- Email: [email protected]
Catalyzed UV oxidation of organic pollutants in biologically treated wastewater effluents
The Science of the Total Environment 277 Ž2001. 87᎐94
Catalyzed UV oxidation of organic pollutants in biologically treated wastewater effluents Gen-Shuh Wang a,U , Huei-Wen Chena , Shyh-Fang Kang b a
Department of Public Health, National Taiwan Uni¨ ersity, 1 Jen-Ai Road, 1st Sec., Room 1432, Taipei, Taiwan b Department of Water Resources and En¨ ironmental Engineering, Tamkang Uni¨ ersity, Taipei, Taiwan Received 28 September 2000; accepted 6 November 2000
Abstract A batch reactor was used to evaluate the efficiency of advanced oxidation process of the organic pollutants in biologically treated wastewater effluents with UVrH 2 O 2 . A 450-W high-pressure mercury vapor lamp was used as the light source. During the degradation process, the concentration of the dissolved organic compounds could be increased by more than twofold due to the decomposition of microorganisms. This increase of the dissolved organic compounds was eliminated if the water was filtered before the photodegradation experiments. It is observed that the UV alone could play a role for the oxidation of the organic pollutants; however, the addition of a small amount of hydrogen peroxide promotes the degradation efficiency of organic compounds in wastewater. The best oxidation efficiency was obtained when the water samples were under acidic conditions ŽpH 5., and the rate of degradation was not enhanced with the increasing H 2 O 2 dosages. The optimum H 2 O 2 dose was between 0.01% and 0.1% for the oxidation processes in this study. The presence of the carbonaterbicarbonate ions in water inhibits the degradation of the organic compounds. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Biologically treated wastewater effluents; Organic pollutants; UV oxidation; Hydrogen peroxide ŽH 2 O 2 .
U
Corresponding author. Tel.: q886-2-23940612; fax: q886-2-23940612. E-mail address: [email protected] ŽG. Wang.. 0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 8 6 5 - 2
88
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
1. Introduction The presence of organic pollutants in water and wastewater has been the cause of public concern in past decades due to their potential health hazards. Depending on their origins, various compositions of organic pollutants have been released into receiving water bodies from wastewater treatment plants and have contaminated the water resources. To meet the strict effluent water quality standards, it has become necessary for wastewater treatment plants to adopt advanced treatment technologies. Advanced oxidation process ŽAOP., which has been proven to be able to successfully destroy numerous organic and halogenated organic chemicals, has been shown to be a powerful treatment alternative for this purpose ŽBeltran ´ et al., 1993; Hofl et al., 1997; Mokrini et al., 1997; Ku et al., 1998.. In general, AOP was applied as an alternative for disinfection of the microorganisms in the wastewater effluents. However, it can also be used to eliminate toxic or hazardous compounds in effluents. When combined with biological treatments, AOP has the advantages of reducing the concentrations of organic contents andror improving the biodegradability. Photochemical methods include direct and catalyzed photolysis, such as the combination of hydrogen peroxide and ultraviolet light ŽUVrH 2 O 2 process., have been attractive for degradation of organic pollutants in the aqueous phase for many years ŽSapach and Viraraghavan, 1997.. In UVrH 2 O 2 process, direct photolysis of H 2 O 2 under UV irradiation produce the very reactive hydroxyl radicals:
mechanism and to determine the primary and overall quantum yields. During the UVrH 2 O 2 process, the concentrations of the organic compounds directly influence the optimum H 2 O 2 dosage for the oxidative performance ŽSapach and Viraraghavan, 1997; Ku et al., 1998.. When the H 2 O 2 is applied below the optimum dosage, increasing its dosage improves the oxidation performance; while if the H 2 O 2 dose is higher than the optimum dose it may result in lower oxidation efficiency. This ‘effective H 2 O 2 dosage’ can be obtained by calculating the concentrations of H 2 O 2 and the organic compounds ŽInce, 1999.. It has been reported that scavengers like carbonate and bicarbonate ions can deteriorate the oxidation efficiency by consuming hydroxyl radicals ŽSapach and Viraraghavan, 1997.. Although the biological wastewater treatment processes are able to remove more than 85% of the BOD Žbiological oxygen demand. in wastewater, there are some persistent organic compounds that are not easily removed by biological treatment and need further purification ŽSchroder, 1998.. In addition to the persistent organic compounds in wastewater effluents, the unsettled activated sludge in treated effluents may cause additional environmental problems. In this study, UVrH 2 O 2 process is used to evaluate its suitability to eliminate the organic pollutants in the non-chlorinated effluents of a secondary wastewater treatment plant. Factors considered to affect the UVrH 2 O 2 process include the unsettled microorganisms from the activated sludge tank, initial H 2 O 2 dosage, pH, and carbonaterbicarbonate ions in the water.
H 2 O 2 q h ª 2OH ⭈ 2. Experimental The hydroxyl radicals generated in water have a very high oxidizing capacity wredox potential s 2.8 V ŽMasten and Davies, 1994.x, and attack the organic compounds relatively non-selectively with rate constants ranging from 10 6 to 10 10 My1 sy1 ŽBuxton et al., 1988.. The photolysis of H 2 O 2 in pure water was studied extensively in the 1950s ŽWeeks and Matheson, 1956; Baxendale and Wilson, 1957. in order to elucidate the reaction
A 10-l stainless-steel batch reactor with a quartz window was used in this study for the photooxidation of organic compounds in treated wastewater effluents. A 450-W high-pressure mercury-vapor lamp ŽHanovia, Ace Glass Co., Vineland, NJ. was used as the light source. The configuration of the reaction chamber used in this study can be seen elsewhere ŽWang et al., 2000..
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
Before each experiment, the UV light was turned on for 10 min to allow the UV energy to become stable. Eight liters of treated wastewater effluent Žbefore chlorination. from the National Taiwan University Hospital Wastewater Treatment Plant ŽTaipei, Taiwan. was then poured into the reactor. Table 1 gives the general water quality parameters for the water used in this study. This biologically treated wastewater effluent has an average chemical oxygen demand ŽCOD. of 34 mgrl, suspended solids Žmainly unsettled activated sludge. of 8.4 mgrl, alkalinity of 125 mgrl as CaCO 3 , and was at pH 7. The mixing in the reactor was achieved with a magnetic stirrer. Two aliquots of 30 ml were used to measure initial organic concentration; duplicate samples of 30 ml were withdrawn from the reactor at various time intervals for analysis of NPDOC Žnon-permeable dissolved organic carbon., H 2 O 2 residuals and UV absorbance at 254 nm. All of the experiments were conducted at pH 7 except if otherwise mentioned in Section 3. The pH of the tested water was adjusted with 1 N H 2 SO4 and NaOH, and the alkalinity of the solutions was adjusted by NaHCO3 ŽNacalai Tesque, Kyoto, Japan.. Samples Ž4 ml each. were filtered through a 0.45-m syringe filter to remove particles and then acidified with one drop of 2 N HCl solution prior to NPDOC analysis. The NPDOC of the water samples was measured by a total organic carbon analyzer ŽShimadzu TOC 5000A., and the UV-visible spectra were measured by a high-precision , dou ble-beam spectrophotom eter ŽShimadzu UV 160A.. The concentration of the hydrogen peroxide was determined spectrophotometrically with potassium titanium ŽIV. oxalate ŽSellers, 1980..
89
Fig. 1. Comparisons of UV oxidation of treated wastewater effluents with Pyrex and quartz filters. ŽwNPDOCx o s 6 mgrl, pHs 7.0, wH 2 O 2 xo s 0%, sample not filtered..
The UV spectrum and energy distribution of the UV lamp used in this study can be seen elsewhere ŽWang et al., 2000.. The major spectral distribution was at wavelengths of 254, 265, 297, 302, 312, 365, 405 and 435 nm when the quartz tube was used as the cooling device. However, the UV light with wavelength less than 302 nm was filtered out when the Pyrex cooling tube was used as the cooling device.
3. Results and discussion Table 1 Water quality of treated wastewater effluents used in this studya pH SS Žmgrl. COD Žmgrl. NPDOC Žmgrl. Alkalinity Žmgrl as CaCO3 . E. coli ŽCFUr100 ml. wlogx a
7.07 8.4 34 6.5 125 4
Values are average water quality obtained in this study.
3.1. Photodegradation of organic compounds without H2 O2 Fig. 1 compares the destruction rate of organic compounds in wastewater irradiated by UV using a Pyrex or quartz filter without hydrogen peroxide. No apparent change on NPDOC or UV254 was observed when a Pyrex filter was used in the system. This result indicates that UVA ŽUV wave-
90
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
length between 320 and 400 nm. alone was unable to oxidize the dissolved organic component in the water. When UVrquartz was used, the NPDOC in the reactor was increased for approximately 140% after 2 h irradiation. This observation indicates that part of the UV energy was used to decompose the microorganisms in the water Žwhich is not measurable in NPDOC analysis due to the filtration., causing an increase of the NPDOC. The increasing NPDOC concentration in the water indicated that the rate of NPDOC formation resulting from decomposition of microorganisms was higher than the rate of NPDOC oxidation by UV light. However, approximately 50% of UV254 was oxidized in the mean time, indicating that the conjugated double bonds in the organic compounds were easily broken by the UV irradiation. The rapid decrease in UV254 demonstrated that UV oxidation is a good alternative to reduce the formation potential of the disinfection byproducts ŽIto et al., 1998.. 3.2. Photodegradation of organic compounds with H2 O2 When hydrogen peroxide was added in the UVrquartz system, a 25% reduction in NPDOC was observed after 2 h of UV irradiation, even when the water was not filtered ŽFig. 2.. Although a minor increase in NPDOC was still observable in the beginning, the aqueous NPDOC concentration began decreasing after 1 h of UV irradiation when H 2 O 2 was added. Compared to the UVrquartz system without H 2 O 2 addition ŽFigs. 1 and 2., a much faster rate of microorganism decomposition and NPDOC mineralization in the water was apparent. The hydroxyl radicals produced in the UVrH 2 O 2 process should play the major role for faster decomposition of microorganisms and their final mineralization. The small organic molecules generated from the decomposed microorganisms were soon mineralized by the hydroxyl radicals, and hence the concentration of the NPDOC in water did not increase as much as those observed in the photolysis experiment without H 2 O 2 . As shown in Fig. 2, a 20% increase in NPDOC was observed after 1 h UV irradiation when H 2 O 2
Fig. 2. Effect of H 2 O 2 on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 6 mgrl, pHs 7.0, sample not filtered..
was added. This increase in NPDOC was much smaller than the NPDOC increase when H 2 O 2 was not added. Since the water samples taken from the wastewater treatment plant were not filtered prior to UV irradiation, the simultaneous reactions of microorganisms disinfectionrdecomposition and NPDOC mineralization were happening in the system. When the microorganisms were decomposed into smaller organic molecules by the UVrH 2 O 2 process, they were detected in NPDOC analysis. Compared with the 20% increase in NPDOC in UVrH 2 O 2 process and the 100% NPDOC increase without H 2 O 2 as observed in Fig. 2, it is apparent that the rate of NPDOC oxidation in Fig. 2 should be faster than the rate of the microorganism decomposition, resulting in the rapid overall NPDOC removal. In order to evaluate the effects of the presence of microorganisms on the NPDOC oxidation by the UVrH 2 O 2 process, the suspended solids Žmainly unsettled microorganisms from the activated sludge tank. were separated by sedimentation from the aqueous phase to evaluate its contribution to the NPDOC concentration. The initial aqueous NPDOC concentration in the treated wastewater sample was 6 mgrl, with 8 mgrl of suspended solids. Before the UVrH 2 O 2 process, 8 l treated wastewater effluent was allowed to settle overnight. The water was removed and the
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
91
1a and Fig. 2. This observation indicates that the hydroxyl radicals generated from H 2 O 2 dissociation were the main responsible species for the oxidation processes. During the UVrquartzr H 2 O 2 process, the results of NPDOC concentration observed in Fig. 2 provide the overall oxidation performance that include the decomposition of the microorganisms and the final mineralization of the organic compounds. 3.3. Effects of initial H2 O2 dosages
Fig. 3. Effects of the suspended microorganisms on UVrquartz oxidation of treated wastewater effluents. ŽpHs 7.0, wH 2 O 2 x o s 0.1%..
settled sludge was diluted with laboratory organic free water ŽNPDOC- 0.1 mgrl.. The diluted water was allowed to settle for another 4 h to ensure the complete separation of sludge and aqueous NPDOC. Laboratory organic free water was used again to mix with the separated suspended solids to conduct the UV irradiation experiment. Fig. 3 shows the results for the UVrH 2 O 2 oxidation of unfiltered treated wastewater, filtered treated wastewater Žwith 1 m filter. and water prepared from the separated suspended solids. As shown in Fig. 3, the trends for NPDOC oxidation in unfiltered and filtered treated wastewater were similar to those shown in Figs. 1 and 2. The NPDOC concentration of the water sample prepared from the suspended solids increased from 0.9 mgrl to 5.5 mgrl after 90 min UV irradiation and then started to decrease. It is clear that the NPDOC measured in the water prepared from the suspended solids can only come from the unsettled activated sludge in the water. This observation indicates that the microorganisms in the water were decomposed into smaller organic species completely after 90 min of UV irradiation with presence of H 2 O 2 . When the H 2 O 2 is not presented in the systems, it will take a much longer contact time for the UV light to decompose the microorganisms, as shown in Fig.
In order to evaluate the effects of initial H 2 O 2 dosages on the oxidation of organic compounds, the water samples were filtered with a 1-m filter to remove the microorganisms and different amounts of H 2 O 2 were added in the system. The initial H 2 O 2 concentration varied from 0 to 0.5% Ž147 mM., and the initial NPDOC concentration was 7 mgrl. As shown in Fig. 4, the increase in NPDOC concentration was far less than the experiments conducted without prior filtration when H 2 O 2 is not added ŽFigs. 1 and 3.. The NPDOC degradation follows first-order kinetics with observed rate constants of k s 0.0054 miny1 ŽH 2 O 2 s 0.01%., 0.0053 miny1 ŽH 2 O 2 s 0.05%., 0.0058 miny1 ŽH 2 O 2 s 0.1%., and 0.0039 miny1 ŽH 2 O 2 s 0.5%.. When H 2 O 2 is combined with
Fig. 4. Effect of initial H 2 O 2 concentrations on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, pHs 7.0, sample filtered with 1-m filter..
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
92
UV radiation, the rate of NPDOC oxidation increases extraordinarily compared to that of direct UV photolysis ŽFig. 4.. The NPDOC oxidation rate increased with the increase of hydrogen peroxide concentration up to 0.01᎐0.1% Ž3᎐29.7 mM. and then decreased with further increases of H 2 O 2 concentration. Since H 2 O 2 is the principal absorber of UV light in dilute aqueous systems, the use of the UV light in the system will result in the production of hydroxyl radicals, which were the main responsible agents for the NPDOC mineralization. At higher concentrations, however, H 2 O 2 itself reacts with these radicals and hence acts as an inhibiting agent for NPDOC oxidation. In addition, H 2 O 2 absorbs the UV energy applied in the system and hence reduces the energy available for oxidation. Fig. 5 shows the experimental data obtained during the irradiation of a 0.1% hydrogen peroxide aqueous solution. The photolysis of hydrogen peroxide follows first-order kinetics with a rate of 0.024 miny1 , no hydrogen peroxide residual was observed after 2 h UV irradiation. 3.4. Effect of pH It has been reported that the best oxidation
Fig. 5. Degradation of hydrogen peroxide during UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%, pHs 7.0, sample filtered with 1-m filter..
Fig. 6. Effect of initial pH on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0%, sample not filtered..
efficiency was obtained at acidic pH for UVrH 2 O 2 process ŽLiao and Gurol, 1995; Mokrini et al., 1997.; however, higher dissociation rate for H 2 O 2 at higher pH have also been reported ŽKu et al., 1998.. The results for oxidation of the organic compounds by UVrH 2 O 2 process at different pH values were given in Figs. 6 and 7. At the cases where the water samples were not filtered and H 2 O 2 was not added, both the NPDOC and UV254 analysis showed that the best oxidation result was obtained at lower pH ŽFig. 6.. However, when H 2 O 2 was added in the water, filtration or not will result in different conclusions. When the samples were filtered with a 1-m filter prior to the UVrH 2 O 2 process, similar results were observed: the best oxidation result was obtained at the lowest pH ŽFig. 7a.. The first-order rate constants were determined as 0.0066 miny1 ŽpH 5., 0.0059 miny1 ŽpH 7., and 0.0047 miny1 ŽpH 9. for the data shown in Fig. 7a. When the samples were not filtered, both the experiments conducted at pH 5 or pH 9 gave
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
93
better oxidation performance than the experiment conducted at pH 7 ŽFig. 7b.. Although the best oxidation performance was obtained at the lowest pH, the experiment conducted at pH 9 did not give an apparent NPDOC increase as those observed at pH 7. When UVrquartz was used as the light source and H 2 O 2 was added, the NPDOC was expected to increase in the initial stage of the oxidation process and then decrease when the microorganisms in the wastewater were completely decomposed. It is inferred that the microorganisms in the wastewater were not decomposed completely by the UVrH 2 O 2 process conducted at pH 9 due to the lower oxidizing capacity. 3.5. Effect of alkalinity 2y in wastewater The presence of HCOy 3 rCO 3 may compete with organic matter and hydrogen peroxide for reaction with hydroxyl radicals. In
Fig. 8. Effect of alkalinity on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%, pHs 7.0, sample filtered with 1-m filter.. 2y on the this study, the effect of HCOy 3 rCO 3 degradation of NPDOC in the UVrH 2 O 2 process was evaluated at initial carbonate concentration of 128᎐378 mgrl as CaCO 3 . The results for the alkalinity effects were given in Fig. 8. Compared to the rate constant Ž0.0051 miny1 . determined for the experiment conducted with 128 mgrl of 2y HCOy 3 rCO 3 , a 22% and 33% reduction of the rate constants for NPDOC removal Ž0.0040 and 0.0034 miny1 , respectively. were observed when the initial bicarbonate and carbonate concentrations were increased to 253 mgrl and 378 mgrl Žas CaCO 3 .. The hydroxyl radical scavenging effect from bicarbonate and carbonate can explain these inhibition effects ᎏ they react readily with hydroxyl radicals that are the primary oxidizing species in the UVrH 2 O 2 process ŽStaehelin and Hoige, 1985; Buxton et al., 1988.. However, the inhibition effects are not so apparent as those observed in single solute systems ŽBaxendale and Wilson, 1957. due to the heterogeneous organic compositions in wastewater.
4. Conclusions Fig. 7. Effect of initial pH and prefiltration on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%..
The rate of organic compounds oxidation in water is greatly increased with the addition of a
94
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
small amount of H 2 O 2 in the UV disinfection systems. However, the NPDOC in the water will be increased due to the decomposition of the unsettled activated sludge in biologically treated wastewater effluents. This effect should be considered when an advanced oxidation process like UVrH 2 O 2 is used in wastewater treatment facilities for control of the organic pollutants. The optimum NPDOC degradation was obtained when the H 2 O 2 was added at 0.01᎐0.1%. Under these experimental conditions, the NPDOC degradation follows first-order kinetics with an observed rate constant of k f 0.005 miny1 . In UVrH 2 O 2 systems, hydrogen peroxide acts as both an initiating and scavenging agent of hydroxyl radicals. The former effect predominates the UVrH 2 O 2 when the initial hydrogen peroxide concentration is lower than 0.01%, the latter at higher concentrations. However, even at high H 2 O 2 concentrations, oxidation of organic compounds is due to hydroxyl radical attack because H 2 O 2 absorbs most of the light. Experimental results indicated that the best organic degradation was obtained at lower pH when H 2 O 2 was not added. When H 2 O 2 was added in the water, filtration or not will result in different conclusions. The presence of bicarbonatercarbonate species has a negative effect due to the scavenging of hydroxyl radicals. Acknowledgements This research was supported by National Science Council, Taiwan, Republic of China, under Grant No. NSC 89-2211-E002-004. References Baxendale J, Wilson J. The photolysis of hydrogen peroxide at high light intensities. Trans Faraday Soc 1957;53:344᎐356. Beltran ´ F, Ovejero G, Acedo B. Oxidation of atrazine in water by UV radiation combined with H 2 O 2 . Water Res 1993; 27:1013᎐1021.
Buxton G, Greenstock C, Helman W, Ross A. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. J Phys Chem Ref Data 1988;17:513᎐886. Hofl C, Sigl G, Specht O, Wurdack I, Wabner D. Oxidative degradation of AOX and COD by different advanced oxidation processes: a comparative study with two samples of pharmaceutical wastewater. Water Sci Technol 1997; 35:257᎐264. Ince N. ‘Critical’ effect of hydrogen peroxide in photochemical dye degradation. Water Res 1999;33:1080᎐1084. Ito K, Jian W, Nishijima W, Baes A, Shoto E, Okada M. Comparison of ozonation and AOPs combined with biodegradation for removal of the precursors in treated sewage effluents. Water Sci Technol 1998;38:179᎐186. Ku Y, Wang L, Shen Y. Decomposition of EDTA in aqueous solution by UVrH 2 O 2 process. J Hazard Mater 1998;60:41᎐45. Liao C, Gurol M. Chemical oxidation by photolytic decomposition of hydrogen peroxide. Environ Sci Technol 1995;29:3007᎐3014. Masten S, Davies S. The use of ozonation to degrade organic contaminants in wastewater. Environ Sci Technol 1994;28:180A᎐185A. Mokrini A, Ousse D, Esplugas S. Oxidation of aromatic compounds with UV radiationrozonerhydrogen peroxide. Water Sci Technol 1997;35:95᎐102. Sapach R, Viraraghavan T. An introduction to the use of hydrogen peroxide and ultraviolet radiation: an advanced oxidation process. J Environ Sci Health A 1997;32: 2355᎐2366. Schroder H. Characterization and monitoring of persistent toxic organics in the aquatic environment. Water Res 1998;38:151᎐158. Sellers R. Spectrophotometric determination of hydrogen peroxide using potassium titanium ŽIV. oxalate. Analyst 1980;105:950᎐954. Staehelin J, Hoige J. Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. Environ Sci Technol 1985;19:1206᎐1213. Wang G, Liao C, Wu F. Photodegradation of humic acids in the presence of hydrogen peroxide. Chemosphere 2000;42:379᎐387. Weeks J, Matheson M. The primary quantum yield of hydrogen peroxide decomposition. J Am Chem Soc 1956;78:1273᎐1279.
Catalyzed UV oxidation of organic pollutants in biologically treated wastewater effluents Gen-Shuh Wang a,U , Huei-Wen Chena , Shyh-Fang Kang b a
Department of Public Health, National Taiwan Uni¨ ersity, 1 Jen-Ai Road, 1st Sec., Room 1432, Taipei, Taiwan b Department of Water Resources and En¨ ironmental Engineering, Tamkang Uni¨ ersity, Taipei, Taiwan Received 28 September 2000; accepted 6 November 2000
Abstract A batch reactor was used to evaluate the efficiency of advanced oxidation process of the organic pollutants in biologically treated wastewater effluents with UVrH 2 O 2 . A 450-W high-pressure mercury vapor lamp was used as the light source. During the degradation process, the concentration of the dissolved organic compounds could be increased by more than twofold due to the decomposition of microorganisms. This increase of the dissolved organic compounds was eliminated if the water was filtered before the photodegradation experiments. It is observed that the UV alone could play a role for the oxidation of the organic pollutants; however, the addition of a small amount of hydrogen peroxide promotes the degradation efficiency of organic compounds in wastewater. The best oxidation efficiency was obtained when the water samples were under acidic conditions ŽpH 5., and the rate of degradation was not enhanced with the increasing H 2 O 2 dosages. The optimum H 2 O 2 dose was between 0.01% and 0.1% for the oxidation processes in this study. The presence of the carbonaterbicarbonate ions in water inhibits the degradation of the organic compounds. 䊚 2001 Elsevier Science B.V. All rights reserved. Keywords: Biologically treated wastewater effluents; Organic pollutants; UV oxidation; Hydrogen peroxide ŽH 2 O 2 .
U
Corresponding author. Tel.: q886-2-23940612; fax: q886-2-23940612. E-mail address: [email protected] ŽG. Wang.. 0048-9697r01r$ - see front matter 䊚 2001 Elsevier Science B.V. All rights reserved. PII: S 0 0 4 8 - 9 6 9 7 Ž 0 0 . 0 0 8 6 5 - 2
88
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
1. Introduction The presence of organic pollutants in water and wastewater has been the cause of public concern in past decades due to their potential health hazards. Depending on their origins, various compositions of organic pollutants have been released into receiving water bodies from wastewater treatment plants and have contaminated the water resources. To meet the strict effluent water quality standards, it has become necessary for wastewater treatment plants to adopt advanced treatment technologies. Advanced oxidation process ŽAOP., which has been proven to be able to successfully destroy numerous organic and halogenated organic chemicals, has been shown to be a powerful treatment alternative for this purpose ŽBeltran ´ et al., 1993; Hofl et al., 1997; Mokrini et al., 1997; Ku et al., 1998.. In general, AOP was applied as an alternative for disinfection of the microorganisms in the wastewater effluents. However, it can also be used to eliminate toxic or hazardous compounds in effluents. When combined with biological treatments, AOP has the advantages of reducing the concentrations of organic contents andror improving the biodegradability. Photochemical methods include direct and catalyzed photolysis, such as the combination of hydrogen peroxide and ultraviolet light ŽUVrH 2 O 2 process., have been attractive for degradation of organic pollutants in the aqueous phase for many years ŽSapach and Viraraghavan, 1997.. In UVrH 2 O 2 process, direct photolysis of H 2 O 2 under UV irradiation produce the very reactive hydroxyl radicals:
mechanism and to determine the primary and overall quantum yields. During the UVrH 2 O 2 process, the concentrations of the organic compounds directly influence the optimum H 2 O 2 dosage for the oxidative performance ŽSapach and Viraraghavan, 1997; Ku et al., 1998.. When the H 2 O 2 is applied below the optimum dosage, increasing its dosage improves the oxidation performance; while if the H 2 O 2 dose is higher than the optimum dose it may result in lower oxidation efficiency. This ‘effective H 2 O 2 dosage’ can be obtained by calculating the concentrations of H 2 O 2 and the organic compounds ŽInce, 1999.. It has been reported that scavengers like carbonate and bicarbonate ions can deteriorate the oxidation efficiency by consuming hydroxyl radicals ŽSapach and Viraraghavan, 1997.. Although the biological wastewater treatment processes are able to remove more than 85% of the BOD Žbiological oxygen demand. in wastewater, there are some persistent organic compounds that are not easily removed by biological treatment and need further purification ŽSchroder, 1998.. In addition to the persistent organic compounds in wastewater effluents, the unsettled activated sludge in treated effluents may cause additional environmental problems. In this study, UVrH 2 O 2 process is used to evaluate its suitability to eliminate the organic pollutants in the non-chlorinated effluents of a secondary wastewater treatment plant. Factors considered to affect the UVrH 2 O 2 process include the unsettled microorganisms from the activated sludge tank, initial H 2 O 2 dosage, pH, and carbonaterbicarbonate ions in the water.
H 2 O 2 q h ª 2OH ⭈ 2. Experimental The hydroxyl radicals generated in water have a very high oxidizing capacity wredox potential s 2.8 V ŽMasten and Davies, 1994.x, and attack the organic compounds relatively non-selectively with rate constants ranging from 10 6 to 10 10 My1 sy1 ŽBuxton et al., 1988.. The photolysis of H 2 O 2 in pure water was studied extensively in the 1950s ŽWeeks and Matheson, 1956; Baxendale and Wilson, 1957. in order to elucidate the reaction
A 10-l stainless-steel batch reactor with a quartz window was used in this study for the photooxidation of organic compounds in treated wastewater effluents. A 450-W high-pressure mercury-vapor lamp ŽHanovia, Ace Glass Co., Vineland, NJ. was used as the light source. The configuration of the reaction chamber used in this study can be seen elsewhere ŽWang et al., 2000..
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
Before each experiment, the UV light was turned on for 10 min to allow the UV energy to become stable. Eight liters of treated wastewater effluent Žbefore chlorination. from the National Taiwan University Hospital Wastewater Treatment Plant ŽTaipei, Taiwan. was then poured into the reactor. Table 1 gives the general water quality parameters for the water used in this study. This biologically treated wastewater effluent has an average chemical oxygen demand ŽCOD. of 34 mgrl, suspended solids Žmainly unsettled activated sludge. of 8.4 mgrl, alkalinity of 125 mgrl as CaCO 3 , and was at pH 7. The mixing in the reactor was achieved with a magnetic stirrer. Two aliquots of 30 ml were used to measure initial organic concentration; duplicate samples of 30 ml were withdrawn from the reactor at various time intervals for analysis of NPDOC Žnon-permeable dissolved organic carbon., H 2 O 2 residuals and UV absorbance at 254 nm. All of the experiments were conducted at pH 7 except if otherwise mentioned in Section 3. The pH of the tested water was adjusted with 1 N H 2 SO4 and NaOH, and the alkalinity of the solutions was adjusted by NaHCO3 ŽNacalai Tesque, Kyoto, Japan.. Samples Ž4 ml each. were filtered through a 0.45-m syringe filter to remove particles and then acidified with one drop of 2 N HCl solution prior to NPDOC analysis. The NPDOC of the water samples was measured by a total organic carbon analyzer ŽShimadzu TOC 5000A., and the UV-visible spectra were measured by a high-precision , dou ble-beam spectrophotom eter ŽShimadzu UV 160A.. The concentration of the hydrogen peroxide was determined spectrophotometrically with potassium titanium ŽIV. oxalate ŽSellers, 1980..
89
Fig. 1. Comparisons of UV oxidation of treated wastewater effluents with Pyrex and quartz filters. ŽwNPDOCx o s 6 mgrl, pHs 7.0, wH 2 O 2 xo s 0%, sample not filtered..
The UV spectrum and energy distribution of the UV lamp used in this study can be seen elsewhere ŽWang et al., 2000.. The major spectral distribution was at wavelengths of 254, 265, 297, 302, 312, 365, 405 and 435 nm when the quartz tube was used as the cooling device. However, the UV light with wavelength less than 302 nm was filtered out when the Pyrex cooling tube was used as the cooling device.
3. Results and discussion Table 1 Water quality of treated wastewater effluents used in this studya pH SS Žmgrl. COD Žmgrl. NPDOC Žmgrl. Alkalinity Žmgrl as CaCO3 . E. coli ŽCFUr100 ml. wlogx a
7.07 8.4 34 6.5 125 4
Values are average water quality obtained in this study.
3.1. Photodegradation of organic compounds without H2 O2 Fig. 1 compares the destruction rate of organic compounds in wastewater irradiated by UV using a Pyrex or quartz filter without hydrogen peroxide. No apparent change on NPDOC or UV254 was observed when a Pyrex filter was used in the system. This result indicates that UVA ŽUV wave-
90
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
length between 320 and 400 nm. alone was unable to oxidize the dissolved organic component in the water. When UVrquartz was used, the NPDOC in the reactor was increased for approximately 140% after 2 h irradiation. This observation indicates that part of the UV energy was used to decompose the microorganisms in the water Žwhich is not measurable in NPDOC analysis due to the filtration., causing an increase of the NPDOC. The increasing NPDOC concentration in the water indicated that the rate of NPDOC formation resulting from decomposition of microorganisms was higher than the rate of NPDOC oxidation by UV light. However, approximately 50% of UV254 was oxidized in the mean time, indicating that the conjugated double bonds in the organic compounds were easily broken by the UV irradiation. The rapid decrease in UV254 demonstrated that UV oxidation is a good alternative to reduce the formation potential of the disinfection byproducts ŽIto et al., 1998.. 3.2. Photodegradation of organic compounds with H2 O2 When hydrogen peroxide was added in the UVrquartz system, a 25% reduction in NPDOC was observed after 2 h of UV irradiation, even when the water was not filtered ŽFig. 2.. Although a minor increase in NPDOC was still observable in the beginning, the aqueous NPDOC concentration began decreasing after 1 h of UV irradiation when H 2 O 2 was added. Compared to the UVrquartz system without H 2 O 2 addition ŽFigs. 1 and 2., a much faster rate of microorganism decomposition and NPDOC mineralization in the water was apparent. The hydroxyl radicals produced in the UVrH 2 O 2 process should play the major role for faster decomposition of microorganisms and their final mineralization. The small organic molecules generated from the decomposed microorganisms were soon mineralized by the hydroxyl radicals, and hence the concentration of the NPDOC in water did not increase as much as those observed in the photolysis experiment without H 2 O 2 . As shown in Fig. 2, a 20% increase in NPDOC was observed after 1 h UV irradiation when H 2 O 2
Fig. 2. Effect of H 2 O 2 on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 6 mgrl, pHs 7.0, sample not filtered..
was added. This increase in NPDOC was much smaller than the NPDOC increase when H 2 O 2 was not added. Since the water samples taken from the wastewater treatment plant were not filtered prior to UV irradiation, the simultaneous reactions of microorganisms disinfectionrdecomposition and NPDOC mineralization were happening in the system. When the microorganisms were decomposed into smaller organic molecules by the UVrH 2 O 2 process, they were detected in NPDOC analysis. Compared with the 20% increase in NPDOC in UVrH 2 O 2 process and the 100% NPDOC increase without H 2 O 2 as observed in Fig. 2, it is apparent that the rate of NPDOC oxidation in Fig. 2 should be faster than the rate of the microorganism decomposition, resulting in the rapid overall NPDOC removal. In order to evaluate the effects of the presence of microorganisms on the NPDOC oxidation by the UVrH 2 O 2 process, the suspended solids Žmainly unsettled microorganisms from the activated sludge tank. were separated by sedimentation from the aqueous phase to evaluate its contribution to the NPDOC concentration. The initial aqueous NPDOC concentration in the treated wastewater sample was 6 mgrl, with 8 mgrl of suspended solids. Before the UVrH 2 O 2 process, 8 l treated wastewater effluent was allowed to settle overnight. The water was removed and the
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
91
1a and Fig. 2. This observation indicates that the hydroxyl radicals generated from H 2 O 2 dissociation were the main responsible species for the oxidation processes. During the UVrquartzr H 2 O 2 process, the results of NPDOC concentration observed in Fig. 2 provide the overall oxidation performance that include the decomposition of the microorganisms and the final mineralization of the organic compounds. 3.3. Effects of initial H2 O2 dosages
Fig. 3. Effects of the suspended microorganisms on UVrquartz oxidation of treated wastewater effluents. ŽpHs 7.0, wH 2 O 2 x o s 0.1%..
settled sludge was diluted with laboratory organic free water ŽNPDOC- 0.1 mgrl.. The diluted water was allowed to settle for another 4 h to ensure the complete separation of sludge and aqueous NPDOC. Laboratory organic free water was used again to mix with the separated suspended solids to conduct the UV irradiation experiment. Fig. 3 shows the results for the UVrH 2 O 2 oxidation of unfiltered treated wastewater, filtered treated wastewater Žwith 1 m filter. and water prepared from the separated suspended solids. As shown in Fig. 3, the trends for NPDOC oxidation in unfiltered and filtered treated wastewater were similar to those shown in Figs. 1 and 2. The NPDOC concentration of the water sample prepared from the suspended solids increased from 0.9 mgrl to 5.5 mgrl after 90 min UV irradiation and then started to decrease. It is clear that the NPDOC measured in the water prepared from the suspended solids can only come from the unsettled activated sludge in the water. This observation indicates that the microorganisms in the water were decomposed into smaller organic species completely after 90 min of UV irradiation with presence of H 2 O 2 . When the H 2 O 2 is not presented in the systems, it will take a much longer contact time for the UV light to decompose the microorganisms, as shown in Fig.
In order to evaluate the effects of initial H 2 O 2 dosages on the oxidation of organic compounds, the water samples were filtered with a 1-m filter to remove the microorganisms and different amounts of H 2 O 2 were added in the system. The initial H 2 O 2 concentration varied from 0 to 0.5% Ž147 mM., and the initial NPDOC concentration was 7 mgrl. As shown in Fig. 4, the increase in NPDOC concentration was far less than the experiments conducted without prior filtration when H 2 O 2 is not added ŽFigs. 1 and 3.. The NPDOC degradation follows first-order kinetics with observed rate constants of k s 0.0054 miny1 ŽH 2 O 2 s 0.01%., 0.0053 miny1 ŽH 2 O 2 s 0.05%., 0.0058 miny1 ŽH 2 O 2 s 0.1%., and 0.0039 miny1 ŽH 2 O 2 s 0.5%.. When H 2 O 2 is combined with
Fig. 4. Effect of initial H 2 O 2 concentrations on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, pHs 7.0, sample filtered with 1-m filter..
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
92
UV radiation, the rate of NPDOC oxidation increases extraordinarily compared to that of direct UV photolysis ŽFig. 4.. The NPDOC oxidation rate increased with the increase of hydrogen peroxide concentration up to 0.01᎐0.1% Ž3᎐29.7 mM. and then decreased with further increases of H 2 O 2 concentration. Since H 2 O 2 is the principal absorber of UV light in dilute aqueous systems, the use of the UV light in the system will result in the production of hydroxyl radicals, which were the main responsible agents for the NPDOC mineralization. At higher concentrations, however, H 2 O 2 itself reacts with these radicals and hence acts as an inhibiting agent for NPDOC oxidation. In addition, H 2 O 2 absorbs the UV energy applied in the system and hence reduces the energy available for oxidation. Fig. 5 shows the experimental data obtained during the irradiation of a 0.1% hydrogen peroxide aqueous solution. The photolysis of hydrogen peroxide follows first-order kinetics with a rate of 0.024 miny1 , no hydrogen peroxide residual was observed after 2 h UV irradiation. 3.4. Effect of pH It has been reported that the best oxidation
Fig. 5. Degradation of hydrogen peroxide during UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%, pHs 7.0, sample filtered with 1-m filter..
Fig. 6. Effect of initial pH on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0%, sample not filtered..
efficiency was obtained at acidic pH for UVrH 2 O 2 process ŽLiao and Gurol, 1995; Mokrini et al., 1997.; however, higher dissociation rate for H 2 O 2 at higher pH have also been reported ŽKu et al., 1998.. The results for oxidation of the organic compounds by UVrH 2 O 2 process at different pH values were given in Figs. 6 and 7. At the cases where the water samples were not filtered and H 2 O 2 was not added, both the NPDOC and UV254 analysis showed that the best oxidation result was obtained at lower pH ŽFig. 6.. However, when H 2 O 2 was added in the water, filtration or not will result in different conclusions. When the samples were filtered with a 1-m filter prior to the UVrH 2 O 2 process, similar results were observed: the best oxidation result was obtained at the lowest pH ŽFig. 7a.. The first-order rate constants were determined as 0.0066 miny1 ŽpH 5., 0.0059 miny1 ŽpH 7., and 0.0047 miny1 ŽpH 9. for the data shown in Fig. 7a. When the samples were not filtered, both the experiments conducted at pH 5 or pH 9 gave
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
93
better oxidation performance than the experiment conducted at pH 7 ŽFig. 7b.. Although the best oxidation performance was obtained at the lowest pH, the experiment conducted at pH 9 did not give an apparent NPDOC increase as those observed at pH 7. When UVrquartz was used as the light source and H 2 O 2 was added, the NPDOC was expected to increase in the initial stage of the oxidation process and then decrease when the microorganisms in the wastewater were completely decomposed. It is inferred that the microorganisms in the wastewater were not decomposed completely by the UVrH 2 O 2 process conducted at pH 9 due to the lower oxidizing capacity. 3.5. Effect of alkalinity 2y in wastewater The presence of HCOy 3 rCO 3 may compete with organic matter and hydrogen peroxide for reaction with hydroxyl radicals. In
Fig. 8. Effect of alkalinity on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%, pHs 7.0, sample filtered with 1-m filter.. 2y on the this study, the effect of HCOy 3 rCO 3 degradation of NPDOC in the UVrH 2 O 2 process was evaluated at initial carbonate concentration of 128᎐378 mgrl as CaCO 3 . The results for the alkalinity effects were given in Fig. 8. Compared to the rate constant Ž0.0051 miny1 . determined for the experiment conducted with 128 mgrl of 2y HCOy 3 rCO 3 , a 22% and 33% reduction of the rate constants for NPDOC removal Ž0.0040 and 0.0034 miny1 , respectively. were observed when the initial bicarbonate and carbonate concentrations were increased to 253 mgrl and 378 mgrl Žas CaCO 3 .. The hydroxyl radical scavenging effect from bicarbonate and carbonate can explain these inhibition effects ᎏ they react readily with hydroxyl radicals that are the primary oxidizing species in the UVrH 2 O 2 process ŽStaehelin and Hoige, 1985; Buxton et al., 1988.. However, the inhibition effects are not so apparent as those observed in single solute systems ŽBaxendale and Wilson, 1957. due to the heterogeneous organic compositions in wastewater.
4. Conclusions Fig. 7. Effect of initial pH and prefiltration on UVrquartz oxidation of treated wastewater effluents. ŽwNPDOCx o s 7 mgrl, wH 2 O 2 x o s 0.1%..
The rate of organic compounds oxidation in water is greatly increased with the addition of a
94
G.-S. Wang et al. r The Science of the Total En¨ ironment 277 (2001) 87᎐94
small amount of H 2 O 2 in the UV disinfection systems. However, the NPDOC in the water will be increased due to the decomposition of the unsettled activated sludge in biologically treated wastewater effluents. This effect should be considered when an advanced oxidation process like UVrH 2 O 2 is used in wastewater treatment facilities for control of the organic pollutants. The optimum NPDOC degradation was obtained when the H 2 O 2 was added at 0.01᎐0.1%. Under these experimental conditions, the NPDOC degradation follows first-order kinetics with an observed rate constant of k f 0.005 miny1 . In UVrH 2 O 2 systems, hydrogen peroxide acts as both an initiating and scavenging agent of hydroxyl radicals. The former effect predominates the UVrH 2 O 2 when the initial hydrogen peroxide concentration is lower than 0.01%, the latter at higher concentrations. However, even at high H 2 O 2 concentrations, oxidation of organic compounds is due to hydroxyl radical attack because H 2 O 2 absorbs most of the light. Experimental results indicated that the best organic degradation was obtained at lower pH when H 2 O 2 was not added. When H 2 O 2 was added in the water, filtration or not will result in different conclusions. The presence of bicarbonatercarbonate species has a negative effect due to the scavenging of hydroxyl radicals. Acknowledgements This research was supported by National Science Council, Taiwan, Republic of China, under Grant No. NSC 89-2211-E002-004. References Baxendale J, Wilson J. The photolysis of hydrogen peroxide at high light intensities. Trans Faraday Soc 1957;53:344᎐356. Beltran ´ F, Ovejero G, Acedo B. Oxidation of atrazine in water by UV radiation combined with H 2 O 2 . Water Res 1993; 27:1013᎐1021.
Buxton G, Greenstock C, Helman W, Ross A. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. J Phys Chem Ref Data 1988;17:513᎐886. Hofl C, Sigl G, Specht O, Wurdack I, Wabner D. Oxidative degradation of AOX and COD by different advanced oxidation processes: a comparative study with two samples of pharmaceutical wastewater. Water Sci Technol 1997; 35:257᎐264. Ince N. ‘Critical’ effect of hydrogen peroxide in photochemical dye degradation. Water Res 1999;33:1080᎐1084. Ito K, Jian W, Nishijima W, Baes A, Shoto E, Okada M. Comparison of ozonation and AOPs combined with biodegradation for removal of the precursors in treated sewage effluents. Water Sci Technol 1998;38:179᎐186. Ku Y, Wang L, Shen Y. Decomposition of EDTA in aqueous solution by UVrH 2 O 2 process. J Hazard Mater 1998;60:41᎐45. Liao C, Gurol M. Chemical oxidation by photolytic decomposition of hydrogen peroxide. Environ Sci Technol 1995;29:3007᎐3014. Masten S, Davies S. The use of ozonation to degrade organic contaminants in wastewater. Environ Sci Technol 1994;28:180A᎐185A. Mokrini A, Ousse D, Esplugas S. Oxidation of aromatic compounds with UV radiationrozonerhydrogen peroxide. Water Sci Technol 1997;35:95᎐102. Sapach R, Viraraghavan T. An introduction to the use of hydrogen peroxide and ultraviolet radiation: an advanced oxidation process. J Environ Sci Health A 1997;32: 2355᎐2366. Schroder H. Characterization and monitoring of persistent toxic organics in the aquatic environment. Water Res 1998;38:151᎐158. Sellers R. Spectrophotometric determination of hydrogen peroxide using potassium titanium ŽIV. oxalate. Analyst 1980;105:950᎐954. Staehelin J, Hoige J. Decomposition of ozone in water in the presence of organic solutes acting as promoters and inhibitors of radical chain reactions. Environ Sci Technol 1985;19:1206᎐1213. Wang G, Liao C, Wu F. Photodegradation of humic acids in the presence of hydrogen peroxide. Chemosphere 2000;42:379᎐387. Weeks J, Matheson M. The primary quantum yield of hydrogen peroxide decomposition. J Am Chem Soc 1956;78:1273᎐1279.