralization of 4-chlorophenol (1.25 mM) in solutions containing Fe-ions (2– 10 mg/l) and H2O2 (10 mM) within 2 h under visible light irradiation. When this solution was pretreated with zero-valent Fe (14 g/70 ml) in Ar atmosphere, the mineralization attained levels of >80% after the second stage photo-assisted Fenton process. Intermediates that could be effectively degraded by photo-Fenton reactions were not attained in the absence of zero-valent Fe pretreatment. The pretreatment by zero-valent Fe under light lead to about 70% of the stoichiometric amount of chloride contained in 4-CP. Partial recovery of chloride ions indicated the formation of chloro-intermediates. These intermediates were experimentally detected by high-pressure liquid chromatograph (HPLC)–MS and the most important intermediates were identified. Fenton photo-assisted processes were effective employing very low concentrations of Fe2+(2–5 mg/l) after the pretreatment stage that do not need to be separated after the 4-CP degradation process. This is important for the practical application of the novel combined heterogeneous–homogeneous process. Evidence for the stable catalytic performance of the coupled process to degrade 4-CP is presented. The effect on the 4-CP degradation of Fe-ion, H2O2, 4-CP concentration and gas atmosphere was systematically investigated. The activation energy (Ea) of 2.66 kJ/mol was found for the abatement of 4-CP. # 2005 Elsevier B.V. All rights reserved. Keywords: Heterogeneous/homogeneous photo-Fenton processes; Zero-valent Fe; 4-Chlorophenol decomposition; TOC reduction; Activation energy; GC– MS of intermediates; Reductive processes

1. Introduction To find better ways to mineralize pollutants of the halocarbon family is a topic of timely interest due to the increased availability of these compounds in water bodies with known toxic and carcinogenic effects [1]. Halocarbons represent the most abundant family of industrial toxic compounds as it has been recently recognized from different * Corresponding author. Tel.: +41 216933621; fax: +41 216934111. E-mail address: [email protected] (J. Kiwi). 0926-3373/$ – see front matter # 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.apcatb.2005.02.014

polluting sources like herbicides, pesticides, chemical and solvent manufacturing, the paint industry [2] and the dechlorination of drinking water [3]. The application of innovative advanced oxidation technologies (AOTs) like the one presented in this study is of interest to abate efficiently and also reduce the treatment cost through destructive techniques mineralizing the halocarbon. The compound 4chlorophenol (4-CP) is a priority toxic pollutants listed in the US EPA Clean Water act [4] and is difficult to remove from water bodies, since its half-life can reach from months to years.

250

P. Raja et al. / Applied Catalysis B: Environmental 59 (2005) 249–257

The dark decomposition of organic compounds on zerovalent Fe [5] has been reported during the last decade. The photocatalytic decomposition of organic pollutants [6,7] and halocarbons [8] is the subject of ongoing research. Alternative methods have been explored for the oxidation of 4-CP involving electrochemical oxidation [9], using H2O2 and O3 [10] and more recently, homogeneous and heterogeneous photo-assisted Fenton mediated processes [11,12] in general and applied for the degradation of 4-CP in particular [13,14]. Even if (AOTs) are effective in wastewater treatment and detoxification, the actual cost is high [15]. More recently, the combination of chemical pretreatments with inexpensive biological processes seems to improve the situation from the economic point of view [16]. The present work intends to explore a novel approach using zero-valent Fe as a low cost material that is readily available that may decrease the treatment cost of 4-CP degradation when using the Fenton reagent. We will present the first evidence for a treatment combining zero-valent Fe with Fenton photo-assisted processes that use Fe-ions in concentrations much lower than the ones required in the Fenton reagent Fe2+/H2O2 to oxidize organic compounds. These Fe-ion concentrations are in the range allowed by the European Economic Community [17], and therefore, the Feions do not need to be separated at the end of the treatment. In practice, the amount of Fe-ions allowed depends on the specific country and vary from 0.5 mg/l to a higher limit of 5 mg/l. This separation process is costly is terms of chemicals, manpower and time. Simulated sunlight will be employed during this study since solar activated processes are important during the degradation of many compounds in the natural cycle and chlorocarbons [18]. This study will also give some insight into some of the reactions leading to rapid 4-CP dechlorination through an innovative process mediated by zero-valent Fe in combination with the Fenton reagent.

2.2. Irradiation procedures and analyses A Suntest solar simulator (Hanau, Germany) with an output of 90 mW/cm2 with a wavelength distribution following the solar spectra and having 7% of the photons between 290 and 400 nm was used as the irradiation source. The cavity of the Suntest simulator was air cooled at 42 8C. The integral radiant flux of the mercury lamp was measured with a power-meter (YSI Corp., Colorado, USA). In this cavity, the cylindrical Pyrex reactors (70 ml each) were placed with appropriate amounts of solutions. The absorption of the solutions was followed in a Hewlett-Packard 38620 N-diode array spectrophotometer. The total organic carbon (TOC) was measured with a Shimadzu 5000 TOC analyzer. The disappearance of 4-CP was monitored in a Varian Corp. high-pressure liquid chromatograph (HPLC) provided for with a 9065 diode. A Phenomenex C-18 inverse phase column was used in the HPLC and the gradient eluent solution consisted of a mobile phase of water (30%) and acetonitrile (70%). The peroxide concentration of the solutions was measured using Merckoquant1 paper from Merck AG, Switzerland. Analysis of Fe(II)-ions in the aqueous solution was carried out with Ferro-Zine1 (Aldrich Cat. No. 16,060-1) with a detection limit of 0.1 mg/l of iron. A 5 ml aliquot of the sample solution was taken and 100 ml of Ferro-Zine1 with 100 ml of ammonium acetate buffer (pH 5.5) were added into the solution. The Ferro-Zine1 complex at l = 562 nm with a magenta color presented a molar extinction coefficient e562 = 24,100 M1 cm1. Reagents for the determination of Cl-ions were obtained from Merck KGaA (OC330408) containing Fe(III)-nitrate and Hg(SCN)2 solutions. The absorbance of the Fe(SCN)2+ at l = 450 nm was determined since the Cl-ion forms the latter complex after displacing the SCN from Hg(SCN)2. The range of Cl-ion determination was valid between 1 and 200 mg/l. 2.3. Analytical determination of by-products

2. Experimental 2.1. Materials The 4-chlorophenol, acid, bases and H2O2 used were Fluka p.a. reagents and used as received. Bi-distilled water was used throughout this work. After an initial screening of the iron powders available from different suppliers: Alfa Aesar, Fluka, Sigma, Fisher Sci. Co., the most active iron sample towards 4-CP degradation was from Riedel-de Haehn, and was selected for this study. The iron powder was 99%, 100 mesh (0.149 mm) particle size, electrolytic grade (Cat. No. 12312) and used as received. This iron powder is a low cost material, readily available having uniform particle size and was used as received. Acid treatment (HCl) to eliminate the possible carbonate adsorbed at the iron surface showed to be unnecessary since the 4-PC mineralization efficiency was not affected by the pretreatment used.

By-products identification was performed by HPLC– UV–MS using a 1050-Ti chromatographic system (Agilent) equipped with a Luna phenyl-hexyl 150 mm 3 mm column Phenomenex and a 1050 series variable wavelength detector set at 220 nm. Samples, injected by a Gilson 234 auto-sampler (Gilson, Middleton, WI, USA) equipped with a 9010 Rheodyne valve and a 20 ml loop, were eluted by a water/methanol 70/30 mixture at 0.4 ml/min. The HPLC– UV was interfaced to an API 165 mass spectrometer (Applied Biosystems/MSD Sciex) equipped with a turbo-ion spray interface. The interface conditions were: nebulizer gas (air) = 1.2 l/min, curtain gas (nitrogen) = 1 l/min, turbo-ion spray gas (nitrogen at 300 8C) = 6 l/min, needle voltage = 4500 V, orifice voltage = 25 V and ring voltage = 200 V. Samples, injected by the auto-sampler, were analyzed by running a gradient, from 5/90/5 (methanol/ water/ammonium acetate 50 mM in methanol) to 85/0/5 in


15 min, which was maintained for 5 min. The flow from the HPLC–UV was split to allow 200 ml/min to enter the turboion spray interface.

3. Results and discussions 3.1. Heterogeneous photocatalytic degradation of 4-CP by zero-valent Fe under light in exclusion of oxygen Fig. 1a shows the TOC reduction under light in Ar purged for 4-CP solutions contacted with zero-valent Fe in solution. Fe0 ! Fe2þ þ 2e ; E0 ¼ 0:44V

(1)

Under the experimental conditions used in Fig. 1a, the concentration of Fe2+-ions in solution at time zero and up to 16 h pretreatment was observed to be 0.5 2.5 mg/l for dark and light processes, given evidence for reaction (1). The reduction of TOC in Fig. 1a under light during 2 h was observed to be 30%. The zero-valent solutions used in the pretreatment had a natural pH of 7 and were adjusted to 3.

251

The solution of 4-CP 1.25 mM was recirculated in the photoreactor containing the Fe powder (14 g/70 ml) for 16 h after purging the initial solution with Ar for 30 min. The Fe-ions were measured at different pretreatment times but the values found were around 2 mg after 16 h of pretreatment time. Using different amounts of zero-valent Fe than the 14 g/ 70 ml did not lead to a TOC reduction that differs more than 10% with respect to the results reported in Fig. 1a. The insert in Fig. 1a shows that Cl of 4-CP evolves only when H2O2 was present and that the chloride evolution attains a plateau after 30 min. This is in line with the result shown in Fig. 1b where only in the presence of H2O2, the complete disappearance of 4-CP occurs within 30 min as determined by HPLC, involving the reaction: Fe0 þ H2 O2 ! Fe2þ þ 2OH

(2)

This H2O2 oxidation of 4-CP in the presence of zero-valent Fe powder can be described as a Fenton reaction started by Fe-ions leached from the zero-valent iron surface. The amount of Cl liberated from the decomposition of 4-CP as presented in the insert in Fig. 1a reaches about 70% of the stoichiometric amount of the initial 4-CP. This is shown in Fig. 1b and occurs within the initial 30 min of reaction. This indicates that only in the presence of peroxides, intermediates are produced that are able to further decompose 4-CP. Fig. 1b also shows the disappearance of 4-CP in the dark takes longer times than under light. In the dark, no significant decrease of the TOC was observed. The e from reaction (1) reacts with 4-CP [18]: C6 H5 ClO þ e ! C6 H5 O þ Cl ; E0 ¼ 0:8 1:2V

(3)

and the chloride detaches itself from 4-CP in the form of Cl radical before reacting with the e in a reductive type process [6,7,10]: Cl þ e ! Cl

(4)

The overall reaction between zero-valent Fe and 4-CP can be stated as: 2C6 H5 ClO þ Fe0 ! Fe2þ þ 2Cl þ 2C6 H5 O

(5)

and the results of the two half-reactions (1) and (3) are possible since it is thermodynamically favored. The degradation of 4-CP (Fig. 1a and b) works only in oxygen free solutions since in the presence of O2, the reaction (6) proceeds the reaction: O2 þ e ! O 2

(6)

Table 1 Fe(II) amounts as a function of time during the pretreatment of 4-CP solutions Fig. 1. (a) TOC reduction and 4-CP disappearance during the pretreatment of a 4-CP (1.25 mM), solution in the presence of H2O2 (10 mM) in contact with 14 g of zero-valent Fe in 70 ml solution at pH 3. The source of light is a Suntest solar simulated light irradiator (90 mW/cm2). The inset shows the Cl evolution under light observed in the presence and absence of H2O2 (10 mM). (b) Disappearance of 4-CP as followed for a solution with the same make-up as used in (a).

Time (min)

Fe(II) (mg/l) Ar

N2

O2

0 30 60 90 120

0.32 2.09 1.40 1.22 0.09

0.23 0.23 0.61 0.64 0.09

0.36 0.36 0.37 0.69 0.06

252


in competition with reaction (4). The presence of air (O2) would also significantly modify the zero-valent Fe upper layers leading to oxy–hydroxy species that are chemically different than the ones existing in anaerobic conditions [18–20]. This is important since in reaction (1), the electron transfer from zero-valent Fe to 4-CP occurs via surface generated Fe2+. The exact catalytic nature of the surface site that leads to the reduction of the chloride in 4-CP to Cl has not been identified [18,19]. Table 1 shows the concentrations of Fe2+ formed in O2, N2 and Ar atmosphere as a function of irradiation time during zero-valent iron pretreatment. It is readily seen from the inspection of Table 1, that the favourable mineralization of 4-CP in Ar atmosphere is due to a higher intermediate concentrations of Fe(II) relative to the concentrations of Fe(II) detected in N2 and O2 atmosphere. The formation of Fe(II) is essential for the decomposition of H2O2 since it generates the OH-radical as discussed below in Eqs. (7)– (13).

3.2. Pretreatment followed by homogeneous photo-assisted Fenton degradation of 4-CP Fig. 2a shows the results of 4-CP solutions degradation pretreated with zero-valent Fe in three different gas atmospheres. Fig. 2a (Trace 1) shows the TOC decrease applying pretreatment under light in a solution purged with O2 solution. Fig. 2a (Trace 2) shows the TOC decrease a zero-valent Fe pretreatment in a solution purged 30 min with N2 gas. Fig. 2a (Trace 3) shows the TOC decrease due to zero-valent pretreatment of a solution purged with Ar for 30 min. The influence of the gas atmosphere during the pretreatment is readily noticed in Fig. 2a and has a significant effect on the concentration of Fe(II)-ions in solution as reported in Table 1. The formation of Fecomposites with N cannot be excluded when the solutions were purged with N2 gas. This reaction has been reported previously for some metals [21]. In Fig. 2a, the degradation of 4-CP due to the homogeneous reaction is due to:

Fig. 2. (a) Fenton photo-assisted degradation of 4-CP in a solution containing H2O2 (10 mM) and Fe2+-ions (2 mg/l). A Suntest solar simulator (90 mW/cm2) is the irradiation source. Trace 1, zero-valent Fe pretreatment, solution purged with O2. Trace 2, zero-valent Fe pretreatment, solution purged with N2. Trace 3, Trace 1, zero-valent Fe pretreatment, solution purged with Ar. (b) Consecutive 4-CP degradation runs as shown previously in Fig. 2a (Trace 3). (c) Homogeneous Fenton photo-assisted degradation of 4-CP as a function of the amount of Fe2+ added in solution under Suntest solar light irradiation (90 mW/ cm2) in the presence of H2O2 (10 mM) at pH 3.


Fe2þ þ H2 O2 ! Fe3þ þ OH þ OH ; k7 ¼ 76M1 s1 (7) Fe3þ þ H2 O2 ! Fe2þ þ Hþ þ HO2 ; k8 ¼ 0:02M1 s1 (8) The results obtained in Fig. 2a do not show the results obtained for the dark degradation of 4-CP. The TOC reduction was much smaller than the reduction observed under light irradiation. Photo-assisted Fenton reactions have been shown recently to be enhanced due to the decomposition of the photo-active Fe(OH)2+ leading to the additional generation of OH-radicals in solution through the reaction: Fe(OH)2+ + hn ! Fe2+ + OH [22]. Concentrations of Fe2+-ions (2– 10 mg/l) added during the pretreatment stage did not induce any variation in the degradation kinetics of 4-CP during the second stage homogeneous photo-Fenton reaction (Fig. 2c). The stable nature of the coupled heterogeneous-homogeneous 4-CP degradation is shown in Fig. 2b. Practically no variation was observed in the 4-CP TOC reduction using the same zero-valent Fe during three consecutive runs followed by the Fenton reagent Fe2+/Fe3+/H2O2 reagent under light. This shows stable catalytic nature of 4-CP degradation used in this study. Fig. 2c shows that the 4-CP Fenton photo-assisted homogeneous degradation kinetics and efficiency does not depend on the amount of Fe2+-ions in solution. The results for the three different concentrations of Fe2+-ions (Fig. 2c) shows that 37% of the total organic carbon of 4-CP reduction occurred within 120 min. This can be explained by

Fig. 3. TOC variation for solutions with a different initial pH as a function of irradiation time. Solutions were pretreated with zero-valent Fe. The second stage photo-assisted Fenton points are shown. Other solution parameters: H2O2 (10 mM) and Fe2+-ions (2 mg/l). The source of light is a Suntest solar simulated light irradiator (90 mW/cm2).

253

the formation of recalcitrant intermediates that preclude further degradation [23,24]. This shows a slower degradation compared to the results employing pretreatment, as reported previously in Fig. 2a. The pH was seen to vary during the Fenton photo-assisted homogeneous degradation of 4-CP under the experimental conditions used in Fig. 2c (2 mg Fe2+/l under Ar atmosphere) and the results are shown in Fig. 3. A 15% decrease in the amount of 4-CP abated was observed when the pH was increased from pH 3 to 5. As it has been documented in the literature [1,2,10,19], iron–oxy–hydroxydes progressively cover the iron beginning at pH 3. This coverage is completed

Fig. 4. (a) TOC reduction as a function of the H2O2 added in a solution of 4CP (1.11 mM) under solar Suntest simulated solar radiation (90 mW/cm2) in the presence of Fe2+ (5 mg/l) at pH 3. (b) Abatement of 4-CP as a function of the H2O2 added in solution. Trace (1) no H2O2 (2) 5 mM H2O2 (3) 10 mM H2O24) 40 mM H2O2. Other experimental conditions as above in Fig. 4a.

254


pH 5 depending on other solution parameters. This is why solutions with pH 6 and beyond do not work in our case. Fig. 3 shows that iron–oxy–hydroxydes are not catalytically active in the pretreatment described in Section 3.1. Fig. 4a presents the decrease in TOC of 4-CP solutions under visible light in the presence of H2O2 and Fe2+-ions at pH 3 that have not been pretreated with zero-valent Fe. It is readily seen in Fig. 4a that a concentration of H2O2 (5 mM) is not enough to induce meaningful TOC decrease and that a concentration of 10 mM produces the most suitable mineralization of 4-CP. Above this concentration and up to H2O2 (40 mM) Fig. 4a shows that the photodegradation slows down since the propagation step is hindered by the excess of H2O2 added scavenging the OH radicals in solution:

assisted Fenton treatment is applied in homogeneous solution in the absence of zero-valent pretreatment. The shapes of the degradation curves point to a complex process taking place that involves long-lived intermediates as will be examined below in Section 3.5. But the general trend indicates that longer times were necessary for the complete destruction of higher concentrations of 4-CP. The reaction did not depend of the stirring rate. This confirms that we worked in the kinetic reaction regime avoiding external diffusion limitations. Fig. 5b presents the 4-CP disappearance using photoassisted Fenton treatment after heterogeneous zero-valent pretreatment under visible light irradiation was carried out

H2 O2 þ OH ! H2 O þ HO2 ; k9 ¼ 3:3 107 M1 s1 (9) Fig. 4b shows the disappearance of 4-CP in a solution as a function of the H2O2 added in solution. From Fig. 4b, it is readily seen that a concentration of H2O2 (10 mM) was the most suitable one to remove 4-CP. The reason for this has been mentioned above in the section related to Fig. 4a. As mentioned above in reaction (7), the Fe2+-ion, catalyses the decomposition of H2O2 resulting in the generation of OHradicals. But these radicals involve a complex reaction sequence that starts with the chain initiation reaction (7). The Fe3+-ions in reaction (7) catalyze H2O2 decomposition a shown above by Eq. (8): Fe3þ þ H2 O2 ! FeOOH2þ ; k10 ¼ 0:001 0:01M1 s1 (10) The reaction of H2O2 with Fe3+-ions involve also the production of O2: Fe3þ þ HO2 ! Fe2þ þ O2 þ Hþ atpH3; k11 ¼ 1:2 106 M1 s1

(11)

Hydroxyl radicals can oxidize organic compounds (RH) by hydrogen abstraction of the organic compound producing the organic radical (R). RH þ OH ! H2 O þ R ; k12 ¼ 107 M1 s1

(12)

The chain termination is generally ascribed to the reaction: OH þ Fe2þ ! Fe3þ þ OH ; k13 ¼ 3:2 108 M1 s1 (13) Since k9 = 3.3 10 M s , while k13 = 3.2 10 M1 s1, the concentration of H2O2 is unimportant only when the ratio [RH][/H2O2] is high, which is not the case for the concentrations of the two reactants in Fig. 4a and b. Therefore, the experimental results confirmed what can be estimated from the reaction rates in Eqs. (7)–(13). 7

1 1

8

3.3. Kinetics of 4-CP disappearance as a function of concentration Fig. 5a shows the degradation kinetics of 4-CP as a function of the initial concentration of 4-CP when photo-

Fig. 5. (a) Disappearance of 4-CP as a function of reactant concentration by assisted photo-Fenton treatment of solutions containing H2O2 (10 mM) and Fe2+ (5 mg/l) at pH 3. Light intensity (90 mW/cm2). (b) Disappearance of 4CP as a function of reactant concentration by assisted photo-Fenton treatment of solutions containing H2O2 (10 mM) and Fe2+ (5 mg/l) at pH 3. The solutions have been previously pretreated with zero-valent Fe as described in Fig. 1a.


for 2 h as described in Fig. 1a. It is readily seen that the 4-CP disappearance kinetics is much faster for the same concentrations of this compound reported previously in Fig. 5a, where no pretreatment was applied. 3.4. Effect of temperature on 4-CP disappearance The disappearance of 4-CP (1.11 mM) by Fenton reactions (H2O210 mM and Fe2+ 2 mg/l) under visible light for the case of zero-valent Fe pretreated solutions was carried out as a function of the reaction temperature (controlled continuously with a thermostat). The increase in temperature had a beneficial effect for the kinetic rate. The Ea was determined by following the concentration of 4-CP by HPLC at 20, 40, 55 and 70 8C. Only the initial tangent slope was used in the estimation of Ea. In this way, the energy input due to the light is kept to a minimum and the 4CP concentration decrease is ascribed to the temperature increase. The value of Ea obtained from ln k versus 1/T (K) is 2.66 kJ/mol. This value suggests that ion–molecule and radical–molecule reactions occur requiring activation energy, besides the radical–radical reactions that need no Ea. The degradation rate increased two-fold when the temperature was increased from 20 to 70 8C. This is a

255

valuable result if solar light should be used to activate this process. Solar irradiation induces an increase in the temperature of the reactor vessel. 3.5. Nature of the chloro-carbon intermediates during the degradation of 4-CP in the absence and presence of zero-valent pretreatment By-products formation during the degradation of 4-CP was also investigated by performing HPLC–MS analysis (also called LC–MS). Table 2 reports the chemical structures of the by-products of 4-CP degradation without and with zero-valent pretreatment, respectively. Experimental data for both treatments are consistent with two main pathways. The first one involves the attack of hydroxyl radicals to benzene rings by substitution reactions (by-products 1 and 2). The second degradation route is consistent with the formation of p-chloro-phenoxy radical that, in turn, reacts with 4-CP giving rise to the formation of by-product 3. Furthermore, the presence of by-product 4, originating from both, the OH substitution and p-chloro-phenoxy radical addition, suggests that the two mechanisms are independently active. By-products with m/z 189, 257 and 221 were also observed but are not presented in Table 2 because there

Table 2 Chemical structures of the by-products detected during 4-CP degradation By-product

Chemical structure

Molecular weight

Without zero-valent pretreatment

With zero-valent pretreatment

1

144

X

X

2

160

X

3

254

X

X

4

270

X

X

256


yield low molecular weight compounds, as reported widely for the oxidation of organic compounds [25].

4. Conclusions A novel combined chemical treatment using zero-valent Fe followed by Fenton reagent was used to decompose and mineralize 4-CP solutions. This combined treatment is shown to be suitable for 4-CP degradation and could be extended to other hydrophobic pollutants. Very low levels of Fe-ion were seen to be sufficient in the second stage homogeneous photo-Fenton treatment to reach suitable degradation kinetics for 4-CP. The Fe-ion concentration used can be retained in the process without causing additional water pollution (or needing costly separation). The iron remediation technique proposed here involves an electron transfer requiring direct contact between zerovalent Fe and 4-CP. Iron was dissolved during the heterogeneous catalytic pretreatment in weak acidic solutions. Very little has been published on iron surface oxidation in the presence of H2O2 but in the exclusion of air. This makes it difficult to model the individual reactions taking place during the pretreatment stage. Fig. 6. (a) Time evolution of by-products during 4-CP degradation without zero-valent pretreatment. By-products structures are reported in Table 1. (b) Time evolution of by-products during 4-CP degradation with zero-valent pretreatment. By-products structures are reported in Table 1.

was no plausible structure consistent with the measured molecular weight. Fig. 6a and b show the temporal formation and decay of the by-products that could be identified by HPLC–MS. The trends reported in Fig. 6a and b show that the production of the intermediate was maximized at reaction times between 8 and 40 min. The fact that intermediates containing Cl-atom are detected in parallel to Cl-ions (see inset in Fig. 1a) can be explained by the following two observations: (1) that the Cl-ions concentration maximizes at reaction times higher than 30 min (see inset of Fig. 1a) while all the by-products, except by-product 2, reached the maximum amount at reaction time between 0 and 20 min. It follows that the byproducts detection is consistent with the Cl-ions evolution reported in the inset of Figs. 1 and (2) on the basis of the 4CP used (1.25 mM) a theoretical maximum concentration of Cl-ions of 44 ppm could be expected. Instead, Cl-ions reach a plateau of about 30 ppm after 30 min reaction (see inset in Fig. 1a). Therefore, even at higher reaction times, there are about 15 ppm of organic chlorine still present into the solution. This is consistent with the detection of by-product 2 at higher reaction times (Fig. 6a). Beyond these times, the range of maximum intermediate production, the intermediates slowly disappear during the later reaction stages. At longer reaction times, no by-products could be detected by HPLC–MS. This behavior suggests the opening of the aromatic ring due to consecutive oxidation reactions lead to

Acknowledgments We gratefully acknowledge the financial support of CTI TOP-NANO 21 under Grant No. 5960.1 (Bern, Switzerland) and of the COST D-19 Program under Grant No. CO 2.0068.

References [1] H. Halmann, Photodegradation of Organic Pollutants, CRC Publishing Co., Boca Raton, FL, 1996. [2] R. Bauer, G. Waldner, S. Hager, S. Malato, P. Maletzky, Catal. Today 53 (1999) 131. [3] U.G. Ahlborg, T.M. Thandberg, CRC Crit. Rev. Toxicol. 7 (1980) 1. [4] EPA, July 2002. http://www.scorecard.org. [5] L. Matheson, G. Tratnyek, Environ. Sci. Technol. 28 (1994) 2045. [6] C. Pulgarin, G. Pajonk, J. Bandara, J. Kiwi, J. Adv. Oxid. Technol. 1 (1996) 94. [7] S. Oh, P. Chiu, B. Kim, D. Cha, Water Res. 37 (2003) 4275. [8] M. Hugull, I. Boz, R. Apak, J. Hazard. Mater. B 64 (1999) 313. [9] O. Comninellis, Stud. Environ. Sci. 59 (1994) 77. [10] Th. Oppenlaender, Photochemical Purification of Water and Air, Wiley-VCH, Weinheim, Germany, 2003. [11] M.P. Ormad, J.L. Ovelleiro, J. Kiwi, Appl. Catal. B 32 (2001) 157. [12] S. Parra, L. Henao, E. Mielczarski, J. Mielczarski, P. Albers, E. Suvorova, J. Guindet, J. Kiwi, Langmuir 20 (2004) 5621. [13] B. Kwon, D. Lee, N. Kang, J. Yoon, Water Res. 33 (1999) 2110. [14] A. Detomaso, A. Lopez, G. Lovecchio, G. Mascolo, R. Curci, Environ. Sci. Pollut. Res. 10 (2003) 379. [15] D.F. Ollis, Emerging Technologies in Hazardous Waste Management III, American Chemical Society, 1993 p. 67. [16] J. Bandara, C. Pulgarin, P. Peringer, J. Kiwi, J. Photochem. Photobiol. A 111 (1997) 253.

P. Raja et al. / Applied Catalysis B: Environmental 59 (2005) 249–257 [17] EEC Council Directives 76/4647, European Economic Community, Brussels, Belgium, 1982. [18] M. Pera-Titus, V. Garcia-Molina, M. Banos, J. Gime´ nez, S. Esplugas, Appl. Catal. B 47 (2004) 219. [19] R. Cornell, U. Schwertmann, The Iron Oxides, Structure, Properties, Reactions, Occurrence and Uses, VCH, Weinheim, Germany, 1996. [20] M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, Cebelcor, Brussels, 1974.

257

[21] J.R. Anderson, The Structure of Metallic Catalysts, Academic Press, New York, 1975. [22] G. Ruppert, R. Bauer, G. Heisler, J. Photochem. Photobiol. A 73 (1993) 75. [23] D.L. Sedlak, A.W. Andren, Environ. Sci. Technol. 25 (1991) 777. [24] F. Lucking, H. Koser, M. Jank, A. Ritter, Water Res. 32 (1998) 2607. [25] G. Mascolo, A. Lopez, H. James, M. Fielding, Water Res. 35 (2001) 1695.