Environ. Sci. Technol. 2001, 35, 2849-2853
Photoassisted Electrochemical Degradation of Organic Pollutants on a DSA Type Oxide Electrode: Process Test for a Phenol Synthetic Solution and Its Application for the E1 Bleach Kraft Mill Effluent R. T. PELEGRINI,† R. S. FREIRE,‡ N . D U R A N , ‡ A N D R . B E R T A Z Z O L I * ,† Faculdade de Engenharia Mecaˆnica, Universidade Estadual de Campinas, Caixa Postal 6122-13083-970, Campinas, SP, Brazil, and Instituto de Quı´mica, Universidade Estadual de Campinas, Caixa Postal 6154-13083-970, Campinas, SP, Brazil
In this paper, the performance of a photoassisted electrolysis process, for the degradation of organic pollutants, is investigated. Results obtained in this work have shown that the thermally prepared anode of titanium, coated with 70TiO2/30RuO2, exhibits photoactivity and may be used for the treatment of effluents. A synthetic phenol aqueous solution and a real paper mill industry effluent were treated. Kinetic analysis showed a synergetic effect of electrolysis and photocatalysis and degradation rates are an order of magnitude greater than the sum of the results reached by using both processes individually. Using a 125 W mercury bulb and 20 mA cm-2, the phenol concentration decayed 85% in 90 min and 70% reduction of TOC was obtained. In the application of the treatment process for the degradation of the E1 bleach Kraft mill effluent, total phenols were practically eliminated in a short period of processing time, and color, usually resistant to biological treatment, was reduced to 10% from its initial value measured in terms of absorbance. Reductions of AOX, COD, and BOD by 25%, 30%, and 35%, respectively, were also observed.
Introduction Biological and chemical conventional oxidative treatments of aqueous effluent streams, containing organic compounds, are often efficient in complying with legislation. However, this feature is not enough nowadays when environmental considerations are permanently present in the agenda. Conventional technology for wastewater chemical treatment demands transportation, storing, and handling of hazardous chemicals and leads to the generation of toxic sludge. Biological digestion is a long-term treatment in big physical areas, which also leads to the generation of a huge amount of nonbiodegradable soluble and cellular residues. Furthermore, high molecular weight fractions present in some types of aqueous effluents tend to be resistant to biodegradation. * Corresponding author e-mail:
[email protected]. † Faculdade de Engenharia Meca ˆ nica, Universidade Estadual de Campinas. ‡ Instituto de Quı ´mica, Universidade Estadual de Campinas. 10.1021/es001784j CCC: $20.00 Published on Web 06/05/2001
2001 American Chemical Society
These questions are encouraging discussions with regards this environmental problem and the adoption of solutions at the source of effluents. In this field, electrochemical and photochemical technologies may offer an efficient means of controlling pollution as they provide the degradation of organic pollutants without the drawbacks observed in conventional treatments. Electrolysis, heterogeneous photocatalysis, or photoassisted electrolysis may be used for organics abatement as a main or supplementary treatment. Indeed, electrons and photons are the only reactants added to the treatment process that generates no byproducts at all. Literature summarizes the principles and mechanisms for the electrochemical treatment of aqueous solutions containing organic compounds with simultaneous oxygen evolution (1, 2). The key for efficient electrolytic treatment is strongly based on the anode material choice. High corrosion resistance, physical and chemical stability under high positive potentials are the main requests. When properties as above are required, dimensionally stable anodes (DSA), as those discovered by Beer in the seventies (3, 4), are the natural candidates. This designation denotes a class of thermally prepared oxide electrodes where a titanium substrate is covered by metallic oxides. Coatings onto titanium include TiO2, IrO2, RuO2, and Ta2O5. Combinations, such as TiO2/ RuO2, are indicated for alkaline medium, while IrO2/Ta2O5 usually shows longer service life in acidic electrolytes. On the other hand, some DSA type oxide electrodes may receive additions of SnO2 and Sb2O5 in concentrations ranging from minor to main components which increase the service life (5, 6). Heterogeneous photocatalysis as a tool for aqueous effluent treatment is based on the oxidation of an organic pollutant on the surface of a semiconductor catalyst, especially the anatase form of TiO2 particles. Photocatalytic process for organics oxidation has been described in the literature demonstrating the dependence of organics mineralization rate on TiO2 concentration; illumination intensity; organic pollutant concentration; temperature; and pH and type of anions in the solution (7-10). As the TiO2 catalyst is kept in suspension during illumination, separation of the solid phase must be carried out after treatment. Although, many studies have been reported in an effort to find the most efficient technique, coagulation with aluminum chloride allows the separation and reuse of the TiO2 catalyst (11). Immobilization of TiO2 catalyst on a conductive substrate has been tried. By cycling potential, anatase has been immobilized onto an optically transparent SnO2 conducting glass electrode for mechanistic studies of oxidation of selected organic compounds and for the determination of oxidation potentials (12). Sputtering has also been used for the immobilization of TiO2, thus obtaining different stoichiometries and crystal structures (13). Although somewhat delayed, it has been noted that the rutile form of TiO2 may also present photocatalytic properties (14) and, when immobilized on a conductive substrate under an applied bias potential, may have improved its oxidative efficiency due to the electron-hole recombination reduction. Similar photoassisted processes of electrolysis have been used for the degradation of reactive dyes (15) and for the oxidation of nitride ion (16) and of phenol (13). This property of the rutilic structure of TiO2 allows the use of some types of DSA oxide electrodes in which this allotropic form of TiO2 is the major component. In this paper, a study of the potentiality of a titanium anode, coated with 70TiO2/30RuO2, for the photoassisted VOL. 35, NO. 13, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2849
TABLE 1. Physical-Chemistry Parameters of the E1 Bleach Kraft Mill Effluent absorbance (465 nm) pH total organic carbon (mg L-1) total phenols (mg L-1) acute toxicity E. coli (% inhibition) AOX (mg L-1) DQO (mg L-1) DBO (mg L-1) conductivity (mS - 30 °C)
FIGURE 1. Schematic view of the photoelectrochemical cell reactor: (1) water jacket for temperature control; (2) Ti/TiO2/RuO2 anode; (3) Ti screen cathode; (4) quartz tube for the lamp; and (5) magnetic bar stirrer. electrolysis process (called from now on as photoelectrochemical process), is investigated. First, in preliminary experiments, a synthetic solution containing phenol was used in order to follow the performance of this electrode. Efficiency of photocatalysis, electrochemical, and photoelectrochemical treatments were compared for phenol abatement and TOC reduction. Then, in a new series of experiments, a real aqueous effluent taken from a pulp and paper industry (Kraft process) was subjected to the same treatment. The E1 bleach Kraft mill effluent, taken from the first stage of the alkaline extraction, presents the highest toxicity when compared to the aqueous streams from the subsequent pulp bleaching stages. It presents high concentrations of macromolecules derived from lignin fragmentation, that usually are refractory to biological digestion (17). High levels of adsorbable organic halogens are also found in this type of aqueous effluent, and these compounds are also responsible for the brown color of the end-of-pipe effluent that, after biological treatment, may get darker (18-20). The effectiveness of the treatment process in the abatement of such toxicity was monitored by following the total phenols concentration decay; inhibition of acute toxicity test; and TOC and color reduction.
Experimental Details Apparatus. Photocatalytic, electrochemical, and photoelectrochemical treatments were carried out in a 4 in. diameter, 500 mL, single compartment, glass cell with an 1 in. diameter quartz tube placed in the center and used as the UV bulb housing. The experimental set up is sketched in Figure 1. A Ti screen cylinder was used as a cathode and mounted around the quartz tube. Concentric to this set, an Ti anode coated with 70TiO2/30 RuO2 was used in such a way that the anode/ cathode gap was 1 mm. The anode area was 100 cm2, and the coating process is described elsewhere (21). The UV radiation was provided by a high-pressure mercury lamp (Philips HPL-N 125 W, fluency rate of 31.1 J m2 s-1) that presented λmax ) 254 nm and from which the cover was removed. Solutions. Synthetic phenol solution used in the preliminary experiments was prepared with 50 mg L-1 of phenol in a 0.1 mol L-1 Na2SO4 supporting electrolyte, pH 7. The real effluent was obtained from a paper mill industry in Campinas city region (Sa˜o Paulo state- Brazil), which processes Eucalyptus grandis pulp wood using the Kraft process. It was taken from the E1 stage of the alkaline extraction, stored at 4 °C in glass flasks, and used without previous filtration or any other teatment. The main characteristics of the real effluent are shown in Table 1. 2850
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 13, 2001
0.71 11.0 1275 37 60 60 2500 760 10.1
Operational Parameters. Combination of 20 mA cm-2 current density with 125 W UV bulb was used. Temperature of the experiments was kept at 30 °C by using an outer water jacket for the cell, where 300 mL of solution were processed in each experiment magnetically stirred. Analytical Control. Phenol concentration in the synthetic solution was determined in a Shimadzu HPLC equipment with a Shin-Pack CLC-ODS column and a UV-vis detector, using a wavelength of 265 nm. The total amount of phenols in the real aqueous effluent was followed by UV-vis colorimetric analysis according to Folin-Ciocalteau standard procedure. In this analysis, an analyte sample of 1 mL is added to a 250 µL carbonate-tartarate buffer and 25 µL of the Folin-Ciocalteau reactant (Aldrich). After 30 min resting time at 20 °C, absorbance was recorded in a U-2000 Hitachi spectrophotometer, at 700 nm. The same equipment was used for color determination with a wavelength of 465 nm with 7.6 pH fixed by a 0.1 mol L-1 phosphate buffer solution. Total organic carbon values were obtained in a TOC 5000 Shimadzu equipment. Profiles of all these parameters were obtained by sampling solutions at predetermined time intervals during the processing. The toxicity of the effluent was evaluated in comparison to the toxicity of irradiated/ozonizated solutions by measuring the inhibition of the respiration of Escherichia coli cultures. The assay consists of the incubation of E. coli cultures at 37 °C with known amounts of the stressing agent. When the CO2 concentration, produced by microbial respiration, reached 0.5 mmol L-1, or approximately 9 × 108 cells mL-1, 45 mL of the E. coli culture was transferred into several flasks, each of which receiving 5 mL of the sample withdrawn at a selected treatment time. As a control, 5 mL of distilled water was introduced into one of the flasks, and the CO2 production was monitored every 20 min using flow injection analysis (FIA). The toxicity test was followed for a maximum period of 120 min. For incubation periods of more than 120 min, there is loss of CO2 to the atmosphere due to CO2 oversaturation (>5 mmol L-1) in the aqueous culture medium. The bacteria (ATCC 25922), used in the respirometric acute toxicity test, was provided by the Laboratory of Microbiology from the University of Campinas Hospital.
Results and Discussion Processing of Phenol Synthetic Solution. Figure 2 presents the results relative to phenol degradation: (a) photocatalytic treatment, using a 125 W bulb; (b) for the electrochemical treatment, using current density of 20 mA cm-2; and (c) for the combination of both, or the photoelectrochemical treatment. Starting from 50 mg L-1 of phenol solution, 20% of the initial concentration was observed after 90 min of the oxide electrode surface illumination. On the other hand, the electrochemical process showed 30% reduction of the phenol concentration under current density of 20 mA cm-2 within the same time interval. The goal of this comparison is not to show which process is more efficient since they are different processes and the result yields are not comparable. However, it is useful to show that when both treatments are simul-
FIGURE 2. Phenol concentration reduction as a function of treatment time for the photocatalytic process (125 W mercury bulb), for the electrochemical process (20 mA cm-2) and for the photoelectrochemical process (125 W, 20 mA cm-2). Volume treated ) 300 mL. C(0) ) 50 mg L-1. Anode area ) 100 cm2.
FIGURE 4. Logarithm of phenol normalized concentration as a function of treatment time for the three processes considered. Data taken from Figure 2.
TABLE 2: Kinetic Constants for Phenol Oxidation for the Three Processa
a
process
kph/m s-1
photcatalytic electrochemical photoelectrochemical
2.9 × 10-3 5.3 × 10-3 2.2 × 10-2
Data taken from the inset of Figure 4.
in which the physiosorbed hydroxyl radical is generated when electrolysis is assisted by photocatalysis. Electrolytic discharge of water and hydroxyl radical formation is more likely to occur on RuO2 conductive phase of the oxide coating, according to ref 22: FIGURE 3. Total organic carbon reduction as a function of treatment time for the three processes. See Figure 1 for the operational conditions. taneously applied, the resulting phenol degradation rate is more than a simple sum of their effects. Figure 2 also shows the curve of phenol concentration decay for the photoelectrochemical process in which 85% of reduction was observed in 90 min. The same effect can be observed in Figure 3, which shows the TOC reduction as a function of the treatment time. Again, the combination of both treatments presented a final result of 70% of TOC reduction that is greater than the sum of the results reached by using both process individually. The synergy observed, that improves the phenol conversion rate, may be better observed by the kinetic analysis. Considering the phenol oxidation as a first-order decay process, mass balance results in
( )
Akph C(t) ) C(0)expt V
(1)
where C(0) is the initial concentration, A is the electrode area, V is the volume of the solution being processed, t the treatment time, and kph is the kinetic constant or the apparent rate constant for phenol oxidation. The slope in the plot of ln[C(t)/C(0)] vs t leads to the kinetic constant. Figure 4 shows this plot, using data taken from Figure 2, for the three processes considered in this investigation from which the kinetic constant for phenol abatement were calculated. Table 2 shows that kinetic constant obtained by using the photoelectrochemical process is 1 order of magnitude greater than those calculated for the individual processes. The probable reason for the improvement in the conversion rate is the generation of a greater number of active sites
H2O + RuO2[ ] f RuO2[OH‚] + H+ + e-
(2)
The physiosorbed species are responsible for the organic (R) oxidation to RO on RuO2 particles:
RuO2 [OH‚] + R f RuO2 [ ] + RO + H+ + e-
(3)
With the illumination of the electrode oxide surface, the TiO2 semiconductor rutile phase generates electron/hole pairs:
TiO2[ ] + hν f e- + h+
(4)
Discharge of water also occurs on the oxidative sites of the electrode oxide surface, particularly on the TiO2 particles, generating hydroxyl radicals as follows
TiO2[ ] + H2O + h+ f TiO2[OH‚] + H+
(5)
and the organic compound is oxidized on TiO2 particles:
TiO2[OH‚] + R f TiO2 [ ] + RO + H+ + e-
(6)
Equations 2-6 suggest a summation of the electrolytic and photocatalytic hydroxyl generation phenomen. However, results depicted in Table 2 show a greater conversion rate than the sum of both process applied individually. This synergy may be due to the reduction of electron/hole recombination rate which makes the semiconductor surface inactive (23). By assisting photocatalysis with electrolysis, photogenerated electrons are promptly withdraw from the anode which reduces recombination rate. The synergic aspect observed by assisting electrolysis with photocatalysis, or vice versa, is important when one considers the need for VOL. 35, NO. 13, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2851
FIGURE 5. Results for the E1 bleach Kraft mill effluent. Color (A), total phenols (B), TOC (C), and acute toxicity (D) reduction as a function of treatment time for the photocatalytic process (125 W mercury bulb), for the electrochemical process (20 mA cm-2) and for the photoelectrochemical process (125 W, 20 mA cm-2). Volume treated ) 300 mL. Anode area ) 100 cm2. degradation of some organic pollutants that usually are recalcitrant to electrochemical or heterogeneous photocalytic treatments. The majority of aqueous effluents containing organic pollutants that demand degradation of TOC and mainly reduction of color greatly benefit from the application of the combined process, as we will see later. Processing of the Real E1 Effluent. As described earlier, the E1 solution (conductivity of 10.1 mS) was processed using current density of 20 mA cm-2 and a 125 W mercury lamp. Results obtained for real effluent processing are shown in Figure 5. Figure 5A shows the evolution of solution discoloration that usually represents a big challenge to conventional biological treatments. Electrochemical and photocatalytical process, when used individually, reduced 50% and 20% of the solution color, respectively. However, 90% of discoloration was reached after 90 min of photoelectrochemical treatment. This result means a reduction of absorbance from the initial value of 0.71 to 0.071 or from a brown color solution to a transparent solution. Ninety minutes of processing was also enough for the elimination of almost 100% of total phenols, as can be seen in Figure 5B. In this case both electrochemical and photoelectrochemical processes showed similar efficiencies, and small differences are within the experimental error. Figure 5C shows total organic carbon (TOC) reduction for the same operational conditions in which up to 40% reduction is observed. Finally, Figure 5D shows the data recorded for the acute toxicity tests using E. coli. Toxicity was reduced to 60% from the initial value in 60 min of the combined process application, that means E. coli growth inhibition from the initial value of 60% to 36%. Taking a general view from Figure 5, synergic effect of photoelectrochemical treatment is clear for discoloration and also for TOC reduction. Photocatalytic process was responsible for 6% of TOC removal, while electrochemical processing reduced 19%. By the combination of both, 39% of TOC initial value was observed. However, total phenols present in this aqueous effluent are surprisingly easy to degrade by the electrochemical technique in a narrow time interval, so that it is difficult to decide whether the photoelectrochemical treatment may contribute to the abatement rate. Actually, the electrochemical process seems to oxidize lignin and its fragments at similar rates when compared to photoelectrochemical process. 2852
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 13, 2001
Values of TOC reduction, by using photoelectrochemical process, are expressive when compared to conventional biological treatment. An activated sludge process is capable of 10-30% of TOC reduction for the same effluent (24). Furthermore, no color reduction is usually observed after 10-15 days of conventional biological treatment (25), and, for the photoelectrochemical process considered here, 90% discoloration was observed. Other parameters used for a treatment process performance evaluation are chemical oxygen demand (COD) and biological oxygen demand (BOD). One of the sources of COD and BOD in the pulp bleaching aqueous effluent is represented by high molecular weigh fractions, and part of them are refractory to biological digestion (26). Results obtained in this work showed that COD and BOD were reduced by 30% and 35%, respectively, related to those initial values shown in Table 1, for 90 min of treatment. Adsorbable organic halogens (AOX) are also responsible for the high toxicity of the effluente (27) and are difficult to degrade. Biological digestion usually takes 10-20 days for degradation of 30% (28). In the present case, 90 min of photoelectrochemical treatment resulted in 25% of reduction. Results obtained in this work have shown that thermally prepared DSA type oxide electrodes exhibit photoactivity and may be used for photoassisted electrolysis for organics degradation in aqueous effluents. This feature enlarges the range of DSA electrode applications, usually restricted to electrolytic processes. By assisting electrolysis with photocatalysis, a significant increase in the phenol degradation rate was observed. Kinetic analysis showed a synergetic effect in which phenol concentration decay, and the TOC reduction rate is 1 order of magnitude greater than the sum of the results reached by using both processes individually. The same important effect was observed when the treatment process was used for the degradation of the E1 bleach Kraft mill effluent. Total phenols were practically eliminated in short processing time and color reduction, usually resistant to biological treatment, greatly benefits from the application of the combined processes. By using a photoelectrochemical process for toxicity removal of a real effluent other parameters as AOX, BOD, and COD were reduced. Results reported in this paper also show that photoelectrochemical process can be used as the main treatment for an effluent stream or integrated to a biological oxidation processes in a wastewater treatment unity. The photoelectrochemical treatment could be used for the oxidation of refractory compounds and higher molecular weight fractions, followed by biological mineralization.
Acknowledgments The authors would like to acknowledge financial support from FAPESP.
Literature Cited (1) Savall, A. Chimia 1995, 49, 23. (2) Comninellis, C. Electrochemical oxidation of organic pollutants for wastewater treatment. In Environmental Oriented Electrochemistry; Sequeira, C. A. C., Ed.; Elsevier: Amsterdam, 1994; pp 77-101. (3) Beer, H. B. U.S. Patent 1972, 3 632 498. (4) De Nora, O.; Nidola, A.;Trisoglio, G.; Bianchi, G. Br. Patent 1973, 1 399 576. (5) Lipp, L.; Pletcher, D. Electrochim. Acta 1997, 42, 7, 1091. (6) Correa-Lozano, B.; Comninellis, Ch.; De Battisti, A. J. Appl. Electrochem. 1997, 27, 8, 970. (7) Mills, A.; Davies, R. H.; Worsley, D. Chem. Soc. Rev. 1993, 22, 6, 417. (8) Serpone, N. Res. Chem. Intermediates 1994, 20, 9, 953. (9) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. Rev. 1995, 95, 69. (10) Mills, A.; LeHunt, S. J. Photochem. Photobiol. A- Chem. 1997, 108, 1, 1.
(11) Kagaya, S.; Shimizu, K.; Hasegawa, K. Water Res. 1999, 33, 7, 1753. (12) Taghizadeh, A.; Lawrence, M. F.; Miller, L.; Anderson, M. A.; Serpone, N. J. Photochem. Photobiol. A- Chem. 2000, 130, 2-3, 145. (13) Dimitriu, D.; Bally, A. R.; Ballif, C.; Hones, P.; Schmid, P. E.; Sanjines, R.; Levy, F.; Parvulescu, V. I. Appl. Catal. B- Environ. 2000, 25, 2-3, 83. (14) Wong, J. C. S.; Linsebigler, A. L.; Lu, G.; Fan, J.; Yates, J. T., Jr. J. Phys. Chem. 1995, 99, 335. (15) Pelegrini, R.; Peralta-Zamora, P.; de Andrade, A. R.; Reyes, J.; Duran, N. Appl. Catal. B- Environ. 1999, 22, 2, 83. (16) Sun, C. C.; Chou, T. C. J. Mol. Catal. A- Chem. 2000, 151, 1-2, 133. (17) Bullock, C. M.; Bicho, P. A.; Saddler, J. N. Water Res. 1996, 30, 5, 1095. (18) Royarcand, L.; Archibald, F. S.; Briere, F. Tappi J. 1991, 74, 9, 211. (19) Royarcand, L.; Archibald, F. S. Water Res. 1993, 27, 5, 873. (20) Parker, W. J.; Hall, E. R.; Farquhar, G. J. Water Environ. Res.
1993, 65, 3, 264. (21) Boodts, J. F. C.; Trasatti, S. J. Electrochem. Soc. 1990, 137, 12, 3784. (22) Simond, O.; Schaller, V.; Comninellis, Ch. Electrochim. Acta 1997, 42, 13-14, 2009. (23) Pelegrini, R.; Reyes, J.; Duran, N.; Peralta-Zamora, P.; De Andrade, A. R. J. Appl. Electrochem. 2000, 30, 953. (24) Cecen, F. Water Sci. Technol. 1999, 40, 11-12, 305. (25) Garg, S. K.; Modi, D. R. Crit. Rev. Biotechnol. 1999, 19, 2, 85. (26) Barker, D. J.; Mannucchi, G. A.; Salvi, S. M. L.; Stuckey, D. C. Water Res. 1999, 33, 11, 2499. (27) Berge, D.; Ratanaweera, H.; Efraimsen, H. Water Sci. Technol. 1994, 29, 5-6, 219. (28) Schnell, A.; Steel, P.; Melcer, H.; Hodson, P. V.; Carey, J. H. Water Res. 2000, 34, 493.
Received for review October 18, 2000. Revised manuscript received March 7, 2001. Accepted April 12, 2001. ES001784J
VOL. 35, NO. 13, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2853
