- Email: [email protected]
Phototransformation of methabenthiazuron in the presence of nitrate and nitrite ions
Chemosphere 60 (2005) 1523–1529 www.elsevier.com/locate/chemosphere
Phototransformation of methabenthiazuron in the presence of nitrate and nitrite ions Moulay A. Malouki, Bernadette Lave´drine, Claire Richard
*
Laboratoire de Photochimie Mole´culaire et Macromole´culaire, UMR n 6505 CNRS, Universite´ Blaise Pascal, F-63177 Aubie`re Cedex, France Received 10 December 2004; received in revised form 16 February 2005; accepted 20 February 2005 Available online 18 April 2005
Abstract The influence of nitrate and nitrite ions on the degradation of methabenzthiazuron upon irradiation using artificial solar light has been investigated. The rate of degradation of methabenzthiazuron (1 lM) was accelerated by NO 3 (0.1 mM) by a factor of 10. The irradiation of methabenzthiazuron (0.1 mM) in the presence of NO 3 (1 mM) or NO 2 (0.1 mM) yielded numerous intermediary photoproducts. Mineralization was achieved after prolonged exposure. Some were identified with the help of LC–ESI–MS and flow injection APCI–MS techniques. Both oxidations of the aromatic ring and of the urea chain were observed. The former started by hydroxylation of the ring. Further oxidation of the ring led to cleavage of the benzenic ring with formation of dialdehydic, diacidic and anhydric compounds. Complete removal of the lateral urea chain took place subsequently to demethylation of the terminal methyl group and loss of the CO–NH2 group. Nitration was a minor process. This work shows that the photodegradation of methabenzthiazuron in the presence of nitrate or nitrite ions is highly non-specific. 2005 Elsevier Ltd. All rights reserved. Keywords: Photoinduction; Aquatic environment; Herbicide; Photoproducts; Oxidation; Mineralization
1. Introduction Methabenzthiazuron (N-(2-benzothiazolyl)-N,N 0 dimethylurea) (MBTU) is an active ingredient of Tribunil (Bayer, France) and Ormet (Phytorus S.A., France). Methabenzthiazuron belongs to the group of urea herbicides, which is used to control a broad spectrum of broad-leaved weeds and grasses in winter corn, spring wheat, grass seed and in nurseries. It is a selective
*
Corresponding author. Tel.: +33 4 73 40 71 42; fax: +33 4 73 40 77 00. E-mail address: [email protected] (C. Richard).
herbicide that is primarily absorbed through the roots. Methabenzthiazuron inhibits photosynthetic electron transport. Its bioactivity persists in the field soil for more than one growing season. Bioconcentration factor is about 10.1 for the pH range 7–10. It is toxic for fish and daphnia: LC50 is 15.9 mg/l (96 h) for rainbow trout and 30.6 mg/l (48 h) for daphnia magna. It is classified as toxic for the environment and very toxic for aquatic organisms (Tomlin, 2000). MBTU is soluble in water at 59 mg l1 (20 C within the pH range 7–10) and may be transferred into the aquatic environment by leaching. Screening surveys in France showed that concentrations of MBTU up to 1 lM may be reached in surface waters close to agriculture areas. MBTU absorbs UVB radiations (Fig. 1);
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.02.080
1524
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529 14
100
3.0x10
14
1.5x10
MBTU
14
1.0x10
5000
ε (M-1. cm-1)
-1 -1
14
wavelength (nm)
photons cm nm s
-1
14
2.0x10
270 280 290 300 310 320 330 340 350 360
10000
-2
2.5x10
-1
extinction coefficient (M cm )
15000
50
13
5.0x10 0.0
0 300
350
400
450
0 200
wavelength (nm)
Fig. 1. UV–visible spectrum of MBTU and ( ) solar flux reaching the earth surface in summer; insert: emission spectrum of sunlamps.
however, it was found to be very slowly photolyzed in solutions when irradiated above 290 nm (Sakriss et al., 1976; Malouki et al., 2003). In contrast 6-hydroxymethabenzthiazuron, its main biodegradation metabolite, undergoes photooxidation in the same exposure conditions (Malouki et al., 2003). As an alternative to direct photolysis, MBTU may be degraded via photoinduced reactions mediated by chromophores present in continental surface waters. Nitrate ions are usually present in the aquatic environment at concentrations depending strongly on the geographic location and on the agricultural activities of the surrounding areas. Concentrations up to 103 M may be reached. Nitrite ions that are produced by photolysis of nitrate ions and by photodegradation of aquatic humic substances (Kieber et al., 1999) are present too, but at lower concentration. Nitrate shows a maximum of absorption at 304 nm (e 7 M1 cm1) (Wagner et al., 1980), while nitrite at 355 nm (e 22 M1 cm1) (Strickler and Kasha, 1963) (Fig. 2). Both absorb solar radiation in the actinic spectrum and can undergo chemical reactions. It is well established that under excitation nitrite ions yields hydroxyl radicals and nitrogen monoxide according to NO þ hm ! NO þ O ð1Þ 2
O
þ H2 O ! OH þ OH
ð2Þ
The quantum yield of process (1) is 0.025 at 355 nm (Zafiriou and Bonneau, 1987). The photolysis of nitrate ions is more complex since two primary processes take place: NO 3 þ hm ! NO2 þ O
NO 3
þ hm ! NO2 þ O
ð3Þ ð4Þ
The quantum yield of process (3) was evaluated to be 1.1 · 103 and the one of process (4) to 9.2 · 103 (War-
250
300
350
400
450
wavelength (nm)
Fig. 2. UV–visible spectra of KNO3 (–––), NaNO2 (——) and HNO2 ( ) in water.
neck and Wurzinger, 1988). The photolysis of nitrite and nitrate ions in the aquatic environment cannot be therefore neglected. Moreover, as sources of hydroxyl radicals, nitrate and nitrite ions are able to photoinduce the phototransformation of organic pollutants (Zepp et al., 1987; Machado and Boule, 1994). The oxidations are mainly attributed to hydroxyl radicals that are strong oxidants, atomic oxygen being not as reactive. Nitration and nitrosation reactions were reported too. The reactions are concentration dependent due to the following processes that may occur too: NO þ NO ! N O ð5Þ 2
2
2
4
þ N2 O4 þ H2 O ! NO 2 þ NO3 þ 2H
ð6Þ
NO 2 þ OH ! NO2 þ OH
8
ð7Þ
1 1
3 1
with k5 = 4.5 · 10 M s and k6 = 10 s (Graetzel et al., 1969) and k7 = 1010 M1 s1 (Buxton et al., 1988). In this report, we describe our research on the photolysis of methabenzthiazuron in the presence of nitrate and nitrite ions. In a first step, we investigated the ability of nitrate (0.1 mM) to photoinduce the degradation of methabenzthiazuron (1 lM), i.e., at levels relevant to environmental conditions. Then, in order to identify intermediary photoproducts, we performed an analytical study using higher concentrations of reactants. Several photoproducts were characterized. This helps to get a better insight into the mechanism of degradation from the first steps of oxidation to complete mineralization. 4
CH3
N
5 6
S 7
2
CH3
N C N O
Structure of MBTU
H
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
2. Materials and methods 2.1. Materials Methabenthiazuron (MBTU) was purchased from Riedel-de-Hae¨n (Pestanal). Authentic sample of 6-OHMBTU was obtained by biodegradation of MBTU as previously described (Malouki et al., 2003). Nitrate and nitrite ions were Fluka products (99.0% purity grade). Water was purified using a Milli-Q device (Millipore). 2.2. Irradiations Polychromatic irradiations were performed using a device equipped with six fluorescent Duke sunlamp GL 20 tubes (SNEE, Aubervilliers, France) emitting within 275–350 nm with a maximum of emission at 313 nm (insert of Fig. 2). The reactor was cylindrical and made in Pyrex glass in order to cut off the wavelengths shorter than 290 nm. In these conditions, samples received radiations in the range 290–350 nm only. MBTU (1 lM or 0.1 mM) was dissolved in Milli-Q water or in aqueous solutions containing KNO3 (0.1 mM or 1 mM), NaNO2 or HNO2 (0.1 mM). Solutions were acidified with HClO4. These solutions were poured into the Pyrex glass reactor and irradiated using the sunlamps. Aliquots of 0.5 ml were removed from irradiated solutions at selected intervals and analysed by HPLC. Dark control experiments showed no loss of compounds by adsorption on glass surfaces or by spontaneous transformation on the time scale of irradiation experiments, irrespective of the aqueous media used (pure water, nitrate, nitrate or nitrous acid solutions).
1525
100% reached at t = 20 min. The electrospray interface (ESI) operated in positive mode. Nitrogen was used as shealth gas (13 l/min, 350 C) and the spray voltage was 4 kV. The pressure of nebulisation was 4 bar. The scan range was 100–500 uma. Another set of analyses was performed on a Hewlett–Packard 5989 MS apparatus equipped with a HP G1075A APCI interface. Methanolic irradiated solutions were injected in the mobile phase using the flow injection technique. The following optimised conditions were used: APCI vaporiser temperature 350 C; capillary temperature 150 C; source voltage 6 kV; capillary voltage 25 V; source current 5 lA; sheath gas flow (N2) 80 arbitrary units. Total organic carbon (TOC) measurements in the liquid phase were carried out with a TOC analyser Schimadzu model TOC-5O5OA. The calibration curves within the range 0–30 mg l1 were obtained by using potassium hydrogen phthalate and sodium hydrogen carbonate for organic and inorganic carbon, respectively.
3. Results 3.1. Kinetic of MBTU photodegradation in the presence of nitrite and nitrate ions We first investigated the photodegradation of MBTU (1 lM) in pure water and in the presence of NO 3 (0.1 mM). These concentrations were chosen because they were relevant of environmental conditions. Results are presented in Fig. 3. MBTU was found to be photolyzed much more slowly in Milli-Q water than in the presence of nitrate. The rates could be treated in
2.3. Analyses 1.0 0.8 0.6 0.0
0.4
-0.2
ln(c/c0)
c/c0
Losses of MBTU were monitored by HPLC–UV using a Waters apparatus equipped with two pumps (model 510), an auto sampler, a photodiode array detector (model 996) and a C18 reversed-phase (4.6 mm · 250 mm, 5 lm) Spherisorb S5 ODS2 Waters column. Eluents were mixtures of water and methanol (40/60, v/v) at a constant flow of 1 ml min1. UV spectra were recorded on a Cary 3 (Varian, Les Ulis, France) spectrophotometer. A 1-cm path quartz cell was used in all the experiments. The reference beam blank was always Milli-Q water. Analyses of MBTU and photoproducts were performed by HPLC–MS using a Hewlett–Packard 1100-MSD apparatus equipped with a Ultisphere HDO C18 (3 mm · 100 mm, 3 lm) column. A mixture of water with formic acid (pH = 2.9) and methanol was used as mobile phase at a constant flow of 0.4 ml min1. The gradient was the following one: 70% of methanol at t = 0 and linear increase until
-0.4
-0.6
0.2
-0.8 0
10
20
30
40
50
time (hour)
0.0 0
10
20
30
40
50
time (hour) Fig. 3. Consumption profiles of MBTU (1 lM) irradiated using sunlamps: (s) in pure water, (d) in presence of NO 3 (0.1 mM); insert: plot of ln(c/c0) versus t, where c represents the MBTU concentration at time t and c0 the MBTU concentration at time 0.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
19.930
16 - 20 mAU 11.613
35
19.506 20.215
15
18.688
14
15.460
11.081
9 12.126
7.700
7 8.874 8.999 9.335 9.570 9.957 10.310
6.604
5
1 4.997
10
1.633
15
12.960 13.390 13.608 13.935 14.184
3
20
17.541 17 .706 18.020 18.355
25
17.340
30
0.927
first-order form (see insert of Fig. 3). We got ln(c/ c0) = k.t, where c represents the MBTU concentration at time t, c0 the MBTU concentration at time 0, and k the apparent first order rate constant of MBTU consumption. In pure water, we computed k = 0.012 ± 0.001 with R2 = 0.977 and in the presence of nitrate k = 0.12 ± 0.01 with R2 = 0.998. The rate of MBTU was thus 10 fold accelerated by NO 3. In a second step, we studied the reaction at a higher MBTU concentration in the goal to monitor mineralization and to undertake an analytical study of photoproducts. The initial rate of MBTU (0.1 mM) photolysis was accelerated by NO 3 (1 mM) by a factor of 2, by NO2 (0.1 mM, pH = 5) by a factor of 3 and by HNO2 (0.1 mM, pH = 2) by a factor of 7.
1.385
1526
0 0
5
10
15
20
Time (min)
3.2. Mineralization of MBTU We monitored the mineralization of MBTU by measuring the TOC remaining in solution in the course of the irradiation. The decrease of organic carbon in the solution was due to the oxidation of the organic material into mineral carbonate species. Fig. 4 gives the consumption profile of MBTU (curve a) and the decrease profile of TOC (curve b) in solutions containing MBTU (0.1 mM) and NO 3 (1 mM). MBTU was completely consumed after 25 h of irradiation. At this stage, 5.5% of initial organic carbon was transformed into carbonate species. The decrease of TOC showed an acceleration when MBTU is disappeared indicating that photoproducts were more easily oxidized than MBTU itself. About 70% of the starting carbon content was mineralized after 192 h of irradiation. 3.3. Identification of photoproducts
1.0
4
12
concentration of MBTU x 10 (M)
The typical HPLC chromatogram of a solution of MBTU (0.1 mM) and NO 3 (1 mM) irradiated until a
TOC (mg l-1)
0.8 8
0.6 0.4
4 0.2 0 0
50
100
150
0.0 200
Irradiation time (hour)
Fig. 4. Irradiation of MBTU (0.1 mM) in presence of NO 3 (1 mM using sunlamps): (s) curve a, consumption profile of MBTU; (d) curve b, decrease of TOC in the solution.
Fig. 5. HPLC chromatogram of a solution containing MBTU 3 M) irradiated using sunlamps until a (104 M) and NO 3 (10 conversion extent of 70%. Numbered peaks refer to compounds for which molecular ions could be obtained by LC–ESI–MS and a structure is proposed (see Table 1).
conversion extent of 70% is given in Fig. 5. Numerous photoproducts are present. Very similar chromatograms were obtained from irradiated solutions of MBTU (0.1 mM) and NO 2 (0.1 mM). Several photoproducts could be identified with the help of LC–MS using ES interface and of MS using APCI interface and the flow injection technique. Table 1 lists their molecular ions and proposed chemical structures. MBTU that shows a molecular weight of 221 was eluted after 21 min and gave two molecular ion peaks at m/z = 222 and 244. The former corresponded to [M + H]+ and the latter to [M + Na]+. All the photoproducts were eluted just before MBTU indicating that they were more polar than it. Peak 16 showed molecular ion peaks at m/z = 238 and 260. Its molecular weight was thus heavier by 16 uma than that of MBTU. By comparison to the authentic standard that was previously synthetized (Malouki et al., 2003), we assigned peak 16 to 6-hydroxy-MBTU. This compound is the main metabolite of MBTU biodegradation. Peaks 17–20 showed the same molecular ion peaks than 16 and close retention times. Oxidation of the aromatic ring is expected to yield four isomers with OH function in position 4, 5, 6 or 7. Three peaks among 17–20 should correspond to the hydroxylated derivatives in position 4, 5 and 7. Phenyl urea herbicides were found to undergo demethylation (Amine-Kodja et al., 2004). This reaction is expected to occur in the case of MBTU too. It should yield photoproducts lighter by 14 uma compared to undemethylated compounds. Peak 19 0 showed molecular ion peaks at m/z = 208 and 220, lighter by 14 than those of MBTU and peak 15 molecular ion peak at m/z = 224, lighter by 14 than that of 6-hydroxy-MBTU.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
1527
Table 1 Molecular peak ions detected by LC–ESI–MS or APCI–MS, retention times in liquid chromatography (see Fig. 5) and proposed structures m/z
Interface
Number* (retention time)
Proposed structure
CH3
N +
+
222/244 [M + H] /[M + Na]
ESI+
S
142/164 [M + H]+/[M + Na]+
ESI+
ESI+
H
O
OHC
N
OHC
S
1 (5.0)
N 165 [M + H]+
CH3
N C N
MBTU (21)
CH3
9 (12.1)
N H
S N 181 [M + H]+
APCI+
a
CH3 N
HO
H
S
CH3
N 208/220 [M + H]+/[M + Na]+
ESI+
19 0 (19.5)
N C NH2 S
O CH3
N 224 [M + H]+
ESI+
15 (15.5)
HO
N C NH2 S
O
N 226 [M + H]+
APCI+
b
CH3
O2 N
N H
S
HO
O CH3
N 228/250 [M + H]+/[M + Na]+
ESI+
14 (14.2)
N C NH2
O S
O
O N
238/260 [M + H]+/[M + 23]+
ESI+
S
HO
ESI+
17–20 (18–20)
+
245/268 [M + H] /[M + Na]
ESI+
APCI+
HO 2C
N
HO 2C
S
CH3 N C NH2 O
O2 N
HO2 C ESI+
S
Number given on HPLC chromatogram of Fig. 5.
N
3
CH3 H
O
CH3
CO 2H
N C N HO2 C
*
CH3 N C N
HO
290/312 [M + H]+/[M + Na]+
H
O
7
c
CH3
N C N
HO
N 283 [M + H]+
H
O
S
+
CH3
CH3
N 238/260 [M + H]+/[M + 23]+
CH3 N C N
16 (17.3)
S
O
H
1528
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
Peaks 19 0 and 15 are thus likely to be demethylated compounds. Two methyl groups may be lost: that of terminal N atom and that of N atom in a-position to benzothiazole group. Molecular ion peak at m/z = 165 detected in LC–ES–MS (peak 9) fits well with a benzothiazole structure substituted on C2 by NHCH3, while molecular ion peak at m/z = 181 (peak a) obtained by APCI–MS may correspond to 6-hydroxybenzothiazole substituted by NHCH3 on carbon C2. No intermediates bearing an NH2 substituent on carbon C2 were detected. It can be thus concluded that demethylation of MBTU and 6-hydroxy-MBTU yielding 19 0 and 15 occurs on the terminal N atom. Peaks b and c that were obtained by APCI–MS showed molecular ion peaks at m/z = 226 and 283, respectively, corresponding to molecular weights of 225 and 282. These masses are heavier by 45 uma than those of products a and 16–20. This is consistent with nitration of the aromatic ring that is expected to occur based on literature data (Machado and Boule, 1995). Products 7 and 14 showed molecular ions peaks that differ by 18 uma (m/z = 246 and 268 and m/z = 228 and 250, respectively). It is in accordance with 7 that losses water to give 14. Product 7 is likely to be a diacidic comN HO
CH 3 N
S
MBTU
N
N
N NO 2
H OH
C
NH 2
O CH 3 N
S
CH 3 N
CH 3
S
NO 2
C O
OH OH
by hydroxyl radicals in the reaction. Numerous products appeared on the chromatograms. The structure of nine could be proposed on the basis of mass spectrometry analyses. The analysis of data revealed that oxidations of benzene ring and of urea chain occurred leading finally to the cleavage of the aromatic ring and to the removal of the urea chain. Nitration of the aromatic ring that generally yields very toxic compounds also took place but this reaction was minor. The primary steps were either hydroxylation or nitration of the aromatic ring or demethylation of the N terminal urea chain. We did not observe dihydroxy-MBTU that are expected to be produced by attack of a second hydroxyl radical on hydroxy-MBTU. This might be explained by a higher oxidizability of dihydroxylated compounds compared to MBTU or mono-hydroxylated compounds. May be they are in too low concentration for characterization. On the other hand, products obtained after ring cleavage and loss of two carbon atoms (diacidic, anhydride and dialdehyde derivatives) were present in irradiated solutions. Removal of the urea chain took place in several steps too: demethylation, loss of COCH3 and loss of NHCH3. Mineralization of MBTU finally occurred.
C O
pound and product 14 the corresponding anhydride. They are produced after cleavage of the aromatic ring. Peaks 1 and 3 correspond to products with a high degree of transformation and their assignment is more speculative. Product 1 showed molecular ion peaks at m/z = 142 and 164. This molecular weight is light suggesting that the benzothiazole ring is destroyed and that only the thiazole structure remains. Product 3 has probably a molecular weight of 289. This heavy mass is explained by a high degree of oxidation. Product 3 may bear two CO2H in place of the aromatic ring and a third one in place of the terminal methyl group. 3.4. Mechanism of reaction MBTU was found to be efficiently photodegraded by NO 3 or NO2 . Very close results were observed using both photoinductors. It indicates the main role played
CH 3 N
ring cleavage or removal of the urea chain
OH
mineralization
OH
H
3.5. Environmental significance On the basis of the present study, it can be concluded that MBTU should undergo phototransformation in aquatic environments containing nitrate and nitrite ions. However, the importance of this reaction path is expected to strongly vary with the water composition (Lam et al., 2003). The first important parameter is the level of nitrate and/or nitrite ions in the medium because it controls the rate of light absorption and thus the rate of hydroxyl radicals production. Another important factor is the concentration of dissolved organic matter that is known to trap hydroxyl radicals. In this way, it will compete with MBTU for their scavenging. Moreover, it absorbs solar light and may act as inner filter too. Lastly, the photoinduced transformation of MBTU may be affected by bicarbonate ions that are able to trap hydroxyl radicals too, resulting in fewer oxidant species.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
In conclusion, nitrate and nitrite ions were found to mineralize methabenzthiazuron upon irradiation within solar radiations. Numerous intermediary photoproducts could be characterized using HPLC–MS. Methabenzthiazuron mainly underwent oxidation reactions via hydroxyl radicals produced upon irradiation of nitrate and nitrite ions. After several oxidation steps, cleavage of the aromatic ring and removal of the urea chain occurred. Nitration was a minor pathway. These reactions are expected to take place in natural waters containing nitrate or nitrite ions and it would be now necessary to investigate the toxicity of photoproducts on aquatic organisms.
References Amine-Kodja, A., Boulkamh, A., Boule, P., 2004. Photochemical behaviour of phenylurea herbicides. Photochem. Photobiol. Sci. 3, 317–321. Buxton, G.V., Greenstock, C.L., Helman, P., Ross, A.B., 1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. J. Phys. Chem. Ref. Data 17, 513–886. Graetzel, M., Henglein, A., Lilie, J., Beck, G., 1969. Pulse radiolytic study of some elementary processes of nitrite ion oxidation and reduction. Ber. Bunsen. Phys. Chem. 73, 646– 653. Kieber, R.J., Li, A., Seaton, P.J., 1999. Production of nitrite from the photodegradation of dissolved organic matter in natural waters. Environ. Sci. Technol. 33, 993–998. Lam, M., Tantuco, K., Mabury, S., 2003. Photo-fate: a new approach in accounting for the contribution of indirect
1529
photolysis of pesticides and pharmaceuticals in surface waters. Environ. Sci. Technol. 37, 899–907. Machado, F., Boule, P., 1994. Phototransformation of phenolic derivatives in aqueous solution induced by excitation of nitrate and nitrite ions. Trends Photochem. Photobiol. 3, 29–37. Machado, F., Boule, P., 1995. Photonitration and photonitrosation of phenolic derivatives induced in aqueous solution by excitation of nitrite and nitrate ions. J. Photochem. Photobiol. A 86, 73–80. Malouki, M., Giry, G., Besse, P., Combourieu, B., Sancelme, M., Bonnemoy, F., Richard, C., Delort, A.M., 2003. Transformation of the herbicide methabenzthiazuron by coupling microbiology and photochemistry. Environ. Toxicol. Chem. 22, 2013–2019. Sakriss, W., Ga¨b, G., Korte, F., 1976. Contribution to ecological chemistry. CXXVIII. Photooxidation reactions of methabenzthiazuron in solution. Chemosphere 5, 339–348. Strickler, S.J., Kasha, M., 1963. Solvent effects on the electronic absorption spectrum of nitrite ion. J. Am. Chem. Soc. 85, 2899–2901. Tomlin, C.D.S., 2000. The Pesticide Manual, 12th ed. British Crop Protection Council, Farnham, UK, p. 110 and 548. Wagner, I., Strehlow, H., Busse, G., 1980. Flash photolysis of nitrate ions in solution. Z. Phys. Chem. 123, 1–33. Warneck, P., Wurzinger, C., 1988. Product quantum yields for the 305-nm photodecomposition of nitrate in aqueous solution. J. Phys. Chem. 92, 6278–6283. Zafiriou, O.C., Bonneau, R., 1987. Wavelength-dependent quantum yield of hydroxyl radical formation from photolysis of nitrite ion in water. Photochem. Photobiol. 45, 723– 727. Zepp, R.G., Hoigne´, J., Bader, H., 1987. Nitrate-induced photooxidation of trace organic chemicals in water. Environ. Sci. Technol. 21, 443–450.
Phototransformation of methabenthiazuron in the presence of nitrate and nitrite ions Moulay A. Malouki, Bernadette Lave´drine, Claire Richard
*
Laboratoire de Photochimie Mole´culaire et Macromole´culaire, UMR n 6505 CNRS, Universite´ Blaise Pascal, F-63177 Aubie`re Cedex, France Received 10 December 2004; received in revised form 16 February 2005; accepted 20 February 2005 Available online 18 April 2005
Abstract The influence of nitrate and nitrite ions on the degradation of methabenzthiazuron upon irradiation using artificial solar light has been investigated. The rate of degradation of methabenzthiazuron (1 lM) was accelerated by NO 3 (0.1 mM) by a factor of 10. The irradiation of methabenzthiazuron (0.1 mM) in the presence of NO 3 (1 mM) or NO 2 (0.1 mM) yielded numerous intermediary photoproducts. Mineralization was achieved after prolonged exposure. Some were identified with the help of LC–ESI–MS and flow injection APCI–MS techniques. Both oxidations of the aromatic ring and of the urea chain were observed. The former started by hydroxylation of the ring. Further oxidation of the ring led to cleavage of the benzenic ring with formation of dialdehydic, diacidic and anhydric compounds. Complete removal of the lateral urea chain took place subsequently to demethylation of the terminal methyl group and loss of the CO–NH2 group. Nitration was a minor process. This work shows that the photodegradation of methabenzthiazuron in the presence of nitrate or nitrite ions is highly non-specific. 2005 Elsevier Ltd. All rights reserved. Keywords: Photoinduction; Aquatic environment; Herbicide; Photoproducts; Oxidation; Mineralization
1. Introduction Methabenzthiazuron (N-(2-benzothiazolyl)-N,N 0 dimethylurea) (MBTU) is an active ingredient of Tribunil (Bayer, France) and Ormet (Phytorus S.A., France). Methabenzthiazuron belongs to the group of urea herbicides, which is used to control a broad spectrum of broad-leaved weeds and grasses in winter corn, spring wheat, grass seed and in nurseries. It is a selective
*
Corresponding author. Tel.: +33 4 73 40 71 42; fax: +33 4 73 40 77 00. E-mail address: [email protected] (C. Richard).
herbicide that is primarily absorbed through the roots. Methabenzthiazuron inhibits photosynthetic electron transport. Its bioactivity persists in the field soil for more than one growing season. Bioconcentration factor is about 10.1 for the pH range 7–10. It is toxic for fish and daphnia: LC50 is 15.9 mg/l (96 h) for rainbow trout and 30.6 mg/l (48 h) for daphnia magna. It is classified as toxic for the environment and very toxic for aquatic organisms (Tomlin, 2000). MBTU is soluble in water at 59 mg l1 (20 C within the pH range 7–10) and may be transferred into the aquatic environment by leaching. Screening surveys in France showed that concentrations of MBTU up to 1 lM may be reached in surface waters close to agriculture areas. MBTU absorbs UVB radiations (Fig. 1);
0045-6535/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.02.080
1524
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529 14
100
3.0x10
14
1.5x10
MBTU
14
1.0x10
5000
ε (M-1. cm-1)
-1 -1
14
wavelength (nm)
photons cm nm s
-1
14
2.0x10
270 280 290 300 310 320 330 340 350 360
10000
-2
2.5x10
-1
extinction coefficient (M cm )
15000
50
13
5.0x10 0.0
0 300
350
400
450
0 200
wavelength (nm)
Fig. 1. UV–visible spectrum of MBTU and ( ) solar flux reaching the earth surface in summer; insert: emission spectrum of sunlamps.
however, it was found to be very slowly photolyzed in solutions when irradiated above 290 nm (Sakriss et al., 1976; Malouki et al., 2003). In contrast 6-hydroxymethabenzthiazuron, its main biodegradation metabolite, undergoes photooxidation in the same exposure conditions (Malouki et al., 2003). As an alternative to direct photolysis, MBTU may be degraded via photoinduced reactions mediated by chromophores present in continental surface waters. Nitrate ions are usually present in the aquatic environment at concentrations depending strongly on the geographic location and on the agricultural activities of the surrounding areas. Concentrations up to 103 M may be reached. Nitrite ions that are produced by photolysis of nitrate ions and by photodegradation of aquatic humic substances (Kieber et al., 1999) are present too, but at lower concentration. Nitrate shows a maximum of absorption at 304 nm (e 7 M1 cm1) (Wagner et al., 1980), while nitrite at 355 nm (e 22 M1 cm1) (Strickler and Kasha, 1963) (Fig. 2). Both absorb solar radiation in the actinic spectrum and can undergo chemical reactions. It is well established that under excitation nitrite ions yields hydroxyl radicals and nitrogen monoxide according to NO þ hm ! NO þ O ð1Þ 2
O
þ H2 O ! OH þ OH
ð2Þ
The quantum yield of process (1) is 0.025 at 355 nm (Zafiriou and Bonneau, 1987). The photolysis of nitrate ions is more complex since two primary processes take place: NO 3 þ hm ! NO2 þ O
NO 3
þ hm ! NO2 þ O
ð3Þ ð4Þ
The quantum yield of process (3) was evaluated to be 1.1 · 103 and the one of process (4) to 9.2 · 103 (War-
250
300
350
400
450
wavelength (nm)
Fig. 2. UV–visible spectra of KNO3 (–––), NaNO2 (——) and HNO2 ( ) in water.
neck and Wurzinger, 1988). The photolysis of nitrite and nitrate ions in the aquatic environment cannot be therefore neglected. Moreover, as sources of hydroxyl radicals, nitrate and nitrite ions are able to photoinduce the phototransformation of organic pollutants (Zepp et al., 1987; Machado and Boule, 1994). The oxidations are mainly attributed to hydroxyl radicals that are strong oxidants, atomic oxygen being not as reactive. Nitration and nitrosation reactions were reported too. The reactions are concentration dependent due to the following processes that may occur too: NO þ NO ! N O ð5Þ 2
2
2
4
þ N2 O4 þ H2 O ! NO 2 þ NO3 þ 2H
ð6Þ
NO 2 þ OH ! NO2 þ OH
8
ð7Þ
1 1
3 1
with k5 = 4.5 · 10 M s and k6 = 10 s (Graetzel et al., 1969) and k7 = 1010 M1 s1 (Buxton et al., 1988). In this report, we describe our research on the photolysis of methabenzthiazuron in the presence of nitrate and nitrite ions. In a first step, we investigated the ability of nitrate (0.1 mM) to photoinduce the degradation of methabenzthiazuron (1 lM), i.e., at levels relevant to environmental conditions. Then, in order to identify intermediary photoproducts, we performed an analytical study using higher concentrations of reactants. Several photoproducts were characterized. This helps to get a better insight into the mechanism of degradation from the first steps of oxidation to complete mineralization. 4
CH3
N
5 6
S 7
2
CH3
N C N O
Structure of MBTU
H
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
2. Materials and methods 2.1. Materials Methabenthiazuron (MBTU) was purchased from Riedel-de-Hae¨n (Pestanal). Authentic sample of 6-OHMBTU was obtained by biodegradation of MBTU as previously described (Malouki et al., 2003). Nitrate and nitrite ions were Fluka products (99.0% purity grade). Water was purified using a Milli-Q device (Millipore). 2.2. Irradiations Polychromatic irradiations were performed using a device equipped with six fluorescent Duke sunlamp GL 20 tubes (SNEE, Aubervilliers, France) emitting within 275–350 nm with a maximum of emission at 313 nm (insert of Fig. 2). The reactor was cylindrical and made in Pyrex glass in order to cut off the wavelengths shorter than 290 nm. In these conditions, samples received radiations in the range 290–350 nm only. MBTU (1 lM or 0.1 mM) was dissolved in Milli-Q water or in aqueous solutions containing KNO3 (0.1 mM or 1 mM), NaNO2 or HNO2 (0.1 mM). Solutions were acidified with HClO4. These solutions were poured into the Pyrex glass reactor and irradiated using the sunlamps. Aliquots of 0.5 ml were removed from irradiated solutions at selected intervals and analysed by HPLC. Dark control experiments showed no loss of compounds by adsorption on glass surfaces or by spontaneous transformation on the time scale of irradiation experiments, irrespective of the aqueous media used (pure water, nitrate, nitrate or nitrous acid solutions).
1525
100% reached at t = 20 min. The electrospray interface (ESI) operated in positive mode. Nitrogen was used as shealth gas (13 l/min, 350 C) and the spray voltage was 4 kV. The pressure of nebulisation was 4 bar. The scan range was 100–500 uma. Another set of analyses was performed on a Hewlett–Packard 5989 MS apparatus equipped with a HP G1075A APCI interface. Methanolic irradiated solutions were injected in the mobile phase using the flow injection technique. The following optimised conditions were used: APCI vaporiser temperature 350 C; capillary temperature 150 C; source voltage 6 kV; capillary voltage 25 V; source current 5 lA; sheath gas flow (N2) 80 arbitrary units. Total organic carbon (TOC) measurements in the liquid phase were carried out with a TOC analyser Schimadzu model TOC-5O5OA. The calibration curves within the range 0–30 mg l1 were obtained by using potassium hydrogen phthalate and sodium hydrogen carbonate for organic and inorganic carbon, respectively.
3. Results 3.1. Kinetic of MBTU photodegradation in the presence of nitrite and nitrate ions We first investigated the photodegradation of MBTU (1 lM) in pure water and in the presence of NO 3 (0.1 mM). These concentrations were chosen because they were relevant of environmental conditions. Results are presented in Fig. 3. MBTU was found to be photolyzed much more slowly in Milli-Q water than in the presence of nitrate. The rates could be treated in
2.3. Analyses 1.0 0.8 0.6 0.0
0.4
-0.2
ln(c/c0)
c/c0
Losses of MBTU were monitored by HPLC–UV using a Waters apparatus equipped with two pumps (model 510), an auto sampler, a photodiode array detector (model 996) and a C18 reversed-phase (4.6 mm · 250 mm, 5 lm) Spherisorb S5 ODS2 Waters column. Eluents were mixtures of water and methanol (40/60, v/v) at a constant flow of 1 ml min1. UV spectra were recorded on a Cary 3 (Varian, Les Ulis, France) spectrophotometer. A 1-cm path quartz cell was used in all the experiments. The reference beam blank was always Milli-Q water. Analyses of MBTU and photoproducts were performed by HPLC–MS using a Hewlett–Packard 1100-MSD apparatus equipped with a Ultisphere HDO C18 (3 mm · 100 mm, 3 lm) column. A mixture of water with formic acid (pH = 2.9) and methanol was used as mobile phase at a constant flow of 0.4 ml min1. The gradient was the following one: 70% of methanol at t = 0 and linear increase until
-0.4
-0.6
0.2
-0.8 0
10
20
30
40
50
time (hour)
0.0 0
10
20
30
40
50
time (hour) Fig. 3. Consumption profiles of MBTU (1 lM) irradiated using sunlamps: (s) in pure water, (d) in presence of NO 3 (0.1 mM); insert: plot of ln(c/c0) versus t, where c represents the MBTU concentration at time t and c0 the MBTU concentration at time 0.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
19.930
16 - 20 mAU 11.613
35
19.506 20.215
15
18.688
14
15.460
11.081
9 12.126
7.700
7 8.874 8.999 9.335 9.570 9.957 10.310
6.604
5
1 4.997
10
1.633
15
12.960 13.390 13.608 13.935 14.184
3
20
17.541 17 .706 18.020 18.355
25
17.340
30
0.927
first-order form (see insert of Fig. 3). We got ln(c/ c0) = k.t, where c represents the MBTU concentration at time t, c0 the MBTU concentration at time 0, and k the apparent first order rate constant of MBTU consumption. In pure water, we computed k = 0.012 ± 0.001 with R2 = 0.977 and in the presence of nitrate k = 0.12 ± 0.01 with R2 = 0.998. The rate of MBTU was thus 10 fold accelerated by NO 3. In a second step, we studied the reaction at a higher MBTU concentration in the goal to monitor mineralization and to undertake an analytical study of photoproducts. The initial rate of MBTU (0.1 mM) photolysis was accelerated by NO 3 (1 mM) by a factor of 2, by NO2 (0.1 mM, pH = 5) by a factor of 3 and by HNO2 (0.1 mM, pH = 2) by a factor of 7.
1.385
1526
0 0
5
10
15
20
Time (min)
3.2. Mineralization of MBTU We monitored the mineralization of MBTU by measuring the TOC remaining in solution in the course of the irradiation. The decrease of organic carbon in the solution was due to the oxidation of the organic material into mineral carbonate species. Fig. 4 gives the consumption profile of MBTU (curve a) and the decrease profile of TOC (curve b) in solutions containing MBTU (0.1 mM) and NO 3 (1 mM). MBTU was completely consumed after 25 h of irradiation. At this stage, 5.5% of initial organic carbon was transformed into carbonate species. The decrease of TOC showed an acceleration when MBTU is disappeared indicating that photoproducts were more easily oxidized than MBTU itself. About 70% of the starting carbon content was mineralized after 192 h of irradiation. 3.3. Identification of photoproducts
1.0
4
12
concentration of MBTU x 10 (M)
The typical HPLC chromatogram of a solution of MBTU (0.1 mM) and NO 3 (1 mM) irradiated until a
TOC (mg l-1)
0.8 8
0.6 0.4
4 0.2 0 0
50
100
150
0.0 200
Irradiation time (hour)
Fig. 4. Irradiation of MBTU (0.1 mM) in presence of NO 3 (1 mM using sunlamps): (s) curve a, consumption profile of MBTU; (d) curve b, decrease of TOC in the solution.
Fig. 5. HPLC chromatogram of a solution containing MBTU 3 M) irradiated using sunlamps until a (104 M) and NO 3 (10 conversion extent of 70%. Numbered peaks refer to compounds for which molecular ions could be obtained by LC–ESI–MS and a structure is proposed (see Table 1).
conversion extent of 70% is given in Fig. 5. Numerous photoproducts are present. Very similar chromatograms were obtained from irradiated solutions of MBTU (0.1 mM) and NO 2 (0.1 mM). Several photoproducts could be identified with the help of LC–MS using ES interface and of MS using APCI interface and the flow injection technique. Table 1 lists their molecular ions and proposed chemical structures. MBTU that shows a molecular weight of 221 was eluted after 21 min and gave two molecular ion peaks at m/z = 222 and 244. The former corresponded to [M + H]+ and the latter to [M + Na]+. All the photoproducts were eluted just before MBTU indicating that they were more polar than it. Peak 16 showed molecular ion peaks at m/z = 238 and 260. Its molecular weight was thus heavier by 16 uma than that of MBTU. By comparison to the authentic standard that was previously synthetized (Malouki et al., 2003), we assigned peak 16 to 6-hydroxy-MBTU. This compound is the main metabolite of MBTU biodegradation. Peaks 17–20 showed the same molecular ion peaks than 16 and close retention times. Oxidation of the aromatic ring is expected to yield four isomers with OH function in position 4, 5, 6 or 7. Three peaks among 17–20 should correspond to the hydroxylated derivatives in position 4, 5 and 7. Phenyl urea herbicides were found to undergo demethylation (Amine-Kodja et al., 2004). This reaction is expected to occur in the case of MBTU too. It should yield photoproducts lighter by 14 uma compared to undemethylated compounds. Peak 19 0 showed molecular ion peaks at m/z = 208 and 220, lighter by 14 than those of MBTU and peak 15 molecular ion peak at m/z = 224, lighter by 14 than that of 6-hydroxy-MBTU.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
1527
Table 1 Molecular peak ions detected by LC–ESI–MS or APCI–MS, retention times in liquid chromatography (see Fig. 5) and proposed structures m/z
Interface
Number* (retention time)
Proposed structure
CH3
N +
+
222/244 [M + H] /[M + Na]
ESI+
S
142/164 [M + H]+/[M + Na]+
ESI+
ESI+
H
O
OHC
N
OHC
S
1 (5.0)
N 165 [M + H]+
CH3
N C N
MBTU (21)
CH3
9 (12.1)
N H
S N 181 [M + H]+
APCI+
a
CH3 N
HO
H
S
CH3
N 208/220 [M + H]+/[M + Na]+
ESI+
19 0 (19.5)
N C NH2 S
O CH3
N 224 [M + H]+
ESI+
15 (15.5)
HO
N C NH2 S
O
N 226 [M + H]+
APCI+
b
CH3
O2 N
N H
S
HO
O CH3
N 228/250 [M + H]+/[M + Na]+
ESI+
14 (14.2)
N C NH2
O S
O
O N
238/260 [M + H]+/[M + 23]+
ESI+
S
HO
ESI+
17–20 (18–20)
+
245/268 [M + H] /[M + Na]
ESI+
APCI+
HO 2C
N
HO 2C
S
CH3 N C NH2 O
O2 N
HO2 C ESI+
S
Number given on HPLC chromatogram of Fig. 5.
N
3
CH3 H
O
CH3
CO 2H
N C N HO2 C
*
CH3 N C N
HO
290/312 [M + H]+/[M + Na]+
H
O
7
c
CH3
N C N
HO
N 283 [M + H]+
H
O
S
+
CH3
CH3
N 238/260 [M + H]+/[M + 23]+
CH3 N C N
16 (17.3)
S
O
H
1528
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
Peaks 19 0 and 15 are thus likely to be demethylated compounds. Two methyl groups may be lost: that of terminal N atom and that of N atom in a-position to benzothiazole group. Molecular ion peak at m/z = 165 detected in LC–ES–MS (peak 9) fits well with a benzothiazole structure substituted on C2 by NHCH3, while molecular ion peak at m/z = 181 (peak a) obtained by APCI–MS may correspond to 6-hydroxybenzothiazole substituted by NHCH3 on carbon C2. No intermediates bearing an NH2 substituent on carbon C2 were detected. It can be thus concluded that demethylation of MBTU and 6-hydroxy-MBTU yielding 19 0 and 15 occurs on the terminal N atom. Peaks b and c that were obtained by APCI–MS showed molecular ion peaks at m/z = 226 and 283, respectively, corresponding to molecular weights of 225 and 282. These masses are heavier by 45 uma than those of products a and 16–20. This is consistent with nitration of the aromatic ring that is expected to occur based on literature data (Machado and Boule, 1995). Products 7 and 14 showed molecular ions peaks that differ by 18 uma (m/z = 246 and 268 and m/z = 228 and 250, respectively). It is in accordance with 7 that losses water to give 14. Product 7 is likely to be a diacidic comN HO
CH 3 N
S
MBTU
N
N
N NO 2
H OH
C
NH 2
O CH 3 N
S
CH 3 N
CH 3
S
NO 2
C O
OH OH
by hydroxyl radicals in the reaction. Numerous products appeared on the chromatograms. The structure of nine could be proposed on the basis of mass spectrometry analyses. The analysis of data revealed that oxidations of benzene ring and of urea chain occurred leading finally to the cleavage of the aromatic ring and to the removal of the urea chain. Nitration of the aromatic ring that generally yields very toxic compounds also took place but this reaction was minor. The primary steps were either hydroxylation or nitration of the aromatic ring or demethylation of the N terminal urea chain. We did not observe dihydroxy-MBTU that are expected to be produced by attack of a second hydroxyl radical on hydroxy-MBTU. This might be explained by a higher oxidizability of dihydroxylated compounds compared to MBTU or mono-hydroxylated compounds. May be they are in too low concentration for characterization. On the other hand, products obtained after ring cleavage and loss of two carbon atoms (diacidic, anhydride and dialdehyde derivatives) were present in irradiated solutions. Removal of the urea chain took place in several steps too: demethylation, loss of COCH3 and loss of NHCH3. Mineralization of MBTU finally occurred.
C O
pound and product 14 the corresponding anhydride. They are produced after cleavage of the aromatic ring. Peaks 1 and 3 correspond to products with a high degree of transformation and their assignment is more speculative. Product 1 showed molecular ion peaks at m/z = 142 and 164. This molecular weight is light suggesting that the benzothiazole ring is destroyed and that only the thiazole structure remains. Product 3 has probably a molecular weight of 289. This heavy mass is explained by a high degree of oxidation. Product 3 may bear two CO2H in place of the aromatic ring and a third one in place of the terminal methyl group. 3.4. Mechanism of reaction MBTU was found to be efficiently photodegraded by NO 3 or NO2 . Very close results were observed using both photoinductors. It indicates the main role played
CH 3 N
ring cleavage or removal of the urea chain
OH
mineralization
OH
H
3.5. Environmental significance On the basis of the present study, it can be concluded that MBTU should undergo phototransformation in aquatic environments containing nitrate and nitrite ions. However, the importance of this reaction path is expected to strongly vary with the water composition (Lam et al., 2003). The first important parameter is the level of nitrate and/or nitrite ions in the medium because it controls the rate of light absorption and thus the rate of hydroxyl radicals production. Another important factor is the concentration of dissolved organic matter that is known to trap hydroxyl radicals. In this way, it will compete with MBTU for their scavenging. Moreover, it absorbs solar light and may act as inner filter too. Lastly, the photoinduced transformation of MBTU may be affected by bicarbonate ions that are able to trap hydroxyl radicals too, resulting in fewer oxidant species.
M.A. Malouki et al. / Chemosphere 60 (2005) 1523–1529
In conclusion, nitrate and nitrite ions were found to mineralize methabenzthiazuron upon irradiation within solar radiations. Numerous intermediary photoproducts could be characterized using HPLC–MS. Methabenzthiazuron mainly underwent oxidation reactions via hydroxyl radicals produced upon irradiation of nitrate and nitrite ions. After several oxidation steps, cleavage of the aromatic ring and removal of the urea chain occurred. Nitration was a minor pathway. These reactions are expected to take place in natural waters containing nitrate or nitrite ions and it would be now necessary to investigate the toxicity of photoproducts on aquatic organisms.
References Amine-Kodja, A., Boulkamh, A., Boule, P., 2004. Photochemical behaviour of phenylurea herbicides. Photochem. Photobiol. Sci. 3, 317–321. Buxton, G.V., Greenstock, C.L., Helman, P., Ross, A.B., 1988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. J. Phys. Chem. Ref. Data 17, 513–886. Graetzel, M., Henglein, A., Lilie, J., Beck, G., 1969. Pulse radiolytic study of some elementary processes of nitrite ion oxidation and reduction. Ber. Bunsen. Phys. Chem. 73, 646– 653. Kieber, R.J., Li, A., Seaton, P.J., 1999. Production of nitrite from the photodegradation of dissolved organic matter in natural waters. Environ. Sci. Technol. 33, 993–998. Lam, M., Tantuco, K., Mabury, S., 2003. Photo-fate: a new approach in accounting for the contribution of indirect
1529
photolysis of pesticides and pharmaceuticals in surface waters. Environ. Sci. Technol. 37, 899–907. Machado, F., Boule, P., 1994. Phototransformation of phenolic derivatives in aqueous solution induced by excitation of nitrate and nitrite ions. Trends Photochem. Photobiol. 3, 29–37. Machado, F., Boule, P., 1995. Photonitration and photonitrosation of phenolic derivatives induced in aqueous solution by excitation of nitrite and nitrate ions. J. Photochem. Photobiol. A 86, 73–80. Malouki, M., Giry, G., Besse, P., Combourieu, B., Sancelme, M., Bonnemoy, F., Richard, C., Delort, A.M., 2003. Transformation of the herbicide methabenzthiazuron by coupling microbiology and photochemistry. Environ. Toxicol. Chem. 22, 2013–2019. Sakriss, W., Ga¨b, G., Korte, F., 1976. Contribution to ecological chemistry. CXXVIII. Photooxidation reactions of methabenzthiazuron in solution. Chemosphere 5, 339–348. Strickler, S.J., Kasha, M., 1963. Solvent effects on the electronic absorption spectrum of nitrite ion. J. Am. Chem. Soc. 85, 2899–2901. Tomlin, C.D.S., 2000. The Pesticide Manual, 12th ed. British Crop Protection Council, Farnham, UK, p. 110 and 548. Wagner, I., Strehlow, H., Busse, G., 1980. Flash photolysis of nitrate ions in solution. Z. Phys. Chem. 123, 1–33. Warneck, P., Wurzinger, C., 1988. Product quantum yields for the 305-nm photodecomposition of nitrate in aqueous solution. J. Phys. Chem. 92, 6278–6283. Zafiriou, O.C., Bonneau, R., 1987. Wavelength-dependent quantum yield of hydroxyl radical formation from photolysis of nitrite ion in water. Photochem. Photobiol. 45, 723– 727. Zepp, R.G., Hoigne´, J., Bader, H., 1987. Nitrate-induced photooxidation of trace organic chemicals in water. Environ. Sci. Technol. 21, 443–450.