Article pubs.acs.org/est
Transformation of Biocides Irgarol and Terbutryn in the Biological Wastewater Treatment Agnessa Luft,† Manfred Wagner,‡ and Thomas A. Ternes*,† †
Federal Institute of Hydrology (BfG), 56068 Koblenz, Germany Max Planck Institute for Polymer Research, 55128 Mainz, Germany
‡
S Supporting Information *
ABSTRACT: The biocides irgarol and terbutryn enter the wastewater treatment plant (WWTP) via combined sewer systems after leaching from coatings and paints of materials. In this study, the biotransformation of irgarol and terbutryn was examined in aerobic batch experiments with activated sludge taken from the nitrification zone of a conventional WWTP, since currently there is no information about the fate of irgarol and terbutryn in biological wastewater treatment. Both, irgarol and terbutryn were transformed into one main transformation product (TP) following pseudo first-order kinetics. The TPs were tentatively identified by high-resolution mass spectrometry (HR-MS) to be irgarol sulfoxide and terbutryn sulfoxide. The final confirmation of the proposed chemical structures of the TPs was achieved by a comparison of mass spectra and nuclear magnetic resonance (NMR) spectra with those of authentic reference standards (e.g., synthesized). An analytical method for the sensitive quantification of irgarol, terbutryn and their TPs in environmental samples was developed based on solid phase extraction (SPE) and LC tandem MS detection. Irgarol sulfoxide and terbutryn sulfoxide were detected in the effluents (average concentrations up to 22 ng L−1 and 65 ng L−1) of all four investigated WWTPs as well as in streams and small rivers (up to 14 ng L−1 and 34 ng L−1). Luminescent bacteria inhibition test with Vibrio f ischeri exhibited that the TPs irgarol sulfoxide and terbutryn sulfoxide feature a similar bacterial toxicity than the parent compounds.
■
INTRODUCTION Irgarol (2-methylthio-4-tert-butylamino-6-cyclopropylamino-striazine) and terbutryn (2-methylthio-4-tert-butylamino-6-ethylamino-s-triazine) belong to the s-triazines. Irgarol is well-known as active agent in antifouling paints applied to the hulls of ships and boats, since the late 1980s it is often used as a substitute for tributyltin (TBT).1 Terbutryn is known as pesticide in the agriculture which is banned for agricultural use since 2003 in the EU. However, both compounds are still being widely used as biocidal compounds in building materials.2,3 According to the Biocidal Products Directive 98/8/EC biocides are “active substances [...] intended to destroy, deter, render harmless, prevent the action of, or otherwise exert a controlling effect on any harmful organism”.4 Irgarol and terbutryn are applied as algicides in surface coatings of buildings, such as paints and renders, to avoid growth of algae.5,6 A continuously leaching of biocides from surface coatings is intended, since it is the prerequisite for their growth-inhibiting effects of algae.2,7 Therefore, irgarol and terbutryn can be discharged directly into surface waters via runoff water of urban areas and in case of irgarol also via leaching from boat coatings. However, the majority of these biocides reaches municipal WWTPs via combined sewer systems.7,8 Due to their incomplete elimination in the WWTPs, both irgarol and terbutryn are released © 2013 American Chemical Society
into surface waters. Irgarol and terbutryn have been detected in WWTP effluents up to 22 ng L−1 and 123 ng L−1 and in surface waters up to 11 ng L−1 and 169 ng L−1,9 respectively. In a sampling period from 2003 to 2006, Quednow et al.10 determined terbutryn in four small rivers in Germany with mean concentrations up to 0.58 μg L−1, although the use of terbutryn in agriculture had been forbidden. As required by the Water Framework Directive (WFD) 2000/60/EC of the European Community environmental quality standards (EQS) are established to achieve a good chemical status of European water bodies. The annual average EQS of irgarol and terbutryn in streams and rivers reported by the EU (Directive 2013/39/EU, 12.08.2013) are as low as 2.5 ng L−1 and 65 ng L−1, respectively. These low EQS are based on the toxicity of irgarol and terbutryn to higher plants, algae, and aquatic epibiotic organism by inhibition of photosynthesis.11−16 Freshwater mesocosm studies indicated the effects of irgarol on chlorophytes (EC50 = 0.34 μg L−1), on species of cyclopoid copepods (EC50 = 0.09 μg L−1) and ostracods (EC50 = Received: Revised: Accepted: Published: 244
August 8, 2013 November 5, 2013 November 8, 2013 December 12, 2013 dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
0.11 μg L−1).11 Furthermore, effects were found on macrophytes with EC50 values of 0.21 μg L−1 for the species Myriophyllum verticillatum.17 Terbutryn exhibits a toxicity to aquatic organisms such as microalgae with 72 h-EC50 values of 2 μg L−1 for the species Selenastrum capricornutum.13 Previous studies highlight the possibility of biodegradation,18,19 photodegradation,20−24 and chemically catalyzed degradation25−27 of irgarol and terbutryn. Irgarol was transformed mainly via the mechanism of N-dealkylation into the stable degradation product M1 (2-methylthio-4-tert-butylamino-6-amino-s-triazine) during biodegradation by the white rot fungus Phanerochaete chrysosporium,18 photodegradation20,22,24 and mercuric chloride-catalyzed hydrolysis26 in water. Further byproducts during photodegradation of irgarol were reported as, for example, hydroxy (2-hydroxy-4-tert-butylamino-6-cyclopropylamino-s-triazine) and sulfone (2-methylsulfonyl-4-tertbutylamino-6-cyclopropylamino-s-triazine) derivatives.22 Much less studies exist for the degradation of terbutryn. Muir et al.19 reported degradation half-lives for terbutryn of 240 and 180 days in water with pond and river sediments under laboratory conditions, respectively. The major degradation product was hydroxy-terbutryn (2-hydroxy-4-tert-butylamino-6ethylamino-s-triazine), and to a minor extent terbutryn sulfoxide (2-methylsulfinyl-4-tert-butylamino-6-ethylamino-striazine) and N-deethylated terbutryn (2-methylthio-4-tertbutylamino-6-amino-s-triazine) were found.19 The latter has the same chemical structure as M1, the main TP known of the degradation of irgarol. In field studies N-deethylated terbutryn was a major product in water and sediment extracts in farm ponds.28 However, to the best of our knowledge, no studies reported the biodegradation of irgarol and terbutryn during biological wastewater treatment. Biodegradation studies of biocides imply a particular challenge because elevated concentrations of biocides have to be avoided as they might lead to an inhibition of the microbial activity. Hence, the limited possibility for the use of elevated concentrations of biocides in batch experiment hinders the quantitative isolation of TPs for their identification. The objective of the current study was to investigate the biodegradation of irgarol and terbutryn in contact with activated sludge at environmentally relevant concentrations. Identification of biotic TPs was accomplished by a two step approach using LC-HR-MS for the elucidation of tentative structures and the comparison with (synthesized) reference standards for final confirmation. Furthermore, the importance of WWTPs for the formation of irgarol and terbutryn TPs and their occurrence in surface waters were determined.
Table 1. Overview of the Practical Experiments TP identification
• • • •
transformation kinetic/mass balance
• batch experiments with activated sludge • quantification of target compounds and TPs via LC-tandem MS, direct injection
environmental samples from WWTPs as well as streams and rivers bacterial screening toxicity test
• quantification of target compounds and TPs via LC-tandem MS after SPE
batch experiments with activated sludge TP identification via HR-MS chemical synthesis of irgarol TP isolation of the synthesized irgarol TP via semipreparative HPLC-UV • confirmation of the chemical structure via NMR
• luminescent bacteria inhibition test with Vibrio f ischeri analyses via thin layer chromatography (TLC)
and methanol were picograde and purchased from LGC Promochem (Wesel, Germany). Iron(III) chloride (>97%), periodic acid, sodium thiosulfate, and ammonium formate (purum grade) were obtained from Sigma-Aldrich (Schnelldorf, Germany). Formic acid (p.a.) was purchased from Merck (Darmstadt, Germany). Ultra pure water (Milli-Q water) was obtained from a Milli-Q system (Millipore, Billerica, MA). Details on the standard solutions are given in the Supporting Information (SI). Batch Experiments of Irgarol and Terbutryn in Activated Sludge. For the batch experiments, sewage sludge was taken from a municipal WWTP (WWTP 5, more details in the SI) as a grab sample from the nitrification zone of the activated sludge unit. The activated sludge (total suspended solids (TSS): 4 gSS L−1, total organic carbon (TOC): 0.3 g gSS−1) was transported to the laboratory, where batch experiments were started on the same day. Sludge (100 mL) was poured into 1 L amber glass bottles and filled up with 900 mL of WWTP effluent (kinetic experiments) or groundwater (TP identification). The used groundwater is free of all target compounds. The sludge was diluted to minimize the effect of sorption. The amber glass bottles were constantly purged with a mixture of CO2 and air at a definite ratio (1:20). This was done to adjust a stable pH (pH 7.1 ± 0.1) and oxic conditions. The sludge was continuously stirred during the experiments at a temperature of 22 °C ± 1 °C. The analytes (concentrations see Table 2) were spiked in the diluted sludge after an equilibration time of 2 h. In addition, analytes were spiked to filtered effluent (0.45 μm cellulose nitrate filters, Sartorius Stedim Biotech, Göttingen, Germany) without sludge (abiotic control). Nonspiked sludge samples (blank samples) were used as reference samples for the identification of TPs by HR-MS as well as for the determination of the initial analyte concentrations. Samples were taken immediately after spiking and at definite time intervals. For measurements (HR-MS) with direct injection, samples were filtered through 0.45 μm syringe filters (cellulose nitrate filters, Spartan, Whatman), and stored in the freezer (−25 °C) until analysis. For the kinetic experiments, samples (50 mL) were taken and filtered through glass fiber filters (GF/6, Whatman), and stored at 4 °C. Within 24 h the samples were extracted using SPE (see below). Identification of TPs via High Resolution Mass Spectrometry. The nonenriched batch samples were analyzed by linear ion trap quadrupole (LTQ)-Orbitrap-MS (LTQ
■
MATERIALS AND METHODS A short overview of the practical experiments is presented in Table 1 listed in chronological order. Chemicals. Irgarol was purchased from Sigma-Aldrich (Schnelldorf, Germany), terbutryn from Dr. Ehrenstorfer GmbH (Augsburg, Germany), terbutryn sulfoxide from MicroCombiChem e.K. (Wiesbaden, Germany) and M1 (common transformation product of irgarol and terbutryn) from Ciba Specialty Chemicals. Irgarol sulfoxide (2-methylsulfinyl-4-tertbutylamino-6-cyclopropylamino-s-triazine) was chemically synthesized. The surrogate standards irgarol-d9 and terbutryn-d5 were purchased from Dr. Ehrenstorfer (Augsburg, Germany). DMSO-d6 (isotopic enrichment 99.96%), used as NMR solvent, was purchased from Deutero GmbH (Kastellaun, Germany). Other solvents like acetonitril, n-heptane, acetone, 245
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Table 2. Overview of the Batch Experiments with Different Initial Concentrations, Their Transformation Kinetic Parameters, And the Applicationsa compound
spike conc. [μg L−1]
n
r2
t1/2 * [d]
irgarol terbutryn irgarol sulfoxide, terbutryn sulfoxide
0.025 0.5 2.0 0.5 2.0 200 200 0.2
1 4 3 4 3 1 1 2
0.945 0.987 0.986 0.988 0.994 nd nd nd
6.2 ± 0.8 10.5 ± 0.5 1.9 ± 0.1 7.7 ± 0.3 2.1 ± 0.1 nd nd nd
irgarol
terbutryn
a
kbiol [L gSS−1 d−1] 0.3 0.2 0.9 0.2 0.8
± 0.1 ± 0.02 ± 0.1 ± 0.03 ± 0.1 nd nd nd
dateb sludge sludge sludge sludge sludge sludge sludge sludge
Oct. 2012 Oct. 2012 Jan. 2011 Oct. 2012 Jan. 2011 Oct. 2010 Oct. 2010 Oct. 2012
application mass balance, mass balance, mass balance, mass balance, mass balance, identification identification stability test
kinetic kinetic kinetic kinetic kinetic
nd: not determined, btime point at which the sludge was taken, * p < 0.0001.
Labortechnik GmbH, Essen, Germany) based on the retention time and signal slope of the peaks in the chromatogram. An aliquot of each fraction was analyzed via HR-LC-MS (LTQ Orbitrap Velos) and HPLC-UV to examine the composition and the purity of the fractions. The purity of the fraction of the synthesized irgarol TP was estimated to be higher than 95%. The product was freeze-dried to complete dryness and was used to prepare a stock solution for quantitative analyses as well as for identification by HR-MS and NMR measurements. Identification of the Irgarol TP via NMR. NMR spectra were recorded at a Bruker AVANCE III 700 instrument (Rheinstetten, Germany). Therefore, about 12 mg of irgarol and 3.7 mg of the synthesized irgarol TP were dissolved in 0.8 mL of DMSO-d6. 1H NMR spectra were measured at 700 MHz, and 13C NMR spectra were measured at 176 MHz. Homo- and heteronuclear chemical shift correlations were determined by homonuclear 2D-1H,1H−COSY (correlated spectroscopy) and heteronuclear 2D-1H,13C-HSQC (hetero single quant correlation). The chemical shifts were reported in ppm relative to tetramethylsilane. All spectra and correlation experiments were performed at 298.3 K. More detailed information is provided in the SI. Environmental Samples from WWTPs as well as Streams and Rivers. Composite samples (24 h) of influent and effluent were collected over a period of three days from four conventional WWTPs. Grab samples of streams and small rivers were collected over an interval of two weeks from different small rivers in Hesse, Germany, in June 2012. Parameters such as pH, temperature, electrical conductivity, and oxygen content were measured on each sampling location (SI Table S1). The samples of streams and small rivers were taken during wet and dry weather periods. All samples were transported in amber and solvent-rinsed glass bottles under cool conditions to the laboratory, where they were immediately filtered through glass-fiber filters (GF/6, Whatman), and stored at 4 °C. The extraction (SPE) was accomplished within 24 h. More information about the four WWTPs and the sample locations of the streams and small rivers is provided in SI. Analysis of Batch Samples and Environmental Samples from WWTPs as well as Streams and Rivers. The batch samples (45 mL), wastewater samples (100 mL influent, 200 mL effluent), and stream and small river samples (500 mL) were enriched with SPE. For this, Oasis HLB cartridges (200 mg, Waters, Milford, MA) were washed and conditioned using 1 × 2 mL n-heptan, 1 × 2 mL acetone, 3 × 2 mL methanol, and 4 × 2 mL groundwater (pH 6.9). The aqueous samples were spiked with 25 ng of the surrogate standards (irgarol-d9 and terbutryn-d5). The samples were
Orbitrap Velos, Thermo Scientific, Bremen, Germany) using electrospray ionization (ESI) in the positive ion mode to obtain exact masses of parent and fragment ions of irgarol, terbutryn and their respective TPs. The LTQ-Orbitrap-MS was coupled to a Thermo Sientific Accela liquid chromatography system (Accela pump and autosampler). The synthesized irgarol TP (see below) and the commercially available terbutryn TP were also analyzed by LTQ-Orbitrap-MS to compare the mass spectra with those of the respective TPs from the batch experiments. More detailed information about the chromatographic conditions, the MS source parameters and the datadependent acquisition parameters is provided in the SI. Chemical Synthesis of Irgarol TP. Due to the biocidal effect of irgarol and terbutryn, large concentrations could not be spiked in the batch experiments. Hence, the amount of formed TPs was too low for their isolation from the biological batch experiments. Fortunately, the proposed terbutryn TP was commercially available, but not the irgarol TP. Therefore, the irgarol TP was chemically synthesized by a method described by Kim et al.29 (Figure 1). Briefly, irgarol (5 mg) and FeCl3
Figure 1. Chemical synthesis of irgarol TP from irgarol.
(∼1.5 mg) were dissolved in acetonitrile (2 mL). H5IO6 (5 mg) was added, and after a reaction time of 4 h the reaction was quenched using a saturated aqueous solution of Na2S2O3 (5 mL). Irgarol TP was chemical synthesized with yield up to 50%. The reaction product was characterized by LTQ Orbitrap Velos. Isolation of the synthesized Irgarol TP via semipreparative HPLC-UV. The reaction product was isolated by semipreparative HPLC-UV. A Knauer HPLC-UV system (Knauer Smartline system, Berlin, Germany) consisting of a Smartline Autosampler 3950, Smartline Manager 5050 (degasser), Smartline Pump 1000, column oven and Smartline PDA Detector 2850 (operated at 254 nm) was used. Isolation of individual analytes was achieved by chromatographic separation on a semipreparative Synergi Fusion-RP 80 Å column (250 × 10 mm, 4 μm) from Phenomenex (Aschaffenburg, Germany). Details on the chromatographic conditions are provided in the SI. Individual fractions were collected using an automated sample collector (Büchi Fraction Collector C-660, Büchi 246
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Figure 2. MS2 spectrum of IGTP269 (A) and TBTP257 (B) determined by LTQ-Orbitrap-MS (ESI positive ion mode). Proposed chemical structures of fragment ions with the respective exact masses and mass error (in parts-per-million) are shown.
passed through the preconditioned cartridges at a flow rate of approximately 5 mL min−1. Then cartridges were dried using a stream of nitrogen. The enriched analytes were eluted with 4 × 2 mL methanol/acetone 60/40 (v/v). Finally, the extracts were evaporated up to 100 μL under a gentle stream of nitrogen and filled to the final volume of 1 mL with acetonitrile. These samples were analyzed by LC-tandem MS using an Agilent HPLC system (Agilent 1200 Series, Agilent Technologies, Waldbronn, Germany) coupled with a triple quadrupole linear ion trap mass spectrometer (Qq-LIT-MS, MDS SCIEX 4000 Q TRAP, Applied Biosystems, Langen, Germany) in positive ionization mode. More information about the chromatographic conditions, the method validation and quantification is provided in the SI.
Bacterial Screening Toxicity Test. Luminescent bacteria inhibition test with Vibrio f ischeri was applied for irgarol, terbutryn, and their TPs to assess the bacterial toxicity of these substances.30,31 Therefore, different amounts (10 ng, 100 ng, and 200 ng) of all four substances and a solvent control were sprayed on the HPTLC plate (Silica gel 60 F 254, Merck, Darmstadt, Germany) using an automatic TLC sampler (ATS 4, CAMAG, Muttenz, Switzerland). The TLC plate was automatically dipped for 2 s at a speed of 30 mm s−1 into a solution with Vibrio f ischeri. Afterward, the bioluminescence was continuously monitored with an exposure time of 60 s using a cooled CCD camera (CAMAG Bioluminizer). 247
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
■
Article
RESULTS AND DISCUSSION Identification of Irgarol TP and Terbutryn TP. Irgarol and terbutryn were transformed in the aerobic batch experiments with activated sludge to one TP each. On the basis of the analyses via HR-MS, proposed structures were developed for both TPs. The final confirmation of the proposed chemical structures of TPs was achieved by the comparison of mass spectra and retention times with reference standards. The reference standard for the irgarol TP was not commercially available and was therefore synthesized. Finally, 1D and 2DNMR experiments were used to confirm its chemical structure. Identification by LTQ-Orbitrap-MS. The determination of the exact masses from the full scan measurements using LTQ-Orbitrap-MS exhibited that irgarol (C11H19N5S) and terbutryn (C10H19N5S) were transformed into one irgarol TP (IGTP269) (C11H19ON5S) and one terbutryn TP (TBTP257) (C10H19ON5S), respectively. Hence, irgarol and terbutryn were oxidized to a TP containing one additional oxygen atom whose position was further elucidated by high resolution MS2 and MS3 experiments. From the MSn experiments it could be concluded that the triazine ring remained unchanged, because only the fragments with elimination of the side chains are visible. The HR-mass spectra of IGTP269 and TBTP257 (Figure 2) show the fragment ions [M+H−C4H8]+ and [M+H−CH2S]+ with a typical leaving group, the tertiary butene group (56.0626 Da) and the methanethial group (45.9877 Da), respectively. This is also obvious in the HR-MS-spectra of irgarol and terbutryn. Furthermore, the HR-mass spectra of IGTP269 and TBTP257 show the fragment ion [M+H−H2O]+ which indicated a cleavage of H2O (18.0107 Da). As the oxygen atom might be located at various positions of the original irgarol and terbutryn, there are several structural possibilities of the TPs formed. If the oxidation took place at the cyclopropyl group of irgarol, in the mass spectrum of the TP a m/z 226 fragment ion [M+H−C2H4O]+ would be expected as described in Lam et al.25 However, this was not the case. An oxidation of the ethyl group of terbutryn was also not observed. As the fragment ion [M+H−C4H8]+ was visible in the HR-MS-spectra of irgarol and terbutryn, it can be ruled out that an oxidation occurred at the tertiary butyl moiety or the nitrogen atoms. However, the oxidation could still occur at the sulfur leading to sulfoxides. The IGTP269 and TBTP257 from the batch experiments were unstable upon the addition of acid (e.g., formic acid), because they react further to the 2-hydroxy derivatives (Figure 3).
available. Hence, irgarol and terbutryn exhibited a similar transformation process which is also reflected in the transformation kinetics of both substances. For methylthio-striazines (simetryn, dimethametryn, and prometryn) a formation of sulfoxides has also been proposed.32 Transformation Kinetics of Irgarol and Terbutryn in Activated Sludge. Irgarol and terbutryn were exponentially dissipated in the aerobic batch experiments with activated sludge (Figure 4). This decrease can be described by pseudo
Figure 4. Degradation [%] of irgarol (A) and terbutryn (B) in batch systems using diluted activated sludge (1:10 with effluent, 0.40 gSS L−1) at different times. Irgarol and terbutryn were spiked in different concentrations.
first-order kinetics.33 As shown below, the dissipation is mainly caused by transformation since sorption is of minor importance. In the batch experiments with initial concentrations of 2 μg L−1 the calculated degradation half-lives (t1/2) of irgarol and terbutryn were 1.9 ± 0.1 and 2.1 ± 0.1 days, respectively. The transformation rate constant (kbiol) was 0.9 ± 0.1 L gSS−1 d−1 for irgarol and 0.8 ± 0.1 L gSS−1 d−1 for terbutryn, suggesting according to Joss et al.33 a removal of about 40−60% in WWTPs. However, in batch experiments with sludge taken from the same WWTP, but at another date (Oct. 2012 instead of Jan. 2011) significantly higher t1/2 and lower kbiol values were achieved (Table 2). Since, no significant difference in the transformation rate was found in the batch experiments with initial concentrations of 0.025 μg L−1 and 0.5 μg L−1 and the sludge taken at the same day, it is very likely that the difference was caused by variations of the sludge characteristics. With sludge taken at the same day, the transformation rate constants of irgarol and terbutryn were always comparable. Thus, a normalization of the transformation rates with those of well-known micropollutants might be a promising solution. Systematic studies are recommended to elucidate the dependency of transformation rate constants on the characteristics of the activated sludge. Mass Balances. Irgarol decreased up to 45% and terbutryn up to 55% within 9 days of incubation in the batch experiments with initial concentrations of 0.5 μg L−1 (Figure 5). The
Figure 3. Proposed transformation pathway of methylthio-s-triazines in aerobic batch experiments with activated sludge.
The chemical structure of IGTP269 and TBTP257 was identified using suitable reference standards (irgarol sulfoxide and terbutryn sulfoxide), as their retention times, exact masses and fragmentation patterns correspond to the TPs from the batch experiments. The reference standard of IGTP269 was synthesized (irgarol sulfoxide), as it was not commercially 248
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
sulfoxides. Dealkylated products such as M1 were not found. According to the low sludge-water distribution coefficient Kd,sec values (irgarol: Kd,sec = 140 ± 10 L kgTSS−1 and terbutryn: Kd,sec = 160 ± 10 L kgTSS−1)34 the sorbed amount of irgarol and terbutryn might increase the mass balances by around 6% (details in the SI) using diluted activated sludge (TSS: 0.4 gSS L−1 caused by dilution 1:10) in the batch experiments. Due to their increased polarity the influence of sorption of the sulfoxide TPs on the mass balance can be expected to be negligible. Since no decrease of the concentrations was observed in the filtered control samples (SI Figure S3), the transformation is most likely associated with a microbial (enzymatic) activity. Similar batch experiments spiked with irgarol sulfoxide and terbutryn sulfoxide exhibited that both TPs are quite persistent as no significant decrease of concentrations were observed within 9 days (SI Figure S4). Although it is reported that methylthio-s-triazines can be transformed to 2-hydroxy derivatives via sulfoxides and sulfones in aerobic and flooded soil32,35 as well as by isolated bacteria (bacterial strain JUN7)32 from soil, a further oxidation of sulfoxides to sulfones was not observed in the batch experiments with activated sludge. Only after acidification (pH 3.5 by adding formic acid) irgarol sulfoxide and terbutryn sulfoxide were transformed into the 2hydroxy derivatives. Furthermore, M1 (major TP of irgarol and terbutryn identified in other degradation studies18,20,28) was not detected in the batch experiments. Figure 6 illustrates irgarol and terbutryn sulfoxide TPs, formed in contact with activated sludge, as essential part of the transformation pathway of irgarol and terbutryn in contact with sediments and by solar irradiation. Environmental Occurrence of Transformation Products. The environmental relevance of the identified polar sulfoxide TPs as well as the known TP M1 was elucidated by examining their formation in conventional WWTPs as well as their occurrence in streams and small rivers containing an elevated portion of treated wastewater of about 10 to 65% (data
Figure 5. Mass balance [%] of irgarol (A) and terbutryn (B) in batch systems using diluted activated sludge (1:10 with effluent, 0.40 gSS L−1). Irgarol and terbutryn were spiked at a concentration of 0.5 μg L−1. Calculations were done on a molar basis (mol L−1).
decrease of irgarol and terbutryn was accompanied by a formation of irgarol sulfoxide and terbutryn sulfoxide, respectively. The irgarol sulfoxide was formed up to 40% and the terbutryn sulfoxide up to 35% of the initial concentration of irgarol and terbutryn (0.5 μg L−1). Hence, the almost closed mass balances of 95% (irgarol) and 80% (terbutryn) indicate a nearly quantitative transformation to the corresponding
Figure 6. Transformation of irgarol and terbutryn to sulfoxide derivatives in contact with activated sludge in comparison with previously reported TPs formed in contact with sediments or by solar irradiation in water. 249
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Table 3. Occurrence of Irgarol, Terbutryn, And Their TPs in Raw and Treated Wastewater, Concentrations of Samples of One Day and of Three Days Togetherd WWTP 1
WWTP 2
WWTP 3
WWTP 4
LOQ
−1 a
influent
[ng L ]
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent effluent
[ng L−1]c [ng L−1] [ng L−1]
irgarol
irgarol sulfoxide
terbutryn
terbutryn sulfoxide
M1
3.4 ± 0.04 2.0 ± 0.3 1.5 ± 0.6 2.3 3.0 ± 1.1 3.0 ± 0.4 2.8 ± 0.2 2.9 8.9 ± 0.8 2.9 ± 0.2 4.3 ± 0.5 5.4 10.2 ± 2.2 9.3 ± 1.0 7.5 ± 5.1 9.0 3.8 ± 0.3 8.0 ± 0.3 4.7 ± 0.5 5.5 5.6 ± 1.3 7.3 ± 1.3 9.7 ± 1.6 7.5 9.2 ± 1.6 26 ± 4 13 ± 3 16 17 ± 4 15 ± 11 11 ± 11 14 2 1
Transformation of Biocides Irgarol and Terbutryn in the Biological Wastewater Treatment Agnessa Luft,† Manfred Wagner,‡ and Thomas A. Ternes*,† †
Federal Institute of Hydrology (BfG), 56068 Koblenz, Germany Max Planck Institute for Polymer Research, 55128 Mainz, Germany
‡
S Supporting Information *
ABSTRACT: The biocides irgarol and terbutryn enter the wastewater treatment plant (WWTP) via combined sewer systems after leaching from coatings and paints of materials. In this study, the biotransformation of irgarol and terbutryn was examined in aerobic batch experiments with activated sludge taken from the nitrification zone of a conventional WWTP, since currently there is no information about the fate of irgarol and terbutryn in biological wastewater treatment. Both, irgarol and terbutryn were transformed into one main transformation product (TP) following pseudo first-order kinetics. The TPs were tentatively identified by high-resolution mass spectrometry (HR-MS) to be irgarol sulfoxide and terbutryn sulfoxide. The final confirmation of the proposed chemical structures of the TPs was achieved by a comparison of mass spectra and nuclear magnetic resonance (NMR) spectra with those of authentic reference standards (e.g., synthesized). An analytical method for the sensitive quantification of irgarol, terbutryn and their TPs in environmental samples was developed based on solid phase extraction (SPE) and LC tandem MS detection. Irgarol sulfoxide and terbutryn sulfoxide were detected in the effluents (average concentrations up to 22 ng L−1 and 65 ng L−1) of all four investigated WWTPs as well as in streams and small rivers (up to 14 ng L−1 and 34 ng L−1). Luminescent bacteria inhibition test with Vibrio f ischeri exhibited that the TPs irgarol sulfoxide and terbutryn sulfoxide feature a similar bacterial toxicity than the parent compounds.
■
INTRODUCTION Irgarol (2-methylthio-4-tert-butylamino-6-cyclopropylamino-striazine) and terbutryn (2-methylthio-4-tert-butylamino-6-ethylamino-s-triazine) belong to the s-triazines. Irgarol is well-known as active agent in antifouling paints applied to the hulls of ships and boats, since the late 1980s it is often used as a substitute for tributyltin (TBT).1 Terbutryn is known as pesticide in the agriculture which is banned for agricultural use since 2003 in the EU. However, both compounds are still being widely used as biocidal compounds in building materials.2,3 According to the Biocidal Products Directive 98/8/EC biocides are “active substances [...] intended to destroy, deter, render harmless, prevent the action of, or otherwise exert a controlling effect on any harmful organism”.4 Irgarol and terbutryn are applied as algicides in surface coatings of buildings, such as paints and renders, to avoid growth of algae.5,6 A continuously leaching of biocides from surface coatings is intended, since it is the prerequisite for their growth-inhibiting effects of algae.2,7 Therefore, irgarol and terbutryn can be discharged directly into surface waters via runoff water of urban areas and in case of irgarol also via leaching from boat coatings. However, the majority of these biocides reaches municipal WWTPs via combined sewer systems.7,8 Due to their incomplete elimination in the WWTPs, both irgarol and terbutryn are released © 2013 American Chemical Society
into surface waters. Irgarol and terbutryn have been detected in WWTP effluents up to 22 ng L−1 and 123 ng L−1 and in surface waters up to 11 ng L−1 and 169 ng L−1,9 respectively. In a sampling period from 2003 to 2006, Quednow et al.10 determined terbutryn in four small rivers in Germany with mean concentrations up to 0.58 μg L−1, although the use of terbutryn in agriculture had been forbidden. As required by the Water Framework Directive (WFD) 2000/60/EC of the European Community environmental quality standards (EQS) are established to achieve a good chemical status of European water bodies. The annual average EQS of irgarol and terbutryn in streams and rivers reported by the EU (Directive 2013/39/EU, 12.08.2013) are as low as 2.5 ng L−1 and 65 ng L−1, respectively. These low EQS are based on the toxicity of irgarol and terbutryn to higher plants, algae, and aquatic epibiotic organism by inhibition of photosynthesis.11−16 Freshwater mesocosm studies indicated the effects of irgarol on chlorophytes (EC50 = 0.34 μg L−1), on species of cyclopoid copepods (EC50 = 0.09 μg L−1) and ostracods (EC50 = Received: Revised: Accepted: Published: 244
August 8, 2013 November 5, 2013 November 8, 2013 December 12, 2013 dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
0.11 μg L−1).11 Furthermore, effects were found on macrophytes with EC50 values of 0.21 μg L−1 for the species Myriophyllum verticillatum.17 Terbutryn exhibits a toxicity to aquatic organisms such as microalgae with 72 h-EC50 values of 2 μg L−1 for the species Selenastrum capricornutum.13 Previous studies highlight the possibility of biodegradation,18,19 photodegradation,20−24 and chemically catalyzed degradation25−27 of irgarol and terbutryn. Irgarol was transformed mainly via the mechanism of N-dealkylation into the stable degradation product M1 (2-methylthio-4-tert-butylamino-6-amino-s-triazine) during biodegradation by the white rot fungus Phanerochaete chrysosporium,18 photodegradation20,22,24 and mercuric chloride-catalyzed hydrolysis26 in water. Further byproducts during photodegradation of irgarol were reported as, for example, hydroxy (2-hydroxy-4-tert-butylamino-6-cyclopropylamino-s-triazine) and sulfone (2-methylsulfonyl-4-tertbutylamino-6-cyclopropylamino-s-triazine) derivatives.22 Much less studies exist for the degradation of terbutryn. Muir et al.19 reported degradation half-lives for terbutryn of 240 and 180 days in water with pond and river sediments under laboratory conditions, respectively. The major degradation product was hydroxy-terbutryn (2-hydroxy-4-tert-butylamino-6ethylamino-s-triazine), and to a minor extent terbutryn sulfoxide (2-methylsulfinyl-4-tert-butylamino-6-ethylamino-striazine) and N-deethylated terbutryn (2-methylthio-4-tertbutylamino-6-amino-s-triazine) were found.19 The latter has the same chemical structure as M1, the main TP known of the degradation of irgarol. In field studies N-deethylated terbutryn was a major product in water and sediment extracts in farm ponds.28 However, to the best of our knowledge, no studies reported the biodegradation of irgarol and terbutryn during biological wastewater treatment. Biodegradation studies of biocides imply a particular challenge because elevated concentrations of biocides have to be avoided as they might lead to an inhibition of the microbial activity. Hence, the limited possibility for the use of elevated concentrations of biocides in batch experiment hinders the quantitative isolation of TPs for their identification. The objective of the current study was to investigate the biodegradation of irgarol and terbutryn in contact with activated sludge at environmentally relevant concentrations. Identification of biotic TPs was accomplished by a two step approach using LC-HR-MS for the elucidation of tentative structures and the comparison with (synthesized) reference standards for final confirmation. Furthermore, the importance of WWTPs for the formation of irgarol and terbutryn TPs and their occurrence in surface waters were determined.
Table 1. Overview of the Practical Experiments TP identification
• • • •
transformation kinetic/mass balance
• batch experiments with activated sludge • quantification of target compounds and TPs via LC-tandem MS, direct injection
environmental samples from WWTPs as well as streams and rivers bacterial screening toxicity test
• quantification of target compounds and TPs via LC-tandem MS after SPE
batch experiments with activated sludge TP identification via HR-MS chemical synthesis of irgarol TP isolation of the synthesized irgarol TP via semipreparative HPLC-UV • confirmation of the chemical structure via NMR
• luminescent bacteria inhibition test with Vibrio f ischeri analyses via thin layer chromatography (TLC)
and methanol were picograde and purchased from LGC Promochem (Wesel, Germany). Iron(III) chloride (>97%), periodic acid, sodium thiosulfate, and ammonium formate (purum grade) were obtained from Sigma-Aldrich (Schnelldorf, Germany). Formic acid (p.a.) was purchased from Merck (Darmstadt, Germany). Ultra pure water (Milli-Q water) was obtained from a Milli-Q system (Millipore, Billerica, MA). Details on the standard solutions are given in the Supporting Information (SI). Batch Experiments of Irgarol and Terbutryn in Activated Sludge. For the batch experiments, sewage sludge was taken from a municipal WWTP (WWTP 5, more details in the SI) as a grab sample from the nitrification zone of the activated sludge unit. The activated sludge (total suspended solids (TSS): 4 gSS L−1, total organic carbon (TOC): 0.3 g gSS−1) was transported to the laboratory, where batch experiments were started on the same day. Sludge (100 mL) was poured into 1 L amber glass bottles and filled up with 900 mL of WWTP effluent (kinetic experiments) or groundwater (TP identification). The used groundwater is free of all target compounds. The sludge was diluted to minimize the effect of sorption. The amber glass bottles were constantly purged with a mixture of CO2 and air at a definite ratio (1:20). This was done to adjust a stable pH (pH 7.1 ± 0.1) and oxic conditions. The sludge was continuously stirred during the experiments at a temperature of 22 °C ± 1 °C. The analytes (concentrations see Table 2) were spiked in the diluted sludge after an equilibration time of 2 h. In addition, analytes were spiked to filtered effluent (0.45 μm cellulose nitrate filters, Sartorius Stedim Biotech, Göttingen, Germany) without sludge (abiotic control). Nonspiked sludge samples (blank samples) were used as reference samples for the identification of TPs by HR-MS as well as for the determination of the initial analyte concentrations. Samples were taken immediately after spiking and at definite time intervals. For measurements (HR-MS) with direct injection, samples were filtered through 0.45 μm syringe filters (cellulose nitrate filters, Spartan, Whatman), and stored in the freezer (−25 °C) until analysis. For the kinetic experiments, samples (50 mL) were taken and filtered through glass fiber filters (GF/6, Whatman), and stored at 4 °C. Within 24 h the samples were extracted using SPE (see below). Identification of TPs via High Resolution Mass Spectrometry. The nonenriched batch samples were analyzed by linear ion trap quadrupole (LTQ)-Orbitrap-MS (LTQ
■
MATERIALS AND METHODS A short overview of the practical experiments is presented in Table 1 listed in chronological order. Chemicals. Irgarol was purchased from Sigma-Aldrich (Schnelldorf, Germany), terbutryn from Dr. Ehrenstorfer GmbH (Augsburg, Germany), terbutryn sulfoxide from MicroCombiChem e.K. (Wiesbaden, Germany) and M1 (common transformation product of irgarol and terbutryn) from Ciba Specialty Chemicals. Irgarol sulfoxide (2-methylsulfinyl-4-tertbutylamino-6-cyclopropylamino-s-triazine) was chemically synthesized. The surrogate standards irgarol-d9 and terbutryn-d5 were purchased from Dr. Ehrenstorfer (Augsburg, Germany). DMSO-d6 (isotopic enrichment 99.96%), used as NMR solvent, was purchased from Deutero GmbH (Kastellaun, Germany). Other solvents like acetonitril, n-heptane, acetone, 245
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Table 2. Overview of the Batch Experiments with Different Initial Concentrations, Their Transformation Kinetic Parameters, And the Applicationsa compound
spike conc. [μg L−1]
n
r2
t1/2 * [d]
irgarol terbutryn irgarol sulfoxide, terbutryn sulfoxide
0.025 0.5 2.0 0.5 2.0 200 200 0.2
1 4 3 4 3 1 1 2
0.945 0.987 0.986 0.988 0.994 nd nd nd
6.2 ± 0.8 10.5 ± 0.5 1.9 ± 0.1 7.7 ± 0.3 2.1 ± 0.1 nd nd nd
irgarol
terbutryn
a
kbiol [L gSS−1 d−1] 0.3 0.2 0.9 0.2 0.8
± 0.1 ± 0.02 ± 0.1 ± 0.03 ± 0.1 nd nd nd
dateb sludge sludge sludge sludge sludge sludge sludge sludge
Oct. 2012 Oct. 2012 Jan. 2011 Oct. 2012 Jan. 2011 Oct. 2010 Oct. 2010 Oct. 2012
application mass balance, mass balance, mass balance, mass balance, mass balance, identification identification stability test
kinetic kinetic kinetic kinetic kinetic
nd: not determined, btime point at which the sludge was taken, * p < 0.0001.
Labortechnik GmbH, Essen, Germany) based on the retention time and signal slope of the peaks in the chromatogram. An aliquot of each fraction was analyzed via HR-LC-MS (LTQ Orbitrap Velos) and HPLC-UV to examine the composition and the purity of the fractions. The purity of the fraction of the synthesized irgarol TP was estimated to be higher than 95%. The product was freeze-dried to complete dryness and was used to prepare a stock solution for quantitative analyses as well as for identification by HR-MS and NMR measurements. Identification of the Irgarol TP via NMR. NMR spectra were recorded at a Bruker AVANCE III 700 instrument (Rheinstetten, Germany). Therefore, about 12 mg of irgarol and 3.7 mg of the synthesized irgarol TP were dissolved in 0.8 mL of DMSO-d6. 1H NMR spectra were measured at 700 MHz, and 13C NMR spectra were measured at 176 MHz. Homo- and heteronuclear chemical shift correlations were determined by homonuclear 2D-1H,1H−COSY (correlated spectroscopy) and heteronuclear 2D-1H,13C-HSQC (hetero single quant correlation). The chemical shifts were reported in ppm relative to tetramethylsilane. All spectra and correlation experiments were performed at 298.3 K. More detailed information is provided in the SI. Environmental Samples from WWTPs as well as Streams and Rivers. Composite samples (24 h) of influent and effluent were collected over a period of three days from four conventional WWTPs. Grab samples of streams and small rivers were collected over an interval of two weeks from different small rivers in Hesse, Germany, in June 2012. Parameters such as pH, temperature, electrical conductivity, and oxygen content were measured on each sampling location (SI Table S1). The samples of streams and small rivers were taken during wet and dry weather periods. All samples were transported in amber and solvent-rinsed glass bottles under cool conditions to the laboratory, where they were immediately filtered through glass-fiber filters (GF/6, Whatman), and stored at 4 °C. The extraction (SPE) was accomplished within 24 h. More information about the four WWTPs and the sample locations of the streams and small rivers is provided in SI. Analysis of Batch Samples and Environmental Samples from WWTPs as well as Streams and Rivers. The batch samples (45 mL), wastewater samples (100 mL influent, 200 mL effluent), and stream and small river samples (500 mL) were enriched with SPE. For this, Oasis HLB cartridges (200 mg, Waters, Milford, MA) were washed and conditioned using 1 × 2 mL n-heptan, 1 × 2 mL acetone, 3 × 2 mL methanol, and 4 × 2 mL groundwater (pH 6.9). The aqueous samples were spiked with 25 ng of the surrogate standards (irgarol-d9 and terbutryn-d5). The samples were
Orbitrap Velos, Thermo Scientific, Bremen, Germany) using electrospray ionization (ESI) in the positive ion mode to obtain exact masses of parent and fragment ions of irgarol, terbutryn and their respective TPs. The LTQ-Orbitrap-MS was coupled to a Thermo Sientific Accela liquid chromatography system (Accela pump and autosampler). The synthesized irgarol TP (see below) and the commercially available terbutryn TP were also analyzed by LTQ-Orbitrap-MS to compare the mass spectra with those of the respective TPs from the batch experiments. More detailed information about the chromatographic conditions, the MS source parameters and the datadependent acquisition parameters is provided in the SI. Chemical Synthesis of Irgarol TP. Due to the biocidal effect of irgarol and terbutryn, large concentrations could not be spiked in the batch experiments. Hence, the amount of formed TPs was too low for their isolation from the biological batch experiments. Fortunately, the proposed terbutryn TP was commercially available, but not the irgarol TP. Therefore, the irgarol TP was chemically synthesized by a method described by Kim et al.29 (Figure 1). Briefly, irgarol (5 mg) and FeCl3
Figure 1. Chemical synthesis of irgarol TP from irgarol.
(∼1.5 mg) were dissolved in acetonitrile (2 mL). H5IO6 (5 mg) was added, and after a reaction time of 4 h the reaction was quenched using a saturated aqueous solution of Na2S2O3 (5 mL). Irgarol TP was chemical synthesized with yield up to 50%. The reaction product was characterized by LTQ Orbitrap Velos. Isolation of the synthesized Irgarol TP via semipreparative HPLC-UV. The reaction product was isolated by semipreparative HPLC-UV. A Knauer HPLC-UV system (Knauer Smartline system, Berlin, Germany) consisting of a Smartline Autosampler 3950, Smartline Manager 5050 (degasser), Smartline Pump 1000, column oven and Smartline PDA Detector 2850 (operated at 254 nm) was used. Isolation of individual analytes was achieved by chromatographic separation on a semipreparative Synergi Fusion-RP 80 Å column (250 × 10 mm, 4 μm) from Phenomenex (Aschaffenburg, Germany). Details on the chromatographic conditions are provided in the SI. Individual fractions were collected using an automated sample collector (Büchi Fraction Collector C-660, Büchi 246
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Figure 2. MS2 spectrum of IGTP269 (A) and TBTP257 (B) determined by LTQ-Orbitrap-MS (ESI positive ion mode). Proposed chemical structures of fragment ions with the respective exact masses and mass error (in parts-per-million) are shown.
passed through the preconditioned cartridges at a flow rate of approximately 5 mL min−1. Then cartridges were dried using a stream of nitrogen. The enriched analytes were eluted with 4 × 2 mL methanol/acetone 60/40 (v/v). Finally, the extracts were evaporated up to 100 μL under a gentle stream of nitrogen and filled to the final volume of 1 mL with acetonitrile. These samples were analyzed by LC-tandem MS using an Agilent HPLC system (Agilent 1200 Series, Agilent Technologies, Waldbronn, Germany) coupled with a triple quadrupole linear ion trap mass spectrometer (Qq-LIT-MS, MDS SCIEX 4000 Q TRAP, Applied Biosystems, Langen, Germany) in positive ionization mode. More information about the chromatographic conditions, the method validation and quantification is provided in the SI.
Bacterial Screening Toxicity Test. Luminescent bacteria inhibition test with Vibrio f ischeri was applied for irgarol, terbutryn, and their TPs to assess the bacterial toxicity of these substances.30,31 Therefore, different amounts (10 ng, 100 ng, and 200 ng) of all four substances and a solvent control were sprayed on the HPTLC plate (Silica gel 60 F 254, Merck, Darmstadt, Germany) using an automatic TLC sampler (ATS 4, CAMAG, Muttenz, Switzerland). The TLC plate was automatically dipped for 2 s at a speed of 30 mm s−1 into a solution with Vibrio f ischeri. Afterward, the bioluminescence was continuously monitored with an exposure time of 60 s using a cooled CCD camera (CAMAG Bioluminizer). 247
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
■
Article
RESULTS AND DISCUSSION Identification of Irgarol TP and Terbutryn TP. Irgarol and terbutryn were transformed in the aerobic batch experiments with activated sludge to one TP each. On the basis of the analyses via HR-MS, proposed structures were developed for both TPs. The final confirmation of the proposed chemical structures of TPs was achieved by the comparison of mass spectra and retention times with reference standards. The reference standard for the irgarol TP was not commercially available and was therefore synthesized. Finally, 1D and 2DNMR experiments were used to confirm its chemical structure. Identification by LTQ-Orbitrap-MS. The determination of the exact masses from the full scan measurements using LTQ-Orbitrap-MS exhibited that irgarol (C11H19N5S) and terbutryn (C10H19N5S) were transformed into one irgarol TP (IGTP269) (C11H19ON5S) and one terbutryn TP (TBTP257) (C10H19ON5S), respectively. Hence, irgarol and terbutryn were oxidized to a TP containing one additional oxygen atom whose position was further elucidated by high resolution MS2 and MS3 experiments. From the MSn experiments it could be concluded that the triazine ring remained unchanged, because only the fragments with elimination of the side chains are visible. The HR-mass spectra of IGTP269 and TBTP257 (Figure 2) show the fragment ions [M+H−C4H8]+ and [M+H−CH2S]+ with a typical leaving group, the tertiary butene group (56.0626 Da) and the methanethial group (45.9877 Da), respectively. This is also obvious in the HR-MS-spectra of irgarol and terbutryn. Furthermore, the HR-mass spectra of IGTP269 and TBTP257 show the fragment ion [M+H−H2O]+ which indicated a cleavage of H2O (18.0107 Da). As the oxygen atom might be located at various positions of the original irgarol and terbutryn, there are several structural possibilities of the TPs formed. If the oxidation took place at the cyclopropyl group of irgarol, in the mass spectrum of the TP a m/z 226 fragment ion [M+H−C2H4O]+ would be expected as described in Lam et al.25 However, this was not the case. An oxidation of the ethyl group of terbutryn was also not observed. As the fragment ion [M+H−C4H8]+ was visible in the HR-MS-spectra of irgarol and terbutryn, it can be ruled out that an oxidation occurred at the tertiary butyl moiety or the nitrogen atoms. However, the oxidation could still occur at the sulfur leading to sulfoxides. The IGTP269 and TBTP257 from the batch experiments were unstable upon the addition of acid (e.g., formic acid), because they react further to the 2-hydroxy derivatives (Figure 3).
available. Hence, irgarol and terbutryn exhibited a similar transformation process which is also reflected in the transformation kinetics of both substances. For methylthio-striazines (simetryn, dimethametryn, and prometryn) a formation of sulfoxides has also been proposed.32 Transformation Kinetics of Irgarol and Terbutryn in Activated Sludge. Irgarol and terbutryn were exponentially dissipated in the aerobic batch experiments with activated sludge (Figure 4). This decrease can be described by pseudo
Figure 4. Degradation [%] of irgarol (A) and terbutryn (B) in batch systems using diluted activated sludge (1:10 with effluent, 0.40 gSS L−1) at different times. Irgarol and terbutryn were spiked in different concentrations.
first-order kinetics.33 As shown below, the dissipation is mainly caused by transformation since sorption is of minor importance. In the batch experiments with initial concentrations of 2 μg L−1 the calculated degradation half-lives (t1/2) of irgarol and terbutryn were 1.9 ± 0.1 and 2.1 ± 0.1 days, respectively. The transformation rate constant (kbiol) was 0.9 ± 0.1 L gSS−1 d−1 for irgarol and 0.8 ± 0.1 L gSS−1 d−1 for terbutryn, suggesting according to Joss et al.33 a removal of about 40−60% in WWTPs. However, in batch experiments with sludge taken from the same WWTP, but at another date (Oct. 2012 instead of Jan. 2011) significantly higher t1/2 and lower kbiol values were achieved (Table 2). Since, no significant difference in the transformation rate was found in the batch experiments with initial concentrations of 0.025 μg L−1 and 0.5 μg L−1 and the sludge taken at the same day, it is very likely that the difference was caused by variations of the sludge characteristics. With sludge taken at the same day, the transformation rate constants of irgarol and terbutryn were always comparable. Thus, a normalization of the transformation rates with those of well-known micropollutants might be a promising solution. Systematic studies are recommended to elucidate the dependency of transformation rate constants on the characteristics of the activated sludge. Mass Balances. Irgarol decreased up to 45% and terbutryn up to 55% within 9 days of incubation in the batch experiments with initial concentrations of 0.5 μg L−1 (Figure 5). The
Figure 3. Proposed transformation pathway of methylthio-s-triazines in aerobic batch experiments with activated sludge.
The chemical structure of IGTP269 and TBTP257 was identified using suitable reference standards (irgarol sulfoxide and terbutryn sulfoxide), as their retention times, exact masses and fragmentation patterns correspond to the TPs from the batch experiments. The reference standard of IGTP269 was synthesized (irgarol sulfoxide), as it was not commercially 248
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
sulfoxides. Dealkylated products such as M1 were not found. According to the low sludge-water distribution coefficient Kd,sec values (irgarol: Kd,sec = 140 ± 10 L kgTSS−1 and terbutryn: Kd,sec = 160 ± 10 L kgTSS−1)34 the sorbed amount of irgarol and terbutryn might increase the mass balances by around 6% (details in the SI) using diluted activated sludge (TSS: 0.4 gSS L−1 caused by dilution 1:10) in the batch experiments. Due to their increased polarity the influence of sorption of the sulfoxide TPs on the mass balance can be expected to be negligible. Since no decrease of the concentrations was observed in the filtered control samples (SI Figure S3), the transformation is most likely associated with a microbial (enzymatic) activity. Similar batch experiments spiked with irgarol sulfoxide and terbutryn sulfoxide exhibited that both TPs are quite persistent as no significant decrease of concentrations were observed within 9 days (SI Figure S4). Although it is reported that methylthio-s-triazines can be transformed to 2-hydroxy derivatives via sulfoxides and sulfones in aerobic and flooded soil32,35 as well as by isolated bacteria (bacterial strain JUN7)32 from soil, a further oxidation of sulfoxides to sulfones was not observed in the batch experiments with activated sludge. Only after acidification (pH 3.5 by adding formic acid) irgarol sulfoxide and terbutryn sulfoxide were transformed into the 2hydroxy derivatives. Furthermore, M1 (major TP of irgarol and terbutryn identified in other degradation studies18,20,28) was not detected in the batch experiments. Figure 6 illustrates irgarol and terbutryn sulfoxide TPs, formed in contact with activated sludge, as essential part of the transformation pathway of irgarol and terbutryn in contact with sediments and by solar irradiation. Environmental Occurrence of Transformation Products. The environmental relevance of the identified polar sulfoxide TPs as well as the known TP M1 was elucidated by examining their formation in conventional WWTPs as well as their occurrence in streams and small rivers containing an elevated portion of treated wastewater of about 10 to 65% (data
Figure 5. Mass balance [%] of irgarol (A) and terbutryn (B) in batch systems using diluted activated sludge (1:10 with effluent, 0.40 gSS L−1). Irgarol and terbutryn were spiked at a concentration of 0.5 μg L−1. Calculations were done on a molar basis (mol L−1).
decrease of irgarol and terbutryn was accompanied by a formation of irgarol sulfoxide and terbutryn sulfoxide, respectively. The irgarol sulfoxide was formed up to 40% and the terbutryn sulfoxide up to 35% of the initial concentration of irgarol and terbutryn (0.5 μg L−1). Hence, the almost closed mass balances of 95% (irgarol) and 80% (terbutryn) indicate a nearly quantitative transformation to the corresponding
Figure 6. Transformation of irgarol and terbutryn to sulfoxide derivatives in contact with activated sludge in comparison with previously reported TPs formed in contact with sediments or by solar irradiation in water. 249
dx.doi.org/10.1021/es403531d | Environ. Sci. Technol. 2014, 48, 244−254
Environmental Science & Technology
Article
Table 3. Occurrence of Irgarol, Terbutryn, And Their TPs in Raw and Treated Wastewater, Concentrations of Samples of One Day and of Three Days Togetherd WWTP 1
WWTP 2
WWTP 3
WWTP 4
LOQ
−1 a
influent
[ng L ]
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent
[ng L−1]c [ng L−1]a
effluent
[ng L−1]c [ng L−1]b
influent effluent
[ng L−1]c [ng L−1] [ng L−1]
irgarol
irgarol sulfoxide
terbutryn
terbutryn sulfoxide
M1
3.4 ± 0.04 2.0 ± 0.3 1.5 ± 0.6 2.3 3.0 ± 1.1 3.0 ± 0.4 2.8 ± 0.2 2.9 8.9 ± 0.8 2.9 ± 0.2 4.3 ± 0.5 5.4 10.2 ± 2.2 9.3 ± 1.0 7.5 ± 5.1 9.0 3.8 ± 0.3 8.0 ± 0.3 4.7 ± 0.5 5.5 5.6 ± 1.3 7.3 ± 1.3 9.7 ± 1.6 7.5 9.2 ± 1.6 26 ± 4 13 ± 3 16 17 ± 4 15 ± 11 11 ± 11 14 2 1
