Ind. Eng. Chem. Res. 2004, 43, 5027-5031
5027
Study of Herbicide Alachlor Removal in a Photocatalytic Process through the Examination of the Reaction Mechanism W. Chu* and C. C. Wong Department of Civil and Structural Engineering, Research Centre for Environmental Technology and Management, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong
The reaction mechanism and intermediates of the photocatalytic degradations of alachlor in TiO2 suspensions using narrow-band UV radiation with a peak at 300 nm were investigated. The direct photolysis of alachlor was found to be an insignificant and slow process, but the addition of TiO2 enhanced the reaction rates by about 12 times. Intermediates were identified using liquid chromatography-electrospray ionization-mass spectrometry and MS/MS. A total of 8 major intermediates were identified. The evolution profiles of different intermediates were quantified and the reaction mechanism was proposed accordingly. The major degradation mechanisms of alachlor photocatalysis include dechlorination, dealkylation, hydroxylation, scission of the C-O bond, and N-dealkylation. Both the decay of alachlor and the total organic carbon (TOC) increased slightly with initial pH levels. The TOC analysis revealed the different degradation stages of the reaction, where a lag phase with very low TOC decay was followed by a mild TOC decay and a fast pseudo-first-order TOC decay. Introduction Photocatalytic reactions using titanium dioxide (TiO2) and ultraviolet (UV) have been shown to be useful for destroying a wide range of environmental contaminants.1-2 Photocatalytic degradation appears to be an effective strategy for degrading and mineralizing chlorinated pesticides.3-7 In general, photoinduced electrons (e-) and positive holes (h+) are produced from TiO2 under the irradiation of UV light (λ < 380 nm), which has an energy greater than the band gap (3.2 eV) of TiO2. The highly oxidizing positive holes, which could be trapped by surface hydroxyl groups to yield hydroxyl radicals, have been considered to be the dominant oxidizing species contributing to the mineralization process resulting from the TiO2 photocatalyst. In addition to the performance of the treatment process, the study of the formation of byproducts is also important. When using the photocatalytic process mediated by TiO2 to degrade potentially toxic compounds, monitoring the disappearance rate of the target compound is not the most appropriate parameter for classifying the efficacy of this process.8 Information on the formation and decomposition of the degradation intermediates or byproducts is critical before a clean technology can be established. For example, the formation of toxic byproducts during photocatalysis8-11 and the photosensitized degradation12 of chlorinated pesticides or organic compounds were often reported. Alachlor [2-chloro-2′,6′-diethyl-N-(methoxymethyl)acetanilide] is a widely used herbicide; it has been detected in groundwater and rivers in the United States,13 and about 26 million lb. were used for agricultural purposes from 1990 to 1993 in the United States. The maximum contaminant level of alachlor for drinking water established by the United States Environmental Protection Agency is 2 µg/L.14 A review of past research reveals that broadband light sources, such as a xenon lamp, * To whom correspondence should be addressed. Tel.: +85227666075. Fax: +852-23346389. E-mail:
[email protected].
were used to study the photolysis or photocatalysis of alachlor15 whereas the use of narrow-band UV irradiation was seldom attempted. Therefore, this study focused on the degradation mechanisms of alachlor in the UV/TiO2 process using TiO2 suspensions illuminated with narrow-band UV radiation with a peak at 300 nm. The pH effect on the degradation rate of alachlor and on the decay of total organic carbon (TOC) during photocatalysis was also investigated. Since information on the reaction mechanism of the photocatalysis of alachlor is very limited, and only a few intermediates were identified due to the limitations of conventional GC-MS,16 the identification of reaction intermediates was performed using the most recent technique: liquid chromatography-mass spectrometry (LC-MS) and MS/ MS. This enables various soluble intermediates to be identified, and reveals a more detailed scheme of the degradation mechanisms and reaction pathways of the process. TOC changes were also monitored simultaneously, so that the relationship between mineralization and alachlor degradation could be established. Methodology Chemicals. Alachlor (99.7% HPLC grade) was purchased from Riedel-de Hae¨n, and the initial concentration of alachlor in all of the experiments was 2.2 × 10-5 mol/L. Titanium dioxide (TiO2-P25; containing 70% anatase and 30% rutile) with an average particle size of 30 nm and BET surface areas of 50 m2/g,17 obtained from Degussa Corp., Japan, was used in this study. Since the initial alachlor concentration was relatively low, the concentration of titanium dioxide used in all experiments was 5 mg/L and not the optimum concentration at 50 mg/L,18 so that a better resolution of reaction mechanism can be realized. Acetonitrile of HPLC grade from LAB-SCAN was used in the preparation of the mobile phase in the HPLC analysis without further purification. All solutions were prepared using water, with a resistivity of 18 MΩ‚cm, that had been distilled-deionized by a Barnsted Nanopure water
10.1021/ie0342356 CCC: $27.50 © 2004 American Chemical Society Published on Web 06/29/2004
5028 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004
Figure 1. Experimental setup.
system. Analytical reagent grade hydrochloric acid and sodium hydroxide were used to adjust the initial pH of the solutions to the predetermined levels. Methods. All direct photolysis and photocatalysis experiments were conducted using an RPR-200 Rayonet photochemical reactor purchased from the South New England Ultraviolet Company. In each experiment, a quartz vessel (52 mm i.d. × 470 mm length) filled with 500 mL of sample solutions was placed inside the reactor and illuminated with narrow-band UV radiation at 300 nm (see Figure 1). A magnetic stirrer was located at the reactor’s base so that a homogeneous TiO2 suspension could be maintained throughout the reaction. A cooling fan was also installed at the reactor base so that the experimental temperature was maintained at 24-25 °C. For UV irradiation, 10 narrow-band “sunlight” UV lamps with 300 nm and a power consumption of 210 W were used in this study. According to the manufacturer’s specifications, the UV lamps have 70% of the radiation energy falling from 275 to 335 nm with a peak centered at around 300 nm, and the total photon flux for 10 lamps is calculated to be 8.62 × 10-6 einstein cm-2 min-1. The reaction was initiated by turning on the UV lamps, and samples were taken from the quartz vessel at different reaction intervals. The collected samples were first centrifuged at 3000 rpm for 10 min, using a Hettich Zentrifugen centrifuge, to remove the TiO2 prior to chemical quantification. The remaining alachlor in the solution was then quantified by LC. The LC system comprised a Spectra System P4000 gradient pump, a Spectra System AS3000 Autosampler with a 20 µL injection loop, a ThermoQuest Hypersil ODS (3 µm × 150 mm × 2.1 mm) column, and a Spectra System UV6000LP photodiode array UV detector. The maximum absorption wavelength (λmax) selected to detect the alachlor was 265 nm. The mobile phase consisted of 70% acetonitrile and 30% distilled-deionized water and was delivered at a flow rate of 0.2 mL/min. The TOC decay was determined using a Shimadzu TOC 5000A equipped with an ASI-5000A autosampler. The study of intermediates in the photocatalysis process was performed using a Finnigan LCQ DUO ion trap mass spectrometer with electrospray ionization (ESI), which was coupled to the LC. As suggested by another researcher,19 the spectrometer was operated in positive ionization mode to increase the detection sensitivity/ionization performance. Nitrogen generated by
Figure 2. Dark reaction, direct photolysis, and photocatalysis of alachlor. The UV wavelength used is narrow-band UV with peak at 300 nm.
the NITROX nitrogen generator was used both as a sheath and as an auxiliary gas. The capillary temperature was 200 °C, and the rates of the flow of sheath gas into the atmospheric pressure ionization region were 0.6 L/min. Helium purchased from Hong Kong Special Gas Co. Ltd. was used as the collision gas during the collision-induced dissociation process in the MS/MS analysis and as the damping gas in the mass analyzer. By direct infusion of alachlor into the mass spectrometer, the ion optics setting was automatically optimized using LCQ Tune plus software. Results and Discussion Direct Photolysis and Photocatalysis of Alachlor. The reaction of alachlor and TiO2 in the dark (i.e., without exposure to UV light) was examined. No significant variations in alachlor concentrations were observed after 30 min of mixing, indicating that no chemical or physical reaction was involved between the two chemicals, as shown in Figure 2. In addition, this also suggests that the adsorption (or loss) of alachlor onto the surface of the TiO2 particles can be neglected in this study. Both the direct photolysis (without TiO2) and photocatalysis (with TiO2) of alachlor by narrow band UV 300 nm were investigated and are shown in Figure 2. It was found that the direct photolysis and photocatalysis processes all followed pseudo-first-order kinetics. Direct photolysis without the use of TiO2 was apparently slow, due to the low molar absorptivity of alachlor at 300 nm and the lower energy of the UV irradiation. However, the presence of TiO2 can significantly improve the reaction rate by 12 times, indicating that alachlor is mainly oxidized by the hydroxyl radicals that are generated through the photocatalysis process. Mechanism of the Photocatalytic Decay of Alachlor. GC-MS analysis has been used a great deal in the intermediate studies; however, a solvent extraction and/or sample derivatization procedure was usually required, and the escaping of some polar intermediates from MS detection was inevitable. Therefore, the identification of intermediates in this study was achieved by using the most recent LC-MS to avoid the problem. Protonated alachlor and its reaction intermediates [M + H+] were monitored throughout the reaction. Under the current experimental setting, an intensity on the
Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5029
Figure 3. Proposed degradation mechanism of alachlor photocatalysis. Table 1. Summary of MS/MS Analysis Showing the Daughter Ions, Relative Abundances, Collision Energy Applied, and Proposed Fragments protonated parent ion [M + H+] 270 286 258 252 244 242 212 180 163
compd no.
[M + H+] of major daughter ions (relative abundances, %)
collision energy (eV)
proposed fragments for first daughter ion
proposed fragments for second daughter ion
alachlor 1 2 3 4 5 6 7 8
238 (100%) 268 (36%), 254 (100%), 242 (45%) 240 (30%), 226 (85%) 234 (6%), 220 (13%), 176 (100%) 226 (95%), 195 (10%) 210 (100%), 198 (46%) 180 (100%), 168 (46%) 163 (20%), 152 (50%), 134 (28%) 147 (100%), 134 (12%)
20 20 26 25 27 27 22 26 26
CH3OH H 2O H 2O H2O H 2O CH3OH CH3 + OH OH CH3
CH3OH CH3OH CH3OH ClCH2 CHOCH3 CH3 + CH2CH3 CH2CH3 CH2CH3
order of 105-106 relative abundance of ion peaks by an MS in full scan mode was achieved to detect the decay of alachlor and the generation/decay of reaction intermediates. A total of eight major intermediates were identified in the alachlor photocatalysis at a narrowband UV of 300 nm by interpreting and organizing the possible degradation mechanisms from analogous studies,15,20 the evolution profiles of different degradation products, and the results of double mass analysis (MS/ MS). The MS/MS was carried out to obtain more
proposed fragments for third daughter ion CHOCH3 HOCH2 + CH2OCH 3
OH + CH2CH3
information about the structure of the intermediates, which aided the identification of intermediates. Alachlor and the eight intermediates were subjected to an MS/ MS analysis. The results, including the mass of protonated ions ([M + H+]) of the daughter compounds, the relative abundances, the collision energy applied, and the proposed fragments, are summarized in Table 1, where the intermediates were presumably subjected to cleavage of the side chains attached to the nitrogen atom and the aromatic ring. The common fragmentation
5030 Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004
Figure 5. Photocatalytic decay of alachlor and TOC at initial pH levels of 3.1, 6.0, and 10.2.
Figure 4. Variations in protonated ion intensity of alachlor and the reaction intermediates during the alachlor photocatalysis, where the initial pH level is 6.
pattern includes the loss of the mass units 18, 28, 32, and 44, corresponding to the loss of H2O, CHCH3, CH3OH, and CHOCH3, respectively. Figure 3 illustrates the proposed degradation mechanism of the photocatalytic process, in which the decay pathways included dechlorination, hydroxylation, dealkylation, scission of the C-O bond, and N-dealkylation. The results have been compared to those from previous studies of analogous degradation mechanisms, such as alachlor photocatalysis,16 and of the photolysis/photocatalysis of a similar herbicide, propachlor,21 which verifies our findings. For specific intermediates, compounds 1 and 3 are considered to be the major primary intermediates in the process. For compound 1, hydroxyl free radicals react rapidly with unsaturated carbons,22 and hydroxyl aromatic products were often found to be the initial oxidation products for the degradation of aromatic compounds in free-radical-based (•OH) oxidation processes.23-24 In addition, as the aromatic ring is electron-rich in nature, the electrophilic attack of the aromatic ring of alachlor by hydroxyl radicals formed compound 1. The subsequent loss of an ethyl group on the aromatic rings of compound 1 formed compound 2. Dealkylation was initiated by hydrogen abstraction and is normally observed in hydroxyl radical reactions.22 The dechlorination and hydroxylation of alachlor formed compound 3, in which the photodechlorination has been suggested to be a major pathway in a UV photolysis process.25 The scission of the C-O bond of compound 2 formed compound 4. The de-ethylation of alachlor formed compound 5, and the subsequent scission of the C-O bond formed compound 6. The dechlorination and N-dealkylation (or scission of the N-C bond) of compounds 6 and 4 were observed; they generated compounds 8 and 7, respectively. An analogous N-dealkylation mechanism was previously reported in the photocatalytic degradation of propachlor,20 pendimethalin,26 and atrazine.27 Verification of the Proposed Degradation Pathways. The proposed mechanism was further verified by Figure 4, which shows the decay and generation of alachlor and of the eight identified reaction intermedi-
ates during the photocatalytic reaction in terms of protonated ion [M + H+] intensity. A study of the evolution profiles offers useful information to support the proposed mechanism. For instance, in Figure 4, the decay of alachlor leads to the generation of primary intermediates such as compounds 1 and 5, whereas the decay of these two primary intermediates generates secondary intermediates such as compounds 2 and 6, respectively. Compounds 7 and 8 are apparently resistant to attacks from free radicals, as their ion intensity is maintained at certain levels at an extended reaction time of 96.5 min. Decay of Alachlor and TOC at Different Initial pH Levels. TOC analysis was carried out along with the decay of alachlor at three different initial pH levels of 3.1, 6.0, and 10.2, and the results are shown in Figure 5. Both the decay of alachlor and TOC increased slightly with initial pH levels. The rate increment is likely due to the increase of hydroxyl ions in the solution, which was suggested to be critical in determining the concentration of hydroxyl radicals (HO•) in a UV photolysis process or an oxidation process.28 An increased amount of OH- at an elevated pH can enhance the generation of free radicals through photo-oxidation by holes, as indicated in eq 2. Since the hydroxyl free radical is the dominant oxidizing species in the photocatalytic process, the photodecay of alachlor is therefore accelerated with the initial pH. At an initial pH of 10.2, about 70% of the TOC was removed (or mineralized) in a reaction time of 130 min. The decay of TOC could be used to verify the conversions (decay/generation) of different molecular size species. In general, the decay of TOC could be divided into three stages. The first stage was the lag phase, where the reduction of TOC was insignificant because the dominant reaction was the decay of alachlor to primary intermediates. These primary intermediates carry high molecular weights, and the benzene rings remain intact. The lag phase was followed by a mild TOC decay, in which about 80% of the alachlor was decayed; the decay of high molecular weight intermediates (to lower ones) was initiated and became the dominant process. The final stage commenced at a reaction time of around 35 min [i.e., ln(t) ) 3.56], where over 90% of the alachlor was removed and the TOC dropped significantly, indicating the breaking of benzene rings. About 60-70% of the TOC was removed (or mineralized) in 130 min.
Ind. Eng. Chem. Res., Vol. 43, No. 17, 2004 5031
Conclusion Both the direct photolysis and photocatalysis of alachlor were found to follow first-order decay kinetics. Direct photolysis was a rather slow process, but the addition of TiO2 enhanced reaction rates by about 12 times using a narrow-band UV of 300 nm. Both the decay of alachlor and TOC increased slightly with initial pH levels. The use of LC-ESI-MS/MS enabled us to determine the polar and nonpolar intermediates directly and hence improve the chances of identifying intermediates. The possible reaction mechanisms associated with the identification of intermediates, end products, and degradation/generation pathways were proposed and verified via examination of the information on LC-MS, MS/MS, and TOC. The TOC analysis revealed the different degradation stages of the reaction. The findings of this study will certainly enrich our understanding about the degradation process of alachlor. Acknowledgment The work described in this paper was supported by a grant from the University Research Fund of The Hong Kong Polytechnic University (GW-045). Literature Cited (1) Ollis, D. F.; Pelizzetti, E.; Serpone, N. Destruction of water contaminants. Environ. Sci. Technol. 1991, 25, 1523. (2) Blount, M. C.; Kim, D. H.; Falconer, J. L. Transparent thinfilm TiO2 photocatalysts with high activity. Environ. Sci. Technol. 2001, 35, 2988. (3) Malato, S.; Ca´ceres, J.; Ferna´ndex-Alba, A. R.; Piedra, L.; Hernando, M. D.; Agu¨era, A.; Vial, J. Photocatalytic treatment of diuron by solar photocatalysis: Evaluation of main intermediates and toxicity. Environ. Sci. Technol. 2003, 37, 2516. (4) Gupta, H.; Tanaka, S. Photocatalytic mineralization of perchloroethylene using titanium dioxide. Water Sci. Technol. 1995, 31, 47. (5) Topalov, A.; Molna´r-Ga´bor, D.; Kosanic´, M.; Abramovic´, B. Photomineralization of the herbicide mecoprop dissolved in water sensitized by TiO2. Water Res. 2000, 34, 1473. (6) Bianco-Prevot, A.; Fabbri, D.; Pramauro, E.; Morales-Rubio, A.; Guardia, M. Continuous monitoring of photocatalytic treatments by flow injection. Degradation of dicamba in aqueous TiO2 dispersions. Chemosphere 2001, 44, 249. (7) Doong, R. A.; Maithreepala, R. A.; Chang, S. M. Heterogeneous and homogeneous photocatalytic degradation of chlorophenols in aqueous titanium dioxide and ferrous ion. Water Sci. Technol. 2000, 42, 253. (8) Jardim, W. F.; Moraes, S. G.; Takiyama, M. K. Photocatalytic degradation of aromatic chlorinated compounds using TiO2: toxicity of intermediates. Water Res. 1997, 31, 1728. (9) Chiron, S.; Fernandez-Alba, A.; Rodriguez, A.; Garcia-Calvo, E. Pesticide chemical oxidation: state-of-the-art. Water Res. 2000, 34, 366. (10) Lu, M. C.; Chen, J. N. Pretreatment of pesticide wastewater by photocatalytic oxidation. Water Sci. Technol. 1997, 36, 117. (11) Manilal, V. B.; Haridas, A.; Alexander, R.; Surender, G. D. Photocatalytic treatment of toxic organics in wastewaters toxicity of photodegradation products. Water Res. 1992, 26, 1035.
(12) Pen˜uela, G. A.; Barcelo´, D. Photosensitized degradation of organic pollutants in water: processes and analytical applications. Trend Anal. Chem. 1998, 17, 605. (13) Potter, T. L.; Carpenter, T. L. Occurrence of alachlor environmental degradation products in groundwater. Environ. Sci. Technol. 1995. 29, 1557. (14) Larson, S. J.; Gilliom, R. J.; Capel, P. D. Pesticides in streams of the United Statessinitial results from the National Water-Quality Assessment Program; U.S. Geological Survey WaterResources Investigations Report 98-4222; U.S. Geological Survey: Sacramento, California, 1999. (15) Moza, P. N.; Hustert, K.; Pal, S.; Sukul, P. Photocatalytic decomposition of pendimethalin and alachlor. Chemosphere 1992, 25, 1675. (16) Pen˜uela, G. A.; Barcelo´, D. Comparative degradation kinetics of alachlor in water by photocatalysis with FeCl3, TiO2 and photolysis, studied by solid-phase disk extraction followed by gas chromatographic techniques. J. Chromatogr. 1996, 754, 187. (17) Nargiello, M.; Herz, T. Physical-chemical characteristics of P-25 making it extremely suited as the catalyst in photodegradation of organic compounds. Proceedings of the 1st international conference on TiO2 photocatalytic purification and treatment of water and air; Elsevier Science: Amsterdam, 1960. (18) Wong, C. C.; Chu, W. The direct photolysis and photocatalytic degradation of alachlor at different TiO2 and UV sources. Chemosphere 2003, 50, 981. (19) Thurman, E. M.; Ferrer, I.; Barcelo´, D. Choosing between atmospheric pressure chemical ionization and electrospray ionization interfaces for the HPLC/MS analysis of pesticides. Anal. Chem. 2001, 73, 5441. (20) Konstantinou, I. K.; Sakka, V. A.; Albanis, T. A. Photocatalytic degradation of propachlor in aqueous TiO2 suspensions. Determination of the reaction pathway and identification of intermediate products by various analytical methods. Water Res. 2002, 36, 2733. (21) Konstantinou, I. K.; Zarkadis, A. K.; Albanis, T. A. Photodegradation of selected herbicides in various natural waters and soils under environmental conditions. J. Environ. Qual. 2001, 30, 121. (22) Draper, W. M.; Crosby, D. G. Solar Photooxidation of pesticides in dilute hydrogen peroxide. J. Agric. Food Chem. 1984, 32, 231. (23) Galindo, C.; Jacques, P.; Kalt, A. Photodegradation of the aminoazobenzene acid orange 52 by three advanced oxidation processes: UV/H2O2, UV/TiO2 and VIS/TiO2 comparative mechanistic and kinetic investigations. J. Photochem. Photobiol. 2000, 130, 35. (24) Wang, Y. B.; Hong, C. S. TiO2-mediated photomineralization of 2-chlorobiphenyl: the role of O2. Water Res. 2000, 34, 2791. (25) Chu, W. Photodechlorination mechanism of DDT in UV/ surfactant system. Environ. Sci. Technol. 1999, 33, 421. (26) Pandit, G. K.; Pal, S.; Das, A. K. Photocatalytic degradation of pendimethalin in the presence of titanium dioxide. J. Agric. Food Chem. 1995, 43, 171. (27) Pelizzetti, E.; Maurino, V.; Minero, C.; Carlin, V.; Pramauro, E.; Zerbinati, O.; Tosato, M. L. Photocatalytic degradation of atrazine and other s-triazine herbicides. Environ. Sci. Technol. 1990, 24, 1559. (28) Chu, W.; Ma, C. W. Quantitative prediction of direct and indirect dye ozonation kinetics. Water Res. 2000, 34, 3153.
Received for review November 4, 2003 Revised manuscript received April 21, 2004 Accepted May 23, 2004 IE0342356
