Environ. Sci. Technol. 2010, 44, 6822–6828
Effect of Halide Ions and Carbonates on Organic Contaminant Degradation by Hydroxyl Radical-Based Advanced Oxidation Processes in Saline Waters JANEL E. GREBEL,† J O S E P H J . P I G N A T E L L O , †,‡ A N D W I L L I A M A . M I T C H * ,† Department of Chemical Engineering, Environmental Engineering Program, Yale University, New Haven, Connecticut 06520, and Department of Environmental Sciences, Connecticut Agricultural Experiment Station, 123 Huntington Street, P.O. Box 1106, New Haven, Connecticut 06504-1106
Received March 31, 2010. Revised manuscript received June 27, 2010. Accepted July 7, 2010.
Advanced oxidation processes (AOPs) generating nonselective hydroxyl radicals (HO•) provide a broad-spectrum contaminant destruction option for the decontamination of waters. Halide ions are scavengers of HO• during AOP treatment, such that treatment of saline waters would be anticipated to be ineffective. However, HO• scavenging by halides converts HO• to radical reactive halogen species (RHS) that participate in contaminant destruction but react more selectively with electron-rich organic compounds. The effects of Cl-, Br-, and carbonates (H2CO3 + HCO3- + CO32-) on the UV/H2O2 treatment of model compounds in saline waters were evaluated. For single target organic contaminants, the impact of these constituents on contaminant destruction rate suppression at circumneutral pH followed the order Br- > carbonates > Cl-. Traces of Br- in the NaCl stock had a greater effect than Cl- itself. Kinetic modeling of phenol destruction demonstrated that RHS contributed significantly to phenol destruction, mitigating the impact of HO• scavenging. The extent of treatment efficiency reduction in the presence of halides varied dramatically among different target organic compounds. Destruction of contaminants containing electron-poor reaction centers in seawater was nearly halted, while 17β-estradiol removal declined by only 3%. Treatment of mixtures of contaminants with each other and with natural organic matter (NOM) was evaluated. Although NOM served as an oxidant scavenger, conversion of nonselective HO• to selective radicals due to the presence of anions enhanced the efficiency of electron-rich contaminant removal in saline waters by focusing the oxidizing power of the system away from the NOM toward the target contaminant. Despite the importance of contaminant oxidation by halogen radicals, the formation of halogenated byproducts was minimal.
* Corresponding author phone: (203)432-4386; fax: (203)432-4387; e-mail:
[email protected]. † Yale University. ‡ Connecticut Agricultural Experiment Station. 6822
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
Introduction Advanced oxidation processes (AOPs) generating hydroxyl radicals (HO•), including the ultraviolet light irradiation of hydrogen peroxide (UV/H2O2), are increasingly attractive. HO• reacts with a wide range of chemicals at near diffusioncontrolled limits, making it an excellent oxidant for broadspectrum contaminant removal. Additionally, HO• often introduces oxygenated functional groups, rendering contaminants more biodegradable (1). Because HO• is a nonselective oxidant, AOPs have often been applied to relatively pure waters to minimize HO• scavenging by matrix components. For example, AOPs have been applied to membrane permeates as a barrier technology during municipal wastewater recycling (2). However, AOPs have been used or evaluated for waters with much more challenging matrices, including landfill leachates and industrial wastewaters (3-5). There has even been recent interest in the application of AOPs to decontaminate membrane concentrates from municipal wastewater recycling operations prior to discharge, due to concerns regarding the impact of the elevated contaminant concentrations on receiving water ecosystems (6, 7). Many of these waters contain high Cl- concentrations (Table SI-1 in the Supporting Information) due to industrial or municipal inputs (8). Although less often measured, these waters also contain Br-, because industrial and food-grade chloride salts generally have Br- impurities (9). Halide ions (X-) are significant HO• scavengers, forming reactive halogen species (RHS; X•, X2•-) (eqs 1-6); rate constants for these reactions are available in Table SI-2 in the Supporting Information. HO• + X- h XOH•-
(1)
XOH•- + H+ h X• + H2O
(2)
XOH•- + X- h X•2 + OH
(3)
X• + X- h X•2
(4)
•X• (or HO•) + HCO3 h CO3 + HX (or H2O)
(5)
2X•2 f 2X2 + 2X
(6)
Although AOP treatment of saline waters would appear inefficient, RHS are also capable of oxidizing contaminants. Halogen atoms (X•) rival HO• in the magnitude of their rate constants with organic compounds, while halogen radical anions (X2•-) are generally less reactive than HO• (10). Reactions characteristic of X• and X2•- include one-electron oxidation, H-abstraction, and addition to unsaturated C-C bonds, whereas HO• reacts almost exclusively by the latter two (10). Hydroxyl adds to the aromatic ring of phenol, while X• and X2•- abstract an electron or H from the OH (11). The conversion of HO• to radical RHS in AOPs is expected to result in greater selectivitysi.e., a broader distribution of reaction rates among the compounds presentswith increasing preference for attack at electron-rich centers. Limited research has been conducted on the impact on AOPs of Cl-, typically the most abundant halide in wastewaters. Pignatello reported that inhibition of the Fenton reaction of 2,4-dichlorophenoxyacetic acid at pH 2.8 was noticeable at concentrations above 0.01 M Cl-, attributable to scavenging of HO• (12). Total organic carbon removal 10.1021/es1010225
2010 American Chemical Society
Published on Web 08/03/2010
decreased with increasing chloride concentrations in Fenton and UV/H2O2 AOPs (13-15). Kiwi et al. (16) saw a decrease in removal of an azo dye in the presence of Cl- when HO• was generated by the Fenton reaction under acidic conditions. The generation of Cl2•- was confirmed by laser spectroscopy at concentrations orders of magnitude greater than HO•, although dye removal slowed overall due to the lower reaction rate constants for Cl2•- with the dye. Additionally, AOP treatment in the presence of Cl- produced chlorinated byproducts (13, 16). Br- is potentially of more significant concern for AOPs. Despite occurring at a 675-fold lower concentration than Cl-, Br- is the most important scavenger of HO• in seawater, removing up to ∼93% (17). Bromination is a greater concern than chlorination, as brominated compounds are generally more cyto- and genotoxic (18). Nonetheless, there have been even fewer studies examining the effect of Br- on AOPs. Fenton degradation of 1,2-dibromoethane was autoinhibitory due to release of Br- and was almost completely stopped by addition of 1 mM NaBr (19). To improve understanding of salinity effects on HO•-based treatment, it is critical to determine the extent to which RHS participate in organic compound degradation. If RHS do not participate significantly, halides would serve mostly as HO• sinks, perhaps reducing treatment efficiency to prohibitive levels. If RHS do participate, their oxidation of organics may offset efficiency losses due to HO• scavenging. The objective of this research was to evaluate the impacts of Cl- and Bron the HO•-based treatment of organic contaminants, using UV/H2O2 as a model AOP treatment technology. Model water matrices were employed to enable the elucidation of the responsible reactions by systematic variation of water constituents. Individual halides and halide mixtures were contrasted against another important HO• scavenger, carbonates. Experimental data were compared against a kinetic model to evaluate the model as a predictive instrument for salinity effects on treatment and to indicate the oxidants responsible for organic degradation. The influence of contaminant structures on treatment efficiency was evaluated, and the formation of halogenated byproduct was assessed.
Materials and Methods Experimental method details are available in the Supporting Information but are summarized below. Phenols, cyclohexanol, and 17β-estradiol were obtained from Sigma Aldrich. Nordic Reservoir NOM was obtained from the International Humic Substances Society. The Specific UV Absorbance at 254 nm (SUVA254) was 4.1 L mg-1 m-1. Halide salts used were as follows: Acros analytical grade NaBr (99.5%) and Fluka ultra grade NaCl, reported to contain H2CO3 (29) (see also Table 1). Note that the previous modeling results indicated that HO• scavenging by Cl- formed ClOH•- as the most important intermediate (Scheme 1). At the circumneutral pH conditions of wastewater concentrates and seawater, reaction 8 is not important. Reactions 13 or 14 are needed to convert ClOH•to X2•-. Although Cl- concentrations exceed Br- concentrations 675-fold in seawater, BrCl•- formation predominates due to the higher rate constant for reaction 14 (Table SI-2). 5 -1 -1 ClOH•- + Cl- h Cl•s 2 + OH k ) 1.0 × 10 M
(13) ClOH•- + Br- h BrCl•- + OH- k ) 1.0 × 109 M-1 s-1 (14) Effect of Mixtures. Within mixtures, the presence of additional constituents is expected to reduce the treatment efficiency of a specific target contaminant by reducing the steady-state concentration of oxidants. Even in mixtures of electron-rich and electron-poor constituents, we expect treatment efficiency to be reduced in the absence of halides, because HO• is a nonselective oxidant. However, for target contaminants with electron-rich reaction centers, the addition of halides to the system should convert nonselective HO• to selective RHS, focusing the oxidizing power of the system on the more reactive target contaminant. As a result, the addition
TABLE 3. Effect of Contaminant Mixtures on UV/H2O2 Treatment of Model Contaminants Phenol and Resorcinol conditiona
k′ (s-1)
% reduction
phosphate no competitor seawater phosphate cyclohexanolb seawater phosphated NOMc seawater
1.22 ((0.14) × 10-3 4.93 ((1.49) × 10-4 1.62 ((0.01) × 10-3 6.11 ((0.65) × 10-4 8.33 × 10-4 2.67 × 10-4
-60 -62 -68
compound
phenol
no competitor resorcinol cyclohexanolb NOMc
phosphate seawater phosphate seawater phosphated seawater
1.58 ((0.03) × 10-3 1.16 ((0.05) × 10-3 2.68 ((0.11) × 10-3 2.23 ((0.11) × 10-3 1.08 ((0.03) × 10-3 1.74 ((0.03) × 10-3
-27 -17 --61
a All solutions contained 1 mM H2O2 and 10 µM organic contaminant except where noted. Phosphate ) 10 mM phosphate buffer, pH 8.1; seawater ) 540 mM Cl-, 0.8 mM Br-, 2.3 mM carbonates, pH 8.1. b [Phenol] or [resorcinol] ) 5 µM and [cyclohexanol] ) 5 µM. c [NOM] ) 4.88 mg-C/L Nordic Reservoir NOM. d The standard condition used for comparison is the solution containing NOM but with unreactive sodium perchlorate rather than halides to maintain constant ionic strength to prevent NOM configuration effects.
of halides would enhance the degradation of the electron-rich target contaminant compared to the same mixture in the absence of halides. To evaluate this possibility, two simple contaminant mixtures were compared in phosphate buffer and simulated seawater: moderately reactive phenol with poorly reactive cyclohexanol and highly reactive resorcinol with cyclohexanol. Additionally, we evaluated the effect of Nordic Reservoir NOM on the treatment of phenol and resorcinol. The SUVA254 of this NOM, 4.1 L mg-1 m-1, is higher than for typical municipal wastewaters and groundwaters (30, 31), indicating a relatively high aromatic content that should be on the higher end of RHS reactivity as compared to these types of waters. In the presence of cyclohexanol, the concentrations of phenol or resorcinol were reduced from 10 µM to 5 µM. This reduction increased the UV fluence, enhancing HO• photoproduction, and resulting in an increase in phenol or resorcinol pseudo-first-order destruction rates in phosphate buffer as compared to the no competitor case (Table 3). In phosphate-buffered water, although NOM absorbs photons at the 254 nm output of the lamps and scavenges oxidants, the presence of 4.88 mg-C/L NOM only reduced phenol and resorcinol destruction rates by 32% in both cases. Significant differences were observed in the degradation behavior of phenol and resorcinol when these same scenarios were evaluated in the presence of seawater halides. With or without competitors, halides reduced phenol removal by ∼60% compared to the phosphate-buffered control (Table 3), indicating that differences in reactivity with RHS between the competitors and phenol were not significant. For resorcinol, which is highly reactive with RHS, the presence of competitors reduced the percentage reduction in resorcinol treatment efficiency in going from the phosphate-buffer control to a seawater halide solution. Resorcinol treatment in seawater was reduced by 27% compared with the phosphate buffer control in the absence of a competitor but by only 17% in the presence of cyclohexanol. In the presence of NOM, resorcinol treatment was actually faster in the presence of seawater scavengers (k′/k′0 > 1). In the phosphatebuffered system, HO• reacts with resorcinol and competitors at similar rates, and HO• reactions with competitors reduce resorcinol degradation. In seawater, the conversion of HO•
to RHS focuses the oxidizing power of the system on the most reactive species (i.e., resorcinol as compared to NOM or cyclohexanol). In fact, the absolute pseudo-first-order destruction rate constant in the presence of seawater and NOM exceeded the value in phosphate buffer without NOM, overcoming the light attenuation and oxidant scavenging effects of NOM. The reason behind this enhancement is not entirely clear, but we are currently evaluating the potential for additional RHS formation from direct reaction of photoexcited NOM with halides. We anticipate that this effect may be even more dramatic for NOM with lower SUVA254, the types of NOM prevalent in groundwaters or municipal wastewaters, because the low aromatic content should exhibit even greater differences in reactivity between HO• and RHS. These results imply that in matrices containing moderate amounts of NOM, treatment of highly electron-rich contaminants would be more efficient in the presence of halides than without. Similarly, using gamma radiolysis as a source of HO•, Grebel et al. (32) found that NOM photobleaching accelerated in the presence of seawater halide ions, as RHS formation targeted the oxidizing power of the system away from less reactive components toward electron-rich chromophores. Formation of Halogenated Byproduct. The importance of RHS reactions suggested the possible formation of halogenated byproducts. To evaluate this possibility, phenol was treated in the seawater and wastewater concentrate matrices and solutions were analyzed periodically over a 35 min reaction time period for chloro- and bromophenols (Figure SI-6). In both cases, halogenated products formed and leveled off, remaining relatively constant. In seawater, the yield of total halophenols (relative to the amount of phenol reacted) was 0.52%, with bromophenols constituting 93-96% of the total. In wastewater concentrate, total halophenols formed at 0.03% yield, with bromophenols constituting 60-100% of the total. Due to scavenging of RHS by carbonate, it might be expected that carbonates would lower halophenol yields; however, in both model waters, the presence or absence of carbonates did not affect halogenated product yields. These results indicate that low levels of halogenated product formation are possible. Whether the halophenols are further oxidized to other halogenated products requires further investigation. Environmental Significance. Halides have often been considered only as scavengers for HO• during AOP treatment, such that treatment of saline waters with AOP would be anticipated to be ineffective. However, HO• scavenging by halides simply converts HO• to more selective oxidants. Accordingly, the efficacy of AOP treatment in saline waters is highly contaminant-specific. Halide scavenging may dramatically reduce the treatment efficiency of electron-poor contaminants that react slowly with RHS, but the extent of the reduction with electron-rich contaminants may be less dramatic. Unfortunately, halides reduce the “broad-spectrum” characteristic which has made AOPs attractive. For the least reactive compounds, reactor residence times might need to be increased to impractical levels. On the other hand, many pharmaceuticals and other contaminants of current interest (e.g., 17β-estradiol) have electron-rich functional groups. For these compounds, AOP treatment under saline conditions, even up to seawater levels, may be more feasible than often assumed. Note that compared to a phosphatebuffered control, even in the absence of competitors like NOM, 17β-estradiol destruction declined by only 43% and 3%, respectively, in simulated wastewater concentrates and seawater, respectively (Table 2). Additionally, halides can play a significant role in the presence of an organic matrix. Although NOM serves as a scavenger of oxidants, conversion of nonselective HO• to selective RHS can enhance the efficiency of electron-rich contaminant removal in organicVOL. 44, NO. 17, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6827
rich waters by focusing the oxidizing power of the system away from the NOM toward the target contaminant. Although halogen radicals are involved in this system, the formation of potentially toxic halogenated byproducts appears to be minimal.
Acknowledgments The authors would like to acknowledge the assistance of Amisha Shah in construction of the UV photoreactor apparatus.
Supporting Information Available Detailed descriptions of materials and methods as well as supporting figures. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Goldstone, J. V.; Pullin, M. J.; Bertilsson, S.; Voelker, B. M. Reactions of hydroxyl radical with humic substances: bleaching, mineralization, and production of bioavailable carbon substrates. Environ. Sci. Technol. 2002, 36, 364–372. (2) Wintgens, T.; Melin, T.; Schafer, A.; Khan, S.; Muston, M.; Bixio, D.; Thoeye, C. The role of membrane processes in municipal wastewater reclamation and reuse. Desalination 2005, 178, 1– 11. (3) U.S. Environmental Protection Agency. Handbook of advanced photochemical oxidation processes; EPA/625/R-98/004; Washington, DC, 1998. (4) Lima, A. A.; Mantalvao, A. F.; Dezotti, M.; Sant’Anna, G. L. Ozonation of a complex industrial effluent: oxidation of organic pollutants and removal of toxicity. Ozone Sci. Eng. 2006, 28, 3–8. (5) Li, G.; An, T.; Chen, J.; Sheng, G.; Fu, J.; Chen, F.; Zhang, S.; Zhao, H. Photoelectrocatalytic decontamination of oilfield produced wastewater containing refractory organic pollutants in the presence of high concentration of chloride ions. J. Hazard. Mater. B 2006, 138, 392–400. (6) Westerhoff, P.; Moon, H.; Minakata, D.; Crittenden, J. Oxidation of organics in retentates from reverse osmosis wastewater reuse facilities. Water Res. 2009, 43, 3992–3998. (7) Benner, J.; Salhi, E.; Ternes, T.; von Gunten, U. Ozonation of reverse osmosis concentrate: Kinetics and efficiency of beta blocker oxidation. Water Res. 2008, 42, 3003–3012. (8) Weber, B.; Juanico, M. Salt reduction in municipal sewage allocated for reuse: the outcome of a new policy in Isreal. Water Sci. Technol. 2004, 50, 17–22. (9) Reddy-Noone, K.; Jain, A.; Verma, K. K. Liquid-phase microextraction-gas chromatography-mass spectrometry for the determination of bromate, iodate, bromide and iodide in highchloride matrix. J. Chromatogr., A 2007, 1148, 145–151. (10) Notre Dame Radiation Laboratory. Radiation Chemistry Data Center, Kinetics Database. 2002. www.rcdc.nd.edu (accessed March 10, 2008). (11) Hasegawa, K.; Neta, P. Rate constants and mechanisms of reaction of Cl2*- radicals. J. Phys. Chem. 1978, 82 (6), 854–857. (12) Pignatello, J. Dark and photoassisted Fe3+-catalyzed degradation of chlorophenoxy herbicides by hydrogen peroxide. Environ. Sci. Technol. 1992, 26, 944–951. (13) Seiss, M.; Gahr, A.; Niessner, R. Improved AOX degradation in UV oxidative waste water treatment by dialysis with nanofiltration membrane. Water Res. 2001, 35, 3242–3248. (14) Moraes, J. E. F.; Quina, F. H.; Nascimento, C. A. O.; Silva, D. N.; Chiavone-Filho, O. Treatment of saline wastewater contaminated with hydrocarbons by the photo-Fenton process. Environ. Sci. Technol. 2004, 38, 1183–1187.
6828
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 17, 2010
(15) Maciel, R.; Sant’Anna, G. L.; Dezotti, M. Phenol removal from high salinity effluents using Fenton’s reagent and photo-Fenton reactions. Chemosphere 2004, 57, 711–719. (16) Kiwi, J.; Lopez, A.; Nadtochenko, V. Mechanism and kinetics of the OH-radical intervention during Fenton oxidation in the presence of a significant amount of radical scavenger (Cl-). Environ. Sci. Technol. 2000, 34, 2162–2168. (17) Mopper, K.; Zhou, X. Hydroxyl radical photoproduction in the sea and its potential impact on marine processes. Science 1990, 250, 661–664. (18) Plewa, M. J.; Wagner, E. D.; Jazwierska, P.; Richardson, S. D.; Chen, P. H.; McKague, A. B. Halonitromethane drinking water disinfection by-products: Chemical characterization and mammalian cell cytotoxicity and genotoxicity. Environ. Sci. Technol. 2004, 38, 62–68. (19) Pignatello, J.; Oliveros, E. Advanced oxidation processes for organic contaminant destruction based on the Fenton reaction and related chemistry. Crit. Rev. Environ. Sci. Technol. 2006, 36, 1–84. (20) Jones, D. R. Improved spectrophotometric method for the determination of low levels of bromide. Anal. Chem. Acta 1993, 271, 315–321. (21) Department of Energy. Handbook of methods for the analysis of various parameters of the carbon dioxide system in seawater; version 2; Dickson, A. G., Goyet, C. E., Eds.; ORNL/CDIAC-74; 1994. http://cdiac.ornl.gov/ftp/cdiac74/ (accessed January 21, 2008). (22) Ianni, J. C. Kintecus, Windows Version 2.80, 2002. www. kintecus.com (accessed November 10, 2007). (23) Crittenden, J. C.; Hu, S.; Hand, D. W.; Green, S. A kinetic model for H2O2/UV process in a completely mixed batch reactor. Water Res. 1999, 33, 2315–2328. (24) Matthew, B. M.; Anastasio, C. A chemical probe technique for the determination of reactive halogen species in aqueous solution: Part 1-bromide solutions. Atmos. Chem. Phys. 2006, 6, 2423–2437. (25) Matthew, B. M.; Anastasio, C. A chemical probe technique for the determination of reactive halogen species in aqueous solution: Part 2-chloride solutions and mixed bromide/ chloride solutions. Atmos. Chem. Phys. Discuss. 2006, 6, 941– 979. (26) Carroll, F. A. Perspectives on Structure and Mechanism in Organic Chemistry; Brooks/Cole Publishing Company: Pacific Grove, CA, U.S.A, 1998. (27) Oppenlander, T. Photochemical Purification of Water and Air; Wiley-VCH: Weinheim, Germany, 2003. (28) Liao, C. H.; Kang, S. F.; Wu, F. A. Hydroxyl radical scavenging role of chloride and bicarbonate ions in the H2O2/UV process. Chemosphere 2001, 44, 1193–1200. (29) Liao, C. H.; Gurol, M. D. Chemical oxidation by photolytical decomposition of hydrogen peroxide. Environ. Sci. Technol. 1995, 29, 3007–3014. (30) Leenheer, J. A.; Rostad, C. E.; Barber, L. B.; Schroeder, R. A.; Anders, R.; Davisson, M. L. Nature and chlorine reactivity of organic constituents from reclaimed water in groundwater, Los Angeles county, California. Environ. Sci. Technol. 2001, 35, 3869– 3876. (31) Krasner, S. W.; Westerhoff, P.; Chen, B.; Rittmann, B. E.; Nam, S.; Amy, G. Impact of wastewater treatment processes on organic carbon, organic nitrogen and DBP precursors in effluent organic matter. Environ. Sci. Technol. 2009, 43, 2911–2918. (32) Grebel, J. E.; Pignatello, J. J.; Song, W.; Cooper, W. J.; Mitch, W. A. Impact of halides on the photobleaching of dissolved organic matter. Mar. Chem. 2009, 115, 134–144.
ES1010225
