- Email: [email protected]
Remediation of s-triazines contaminated water in a laboratory scale apparatus using zero-valent iron powder
Chemosphere 41 (2000) 1835±1843
Remediation of s-triazines contaminated water in a laboratory scale apparatus using zero-valent iron powder Antoine Ghauch *, Joel Suptil ESIGEC-Laboratory of Chemistry and Environmental Engineering LCIE, University of Savoie, Scienti®c Campus, Savoie Technolac, 73376 le Bourget du Lac Cedex, France Received 28 February 2000; accepted 6 April 2000
Abstract Atrazine, propazine and simazine were tested separately and in mixture by batch procedure in a laboratory-constructed apparatus. 3.75 l of a buered s-triazines pesticide solution was treated at room temperature by 325-mesh zerovalent iron powder (ZVIP) (20 g/l). High performance liquid chromatography was used to separate by-products and study the decline in the pesticideÕs concentrations. Results obtained show that the order of degradation was simazine, atrazine and then propazine. The half-lives (t1=2 ) of the s-triazines pesticides are, respectively, 7.4, 9.0 and 10.6 min when they are treated separately, and 9.8, 11.2 and 13.7 min when they are treated together under the same conditions. The ®nal by-product obtained after 50 min of contact of simazine with ZVIP shows a shift to longer wavelength in its UV spectrum. A similar phenomenon is shown for atrazine and propazine. Identical primary by-products are produced and subsequently degraded to 4,6-(diamino)-s-triazine, which seems to be the major by-product of the reductive treatment process. Pathways for the degradation of the studied s-triazines by ZVIP are proposed. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Iron; Pesticides; Reduction; Treatment
1. Introduction The ubiquitous use of triazine pesticides in agriculture applications has resulted in extensive groundwater contamination. The triazines are a group of chemically similar herbicides including atrazine, cyanazine, propazine and simazine primarily used to control broadleaf weeds. Triazines are characterized by a symmetrical six-membered ring consisting of alternating carbon and nitrogen atoms. Atrazine currently is one of the two most widely used agriculture pesticides in the US and Europe (EPA, 1999). About three-fourths of ®eld corn
*
Tel.: +33-4-7975-8844; fax: +33-4-7975-8843. E-mail address: [email protected] (A. Ghauch).
and sorghum acres are treated with atrazine annually for weed control, which accounts for most of the several million pounds used per year. In 1994, for example, about 21±34 million pounds of cyanazine were applied annually (US), with about 85±90% used to control weeds on ®eld corn. After December 2002, cyanazine cannot be used. Roughly 5±7 million pounds of simazine are used per year. About one-third of that amount is applied to ®eld corn, one-third to orchard fruits and nuts, and the remainder is applied to noncrop sites such as lawns (EPA, 1999). Approximately, 20±40 thousand pounds of propazine are used per year on sorghum. The EPA has received and reviewed many studies on triazines, most which were required of the registrants as a condition of initial and continued registration. Based on the results of these studies, the Agency is concerned about potential cancer risks
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 1 3 3 - 8
1836
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
Fig. 1. Structure of the triazine compounds.
Fig. 2. Schematic representation of the cylindro-conical pilot treatment process. Capacity: 3.75 l, material: clear Poly-Vinyl Chloride (PVC), weight: 1 kg.
resulting from exposure to the triazines. Because of their persistent in water and mobile soil, triazines are among the most frequently detected pesticides in groundwater (EPA, 1999). In order to remediate pesticides from water, several researchers have studied elimination of these contaminants using dierent method of treatment. With the exception of granular activated carbon (Duguet et al., 1992) and oxidation processes (De Laat et al., 1995), neither of which is easily adaptable at low cost applications, relatively little attention has been focussed on reductive degradation processes (Sweeny, 1981; Vogel et al., 1987; Jeers et al., 1989; Matheson and Tratnyek, 1994; Roberts et al., 1996; Siantar et al., 1996) using zero-valent iron powder (ZVIP). Cao et al. (1999) have investigated the reducing degradation of azo dye by zero-valent iron in aqueous solution and have shown an eective treatment with half-lives of several minutes for all studied organic compounds.
Fig. 3. UV-Vis absorption spectra of simazine (1.25 mg/l, 22°C, phosphate buered solution pH 6.6) during treatment with acidic pretreated ZVIP (20 g/l). (1) 0 min; (2) 5 min; (3) 10 min; (4) 20 min; (5) 30 min.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
Recently, Ghauch et al. (1999) investigated the treatment of water contaminated with atrazine and parathion using ZVIP. They found that a rapid decreasing in pesticides concentration from 1 ppm (atrazine) and 0.1 ppm (parathion) to undetectable quantities requires 30 min. They also carried out reductive treatment of an industrial euent contaminated by pesticides without the identi®cation of by-products. More recently, Ghauch et al. (2000a) studied for the ®rst time the degradation of carbaryl insecticide (carbamate group) using ZVIP by solid surface room
1837
temperature phosphorescence and liquid chromatography (Ghauch, 2000; Ghauch et al., 2000b,c). They found that the half-life of carbaryl being close to several minutes, and the degradation device with acidic pre-treated ZVIP stills better than with unpretreated iron. The concentration of iron cations Fe3 in the treated solution was determined by diuse re¯ectance spectroscopy (Ghauch et al., 2000d). As a consequence of the previous highly encouraging tests, the present study was undertaken to evaluate the rate of degradation of the triazine pesticides (Fig. 1) in
Fig. 4. (a) S1 -HPLC chromatogram and UV absorption spectrum of simazine. Rt 4.9 min. Maximum absorption peak is at 223.8 nm. S2 -HPLC chromatogram and UV absorption spectra of by-product at 60 min of treatment with ZVIP (20 g/l). (b) A1 -HPLC chromatogram and UV absorption spectrum of atrazine. Rt 7.19 min. Maximum absorption peak is at 223.8 nm. A2 -HPLC chromatogram and UV absorption spectra of by-products at 60 min of treatment with ZVIP (20 g/l). (c) P1 -HPLC chromatogram and UV absorption spectrum of propazine. Rt 11.4 min. Maximum absorption peak is at 223.8 nm. P2 -HPLC chromatogram and UV absorption spectra of by-products at 60 min of treatment with ZVIP (20 g/l).
1838
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
the presence of ZVIP in water. Tests were conducted to con®rm the feasibility of the process, and to understand the mechanism of reduction device. Though not exhaustive, reduction pathways were examined to determine the breakdown products of the degradation process.
Table 1 The half-lives (t1=2 ) of simazine (1.25 mg/l), atrazine (1.25 mg/l) and propazine (1.25 mg/l) transformation using 20 g/l of ZVIP (325 mesh) in phosphate buered solution (pH 6.6) and at room temperature (22°C)a ;b
2. Experimental 2.1. Instrument A Hewlett Packard UV-VIS diode array spectrophotometer was used to record the evolution of the UV spectra throughout the experiments. HPLC analysis of pesticides was performed by a Diode Array Controller (Waters TM 996), equipped with a diode array detector, and supplemented with a programmable multiwavelength UV/VIS detector. Twenty ll samples taken from the reaction bottles were injected automatically via a C18 nonpolar column (Supelco, Discovery 250 4:6 mm, porosity 5 lm). The mobile phase was a mixture of acetonitrile/water (60:40, v/v). 2.2. Pilot A laboratory-constructed pilot (Fig. 2) was developed in order to study the degradation of the triazine pesticides using ZVIP as reductive agent. ZVIP was treated for 10 min with acidic solution (HCl 1M), washed several times with distilled water, and then manually introduced into the pilot. The processing pilot is equipped with a propeller located at middle height, allowing the aspiration of ZVIP through a 5-mm diameter tube, from the bottom to the top surface, for ZVIP to be propagated in the whole reactive zone. Once settled, the ZVIP is aspirated again continuously. The pilot was built to be used in batch, semi-batch, or in continuous mode.
Half-lives (min)
Simazine
Atrazine
Propazine
t1=2 c t1=2 d
7.41 0.39 9.82 0.49
9.04 0.48 11.21 0.56
10.62 0.54 13.71 0.68
a
Errors are 95% con®dence intervals. t1=2 was calculated by plotting (C/C0 ) versus (time of reaction). c Separately. d In mixture. b
2.3. Reagents Iron powder (purity > 99%, 325-mesh nitrogen ¯ushed) was purchased from Acros (Geel, Belgium), and KH2 PO4 and Na2 HPO4 .12H2 O from Prolabo (France). The triazines were obtained from Riedel-de-Haen (Germany) at the highest purity available. Acetonitrile was spectroscopic grade and purchased from Acros. Deionised water was used throughout the experiments. 2.4. Experimental procedures All experiments were carried out at room temperature 22°C 1°C in phosphate buered solutions (0.025 M; pH 6.6). The ZVIP was pretreated with 200 ml 1M HCl for 10 min, then washed with distilled water (400 ml) for four times to remove the residual HCl and Fe2 . Pesticide stock solutions were prepared in aqueous phosphate buered solution as follow: atrazine 20 mg/l, simazine 5 mg/l, propazine 5 mg/l. Working solutions were prepared by an appropriate dilution giving 1.25 mg/l for atrazine, simazine and propazine. 75 g of acid-washed ZVIP were manually added to the pilot
Fig. 5. Comparison of the s-triazines transformation when treated separately (a) and in a mixture (b) at the same concentration of ZVIP (20 g/l), 0.025 M phosphate buered deionised water (pH 6.6) at 22°C.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
1839
Fig. 6. HPLC chromatogram and UV absorption spectra of the triazine solutions treated with 20 g/l ZVIP in 0.025 M phosphate buered deionised water (pH 6.5) at 22°C. (a) SAP1 -siamzine (1.25 mg/l), atrazine (1.25 mg/l) and propazine (1.25 mg/l). Rt of the three pesticides are, respectively, 4.79, 7.07 and 11.4 min. Maximum absorption peaks are at 223.8 nm. (b) SAP2 -treated triazines solution at 30 min of contact with ZVIP. Appearance of all by-products, which represent a maximum absorbency at 223.8 nm, 228.5 nm, 214.5 nm and 219.2 nm, respectively. (c) SAP3 -treated triazines solution at 60 min of contact with ZVIP. All by-products present retention times between 2 and 4 min and a maximum of absorbency at 214.5, 219.2, 223.8 and 228.5 nm, respectively.
Table 2 The rate constant of simazine, atrazine and propazine transformation using 20 g/l of ZVIP (325 mesh) in phosphate buffered solution (pH 6.6) and at room temperature (22°C)a; b
a
Pesticides
Kobs c (minÿ1 )
Kobs d (minÿ1 )
Simazine Atrazine Propazine
0.093 0.004 0.077 0.004 0.065 0.003
0.070 0.003 0.061 0.003 0.050 0.002
Errors are 95% con®dence intervals. The observed transformation rate constant Kobs of the triazine pesticides were calculated by plotting Ln [% (C/C0 )] versus (time of reaction). c Separately. d In mixture.
containing 3.75 l of triazine solution. At intervals, 3 ml were taken with a syringe in order to record the UV-Vis spectra of the treated solution and carry out HPLC chromatography during the reductive process.
3. Results and discussion 3.1. Evolution of the UV absorption spectra of the triazine pesticides
b
The UV absorption spectra of the triazine solutions under our experimental conditions show a maximum at 223.8 nm. Fig. 3 shows the evolution of the UV spectra of simazine during the reducing degradation using
1840
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
ZVIP, with a shift in the maximum wavelength to 228 nm that correlates with the appearance of new byproducts. The same phenomenon was observed for the UV evolution spectra of atrazine and propazine, which are not presented in this paper. In order to understand the global reductive reaction, we proceeded to study the degradation of each of the triazine pesticides by using HPLC to separate by-products and to monitor the disappearance of the starting compounds and the appearance of new by-products. The estimated reductive degradation pathway presented at the end of this paper is based on the basic reductive reaction in organic chemistry coupled with information given by the retention time of by-products and their UVVis absorption spectra. 3.2. Evolution of the s-triazine pesticides chromatograms Fig. 4 shows chromatograms of simazine, atrazine and propazine at the same concentration (1.25 mg/l) before and after treatment with ZVIP. As can be seen, after 60 min of contact with ZVIP, a complete disappearance of the triazines is observed. For example, simazine, which has a retention time of 4.9 min, shows the appearance of four by-products. Three of them disappear completely at the end of the treatment. However the by-product exhibiting a retention time of 2.4 min and a maximum absorption band at 228.5 nm seems to be the ®nal by-product. (Fig. 4(a)). For atrazine (Rt 7.2 min), we observe the appearance of two by-products at Rt 2.4 and 3.3 min presenting a maximum peak at 228.5 nm (Fig. 4(b)). For propazine (Rt 11.4 min), we perceive the appearance of three byproducts at Rt 2.4, 3.2 and 3.74 min presenting maxima peaks at 228.5, 214.5 and 228.5 nm, respectively (Fig. 4(c)). Under aerobic conditions, the triazines were
reduced by ZVIP within a few minutes of half-lives (Fig. 5 and Table 1). The eect of dissolved oxygen (DO) is not studied here, but the consequence of the DO on the ®rst-order rate of the triazines transformation should be non-negligible. This phenomenon was studied by Siantar et al. (1996) who found that increasing the amount of DO from 0 to 41.6 mg/l (saturation concentration of O2 in water at 22°C) decreases linearly the pseudo-®rst-order rate reduction of 1,2-dibromo-3-chloropropane (DBCP) to propane from 0.28 0.03 to 0.07 0.02 min. They concluded that this observation may indicate that the DBCP competes with O2 for iron surface active sites and/or that the O2 deactivates the surface by forming non-reactive iron oxide coatings. In the O2 concentration range relevant for deionised water (under the conditions of the current experiment, 8 mg/l), the decrease in s-triazines transformation rate constant should match with that obtained for DBCP by a maximum of 17%. Increasing the s-triazines concentration in the same reactive media (Fig. 6) has a small eect on the constant degradation rate of the reductive treatment. Table 2 shows that Kobs decreased from 0:093 0:004 to 0:070 0:003 for simazine, 0:077 0:004 to 0:061 0:003 for atrazine and 0:065 0:003 to 0:050 0:002 for propazine when the s-triazines are treated together under the same conditions or when treated separately. The concentration evolution of all by-products when the reductive degradation treatment is carried out in a solution containing simazine (1.25 mg/l) or atrazine (1.25 mg/l) is presented in Fig. 7. (The evolution of the propazine by-products is not presented in this paper.) The number of the intermediate by-products pending from simazine and propazine is less than those from atrazine. This fact can be explained by the symmetrical position of the two ethyl±amino groups on the
Fig. 7. (a) Product distribution of simazine (1.25 mg/l) transformation into 100% 4,6-(amino)-s-triazine via products shown in the estimated simazine pathway map (see Fig. 8(a)). (b) Product distribution of atrazine (1.25 mg/l) transformation into 100% 4,6-(amino)s-triazine via products shown in the estimated atrazine pathway map (see Fig. 8(b)).
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
1841
Fig. 8. Reductive degradation pathway map for simazine (a) and for atrazine (b). The pathway map of propazine is similar. Chemical nomenclature: a-4,6-diethyl-amino-s-triazine; b-4-ethyl amino 6-deethyl amino simazine; c-4-deethyl amino 6-diethyl-amino simazine; d-4-ethyl amino-6-deethyl amino-s-triazine; e-4,6-deethyl-amino simazine; f-4-isopropyl amino 6-deethyl amino propazine; g-4-isopropyl amino 6-ethyl-amino propazine; h-4-isopropyl amino-6-deethyl amino-s-triazine; i-4-deisopropyl amino 6-ethyl amino-s-triazine; j-4,6-diamino-s-triazine.
simazine molecule and the two isopropyl±amino groups on the propazine molecule. On the other hand, atrazine has one ethyl±amino group and one isopropyl±amino group that makes possible the reductive dealkylation from both sides of the molecule. The appearance of 2-deethylamino or 4deethylamino, and 2-deisopropylamino or 4-deisopropylamino triazine, occurs by increasing the number of intermediate products. For all compounds studied, the by-product that appeared at Rt 2.4 min seems to be the major one. It corresponds to the reduced triazine pesticides obtained by the contact with ZVIP. One hypothesis can be envisaged and should correspond to the reductive dechlorination of the triazine ring followed by the reductive dealkylation of the secondary amines. This
can be explained by the elimination of the chlorine atom followed by the elimination of the alkyl group and the formation of the corresponding alkane. In order to fully understand the reductive pathway of the s-triazines pesticides, we investigated the potential presence of compounds recognized as natural by-products obtained by microorganism reductive degradation in soil. For example, comparisons made between a standard solution of hydroxyatrazine and a sample taken from the pilot at dierent reaction times shows an absence of hydroxyatrazine (HPLC chromatograms and UV-Vis spectra analyses). The later is also a by-product obtained when the oxidation process is used to remove atrazine from water (Duguet et al., 1992; De Laat et al. 1995). Added tests were carried out with the intention of
1842
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
studying the possibility of establishing breakdown products such as deisopropylatrazine or deethylatrazine, but results shows no evidence of these molecules. This fact can be explained by the rapid dechlorination of the s-triazines molecules followed by the reductive dealkylation reactions. For all studied s-triazines, we did not observe any byproduct that is obtained through an oxidation process. This can be explained by the fact that the chlorine atoms in the s-triazines pesticide molecules are reduced to hydrogen atoms. Thus, all the intermediate breakdown products do not have a chlorine atom in their structure. Fig. 8 shows the reduction pathway of simazine and atrazine by ZVIP. The reductive pathway of propazine is not presented but is similar to the simazine map by replacing the ethyl±amino group by the isopropyl±amino group.
4. Conclusions In this study we have established that ZVIP appears to be a very good metallic system (non-toxic substance and low-cost material) for the rapid treatment of water contaminated with simazine, atrazine and propazine. The half-lives of reductive reactions are close to several minutes. Simazine undergoes more rapidly than atrazine and propazine due to steric eects. The observed constant rate of degradation Kobs is slightly aected by the concentration of the molecules in the reactive media. Indeed, the existence of the propeller bringing back settled ZVIP in the bottom of the pilot to the surface by agitating it constantly, facilitates mass transport of the target molecules to the surface layer of ZVIP from the bulk of the solution. The major ®nal by-product is the 4,6-diamino-s-triazine. The eect of pH and DO has not been studied. The development of a laboratory-constructed cylindro-conical pilot plant equipped with a propeller to diuse ZVIP in the reacting media, suggests the expansion of this technology to in situ applications. Some problems concerning the settling of ZVIP was observed due to the cylindro-conical form. This has urged us to conceive a new conical apparatus that is under construction in the laboratory. Results on new pesticide treatment will be presented and discussed soon. However, ZVIP continues to receive the signi®cant attention and has begun to be an exciting technology which can be expanded to in situ implementation. Finally, since treatment of organochlorinated compounds, nitrates, Cr (VI), Uranium and Technetium by ZVIP has already been demonstrated, the ZVIP will be a potential key to aquifer restoration in the future by cleaning underground water of pesticides, and consequently by lowering their rate of toxicity.
Acknowledgements The author would like to thank Prof. Rima of the Lebanese University, Prof. Martin-Bouyer of the LCIE laboratory, and Dr. Gallet of the laboratory of Dynamics of Altitude Ecosystems, University of Savoie, for their helpful cooperation.
References Cao, J., Wei, L., Huang, Q., Wang, L., Han, S., 1999. Reducing degradation of azo-dye by zero-valent iron in aqueous solution. Chemosphere 38, 565±571. De Laat, J., Dore, M., Suty, H., 1995. Oxidation of s-triazines by advanced oxidation processes. By-products and kinetic rate constants. Rev. Sci. Eau 8 (1), 23±42. Duguet, J.P., Wable, O., Richard, Y., Toani, G., Dalga, N., 1992. Evaluation technico-economique de lÕelimination de lÕatrazine par le couplage ozone-peroxyde dÕhydrogen/charbon actif en grains sur la station de traitement du Mont Valerien. Water supply, Florence 10, 105±110. Environmental Protection Agency (EPA), Oce of Pesticide Programs August 1999. The triazine Pesticides: Atrazine, Cyanazine, Simazine, and Propazine. Ghauch, A., Rima, J., Amine, C., Martin-Bouyer, M., 1999. Rapid treatment of water contaminated with atrazine and parathion with zero-valent iron. Chemosphere 39, 1309± 1315. Ghauch, A., 2000. Determination of carbaryl and biphenyl through optical ®ber CCD-assisted ¯ash-lamp induced room temperature phosphorescence, Fresenius, J. Anal. Chem., in press. Ghauch, A., Gallet, C., Rima, J., Charef, A., Martin-Bouyer, M., 2000a. Use of zero-valent iron in reductive degradation treatment of water contaminated with carbaryl. Chemosphere, in press. Ghauch, A., Rima, J., Charef, A., Martin-Bouyer, M., 2000b. Use of an imaging spectrograph system with a ®ber optical sensor and a two-dimensional cooled CCD detector for solid surface room temperature phosphorescence of pesticides. Anal. Lett. 33 (4), 709±728. Ghauch, A., Rima, J., Fachinger, C., Suptil, J., MartinBouyer, M., 2000c. Room temperature phosphorescence analyses of polycyclic aromatic hydrocarbons using an imaging sensing system combined with a bifurcated optical ®ber and a cooled charge coupled device detector. Talanta 51, 807±816. Ghauch, A., Turnar, C., Martin-Bouyer, M., Rima, J., Charef, A., Fachinger, C., Suptil, J., 2000d. Use of diuse re¯ectance spectrometry in spot test reactions for quantitative determination of metals in water. Chemosphere 40, 101± 107. Jeers, P.M., Woytowitch, L.M., Wolfe, N.L., 1989. Homogeneous hydrolysis rate constants for selected chlorinated methanes, ethanes, ethenes and propanes. Environ. Sci. Technol. 23, 2198±2206. Matheson, L.J., Tratnyek, P.G., 1994. Reductive dehalogenation of chlorinated methanes by iron metal. Environ. Sci. Technol. 28, 2045±2053.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843 Roberts, A.L., Totten, L.A., Arnold, W.A., 1996. Reductive elimination of chlorinated ethylenes by zero-valent metals. Environ. Sci. Technol. 30, 2654±2659. Siantar, D., Schreier, C., Chou, C.S., Reinhard, M., 1996. Treatment of 1, 2-bibromo-3-chloropropane and nitratecontamined water with zero-valent iron or hydrogen/palladium catalysts. Water Res. 30, 2315±2322.
1843
Sweeny, K.H., 1981. The reductive treatment of industrial wastewater: II process applications. In: Bennett, G.G., (Ed.), American Institute of Chemical Engineers, Symposium Series, Water, 209, 77, 72±78. Vogel, T.M., Criddle, C.S., McCarty, P.L., 1987. Transformation of halogenated aliphatic compounds. Environ. Sci. Technol. 21, 722±736.
Remediation of s-triazines contaminated water in a laboratory scale apparatus using zero-valent iron powder Antoine Ghauch *, Joel Suptil ESIGEC-Laboratory of Chemistry and Environmental Engineering LCIE, University of Savoie, Scienti®c Campus, Savoie Technolac, 73376 le Bourget du Lac Cedex, France Received 28 February 2000; accepted 6 April 2000
Abstract Atrazine, propazine and simazine were tested separately and in mixture by batch procedure in a laboratory-constructed apparatus. 3.75 l of a buered s-triazines pesticide solution was treated at room temperature by 325-mesh zerovalent iron powder (ZVIP) (20 g/l). High performance liquid chromatography was used to separate by-products and study the decline in the pesticideÕs concentrations. Results obtained show that the order of degradation was simazine, atrazine and then propazine. The half-lives (t1=2 ) of the s-triazines pesticides are, respectively, 7.4, 9.0 and 10.6 min when they are treated separately, and 9.8, 11.2 and 13.7 min when they are treated together under the same conditions. The ®nal by-product obtained after 50 min of contact of simazine with ZVIP shows a shift to longer wavelength in its UV spectrum. A similar phenomenon is shown for atrazine and propazine. Identical primary by-products are produced and subsequently degraded to 4,6-(diamino)-s-triazine, which seems to be the major by-product of the reductive treatment process. Pathways for the degradation of the studied s-triazines by ZVIP are proposed. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Iron; Pesticides; Reduction; Treatment
1. Introduction The ubiquitous use of triazine pesticides in agriculture applications has resulted in extensive groundwater contamination. The triazines are a group of chemically similar herbicides including atrazine, cyanazine, propazine and simazine primarily used to control broadleaf weeds. Triazines are characterized by a symmetrical six-membered ring consisting of alternating carbon and nitrogen atoms. Atrazine currently is one of the two most widely used agriculture pesticides in the US and Europe (EPA, 1999). About three-fourths of ®eld corn
*
Tel.: +33-4-7975-8844; fax: +33-4-7975-8843. E-mail address: [email protected] (A. Ghauch).
and sorghum acres are treated with atrazine annually for weed control, which accounts for most of the several million pounds used per year. In 1994, for example, about 21±34 million pounds of cyanazine were applied annually (US), with about 85±90% used to control weeds on ®eld corn. After December 2002, cyanazine cannot be used. Roughly 5±7 million pounds of simazine are used per year. About one-third of that amount is applied to ®eld corn, one-third to orchard fruits and nuts, and the remainder is applied to noncrop sites such as lawns (EPA, 1999). Approximately, 20±40 thousand pounds of propazine are used per year on sorghum. The EPA has received and reviewed many studies on triazines, most which were required of the registrants as a condition of initial and continued registration. Based on the results of these studies, the Agency is concerned about potential cancer risks
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 1 3 3 - 8
1836
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
Fig. 1. Structure of the triazine compounds.
Fig. 2. Schematic representation of the cylindro-conical pilot treatment process. Capacity: 3.75 l, material: clear Poly-Vinyl Chloride (PVC), weight: 1 kg.
resulting from exposure to the triazines. Because of their persistent in water and mobile soil, triazines are among the most frequently detected pesticides in groundwater (EPA, 1999). In order to remediate pesticides from water, several researchers have studied elimination of these contaminants using dierent method of treatment. With the exception of granular activated carbon (Duguet et al., 1992) and oxidation processes (De Laat et al., 1995), neither of which is easily adaptable at low cost applications, relatively little attention has been focussed on reductive degradation processes (Sweeny, 1981; Vogel et al., 1987; Jeers et al., 1989; Matheson and Tratnyek, 1994; Roberts et al., 1996; Siantar et al., 1996) using zero-valent iron powder (ZVIP). Cao et al. (1999) have investigated the reducing degradation of azo dye by zero-valent iron in aqueous solution and have shown an eective treatment with half-lives of several minutes for all studied organic compounds.
Fig. 3. UV-Vis absorption spectra of simazine (1.25 mg/l, 22°C, phosphate buered solution pH 6.6) during treatment with acidic pretreated ZVIP (20 g/l). (1) 0 min; (2) 5 min; (3) 10 min; (4) 20 min; (5) 30 min.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
Recently, Ghauch et al. (1999) investigated the treatment of water contaminated with atrazine and parathion using ZVIP. They found that a rapid decreasing in pesticides concentration from 1 ppm (atrazine) and 0.1 ppm (parathion) to undetectable quantities requires 30 min. They also carried out reductive treatment of an industrial euent contaminated by pesticides without the identi®cation of by-products. More recently, Ghauch et al. (2000a) studied for the ®rst time the degradation of carbaryl insecticide (carbamate group) using ZVIP by solid surface room
1837
temperature phosphorescence and liquid chromatography (Ghauch, 2000; Ghauch et al., 2000b,c). They found that the half-life of carbaryl being close to several minutes, and the degradation device with acidic pre-treated ZVIP stills better than with unpretreated iron. The concentration of iron cations Fe3 in the treated solution was determined by diuse re¯ectance spectroscopy (Ghauch et al., 2000d). As a consequence of the previous highly encouraging tests, the present study was undertaken to evaluate the rate of degradation of the triazine pesticides (Fig. 1) in
Fig. 4. (a) S1 -HPLC chromatogram and UV absorption spectrum of simazine. Rt 4.9 min. Maximum absorption peak is at 223.8 nm. S2 -HPLC chromatogram and UV absorption spectra of by-product at 60 min of treatment with ZVIP (20 g/l). (b) A1 -HPLC chromatogram and UV absorption spectrum of atrazine. Rt 7.19 min. Maximum absorption peak is at 223.8 nm. A2 -HPLC chromatogram and UV absorption spectra of by-products at 60 min of treatment with ZVIP (20 g/l). (c) P1 -HPLC chromatogram and UV absorption spectrum of propazine. Rt 11.4 min. Maximum absorption peak is at 223.8 nm. P2 -HPLC chromatogram and UV absorption spectra of by-products at 60 min of treatment with ZVIP (20 g/l).
1838
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
the presence of ZVIP in water. Tests were conducted to con®rm the feasibility of the process, and to understand the mechanism of reduction device. Though not exhaustive, reduction pathways were examined to determine the breakdown products of the degradation process.
Table 1 The half-lives (t1=2 ) of simazine (1.25 mg/l), atrazine (1.25 mg/l) and propazine (1.25 mg/l) transformation using 20 g/l of ZVIP (325 mesh) in phosphate buered solution (pH 6.6) and at room temperature (22°C)a ;b
2. Experimental 2.1. Instrument A Hewlett Packard UV-VIS diode array spectrophotometer was used to record the evolution of the UV spectra throughout the experiments. HPLC analysis of pesticides was performed by a Diode Array Controller (Waters TM 996), equipped with a diode array detector, and supplemented with a programmable multiwavelength UV/VIS detector. Twenty ll samples taken from the reaction bottles were injected automatically via a C18 nonpolar column (Supelco, Discovery 250 4:6 mm, porosity 5 lm). The mobile phase was a mixture of acetonitrile/water (60:40, v/v). 2.2. Pilot A laboratory-constructed pilot (Fig. 2) was developed in order to study the degradation of the triazine pesticides using ZVIP as reductive agent. ZVIP was treated for 10 min with acidic solution (HCl 1M), washed several times with distilled water, and then manually introduced into the pilot. The processing pilot is equipped with a propeller located at middle height, allowing the aspiration of ZVIP through a 5-mm diameter tube, from the bottom to the top surface, for ZVIP to be propagated in the whole reactive zone. Once settled, the ZVIP is aspirated again continuously. The pilot was built to be used in batch, semi-batch, or in continuous mode.
Half-lives (min)
Simazine
Atrazine
Propazine
t1=2 c t1=2 d
7.41 0.39 9.82 0.49
9.04 0.48 11.21 0.56
10.62 0.54 13.71 0.68
a
Errors are 95% con®dence intervals. t1=2 was calculated by plotting (C/C0 ) versus (time of reaction). c Separately. d In mixture. b
2.3. Reagents Iron powder (purity > 99%, 325-mesh nitrogen ¯ushed) was purchased from Acros (Geel, Belgium), and KH2 PO4 and Na2 HPO4 .12H2 O from Prolabo (France). The triazines were obtained from Riedel-de-Haen (Germany) at the highest purity available. Acetonitrile was spectroscopic grade and purchased from Acros. Deionised water was used throughout the experiments. 2.4. Experimental procedures All experiments were carried out at room temperature 22°C 1°C in phosphate buered solutions (0.025 M; pH 6.6). The ZVIP was pretreated with 200 ml 1M HCl for 10 min, then washed with distilled water (400 ml) for four times to remove the residual HCl and Fe2 . Pesticide stock solutions were prepared in aqueous phosphate buered solution as follow: atrazine 20 mg/l, simazine 5 mg/l, propazine 5 mg/l. Working solutions were prepared by an appropriate dilution giving 1.25 mg/l for atrazine, simazine and propazine. 75 g of acid-washed ZVIP were manually added to the pilot
Fig. 5. Comparison of the s-triazines transformation when treated separately (a) and in a mixture (b) at the same concentration of ZVIP (20 g/l), 0.025 M phosphate buered deionised water (pH 6.6) at 22°C.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
1839
Fig. 6. HPLC chromatogram and UV absorption spectra of the triazine solutions treated with 20 g/l ZVIP in 0.025 M phosphate buered deionised water (pH 6.5) at 22°C. (a) SAP1 -siamzine (1.25 mg/l), atrazine (1.25 mg/l) and propazine (1.25 mg/l). Rt of the three pesticides are, respectively, 4.79, 7.07 and 11.4 min. Maximum absorption peaks are at 223.8 nm. (b) SAP2 -treated triazines solution at 30 min of contact with ZVIP. Appearance of all by-products, which represent a maximum absorbency at 223.8 nm, 228.5 nm, 214.5 nm and 219.2 nm, respectively. (c) SAP3 -treated triazines solution at 60 min of contact with ZVIP. All by-products present retention times between 2 and 4 min and a maximum of absorbency at 214.5, 219.2, 223.8 and 228.5 nm, respectively.
Table 2 The rate constant of simazine, atrazine and propazine transformation using 20 g/l of ZVIP (325 mesh) in phosphate buffered solution (pH 6.6) and at room temperature (22°C)a; b
a
Pesticides
Kobs c (minÿ1 )
Kobs d (minÿ1 )
Simazine Atrazine Propazine
0.093 0.004 0.077 0.004 0.065 0.003
0.070 0.003 0.061 0.003 0.050 0.002
Errors are 95% con®dence intervals. The observed transformation rate constant Kobs of the triazine pesticides were calculated by plotting Ln [% (C/C0 )] versus (time of reaction). c Separately. d In mixture.
containing 3.75 l of triazine solution. At intervals, 3 ml were taken with a syringe in order to record the UV-Vis spectra of the treated solution and carry out HPLC chromatography during the reductive process.
3. Results and discussion 3.1. Evolution of the UV absorption spectra of the triazine pesticides
b
The UV absorption spectra of the triazine solutions under our experimental conditions show a maximum at 223.8 nm. Fig. 3 shows the evolution of the UV spectra of simazine during the reducing degradation using
1840
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
ZVIP, with a shift in the maximum wavelength to 228 nm that correlates with the appearance of new byproducts. The same phenomenon was observed for the UV evolution spectra of atrazine and propazine, which are not presented in this paper. In order to understand the global reductive reaction, we proceeded to study the degradation of each of the triazine pesticides by using HPLC to separate by-products and to monitor the disappearance of the starting compounds and the appearance of new by-products. The estimated reductive degradation pathway presented at the end of this paper is based on the basic reductive reaction in organic chemistry coupled with information given by the retention time of by-products and their UVVis absorption spectra. 3.2. Evolution of the s-triazine pesticides chromatograms Fig. 4 shows chromatograms of simazine, atrazine and propazine at the same concentration (1.25 mg/l) before and after treatment with ZVIP. As can be seen, after 60 min of contact with ZVIP, a complete disappearance of the triazines is observed. For example, simazine, which has a retention time of 4.9 min, shows the appearance of four by-products. Three of them disappear completely at the end of the treatment. However the by-product exhibiting a retention time of 2.4 min and a maximum absorption band at 228.5 nm seems to be the ®nal by-product. (Fig. 4(a)). For atrazine (Rt 7.2 min), we observe the appearance of two by-products at Rt 2.4 and 3.3 min presenting a maximum peak at 228.5 nm (Fig. 4(b)). For propazine (Rt 11.4 min), we perceive the appearance of three byproducts at Rt 2.4, 3.2 and 3.74 min presenting maxima peaks at 228.5, 214.5 and 228.5 nm, respectively (Fig. 4(c)). Under aerobic conditions, the triazines were
reduced by ZVIP within a few minutes of half-lives (Fig. 5 and Table 1). The eect of dissolved oxygen (DO) is not studied here, but the consequence of the DO on the ®rst-order rate of the triazines transformation should be non-negligible. This phenomenon was studied by Siantar et al. (1996) who found that increasing the amount of DO from 0 to 41.6 mg/l (saturation concentration of O2 in water at 22°C) decreases linearly the pseudo-®rst-order rate reduction of 1,2-dibromo-3-chloropropane (DBCP) to propane from 0.28 0.03 to 0.07 0.02 min. They concluded that this observation may indicate that the DBCP competes with O2 for iron surface active sites and/or that the O2 deactivates the surface by forming non-reactive iron oxide coatings. In the O2 concentration range relevant for deionised water (under the conditions of the current experiment, 8 mg/l), the decrease in s-triazines transformation rate constant should match with that obtained for DBCP by a maximum of 17%. Increasing the s-triazines concentration in the same reactive media (Fig. 6) has a small eect on the constant degradation rate of the reductive treatment. Table 2 shows that Kobs decreased from 0:093 0:004 to 0:070 0:003 for simazine, 0:077 0:004 to 0:061 0:003 for atrazine and 0:065 0:003 to 0:050 0:002 for propazine when the s-triazines are treated together under the same conditions or when treated separately. The concentration evolution of all by-products when the reductive degradation treatment is carried out in a solution containing simazine (1.25 mg/l) or atrazine (1.25 mg/l) is presented in Fig. 7. (The evolution of the propazine by-products is not presented in this paper.) The number of the intermediate by-products pending from simazine and propazine is less than those from atrazine. This fact can be explained by the symmetrical position of the two ethyl±amino groups on the
Fig. 7. (a) Product distribution of simazine (1.25 mg/l) transformation into 100% 4,6-(amino)-s-triazine via products shown in the estimated simazine pathway map (see Fig. 8(a)). (b) Product distribution of atrazine (1.25 mg/l) transformation into 100% 4,6-(amino)s-triazine via products shown in the estimated atrazine pathway map (see Fig. 8(b)).
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
1841
Fig. 8. Reductive degradation pathway map for simazine (a) and for atrazine (b). The pathway map of propazine is similar. Chemical nomenclature: a-4,6-diethyl-amino-s-triazine; b-4-ethyl amino 6-deethyl amino simazine; c-4-deethyl amino 6-diethyl-amino simazine; d-4-ethyl amino-6-deethyl amino-s-triazine; e-4,6-deethyl-amino simazine; f-4-isopropyl amino 6-deethyl amino propazine; g-4-isopropyl amino 6-ethyl-amino propazine; h-4-isopropyl amino-6-deethyl amino-s-triazine; i-4-deisopropyl amino 6-ethyl amino-s-triazine; j-4,6-diamino-s-triazine.
simazine molecule and the two isopropyl±amino groups on the propazine molecule. On the other hand, atrazine has one ethyl±amino group and one isopropyl±amino group that makes possible the reductive dealkylation from both sides of the molecule. The appearance of 2-deethylamino or 4deethylamino, and 2-deisopropylamino or 4-deisopropylamino triazine, occurs by increasing the number of intermediate products. For all compounds studied, the by-product that appeared at Rt 2.4 min seems to be the major one. It corresponds to the reduced triazine pesticides obtained by the contact with ZVIP. One hypothesis can be envisaged and should correspond to the reductive dechlorination of the triazine ring followed by the reductive dealkylation of the secondary amines. This
can be explained by the elimination of the chlorine atom followed by the elimination of the alkyl group and the formation of the corresponding alkane. In order to fully understand the reductive pathway of the s-triazines pesticides, we investigated the potential presence of compounds recognized as natural by-products obtained by microorganism reductive degradation in soil. For example, comparisons made between a standard solution of hydroxyatrazine and a sample taken from the pilot at dierent reaction times shows an absence of hydroxyatrazine (HPLC chromatograms and UV-Vis spectra analyses). The later is also a by-product obtained when the oxidation process is used to remove atrazine from water (Duguet et al., 1992; De Laat et al. 1995). Added tests were carried out with the intention of
1842
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843
studying the possibility of establishing breakdown products such as deisopropylatrazine or deethylatrazine, but results shows no evidence of these molecules. This fact can be explained by the rapid dechlorination of the s-triazines molecules followed by the reductive dealkylation reactions. For all studied s-triazines, we did not observe any byproduct that is obtained through an oxidation process. This can be explained by the fact that the chlorine atoms in the s-triazines pesticide molecules are reduced to hydrogen atoms. Thus, all the intermediate breakdown products do not have a chlorine atom in their structure. Fig. 8 shows the reduction pathway of simazine and atrazine by ZVIP. The reductive pathway of propazine is not presented but is similar to the simazine map by replacing the ethyl±amino group by the isopropyl±amino group.
4. Conclusions In this study we have established that ZVIP appears to be a very good metallic system (non-toxic substance and low-cost material) for the rapid treatment of water contaminated with simazine, atrazine and propazine. The half-lives of reductive reactions are close to several minutes. Simazine undergoes more rapidly than atrazine and propazine due to steric eects. The observed constant rate of degradation Kobs is slightly aected by the concentration of the molecules in the reactive media. Indeed, the existence of the propeller bringing back settled ZVIP in the bottom of the pilot to the surface by agitating it constantly, facilitates mass transport of the target molecules to the surface layer of ZVIP from the bulk of the solution. The major ®nal by-product is the 4,6-diamino-s-triazine. The eect of pH and DO has not been studied. The development of a laboratory-constructed cylindro-conical pilot plant equipped with a propeller to diuse ZVIP in the reacting media, suggests the expansion of this technology to in situ applications. Some problems concerning the settling of ZVIP was observed due to the cylindro-conical form. This has urged us to conceive a new conical apparatus that is under construction in the laboratory. Results on new pesticide treatment will be presented and discussed soon. However, ZVIP continues to receive the signi®cant attention and has begun to be an exciting technology which can be expanded to in situ implementation. Finally, since treatment of organochlorinated compounds, nitrates, Cr (VI), Uranium and Technetium by ZVIP has already been demonstrated, the ZVIP will be a potential key to aquifer restoration in the future by cleaning underground water of pesticides, and consequently by lowering their rate of toxicity.
Acknowledgements The author would like to thank Prof. Rima of the Lebanese University, Prof. Martin-Bouyer of the LCIE laboratory, and Dr. Gallet of the laboratory of Dynamics of Altitude Ecosystems, University of Savoie, for their helpful cooperation.
References Cao, J., Wei, L., Huang, Q., Wang, L., Han, S., 1999. Reducing degradation of azo-dye by zero-valent iron in aqueous solution. Chemosphere 38, 565±571. De Laat, J., Dore, M., Suty, H., 1995. Oxidation of s-triazines by advanced oxidation processes. By-products and kinetic rate constants. Rev. Sci. Eau 8 (1), 23±42. Duguet, J.P., Wable, O., Richard, Y., Toani, G., Dalga, N., 1992. Evaluation technico-economique de lÕelimination de lÕatrazine par le couplage ozone-peroxyde dÕhydrogen/charbon actif en grains sur la station de traitement du Mont Valerien. Water supply, Florence 10, 105±110. Environmental Protection Agency (EPA), Oce of Pesticide Programs August 1999. The triazine Pesticides: Atrazine, Cyanazine, Simazine, and Propazine. Ghauch, A., Rima, J., Amine, C., Martin-Bouyer, M., 1999. Rapid treatment of water contaminated with atrazine and parathion with zero-valent iron. Chemosphere 39, 1309± 1315. Ghauch, A., 2000. Determination of carbaryl and biphenyl through optical ®ber CCD-assisted ¯ash-lamp induced room temperature phosphorescence, Fresenius, J. Anal. Chem., in press. Ghauch, A., Gallet, C., Rima, J., Charef, A., Martin-Bouyer, M., 2000a. Use of zero-valent iron in reductive degradation treatment of water contaminated with carbaryl. Chemosphere, in press. Ghauch, A., Rima, J., Charef, A., Martin-Bouyer, M., 2000b. Use of an imaging spectrograph system with a ®ber optical sensor and a two-dimensional cooled CCD detector for solid surface room temperature phosphorescence of pesticides. Anal. Lett. 33 (4), 709±728. Ghauch, A., Rima, J., Fachinger, C., Suptil, J., MartinBouyer, M., 2000c. Room temperature phosphorescence analyses of polycyclic aromatic hydrocarbons using an imaging sensing system combined with a bifurcated optical ®ber and a cooled charge coupled device detector. Talanta 51, 807±816. Ghauch, A., Turnar, C., Martin-Bouyer, M., Rima, J., Charef, A., Fachinger, C., Suptil, J., 2000d. Use of diuse re¯ectance spectrometry in spot test reactions for quantitative determination of metals in water. Chemosphere 40, 101± 107. Jeers, P.M., Woytowitch, L.M., Wolfe, N.L., 1989. Homogeneous hydrolysis rate constants for selected chlorinated methanes, ethanes, ethenes and propanes. Environ. Sci. Technol. 23, 2198±2206. Matheson, L.J., Tratnyek, P.G., 1994. Reductive dehalogenation of chlorinated methanes by iron metal. Environ. Sci. Technol. 28, 2045±2053.
A. Ghauch, J. Suptil / Chemosphere 41 (2000) 1835±1843 Roberts, A.L., Totten, L.A., Arnold, W.A., 1996. Reductive elimination of chlorinated ethylenes by zero-valent metals. Environ. Sci. Technol. 30, 2654±2659. Siantar, D., Schreier, C., Chou, C.S., Reinhard, M., 1996. Treatment of 1, 2-bibromo-3-chloropropane and nitratecontamined water with zero-valent iron or hydrogen/palladium catalysts. Water Res. 30, 2315±2322.
1843
Sweeny, K.H., 1981. The reductive treatment of industrial wastewater: II process applications. In: Bennett, G.G., (Ed.), American Institute of Chemical Engineers, Symposium Series, Water, 209, 77, 72±78. Vogel, T.M., Criddle, C.S., McCarty, P.L., 1987. Transformation of halogenated aliphatic compounds. Environ. Sci. Technol. 21, 722±736.