PII: S0043-1354(00)00460-7
Wat. Res. Vol. 35, No. 8, pp. 1994–2002, 2001 # 2001 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/01/$ - see front matter
DEGRADATION OF 2,4-DICHLOROPHENOL BY IMMOBILIZED IRON CATALYSTS S. SABHI and J. KIWI* Institute of Physical Chemistry II, Swiss Federal Institute of Technology (EPFL), 1015 Lausanne, Switzerland (First received 23 May 2000; accepted in revised form 9 August 2000) Abstract}The degradation of 2,4-Dichlorophenol (from now on 2,4-DCP) has been carried out on Nafion–Fe (1.78%) in the presence of H2O2 under visible light irradiation. A solution containing 2,4-DCP (TOC 72 mg C/L)) is seen to be mineralized in 1 h in the presence of H2O2 (10 mM) under solar simulated visible light (80 mW cm2) at pH values between 2.8 and 11. Homogeneous photo-assisted Fenton reactions were capable of mediating 2,4-DCP degradation only up to pH 5.4. The degradation kinetics of 2,4-DCP on Nafion–Fe membranes was more favorable than the one observed during Fenton photoassisted processes at pH 2.8. The degradation of 2,4-DCP was investigated as a function of the substrate, oxidant concentration and applied light intensity. The Nafion–Fe was seen to be effective over many cycles during the photo-catalytic degradation of 2,4-DCP showing an efficient and stable performance during 2,4-DCP degradation without leaching out Fe3+-ions into the solution. Evidence is presented that the degradation at the surface of the Nafion–Fe membrane seems to be controlled by mass transfer and not by chemical reaction of the species in solution. The approach used to degrade 2,4-DCP is shown to be valid for other chloro-carbons like 4-chlorophenol, 2,3-chlorophenol and 2,4,5-trichlorophenol. # 2001 Elsevier Science Ltd. All rights reserved Key words}2,4-DCP degradation and photo-degradation, immobilized Fenton catalysis, chlorophenols degradation
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 type of compounds in water bodies and their known toxic and carcinogenic effects (Halmann, 1996). Halocarbons represent the most abundant family of industrial toxic compounds as it has been recently recognized from different polluting sources like herbicides, pesticides, chemical and solvent manufacturing and the paint industry (Bauer et al., 1999). The application of innovative Advanced Oxidation Technologies (AOT’s) 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. Fenton processes have been used as a powerful source of oxidative radicals generating OH-radicals from H2O2 in the presence of added Fe3+-ions (Walling, 1975). This process has been shown recently to be enhanced by light due to the decomposition of the photo-active Fe(OH)2+
*Author to whom all correspondence should be addressed. Tel.: +41-21-693-3621; fax: +41-21-693-3621; e-mail: john.kiwi@epfl.ch
leading to important additional OH-radicals in solution (Ruppert et al., 1993; Nadtochenko and Kiwi, 1998). This is due to Fe2+ cyclic regeneration in solution through the light induced reaction FeðOHÞ2þ þ hn ! Fe2þ þ OH 3+
ð1Þ
But Fe /H2O2 homogeneously catalyzed reactions need up to 50–80 ppm of Fe-ions in solution which is well above the European Economic Community directives allowing 2 ppm Fe3+ in treated water to be discharged directly into the environment (EEC, 1982). To remove sizable amounts of Fe-ions from solution at the end of the treatment, precipitation and re-dissolution of the Fe-ions are necessary with the added cost in chemicals and labor. To avoid this two drawbacks, Nafion1 perfluorinated membranes have been recently developed where the Fe-ions cluster have been fixed and remain active in H2O2 decomposition (Fernandez et al., 1999). These membranes have been reported to resist attack by the highly oxidative radical (OH/OH E8 1.90 V NHE) and do not allow leaching out of the Fe exchanged on the sulfonic Nafion groups within the 3000 h testing period. The compound under study (2,4-DCP) is a key intermediate in the synthesis of the herbicide 2,4-D
1994
Chlorophenols degradation on Nafion–Fe films
(2,4-dichlorophenoxyacetic acid). The chloroaromatic under study has been reported to have a removal rate 515 (mg g1 h1). It is considered not degradable by acclimated mixed culture (Pitter and Chudoba, 1991) enabling the 2,4-DCP to go through waste water treatment stations without being abated. Since 1992 several studies involving illuminated TiO2 suspensions of 2,4-DCP have appeared in the scientific literature (Ku and Hsieh, 1992; Minero et al., 1995; Malato et al., 1997; Jardim et al., 1997; Gime´nez et al., 1999). The degradation of 2,4-DCP by way of Fenton reactions has been reported some time ago in closed reaction vessels (Barbeni et al., 1983) and later by coupled chemical–biological Fenton reactions focusing on the nature of the bacterial species of the biological ensuing 2,4-DCP destruction (Koyama et al., 1995). The present study intends (a) to investigate the solution parameters intervening in the degradation of 2,4-DCP at a kinetic acceptable rate by Nafion–Fe (1.78%), (b) to compare the degradation of 2,4-DCP with Fenton photo-assisted homogeneous processes, (c) to see the influence of the variables during the degradation mediated by Nafion–Fe (1.78%) in solution, (d) to see whether the degradation of 2,4DCP on Nafion–Fe (1.78%) can be carried out at more favorable pH values than in the homogeneous Fenton case in view of ulterior biological treatment and finally (e) to see if the Nafion–Fe (1.78%) can be used advantageously in the case of chlorophenols other than the 2,4-DCP chosen as model compound. MATERIALS AND METHODS
1995
l 290 nm) each containing 40 ml of reagent solution. The Suntest solar simulator had an intensity of 80 mW cm2 (1.6 photons s1 cm2) within the wavelength range of 290 and 800 nm. The short wavelength UV radiation from the Suntest solar simulator (l5310 nm) was removed from reaching the samples by the Pyrex wall of the reaction vessels. The radiant flux reaching the solutions in the photolysis vessels (60 ml Pyrex) was set at 50 and 80 mW cm2. The Suntest lamp had a wavelength (l) distribution with about 7% of the emitted photons between 290 and 400 nm. The profile of the photons emitted between l ¼ 400 and 800 nm followed closely the solar spectrum. The radiant flux in mW cm2 was measured with a power meter of LSI Corp, Yellow Springs, CO, USA. The Nafion catalysts strips of 48 cm2 were placed immediately behind the wall of the reaction vessel to act as the only light absorber in this two-phase system. Analysis of the irradiated solutions The detection of the dissolved iron in solution was carried out complexing Fe-ions with cyanate (Fernandez et al., 1999). Spectrophotometric analyses and degradation measurements of the solutions were carried out by way of a Hewlett-Packard 8452 diode array spectrophotometer. The total organic carbon (TOC) was monitored via a Shimadzu 500 instrument equipped with an ASI automatic sample injector. Modeling The modeling of the interacting species in the solution was carried out via the Acuchem IV program written by Braun, Herron, Kahaner (National Bureau of Standards, GA, Md 20899). The fitting was carried out with a program written of the MatLab 4.2 by Volker and Hug of EAWAG (Zu¨rich) in conjunction with Acuchem software. The ChemEQL (version 2.0) Program used was written by Beat Mueller, Kastanienbaum, Switzerland. The ionization of 2,4-DCP as a function of pH was calculated using the pKa values in equation (3) to compute at each pH value the ionized fraction via the ChemEQL program.
Chemicals The FeCl3, H2O2, acid, bases and chlorophenols in this study were p. a. and used as received. Membrane preparation Experiments were conducted with Nafion-perfluorinated membrane (Dupont 117, 0.007 in. thick, Aldrich #7.467-4) containing hydrophilic sulfonate groups immobilized on the fluorocarbon matrix. An improved procedure was used compared to the one reported recently by our laboratory (Fernandez et al., 1999) consisting of three steps: (a) The Nafion membrane was washed with H2O2 (10%) and subsequently with H2SO4 (5 N) to get a perfectly transparent material, (b) this membrane is immersed in FeCl3 (0.1 M) for 30 min and washed repeatedly until all Fe3+ and Cl-ion are removed and (c) the membrane is immersed in NaOH (0.01 M) for 30 min to convert the exchanged Fe3+ to its hydrated form. The Fe-content in the Nafion was determined after digesting the membrane in concentrated HNO3 (Tefloncoated autoclave) under pressure and temperature. This solution was subsequently diluted and the Fe-content measured by atomic absorption spectroscopy (AAS) in a Philips 20 AS instrument provided with a flame detector. The membranes used throughout this work had a loading of 1.78% by weight. Photoreactor and irradiation procedures The irradiation vessels used were 60 ml cylindrical Pyrex flasks (cutoff
RESULTS AND DISCUSSION
Degradation of 2,4-DCP solutions by way of photoassisted Fenton reactions Figure 1 shows the spectra of 2,4-DCP in the presence of various concentrations of Fe3+-ions. The increase in optical density in Fig. 1 is due to the increasing concentrations of Fe3+-ions. But no overall changes in the absorption spectrum of 2,4DCP with a peak at l ¼ 285 nm were observed since no complex formation between Fe3+ and 2,4-DCP occurs. Figure 2 shows the TOC reduction in mg C/L of 2,4-DCP during photo-assisted Fenton reactions at different initially pH values. It is readily seen that reactions proceed only up to pH 5.4. These results are consequent with the observations that accelerated Fenton-mediated degradations of chlorocarbons proceed in acid media (Bauer et al., 1999; Halmann, 1996). The observed rate of 2,4-DCP destruction in the presence of H2O2 alone was 55% compared with the rate of photo-degradation observed in the presence of Fe3+ as reported in Fig. 2.
1996
S. Sabhi and J. Kiwi
Fig. 1. Absorption spectra of 2,4-DCP solutions (1 mM) at pH 5.4 as a function of added concentration of FeCl36H2O.
Fig. 2. Degradation of 2,4-DCP (1 mM) as a function of pH in the presence of FeCl3 (0.46 mM or 26 mg Fe3+ L1) and H2O2 (10 mM) under Suntest irradiation (80 mW cm2) in homogeneous solution.
Chlorophenols degradation on Nafion–Fe films
The overall stochiometry of 2,4-DCP mineralization reported in Fig. 2 can be stated as HO2C6 H3 2Cl2 þ 4 OH þ 5O2 ! 6CO2 þ3H2 O þ 2HCl
ð2Þ
The ionization of 2,4-DCP as a function of pH is shown in Fig. 3. The matrix takes into consideration the pKa value of 7.85 for the deprotonation of the 2,4-DCP OH-group. This is the intersection point in Fig. 3. Up to pH 7.95, the 2,4-DCP (denoted by HL) is the dominating species in solution and above this pH the L form predominates according to the equilibrium HL Ð Hþ þ L
ð3Þ
Degradation of 2,4-DCP on Nafion–Fe membranes. Cyclic nature of the process Figure 4 presents the results of the degradation of 2,4-DCP solutions on Nafion–Fe (1.78%) membranes in the dark and under light irradiation in the presence of H2O2 as shown respectively in traces 6 and 7. If H2O2 is added to solutions of 2,4-DCP no mineralization was observed as seen in traces 1 and 2. Photolysis of 2,4-DCP does not lead to the photo-
1997
degradation of this halocarbon (trace 3). Nafion–Fe membranes in the absence of H2O2 were seen to be ineffective in 2,4-DCP degradation (traces 4 and 5). The results presented in Fig. 4 indicate that only Nafion–Fe membranes in the presence of H2O2 under light irradiation were effective in the degradation of 2,4-DCP in kinetically acceptable times since they were able to overcome the non-degradable intermediates formed under the same experimental conditions in the dark. Fig. 5 shows the catalytic nature of the abatement of 2,4-DCP carried out under the experimental conditions reported in Fig. 4 (trace 7). At the end of each cycle the halocarbon and H2O2 are added to the solution and the irradiation with the Suntest is carried out again. These results show the catalytic nature of the degradation reaction on the Nafion–Fe membrane. pH variation observed during 2,4-DCP degradation. Mechanistic and practical implications Figure 6 presents the results of the degradation of 2,4-DCP in the presence of H2O2 under Suntest light irradiation. It is seen that (a) in a wide pH-range the photo-degradation is possible and (b) the kinetics and efficiency of the photo-degradation are about the
Fig. 3. Distribution of 2,4-DCP species in solution as a function of pH.
1998
S. Sabhi and J. Kiwi
same within the pH range 2.8–11. This experimental result differs considerably from the experimental results presented in Fig. 2 showing that in homogeneous systems the photo-degradation proceeds
only up to pH 5.4 under with a much slower kinetics. The nature of the catalysis and the intermediates in Nafion–Fe membranes (Fernandez et al., 1999; Balanosky et al., 1999) seems to be different compared to homogeneous Fenton processes (Nadtochenko and Kiwi, 1998; Pignatello et al., 1999). Figure 7 present the shift of pH of 2,4-DCP solutions with a different initial pH under Suntest light irradiation. The degradation is mediated by the Nafion–Fe membrane in the presence of added H2O2. The pH values coalesce after about 1 h reaction to a value of around 4. The shift of the initial solution without any pH adjustment is seen to move to more acidic values. This pH drop from 5.4. to 4.0 corresponds to an increase in the [H+] concentration by a factor >13. This pH change suggests the generation of HO2 radicals Fe3þ þ H2 O2 ! Fe2þ þ HO2 þ Hþ
ð4Þ
The superoxyde radical HO2 in equation (4) has a one-electron standard potential HO2/ O 2 E8 ¼ 0:75 V vs NHE and the active form in solution of this radical is HO2 since the pKa of HO2 is 4.8 Fig. 4. Degradation of 2,4-DCP solutions. Initial pH 5.4. Suntest irradiation (80 mW cm2) where pertinent. The full points refer to dark degradations and the open points relate to photo-catalytic reactions. In traces 1 and 2 only H2O2 (10 mM) has been added to 2,4-DCP solutions. Trace 3 shows photolysis of 2,4 DCP. Traces 4 and 5 show 2,4-DCP solutions in the presence of Nafion–Fe membranes without addition of H2O2. Traces 6 and 7 show the degradation of the complete system: 2,4-DCP, Nafion-membrane and H2O2 (10 mM).
HO2 Ð Hþ þO 2
pKa ð4:8 0:1Þ
ð5Þ
A channel responsible for the increased acid character of the solution during degradation has already been mentioned in equation (2). In addition Fig. 7 shows the pH during the degradation coalesces to a pH4 after about 1 h reaction for the different initial solution pH values tried. The HO2 in equation (5) is suggested as a radical species in solution during
Fig. 5. Catalytic decomposition of 2,4-DCP in the presence of H2O2 (10 mM) at an initial pH 5.4 under Suntest irradiation (80 mW cm2).
Chlorophenols degradation on Nafion–Fe films
1999
Fig. 6. Degradation of 2,4-DCP solutions under Suntest light irradiation as a function of initial pH in the presence of Nafion–Fe (1.78%) and H2O2 (10 mM).
2,4-DCP oxidation through its dissociated form at pH>4.8. But this not the only oxidative radical species present in the solution. Within 1 h most of the 2,4-DCP is mineralized as seen from Fig. 6. The rather complex shape of the degradation as a function of time in Fig. 7 also suggests the presence of pH dependent Fe-aqueous complexes with different structures and reactivity. The results presented in Fig. 7 suggests species like: Fe(H2O)5(OH)2+, + Fe(H2O)4(OH)+ 2 and [Fe2(H2O)4(OH)2 ] to be present in solution at pH>3 (Gallard et al., 1999). The degradation of 2,4-DCP on immobilized Nafion–Fe systems has the advantage of avoiding the costly initial pH adjustment to acidic pH values required during the homogeneous Fenton degradation of phenols (Ruppert et al., 1993; Halmann, 1996). For the coupling of the pretreated solutions of 2,4-DCP with low cost biological degradation, the adjustment to pH-values ranging from 6 to 8 is still necessary but the initial step involving costly acidification is not necessary. Dependence of the photo-degradation rate on 2,4-DCP concentration Figure 8 presents the decrease in TOC of 2,4-DCP mediated by Nafion–Fe membranes at four different halocarbon concentrations. The inset presents the decrease in optical absorbance at the 2,4-DCP peak. The steeper decline in TOC at higher concentrations is due to the mass transfer taking place between 2,4DCP in solution and the membrane surface during
the degradation process. In this case the mass transfer between solution and the Nafion–Fe membrane will be directly proportional to the difference in 2,4-DCP concentration existing in the diffusion layer between the bulk of the solution and the membrane interface. The diffusion distance (x) of the HO2radical away from the Nafion–Fe membrane can be estimated from the Smoluschoski diffusion relation. x2 ¼ Dt
ð6Þ
Taking the approximate values in equation (6) of D 105 cm2 s1 and t109 s for the HO2 or any other oxidative radical (s), a diffusion distance of 1 nm can be estimated for the diffusion layer. Supposing that the concentration of radicals away from the Nafion–Fe follows a smooth function, the decrease in the concentration of radicals can be estimated from the relation d½oxid rad =dt ¼ k½c ½HO2
ð7Þ
Substituting the numerical values in equation (7) of the radicals (c 1012 M) with 2,4-DCP (c ¼ 0:25 mM) and assuming the diffusion controlled rate k 6 109 M1 s1 (Halmann, 1996), then d[oxid-rad]/dt can be estimated at 1.5 106 M s1. Influence of the H2O2 concentration and the intensity of the applied light during 2,4-DCP degradation Figure 9 presents the photo-degradation of 2,4DCP under Suntest light irradiation mediated by the Nafion–Fe membrane. The abatement of 2,4-DCP is
2000
S. Sabhi and J. Kiwi
Fig. 7. Shift of solution pH during Nafion–Fe mediated photo-degradation of 2,4 DCP in a solution H2O2 (10 mM) under Suntest light irradiation.
Fig. 8. Degradation of different concentrations of 2,4-DCP as a function of time in solutions with an initial pH 5.4 irradiated with a Suntest (80 mW cm2) in the presence of H2O2 (10 mM). The optical absorption for the different 2,4-DCP concentrations is shown in the inset.
Chlorophenols degradation on Nafion–Fe films
Fig. 9. Degradation of 2,4-DCP as a function of H2O2 concentration in the presence of Nafion–Fe membrane under Suntest light irradiation. The concentrations of H2O2 used are noted in the figure captions.
2001
Fig. 10. Photo-degradation of 4-chlorophenol (4-CP); 2,3chlorophenol (2,3-DCP) and 2,4,5-trichlorophenol (2,4,5TCP) mediated by Nafion–Fe in solutions with H2O2 (10 mM) under Suntest irradiation. Full points refer to experiments in the dark and open points are the results of experiments under light irradiation.
Table 1. Photodegradation of chlorophenols mediated by Nafion–Fe membranes under Suntest light irradiation in the presence of H2O2 (10 mM) Chloro-compound Light exposure: Suntest 80 mW cm2 4-Chlorophenol 2,3-Dichlorophenol 2,4,5-Trichlorophenol Dark exposure 4-Chlorophenol 2,3-Dichlorophenol 2,4,5-Trichlorophenol
Initial TOC
Exposure (h)
Decomposition (%)
73 74 72
2 2 2
90 90 90
71 75 68
2 2 2
35 34 25
more favorable as the concentration of the oxidant increases in solution. The visible light of the Suntest irradiator was adjusted to values of 50 and 80 mW cm2. The TOC decrease was faster in the second case confirming the influence of the intensity of the applied light on the degradation rate. Photodegradation of other chlorophenols on Nafion–Fe membranes Figure 10 presents the degradation of some model chlorophenols as a function of time under Suntest light irradiation on Nafion–Fe membranes. The degradation kinetics is seen to be very close for the three chlorophenols presented in Fig. 10. These results lend support to the claim that Nafion–Fe membranes in oxidative media are capable of degrading different chlorophenols with about the same kinetics and independently of the Cl-substitu-
ents of the particular halocarbons as observed during the photo-catalytic degradation mediated by TiO2 dispersions (Halmann, 1996; Barbeni et al., 1998). More important was the observation that the three chlorophenols degraded with about the same kinetics within the pH range under the conditions used in Fig. 10. This result is close to the result reported in Fig. 6 for 2,4-DCP and may be useful for degradations at biological pH. Results in Table 1 show the effect of light on the decomposition of the three halocarbons and phenol taken as reference compound after a 2 h irradiation period.
CONCLUSIONS
The Nafion–Fe catalytic membrane used to degrade 2,4-DCP during this study shows adequate kinetics and chemical stability during long-term
2002
S. Sabhi and J. Kiwi
operation. The degradation kinetics when Nafion–Fe membranes were used was similar and even faster than in the case of homogeneous photo-assisted Fenton reactions. The Nafion membrane showed stable performance during long-term repetitive catalytic cycles. In the absence of visible light irradiation, the degradation intermediates seem to preclude further degradation of 2,4-DCP in solution. Nafion–Fe mediated photo-catalysis allows to carry out the degradation up to pH 11 avoiding the costly pH adjustment necessary during homogeneous photoassisted Fenton degradation. The degradation of 2,4DCP seems to be controlled by mass transfer as shown by the dependency of the TOC values as a function of the initial 2,4-DCP concentration. An estimate for the width of HO2-radical diffusion layer next to the membrane is presented. The influence of the oxidant concentration and the applied light intensity on the degradation of 2,4-DCP have also been investigated. REFERENCES
Balanosky E., Herrera F., Lopez A. and Kiwi J. (1999) Degradation of textile waste waters. Modeling reaction performance. Water. Res. 34, 2225–2237. Barbeni M., Minero C., Pelizzetti E. and Serpone N. (1983) Chemical degradation of chlorophenols with Fentons reagent. Chemosphere 16, 2225–2237. Bauer R., Waldner G., Hager S., Malato S. and Maletzky P. (1999) The photo–Fenton reaction and the TiO/UV process for wastewater treatment}novel developments. Catal. Today 53, 131–144. EEC List of Council Directives 76/4647. European Economic Community. Brussels, Belgium, 1982. Fernandez J., Bandara J., Lopez A., Buffat Ph. and Kiwi J. (1999) Photo-assisted Fenton degradation of non-biodegradable azo-dye (Orange II) in Fe-free solutions mediated by cation transfer membranes. Langmuir 15, 185–192.
Gime´nez J., Curco D. and Queral M. (1999) Photocatalytic treatment of phenol and 2,4-dichlorophenol in a solar plant in the way to scaling. Catal. Today 54, 229–243. Gallard H., DeLaat J. and Legube B. (1999) Influence du pH sur la vitesse d’oxydation de compose´s organiques par FeII/H2O2. Me´chanism re´actionnels et mode´lisation. New J. Chem. 263–268. Halmann M. (1996) Photodegradation of Water Pollutants. CRC Pub Co, Boca Raton, FL. Jardim F., Moraes S. and Takiyama K. (1997) Photocatalytic degradation of aromatic chlorinated compounds using TO2: toxicity of intermediates. Water Res. 31, 1728–1732. Koyama O., Kamagata Y. and Nakamura K. (1995) Degradation of chlorinated aromatics by Fenton Oxidation and methanogenic digester sludge. Water Res. 24, 895–899. Ku Y. and Hsieh C. (1992) Photocatalytic decomposition of 2,4-dichlorophenol in aqueous TiO2 suspensions. Water Res. 26, 1451–1456. Malato S., Blanco J., Richter C., Curco D. and Gimenez J. (1997) Low-concentrating CPC collectors for photocatalytic water detoxification. Environ. Sci. Technol 35, 157–164. Minero C., Pelizzetti E., Pichat P., Sega M. and Vincenti M. (1995) The photocatalytic degradation of dichlorophenols on TiO2. Environ. Sci. Technol. 29, 2226–2234. Nadtochenko V. and Kiwi J. (1998) Photolysis of FeOH+ 2 and FeCl+ in aqueous solution. Photodissocia2 tion kinetics and quantum yields. Inorg. Chem. 37, 5233–5238. Pignatello J., Liu D. and Huston P. (1999) Evidence for an additional oxidant in the photo-assisted Fenton reaction. Environ. Sci. Technol. 33, 1832–1839. Pitter P. and Chudoba J. (1991) Biodegradability of Organic Substances in the Aqueous Environment. CRC Press, Boca Raton. FL. USA. Ruppert G., Bauer R. and Heisler G. (1993) The photo–Fenton reaction, an effective photochemical wastewater treatment process. Photochem. Photobiol. A 73, 75–78. Walling Ch. (1975) Fenton’s reagent revisited. Acc. Chem. Res. 6, 125–131.