- Email: [email protected]
UV-C processes; Triclosan as a model compound
Science of the Total Environment 601–602 (2017) 397–404
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Photodegradation of micropollutants using V-UV/UV-C processes; Triclosan as a model compound Alfiya Yuval a, Friedler Eran a, Westphal Janin b, Olsson Oliver b, Dubowski Yael a,⁎ a b
Faculty of Civil and Environmental Engineering, Technion, Israel Institute of Technology, Haifa, Israel Institute of Sustainable and Environmental Chemistry, Leuphana University of Lüneburg, Germany
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Triclosan (TCS) photodegradation was faster in aqueous solution than in dry thin films. • Degradation kinetics of aqueous TCS was faster under combined V-UV/UV-C radiation. • Transformation products were identified and their formation/degradation kinetics was quantified. • Toxicity and biodegradability were predicted by QSAR.
a r t i c l e
i n f o
Article history: Received 3 April 2017 Received in revised form 18 May 2017 Accepted 19 May 2017 Available online xxxx Editor: Jay Gan Keywords: Decentralized wastewater treatment Micropollutants Advanced oxidation process Triclosan Transformation products Vacuum ultra-violet
⁎ Corresponding author. E-mail address: [email protected] (D. Yael).
http://dx.doi.org/10.1016/j.scitotenv.2017.05.172 0048-9697/© 2017 Elsevier B.V. All rights reserved.
a b s t r a c t Non-potable reuse of treated wastewater is becoming widespread as means to address growing water scarcity. Removal of micropollutants (MPs) from such water often requires advanced oxidation processes using •OH radicals. •OH can be generated in-situ via water photolysis under vacuum-UV (λ b 200 nm) irradiation. The aim of this study was to investigate the potential of unmasking V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. Triclosan, a biocide of limited biodegradability, served as a model compound for MPs that are not very biodegradable. Its degradation kinetics and identification of intermediate products were investigated under 254 nm and under combined 254/185 nm irradiation both in dry thin films and in aqueous solutions. In the latter, degradation was faster under combined 254/185 nm radiation, although the 185 nm radiation accounted for only 4% of the total UV light intensity. In contrast, triclosan photodegradation in dry film did not show significant differences between these irradiation wavelengths, suggesting that the enhanced degradation of dissolved triclosan under combined radiation is mainly due to oxidation by •OH formed via water photolysis under V-UV. This conclusion was supported by slower TCS degradation in aqueous solution when methanol was added as •OH scavenger. Under both irradiation types (254, 254/185 nm) three transformation products (TPs) were identified: 2,8dichlorodibenzo-p-dioxin, 5-chloro-2-(4- or 2-chlorophenoxy)phenol, and 2-hydroxy-8-chlorodibenzodioxin. In-silico QSAR toxicity assessment predicted potential toxicity and moderate-to-low biodegradability of these TPs. Removal of these TPs was faster under 254/185 nm irradiation. Considering the low cost, simple operation (i.e. no chemicals addition) and small size of such low-pressure mercury lamps, this is a promising direction. Further investigation of the process in flow-through reactors and real
398
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
wastewater/greywater effluent is needed for its future implementation in small on-site systems for posttreatment of persistent pollutants. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Growing demand for freshwater together with depleting natural water sources increase the reuse of treated wastewater (e.g., greywater, municipal wastewater) for non-potable applications. Wastewater reuse can be implemented either by centralized municipal wastewater treatment plants or through decentralized facilities, where greywater (GW) reuse is a preferred alternative (Maimon et al., 2010). A major challenge of both decentralized and centralized wastewater treatment, is the removal of micropollutants (MPs) that are not efficiently removed by conventional treatment. The most widely spread MPs in wastewater are pharmaceuticals and personal care products (PCP's) (e.g. shampoos, soaps, toothpastes and creams) that contain preservatives, biocides, musks and UV-filters. As some of the MPs found are toxic/endocrine disrupters and/or persistent in the environment, water containing them needs to be properly treated prior to its reuse or release to the environment. Advanced oxidation processes (AOPs) seem promising methods to eliminate MPs residues that were not removed during conventional treatment (De la Cruz et al., 2012; Herrmann et al., 2015). In wastewater treatment by AOPs, •OH radicals are usually generated via photolysis of H2O2 under 254 nm light (UV-C). However, •OH and other radicals can also be generated in situ via photolysis of water and dissolved O2 using vacuum-UV (V-UV, λ b 200 nm) (Gonzalez et al., 2004; Zoschke et al., 2014 and reference therein). Such generation of •OH (see detailed reactions in Supplementary Material S1) eliminates the need of chemicals (H2O2) or catalysts (TiO2 or Fe3+) addition. This is very important for decentralized and small-scale wastewater treatment units (e.g., greywater treatment), as it simplifies their operation and maintenance (Zoschke et al., 2014). Furthermore, decentralized systems that treat small volumes of water can easily overcome the limited penetration of V-UV light in water, resulting from the strong absorption of V-UV by the water itself. As many organic compounds are likely to absorb V-UV radiation (Gonzalez et al., 2004), UV light may also lead to direct photolysis of MPs. However, very little information is currently available regarding the spectral properties and quantum yields of MPs in V-UV range. In recent years there is growing interest in using V-UV irradiation for MPs and natural organic matter removal from water (Zoschke et al., 2014 and references therein), but there are still many knowledge gaps regarding it. Studies on the removal of PCP-related MPs by V-UV are very scarce. Kim et al. (2009) compared the removal of 30 PCPs and pharmceticals in water by UV-C and by combined V-UV/UV-C irridation. The study revealed that the combined irradiation was more efficient than UV-C alone, but was unable to determine if this was a result of fast direct photolysis under V-UV irradiation and/or oxidation by •OH that formed only under V-UV irradiation. It should be noted that as long as full mineralization is not achieved, AOPs may form transformation products (TPs) that may continue to pose environmental or health concerns (Fatta-Kassinos et al., 2011; Khaleel et al., 2017). Thus, special attention should be given also to the toxicity and persistence of such products (which in some cases are more toxic than their parent compound). However, many of the previous studies on V-UV water treatment have not addressed this issue (see review by Zoschke et al., 2014). Triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol) (TCS) is a biocide added to PCPs (e.g. antibacterial soaps, deodorants, toothpastes) and used in several other household products (e.g. kitchenware or functional clothing). Hence, it occurs in wastewater effluent, and due to it is
persistence, it is frequently detected in surface waters and sediments (Bock et al., 2010; Cantwell et al., 2010; Kolpin et al., 2002; Singer et al., 2002). The median effective concentration (EC50) of TCS lies between 0.96 and 390 μg L−1 (depending on the species tested) (Orvos et al., 2002). The major transformation process affecting TCS fate in surface waters is likely to be direct photolysis (Anger et al., 2013; Kliegman et al., 2013; Lindström et al., 2002; Singer et al., 2002), as biodegradation and hydrolysis are ineffective for TCS (Singer et al., 2002). In water treatment, TCS was shown to be efficiently removed via direct photolysis under UV-C light (254 nm) as well as via oxidation by •OH (Gao et al., 2014; Sirés et al., 2007; Wong-Wah-Chung et al., 2007). Nevertheless, its photolysis under such UV-C radiation, was accompanied by formation of undesired intermediates, such as dibenzo-p-dioxin (Anger et al., 2013; Kliegman et al., 2013; Latch et al., 2003; Mezcua et al., 2004; Wong-Wah-Chung et al., 2007). The aim of this study was to investigate the potential of unmasking the V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. TCS served as a model compound for MPs that are not very biodegradable. Its photodegradation under UV-C and combined V-UV/UV-C irradiation was investigated, including both kinetics and intermediate products identification. In order to understand whether V-UV contributed to TCS degradation by direct photolysis or by reaction with photo-induced radicals (generated via water photolysis), its removal was examined both in aqueous solution and in dry thin films (in which •OH are not formed). Finally, established Quantitative structure-activity-relationship (QSAR) models were used for in-silico analysis in order to provide preliminary indication for fate and toxicity of new and unknown TPs of TCS.
2. Material and methods 2.1. Chemicals and reagents All chemicals used were analytical or HPLC grade: 5-Chloro-2-(2,4dichlorophenoxy)phenol (TCS, N 97%) was purchased from Sigma (Steinheim, Germany); methanol (N99.8) and hydrochloric acid (32%) from Bio-Lab ltd. (Jerusalem, Israel); acetone (N98%) and sulfuric acid (N98%) from Gadot ltd. (Netanya, Israel); N2 gas (N99.999%) and air from Maxima (Israel). Deionized (DI) water 18.2 MΩ was obtained by MilliPore system (Direct Q3, Merck, Germany).
2.2. UV-C and V-UV light source Low-pressure mercury lamp (λ = 254 nm or λ = 254 + 185 nm) were used (Jelight; 7″, double-bore, PN: 78–2046-7 or 81–3306-7, respectively). According to the manufacturer, both lamps have an input power of 6 W and output intensity (on lamp's surface) of ~ 30 mW cm− 2 at 254 nm and (for the combined lamp) 1.3 mW cm−2 at 185 nm. UV-C irradiation intensity was measured using spectroradiometer (USB2000 + UV-VIS-ES, Ocean Optics, USA); whereas photon flux at 185 nm was measured in batch experiments using methanol photolysis as an actinometer (Oppenländer and Schwarzwälder, 2002). Samples of 80 mM of methanol were placed in the temperaturecontrolled wells (as above), and exposed to 254/185 nm UV-C/V-UV. The photon flux was calculated from the methanol photodegradation rate.
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
2.3. Experimental procedures Triclosan (TCS) degradation under UV-C (254 nm) and V-UV (185 nm) radiation was investigated in batch settings. In order to isolate the various parameters that affect its degradation (i.e., direct vs. indirect photolysis) the batch experiments were conducted both with TCS dissolved in aqueous solution and adsorbed on glass slides as a dry thin film. 2.3.1. Photodegradation of TCS thin films Dry thin films of TCS were obtained by applying 20 μL of 100 mg L−1 TCS solution in acetone on clean glass slides (Duran, Germany) and letting the solvent evaporate. For each experiment six replicates were prepared; five film-coated slides were placed in round sealed glass chamber (60 mm inner diameter; 200 mm length), the sixth slide was immediately extracted in order to evaluate initial TCS concentration. The UV lamp (254 nm or 254/185 nm, see details above) was placed 1 cm above the samples. By altering the inflow gas mixture and radiation wavelength different photo-oxidative environments were obtained, as summarized in Table 1. Ozone was generated via O2 photolysis under 185 nm light (in situ or upstream of the glass chamber). Addition of humidity to the irradiated air resulted in formation of •OH radicals (Petrick et al., 2013). The chamber was purged (200 mL min−1) by either N2 or air. When needed, inflow gas was humidified by bubbling it through DI water upstream of the reactor. In all experiments, the UV lamp was allowed to warm up for 15 min prior to film exposure. Based on the geometry of the films and the amount applied, surface residues generated in our experiments are likely to be multilayer films. In order to avoid changes in reactivity due to changes in film thickness, much attention was paid to the uniformity of film generation procedure. The impact of areal concentration on reactivity of TCS films was tested in a separate set of experiments, in which TCS was deposited in shallow glass caps of two diameters: 1.3 and 0.9 mm. The caps were filled with TCS dissolved in acetone and let to dry. In each run two small diameter caps and two large diameter ones were exposed to 254/185 nm radiation. This was performed in a wooden chamber (details below), in which the samples were placed 7 cm below the lamp. 2.3.2. Photodegradation of TCS dissolved in deionized water These experiments were performed in a wooden chamber reactor (width: 150 mm; length: 200 mm; Height: 180 mm), equipped with a metal shatter under which the samples (0.8 mL of 2.5 mg L−1 TCS in DI water at pH of 5.6–5.7) were placed in four wells (15 mm diameter × 3 mm deep) in a temperature controlled aluminum block. TCS was dissolved directly in DI water, without a co-solvent. In one set of experiments methanol (150 mM) was used as •OH scavenger. The UV lamp (either 254 nm or 254/185 nm) was placed 7 cm above the block. Exposure durations to UV light were controlled using the shatter. A thermocouple located on the metal block provided real time temperature monitoring. 2.4. Analytical analysis TCS was extracted from glass slides in 1–3 mL acetone for 10 min at 30 °C by sonication (MRC LTD, Israel). Extracts were analyzed by High Performance-Liquid Chromatography (HPLC) with Photo-Diode Array
399
(PDA) detection (Agilent 230 solvent delivery, 410 autosampler, 335 PDA detector). 50 μL of samples were injected onto a C18 column (Halo®, 4.6 × 150 mm, 2.7 μm particles). A bisolvent system (A and B) was used at 0.4 mL min−1 flow rate. For initial 14 min solvent mixture was 80% of A (10 mM ammonium acetate in DIW) and 20% of B (acetonitrile), then varied to 3% A: 97% B within 8 min, and back to 80:20 ratio over additional 9 min. Molar absorption coefficient (ε) for TCS was calculated based on absorbance spectra (200–400 nm wavelength) of seven different concentrations of TCS (0–2.6·10− 5 M) measured with spectrophotometer (Genesys 10uv scanning, Thermo Fisher Scientific, USA). Methanol was measured using gas chromatography coupled with flame ionization detector (Agilent 3800). Transformation products were detected using a Dionex Ultimate 3000 UHPLC system (Thermo Scientific, Bremen, Germany) equipped with a solvent rack (SRD-3600), a binary pump (HPG-3400-RS), an autosampler (WPS-3000-TRS), column oven (TCC-3000), UV detector (VWD-3400RS) coupled to a high-resolution mass spectrometer (HRMS; LTQ Orbitrap XL, Thermo Scientific, Dreieich, Germany). The chromatographic separation was carried out on a Nucleoshell RP 18 column (150/3 mm, 2.7 μm, Macherey Nagel, Düren, Germany). The mobile phase consisted of 0.1% formic acid and acetonitrile at a flow rate of 0.3 mL min−1. The gradient for solvent B (acetonitrile) was as follows: 10% B (up to 2 min), increasing to 95% B within 8 min, held at 95% B for 10 min, decreasing back to initial concentration within 2 min, reconditioning up to the end for 8 min. The oven temperature was 20 °C and the detector wavelength were 254 nm and 280 nm for TCS. The MS ionization was done with atmospheric pressure chemical ionization (APCI) in negative ion mode. The parameters for the APCI source for sheath and auxiliary gas flow rates were 40.5 and 5.0 arbitrary units, respectively. The source voltage was set to 6 kV. The vaporizer temperature and the capillary temperature were set at 225 °C and 275 °C, respectively. 3. Results and discussion In order to distinguish between direct photodegradation of TCS and its oxidation by radicals, TCS degradation was investigated both in aqueous solution and in thin dry films. The dry films enable to test separately direct photolysis and photo-induced degradation by radicals. 3.1. Photodegradation of TCS dissolved in DI water Fig. 1 shows the degradation of TCS dissolved in DI water upon exposure to 254 nm radiation (UV-C) and to 254/185 nm radiation (V-UV/ UV-C). TCS degradation under the combined lamp was much faster than under the UV-C lamp, despite the low intensity of 185 nm radiation in this lamp (only ~4% of the energy emitted at 254 nm). This may be the result of (1) faster direct photolysis of TCS under V-UV and (2) oxidation of TCS by •OH generated via water photolysis at 185 nm (i.e., indirect photolysis). The enhancement in TCS degradation rate under combined radiation vanished when methanol was added to solution as an •OH scavenger (Fig. 1), suggesting it was mainly due to TCS oxidation.
Table 1 Photodegradation of TCS on dry thin film, using different experimental conditions. Exp. number
Experiment name
Carrier gas
Radiation (λ) (nm)
Degradation process
1 2 3 4 5 6
254_N2 254/185_N2 254/185 + O3 254/185 + •OH 254/185 + O3 + •OH O3
Dry N2 Dry N2 Dry air Humidified N2 Humidified air Dry air
254 254 254 254 254 −
Direct photolysis Direct photolysis Direct photolysis + heterogeneous oxidation by O3 Direct photolysis + heterogeneous oxidation by •OH Direct photolysis + heterogeneous oxidation by O3 and •OH Heterogeneous oxidation by O3
+ + + +
185 185 185 185
400
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
under the two types of lamps. In addition, its reactivity towards the relevant oxidants (ozone and •OH) was also studied.
3.2. Photodegradation of TCS dry thin films
Fig. 1. TCS (dissolved in water) relative concentration (C/C0) as function of UV dose for the two types of lamps: UV-C emitting at 254 nm (triangles); combined V-UV/UV-C lamp emitting at 254 nm and 185 nm (circles). For the latter, experiments were repeated also in the presence of methanol which acts as •OH scavenger (open green circles). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Quantum yield (Φ(λ)) for TCS photolysis under the two different lamps was calculated using Eq. (1): ΦðλÞ ¼
k C0 V I0 ðλÞ 1−10−εðλÞC0 l A
ð1Þ
where: k - is observed reaction rate constant for TCS degradation (s−1), C0 - initial TCS concentration (M), V - volume of irradiated solution (L), A - irradiated surface area of the solution (cm2) I0 - incident photon flux (Einstein cm−2 s−1), ε - TCS molar absorption coefficient (M−1 cm−1), and l - optical path length (cm). For 254 nm irradiation a quantum yield of 0.29 was obtained, which is in very good agreement with the value 0.3 reported previously (Carlson et al., 2015). However, for the combined lamp the calculated Φ was 1.27; N 4 times higher (Table 2). This extremely high yield supports the assumption that TCS degradation under V-UV/UV-C irradiation involves also indirect photolysis, such as chain reactions with photo-induced radicals. It should be noted that for the combined lamp (254/185 nm) the 185 nm photon flux was neglected in the quantum yield calculation because it contributes only 4% of the radiation intensity and the absorption coefficient of TCS at this wavelength is still unknown. In order to verify that the enhanced TCS degradation observed under combined V-UV/UV-C radiation is not due to faster direct photolysis under 185 nm, TCS photodegradation was also studied as thin dry film
In contrast with the results of the aqueous solution (in the absence of methanol), TCS in dry films did not show significant difference in degradation kinetics under the 254 nm lamp (254_N2) and the combined 254/185 nm lamp (254/185_N2) (Table 2; Fig. S1). This indicates that indeed the small amount of 185 nm radiation from the combined lamp is not sufficient to affect TCS direct photolysis, and the enhanced TCS photodegradation in aqueous solution under the combined radiation is due to its oxidation by photo-induced •OH (i.e. indirect photolysis). Interestingly, the presence of •OH and/or ozone in the gas phase did not increase TCS degradation rates in the dry films (Table 2). While ozone reactivity is not expected to be high towards saturated compounds as TCS, •OH should react with it very quickly (Finlayson-Pitts and Pitts, 2000). Probable explanation for the observed negligible effect of gaseous •OH is that these radicals were formed near the lamp surface, 1 cm away from the TCS dry films. Due to the very short life-time of •OH, only very low •OH flux reached the films' surface to react with TCS. Presence of high ozone levels (treatments III and V, Table 2) resulted in reduced TCS degradation rate, most likely due to absorption of the 254 nm radiation by ozone and its low reactivity towards TCS. In the same line, the slower TCS degradation observed when •OH were generated by irradiating humidified air (treatment V, Table 2) rather than humidified N2 (treatment IV, Table 2), results from ozone absorption of 254 nm radiation and conversion of •OH to less reactive •HO2 (Finlayson-Pitts and Pitts, 2000). The possible positive impact of oxidants presence on TCS degradation was observed at the lower irradiation doses (b 8 J/cm2), i.e. shorter exposure times (Fig. S1). Based on the data obtained for the thin films, it seems that under all treatments involving the combined lamp the reaction rate constants decrease above UV dose of 8 J/cm2. This rate reduction at high doses may suggest the formation of a “passivation layer” of TCS's TPs at the surface, protecting underlying TCS molecules from interacting with gaseous oxidants and maybe even from the UV radiation. The degradation of TCS in dry films was tested under different areal concentrations (μg cm−2) in order to check its impact of film reactivity. Varying surface concentrations was obtained in two ways: (1) applying the same amount of TCS (10 μg) on different slide areas; (2) applying different amount of TCS on the same slide area. Fig. 2a depicts the observed degradation of dry TCS films applied on two slide areas (0.69 and 1.36 cm2) corresponding to two areal concentrations (14.5 and 7.35 μg·cm−2, respectively) upon exposure to 254/ 185 nm light under ambient air. Indeed, faster degradation rate was observed when TCS was spread on larger area (i.e., lower surface concentration); k = (1.81 ± 0.11) × 10−4 and (2.61 ± 0.19) × 10−4 s−1, for 0.69 and 1.36 cm2 slide area, respectively. When varying masses of TCS (23 vs. 31 μg) were applied on the same area (0.69 cm2; 33.3 and 44.9 μg·cm−2, respectively), TCS degradation rate was faster in the films with the lower mass (Fig. 2b). Worth noting that the maximum removal of TCS in both cases did not exceed 60%,
Table 2 Summary of reaction rate constants (k), time required for 50% TCS removal and quantum yields in the different photodegradation experiments. #
Treatment
Lamp type
TCS (dry/dissolved in DI water)
k (s−1)
Time for C/C0 = 50% (s)
Quantum yield
I II III IV V VI VII
254_N2 254/185_N2 254/185 + O3 254/185 + •OH 254/185 + O3 + •OH 254 254/185
254 254 254 254 254 254 254
Dry film Dry film Dry film Dry film Dry film Aqueous solution Aqueous solution
1.57·10−3 (±7·10−5) 1.61·10−3 (±7·10−5) 1.15·10−3 (±5·10−5) 1.66·10−3 (±1·10−4) 1.22·10−3 (±8·10−5) 3.70·10−3 (±3·10−4) 1.61·10−2 (±8·10−4)
442 430 605 419 567 187 43
0.29 1.27⁎
⁎ Effective quantum yield (see text).
nm + 185 nm + 185 nm + 185 nm + 185 nm nm + 185 nm
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
401
Fig. 2. TCS (dry film) relative concentration (C/C0) as function of UV dose using 245/185 lamp: a) Different areal concentration (different slide areas) and constant TCS mass (10 μg); b) Different TCS mass on the same slide area (0.69 cm2).
suggesting that high areal TCS concentration may have limited the maximum TCS removal. Fig. 3 summarizes the observed photo-oxidation rate constants for TCS film under combined 254/185 nm radiation as function of areal concentration of the film, showing a negative non-linear correlation (i.e., larger areal concentration induces slower reaction rate). As the processes involved occur at the solid-gas interface, there is high importance to the film's exposed-surface to volume ratio. Higher areal concentration (i.e. lower surface to volume ratio) implies that lower proportion of the film was exposed to radiation and gaseous oxidants. Furthermore, as the photo-oxidation processes continue, a “passivation layer” of TPs may form on the surface of the film, protecting its deeper parts from oxidation (supporting the hypothesis portrayed above). 3.3. Formation of photo transformation products (TPs) Three photo-TPs were detected and identified following TCS photolysis in water. The structure prediction of the formed TPs was based on the LC-HRMS analysis. The sum formula and the ring double bond
equivalent (RDB) were predicted by Xcalibur Qual browser (Xcalibur 2.2, Thermo Fisher Scientific Inc.) from the acquired accurate masses of the TPs. The mass accuracy of the used mass spectrometer was found to be better than 1 mmu and therefore allows the determination of the exact mass of TPs. Furthermore, the mass from the predicted sum formula and the mass error to the acquired mass was calculated. Results are shown in Table 3.
Table 3 Retention time, acquired mass, predicted formula, ring double bond (RDB) equivalent, calculated mass and mass error of the photo transformation products. ID
Retention time [min]
Acquired mass [m/z]
Predicted formula
RDB
Calculated mass
Mass error [mmu]
TCS TP-1 TP-2 TP-3
9.1 8 8.6 6.3
286.94384 252.98303 252.98214 235.01674
C12H7O2Cl3 C12H6O2Cl2 C12H7O2Cl2 C12H8O3Cl
8.5 8.5 8.5 8.5
286.94389 252.98286 252.98286 235.01675
−0.05 0.17 −0.72 −0.01
Table 4 Predicted structure, ion, logP, in source fragments and MS2 fragments of the TPs.
Fig. 3. Observed reaction rate constant of TCS photo-oxidation under 254/185 nm light as function of the areal concentration of the TCS film. Error bars represent standard error (R2 = 0.745; n = 96).
402
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
From the predicted sum formula, the RDB and the MS2, fragments structure prediction was done, and the results are listed in Table 4. Log P was additionally calculated to confirm the order of the retention times of the predicted products. TP-1 was found to be 2,8-dichlorodibenzo-p-dioxin (2,8-DCDD), which was reported as photoproduct of TCS under UV-C irradiation in several other studies (e.g. Gao et al., 2014; Kliegman et al., 2013; Latch et al., 2003; Qiao et al., 2014). TP-2 is a dechlorinated derivate of TCS. The possible products can be (TP-2a) 5-chloro-2-(4-chlorophenoxy)phenol and (TP-2b) 5-chloro-2(2-chlorophenoxy)phenol due to the MS2 fragmentation to 141.984 m/z. The fragments 141,983 m/z can be formed only when having the chlorhydroxyphenyl substructure in the molecule (Fig. S2). Therefore, dechlorination only happened on the dichlorophenoxy part of the molecule. TP-3 is 2-hydroxy-8-chlorodibenzodioxine and is likely a dechlorination product of TP-1. The temporal change in concentrations of the TPs and parent compound (TCS) during irradiation is shown in Fig. 4. 92% and 99% removal of TCS was observed upon exposure to 1.1 J cm−2 dose of 254 nm and 254/185 nm light, respectively. In line with the faster degradation of TCS, the formation kinetics of the TPs was faster under the combined radiation. Moreover, all TPs also degraded faster under this radiation (via
reaction with •OH and/or direct photolysis by 185 nm radiation), leading to overall lower concentrations in solution after exposure to 1.1 J cm−2 irradiation dose. During TCS photolysis, TP-3 reaches its maximum concentration after TP-1 and TP-2, (Fig. 4) supporting the assumption it is a secondary TP arising from degradation of TP-1. Due to the predicted sum formula a substitution of one Cl-atom of 2,8-DCDD with a hydroxyl group can be assumed. A proposed reaction pathway for the formation of the three detected TPs is given in Fig. 5. Conversion of TCS to the more toxic compound 2,8- dichlorodibenzop-dioxins (2,8- DCDD) has focused much of the concern regarding the environmental fate of TCS (e.g., (Anger et al., 2013; Latch et al., 2003; Mezcua et al., 2004). Toxicological data for TP-2 (5-chloro-2-(4chlorophenoxy)phenol/5-chloro-2-(2-chlorophenoxy)phenol) is scarce. Yet, its reported EC50 (460 μg L−1, Melnikov et al., 2016) is similar to that of TCS (0.96–390 μg L−1, Orvos et al., 2002) and both are classified as very toxic to aquatic life with long lasting effects (ECHA, 2017; MERCK, 2017). To the best of our knowledge, no toxicological information is available for TP-3, 2-hydroxy-8-chlorodibenzodioxin toxicity. Hence a first estimation of its biodegradability and toxicity as compared with that of TCS and the other two TPs was obtained by QSAR analysis using CASE Ultra (v. 1.5.2.0, MultiCASE Inc.) and Oasis Catalogic software (V.5.11.6 TB) (further details in Text S1.2). The prediction of readily/
Fig. 4. Temporal changes in TCS and TP's concentrations (aqueous) as function of UV dose. For each TP, concentrations were normalized to its highest detected concentration (i.e., peak area ratios).
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
403
Fig. 5. Proposed reaction pathway of TCS photodegradation.
non-readily biodegradable molecules under different conditions was made based on data from the OECD 301C MITI-I test (MITI, 2006). The statistical Case Ultra AU6 model and the Catalogic OECD 301C model for readily biodegradability predicted that TCS and its TPs should be not readily biodegradable (Table S1) and, thus, could persist as a dead-end products in the environment. TCS, TP-2b and TP-3 showed positive alert in the microtox toxicity to environmental bacteria model (Case Ultra) (Table S2). The three acute Vibrio fischeri and Daphnia magna models (Oasis Catalogic; Table S2) predicted that TP-1 (2,8DCDD) is more toxic than TCS and the other TPs as reflected by its lower IC50 and EC50 values. These predictions fall in line with literature data (see above). TP-2 was predicted to be less toxic than TCS. However, its toxicity was reported as similar to that of TCS and both are classified as very toxic to aquatic life with long lasting effects (see above). The acute toxicity of TP-3 towards Vibrio fischeri and Daphnia magna was predicted as similar to that of TCS, yet it was predicted as less biodegradable than TCS. The in-silico predicted toxicity and low biodegradability of the TPs suggest potential increase in the aquatic toxicity.
4. Conclusions The goal of this study was to investigate the potential of unmasking the V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. TCS served as a model compound for MPs that are not very biodegradable. In-order to characterize TCS photo-oxidation process different experiments were conducted, using different types of lamps, experimental conditions (gases, humidity
etc.) and even the matrix in which the TCS was present (dry film vs. dissolve in DI water). In aqueous solution TCS photolysis was much faster when combined 254/185 nm radiation was applied rather than 254 nm alone, even though the added V-UV radiation (i.e., 185 nm) accounted for only 4% of the lamp UV-C emission. This significant impact of 185 nm radiation was not observed during photolysis of TCS in solution containing methanol nor in dry films, suggesting that hydroxyl radicals, which are produced via water photolysis under V-UV radiation, play a major role in TCS degradation. Photodegradation of TCS in dry films was slower than in aqueous solution, and was shown to be reversibly correlated to the surface concentration of the film (indicating the importance of surface to volume ratio in such conditions). Under both irradiation conditions (i.e., 254 nm and 254/185 nm) the same three TPs were identified in aqueous solutions: TP-1: 2,8dichlorodibenzo-p-dioxin (DCDD), TP-2a/b: 5-chloro-2-(4chlorophenoxy)phenol, 5-chloro-2-(2-chlorophenoxy)phenol and TP3: 2-hydroxy-8-chlorodibenzodioxin. The degradation kinetics suggests that TP-1 and TP-2 are first formed, while TP-3 is formed by photodegradation of TP-1. It should be noted that under the combined irradiation (254/185 nm) both the maximum concentrations and the persistence against photodegradation of these intermediates was lower than under 254 nm radiation alone. Overall, using combined UV-C and V-UV radiation results in faster photo-oxidation of both TCS and its TPs. In silico QSAR toxicity assessment performed in this study provides a first indication that, additionally to the concerns about 2,8dichlorodibenzo-p-dioxin, photo-degradation of TCS might lead to a not readily biodegradable and similar ecotoxic 2-hydroxy-8chlorodibenzodioxin. The fate and effects of these byproducts in aquatic
404
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
environments cannot be ignored and deserves further investigations. This is especially true for TP-2 and TP-3 for which hardly any empirical data exist. UV-based AOP of MPs is likely to exhibit significant disinfection efficiency, as the latter often requires lower UV doses (Zoschke et al., 2014). It should be noted however that treated wastewater or greywater contain substances that absorb UV irradiation or scavenge radicals, which would reduce the process efficiency compared with the current findings. Hence, in order to evaluate the required UV dose (controlled by radiation flux and retention time) the process should be further studied in flow-through reactors treating wastewater or greywater effluent. Considering the low cost and small size of such low-pressure mercury lamps, the process can be implemented in the future in small on-site systems, such as decentralized wastewater and greywater treatment units, for chemical-free post treatment of persistent pollutants. Acknowledgement Financial support from the State of Lower Saxony, Ministry of Science and Culture, and by the Volkswagen Foundation (Grant No. VWZN2830) is acknowledged. The authors would like to thank Multicase Inc. for kindly providing the CASE Ultra software. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2017.05.172. References Anger, C.T., Sueper, C., Blumentritt, D.J., McNeill, K., Engstrom, D.R., Arnold, W.A., 2013. Quantification of triclosan, chlorinated triclosan derivatives, and their dioxin photoproducts in lacustrine sediment cores. Environ. Sci. Technol. 47:1833–1843. http:// dx.doi.org/10.1021/es3045289. Bock, M., Lyndall, J., Barber, T., Fuchsman, P., Perruchon, E., Capdevielle, M., 2010. Probabilistic application of a fugacity model to predict triclosan fate during wastewater treatment. Integr. Environ. Assess. Manag. 6:393–404. http://dx.doi.org/10.1002/ ieam.134. Cantwell, M.G., Wilson, B.A., Zhu, J., Wallace, G.T., King, J.W., Olsen, C.R., Burgess, R.M., Smith, J.P., 2010. Temporal trends of triclosan contamination in dated sediment cores from four urbanized estuaries: evidence of preservation and accumulation. Chemosphere 78:347–352. http://dx.doi.org/10.1016/j.chemosphere.2009.11.021. Carlson, J.C., Stefan, M.I., Parnis, J.M., Metcalfe, C.D., 2015. Direct UV photolysis of selected pharmaceuticals, personal care products and endocrine disruptors in aqueous solution. Water Res. 84:350–361. http://dx.doi.org/10.1016/j.watres.2015.04.013. De la Cruz, N., Giménez, J., Esplugas, S., Grandjean, D., De Alencastro, L.F., Pulgarín, C., 2012. Degradation of 32 emergent contaminants by UV and neutral photo-fenton in domestic wastewater effluent previously treated by activated sludge. Water Res. 46:1947–1957. http://dx.doi.org/10.1016/j.watres.2012.01.014. ECHA, 2017. European chemical agency [WWW document]. URL https://echa.europa.eu/ home Fatta-Kassinos, D., Hapeshi, E., Achilleos, A., Meric, S., Gros, M., Petrovic, M., Barcelo, D., 2011. Existence of pharmaceutical compounds in tertiary treated urban wastewater that is utilized for reuse applications. Water Resour. Manag. 25:1183–1193. http:// dx.doi.org/10.1007/s11269-010-9646-4. Finlayson-Pitts, B.J., Pitts, J.N., 2000. Chemistry of the upper and lower atmosphere: theory, experiments, and applications. Chemistry of the Upper and Lower Atmosphere. Academic Press http://dx.doi.org/10.1016/B978-0-12-257060-5.50027-7. Gao, Y., Ji, Y., Li, G., An, T., 2014. Mechanism, kinetics and toxicity assessment of •OHinitiated transformation of triclosan in aquatic environments. Water Res. 49: 360–370. http://dx.doi.org/10.1016/j.watres.2013.10.027.
Gonzalez, M.G., Oliveros, E., Wörner, M., Braun, A.M., 2004. Vacuum-ultraviolet photolysis of aqueous reaction systems. J Photochem Photobiol C: Photochem Rev 5:225–246. http://dx.doi.org/10.1016/j.jphotochemrev.2004.10.002. Herrmann, M., Menz, J., Olsson, O., Kümmerer, K., 2015. Identification of phototransformation products of the antiepileptic drug gabapentin: biodegradability and initial assessment of toxicity. Water Res. 85:11–21. http://dx.doi.org/10.1016/j. watres.2015.08.004. Khaleel, N.D.H., Mahmoud, W.M.M., Olsson, O., Kümmerer, K., 2017. Initial fate assessment of teratogenic drug trimipramine and its photo-transformation products – role of pH, concentration and temperature. Water Res. 108:197–211. http://dx.doi. org/10.1016/j.watres.2016.10.078. Kim, I., Yamashita, N., Tanaka, H., 2009. Photodegradation of pharmaceuticals and personal care products during UV and UV/H2O2 treatments. Chemosphere 77:518–525. http://dx.doi.org/10.1016/j.chemosphere.2009.07.041. Kliegman, S., Eustis, S.N., Arnold, W.A., McNeill, K., 2013. Experimental and theoretical insights into the involvement of radicals in triclosan phototransformation. Environ. Sci. Technol. 47:6756–6763. http://dx.doi.org/10.1021/es3041797. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., Buxton, H.T., 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000: a national reconnaissance. Environ. Sci. Technol. 36: 1202–1211. http://dx.doi.org/10.1021/es011055j. Latch, D.E., Packer, J.L., Arnold, W.A., McNeill, K., 2003. Photochemical conversion of triclosan to 2,8-dichlorodibenzo-p-dioxin in aqueous solution. J. Photochem. Photobiol. A Chem. 158:63–66. http://dx.doi.org/10.1016/S1010-6030(03)00103-5. Lindström, A., Buerge, I.J., Poiger, T., Bergqvist, P.A., Müller, M.D., Buser, H.R., 2002. Occurrence and environmental behavior of the bactericide triclosan and its methyl derivative in surface waters and in wastewater. Environ. Sci. Technol. 36:2322–2329. http:// dx.doi.org/10.1021/es0114254. Maimon, A., Tal, A., Friedler, E., Gross, A., 2010. Safe on-site reuse of greywater for irrigation - a critical review of current guidelines. Environ. Sci. Technol. 44, 3213–3220. Melnikov, F., Kostal, J., Voutchkova-Kostal, A., Zimmerman, J.B., Anastas, Paul T., 2016. Assessment of predictive models for estimating the acute aquatic toxicity of organic chemicals. Green Chem. 18:4432–4445. http://dx.doi.org/10.1039/C6GC00720A. MERCK, 2017. The Merck index online - chemicals, drugs and biologicals [WWW document]. URL https://www.rsc.org/Merck-Index/ (accessed 2.9.17). Mezcua, M., Gómez, M.J., Ferrer, I., Aguera, A., Hernando, M.D., Fernández-Alba, A.R., 2004. Evidence of 2,7/2,8-dibenzodichloro-p-dioxin as a photodegradation product of triclosan in water and wastewater samples. Anal. Chim. Acta 524:241–247. http://dx. doi.org/10.1016/j.aca.2004.05.050. MITI, 2006. OECD 2006. Ready Biodegradability: Modified Miti Test (I). Oppenländer, T., Schwarzwälder, R., 2002. Vacuum-UV oxidation (H2O-VUV) with a xenon excimer flow-trough lamp at 172 nm: use of methanol as Actinometer for VUV intensity measurement and as reference compound for •OH-radical competition kinetics in aqueous systems. J. Adv. Oxid. Technol. 5, 155–163. Orvos, D.R., Versteeg, D.J., Inauen, J., Capdevielle, M., Rothenstein, A., Cunningham, V., 2002. Aquatic toxicity of triclosan. Environ. Toxicol. Chem. 21:1338–1349. http://dx. doi.org/10.1002/etc.5620210703. Petrick, L.M., Sabach, S., Dubowski, Y., 2013. Degradation of VX surrogate profenofos on surfaces via in situ photo-oxidation. Environ. Sci. Technol. 47:8751–8758. http://dx. doi.org/10.1021/es4016537. Qiao, X., Zheng, X., Xie, Q., Yang, X., Xiao, J., Xue, W., Chen, J., 2014. Faster photodegradation rate and higher dioxin yield of triclosan induced by cationic surfactant CTAB. J. Hazard. Mater. 275:210–214. http://dx.doi.org/10.1016/j.jhazmat.2014. 05.012. Singer, H., Müller, S., Tixier, C., Pillonel, L., 2002. Triclosan: occurrence and fate of a widely used biocide in the aquatic environment: field measurements in wastewater treatment plants, surface waters, and lake sediments. Environ. Sci. Technol. 36: 4998–5004. http://dx.doi.org/10.1021/es025750i. Sirés, I., Oturan, N., Oturan, M.A., Rodríguez, R.M., Garrido, J.A., Brillas, E., 2007. ElectroFenton degradation of antimicrobials triclosan and triclocarban. Electrochim. Acta 52:5493–5503. http://dx.doi.org/10.1016/j.electacta.2007.03.011. Wong-Wah-Chung, P., Rafqah, S., Voyard, G., Sarakha, M., 2007. Photochemical behaviour of triclosan in aqueous solutions: kinetic and analytical studies. J. Photochem. Photobiol. A Chem. 191:201–208. http://dx.doi.org/10.1016/j.jphotochem.2007.04. 024. Zoschke, K., Börnick, H., Worch, E., 2014. Vacuum-UV radiation at 185 nm in water treatment–a review. Water Res. 52:131–145. http://dx.doi.org/10.1016/j.watres. 2013.12.034.
Contents lists available at ScienceDirect
Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv
Photodegradation of micropollutants using V-UV/UV-C processes; Triclosan as a model compound Alfiya Yuval a, Friedler Eran a, Westphal Janin b, Olsson Oliver b, Dubowski Yael a,⁎ a b
Faculty of Civil and Environmental Engineering, Technion, Israel Institute of Technology, Haifa, Israel Institute of Sustainable and Environmental Chemistry, Leuphana University of Lüneburg, Germany
H I G H L I G H T S
G R A P H I C A L
A B S T R A C T
• Triclosan (TCS) photodegradation was faster in aqueous solution than in dry thin films. • Degradation kinetics of aqueous TCS was faster under combined V-UV/UV-C radiation. • Transformation products were identified and their formation/degradation kinetics was quantified. • Toxicity and biodegradability were predicted by QSAR.
a r t i c l e
i n f o
Article history: Received 3 April 2017 Received in revised form 18 May 2017 Accepted 19 May 2017 Available online xxxx Editor: Jay Gan Keywords: Decentralized wastewater treatment Micropollutants Advanced oxidation process Triclosan Transformation products Vacuum ultra-violet
⁎ Corresponding author. E-mail address: [email protected] (D. Yael).
http://dx.doi.org/10.1016/j.scitotenv.2017.05.172 0048-9697/© 2017 Elsevier B.V. All rights reserved.
a b s t r a c t Non-potable reuse of treated wastewater is becoming widespread as means to address growing water scarcity. Removal of micropollutants (MPs) from such water often requires advanced oxidation processes using •OH radicals. •OH can be generated in-situ via water photolysis under vacuum-UV (λ b 200 nm) irradiation. The aim of this study was to investigate the potential of unmasking V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. Triclosan, a biocide of limited biodegradability, served as a model compound for MPs that are not very biodegradable. Its degradation kinetics and identification of intermediate products were investigated under 254 nm and under combined 254/185 nm irradiation both in dry thin films and in aqueous solutions. In the latter, degradation was faster under combined 254/185 nm radiation, although the 185 nm radiation accounted for only 4% of the total UV light intensity. In contrast, triclosan photodegradation in dry film did not show significant differences between these irradiation wavelengths, suggesting that the enhanced degradation of dissolved triclosan under combined radiation is mainly due to oxidation by •OH formed via water photolysis under V-UV. This conclusion was supported by slower TCS degradation in aqueous solution when methanol was added as •OH scavenger. Under both irradiation types (254, 254/185 nm) three transformation products (TPs) were identified: 2,8dichlorodibenzo-p-dioxin, 5-chloro-2-(4- or 2-chlorophenoxy)phenol, and 2-hydroxy-8-chlorodibenzodioxin. In-silico QSAR toxicity assessment predicted potential toxicity and moderate-to-low biodegradability of these TPs. Removal of these TPs was faster under 254/185 nm irradiation. Considering the low cost, simple operation (i.e. no chemicals addition) and small size of such low-pressure mercury lamps, this is a promising direction. Further investigation of the process in flow-through reactors and real
398
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
wastewater/greywater effluent is needed for its future implementation in small on-site systems for posttreatment of persistent pollutants. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Growing demand for freshwater together with depleting natural water sources increase the reuse of treated wastewater (e.g., greywater, municipal wastewater) for non-potable applications. Wastewater reuse can be implemented either by centralized municipal wastewater treatment plants or through decentralized facilities, where greywater (GW) reuse is a preferred alternative (Maimon et al., 2010). A major challenge of both decentralized and centralized wastewater treatment, is the removal of micropollutants (MPs) that are not efficiently removed by conventional treatment. The most widely spread MPs in wastewater are pharmaceuticals and personal care products (PCP's) (e.g. shampoos, soaps, toothpastes and creams) that contain preservatives, biocides, musks and UV-filters. As some of the MPs found are toxic/endocrine disrupters and/or persistent in the environment, water containing them needs to be properly treated prior to its reuse or release to the environment. Advanced oxidation processes (AOPs) seem promising methods to eliminate MPs residues that were not removed during conventional treatment (De la Cruz et al., 2012; Herrmann et al., 2015). In wastewater treatment by AOPs, •OH radicals are usually generated via photolysis of H2O2 under 254 nm light (UV-C). However, •OH and other radicals can also be generated in situ via photolysis of water and dissolved O2 using vacuum-UV (V-UV, λ b 200 nm) (Gonzalez et al., 2004; Zoschke et al., 2014 and reference therein). Such generation of •OH (see detailed reactions in Supplementary Material S1) eliminates the need of chemicals (H2O2) or catalysts (TiO2 or Fe3+) addition. This is very important for decentralized and small-scale wastewater treatment units (e.g., greywater treatment), as it simplifies their operation and maintenance (Zoschke et al., 2014). Furthermore, decentralized systems that treat small volumes of water can easily overcome the limited penetration of V-UV light in water, resulting from the strong absorption of V-UV by the water itself. As many organic compounds are likely to absorb V-UV radiation (Gonzalez et al., 2004), UV light may also lead to direct photolysis of MPs. However, very little information is currently available regarding the spectral properties and quantum yields of MPs in V-UV range. In recent years there is growing interest in using V-UV irradiation for MPs and natural organic matter removal from water (Zoschke et al., 2014 and references therein), but there are still many knowledge gaps regarding it. Studies on the removal of PCP-related MPs by V-UV are very scarce. Kim et al. (2009) compared the removal of 30 PCPs and pharmceticals in water by UV-C and by combined V-UV/UV-C irridation. The study revealed that the combined irradiation was more efficient than UV-C alone, but was unable to determine if this was a result of fast direct photolysis under V-UV irradiation and/or oxidation by •OH that formed only under V-UV irradiation. It should be noted that as long as full mineralization is not achieved, AOPs may form transformation products (TPs) that may continue to pose environmental or health concerns (Fatta-Kassinos et al., 2011; Khaleel et al., 2017). Thus, special attention should be given also to the toxicity and persistence of such products (which in some cases are more toxic than their parent compound). However, many of the previous studies on V-UV water treatment have not addressed this issue (see review by Zoschke et al., 2014). Triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol) (TCS) is a biocide added to PCPs (e.g. antibacterial soaps, deodorants, toothpastes) and used in several other household products (e.g. kitchenware or functional clothing). Hence, it occurs in wastewater effluent, and due to it is
persistence, it is frequently detected in surface waters and sediments (Bock et al., 2010; Cantwell et al., 2010; Kolpin et al., 2002; Singer et al., 2002). The median effective concentration (EC50) of TCS lies between 0.96 and 390 μg L−1 (depending on the species tested) (Orvos et al., 2002). The major transformation process affecting TCS fate in surface waters is likely to be direct photolysis (Anger et al., 2013; Kliegman et al., 2013; Lindström et al., 2002; Singer et al., 2002), as biodegradation and hydrolysis are ineffective for TCS (Singer et al., 2002). In water treatment, TCS was shown to be efficiently removed via direct photolysis under UV-C light (254 nm) as well as via oxidation by •OH (Gao et al., 2014; Sirés et al., 2007; Wong-Wah-Chung et al., 2007). Nevertheless, its photolysis under such UV-C radiation, was accompanied by formation of undesired intermediates, such as dibenzo-p-dioxin (Anger et al., 2013; Kliegman et al., 2013; Latch et al., 2003; Mezcua et al., 2004; Wong-Wah-Chung et al., 2007). The aim of this study was to investigate the potential of unmasking the V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. TCS served as a model compound for MPs that are not very biodegradable. Its photodegradation under UV-C and combined V-UV/UV-C irradiation was investigated, including both kinetics and intermediate products identification. In order to understand whether V-UV contributed to TCS degradation by direct photolysis or by reaction with photo-induced radicals (generated via water photolysis), its removal was examined both in aqueous solution and in dry thin films (in which •OH are not formed). Finally, established Quantitative structure-activity-relationship (QSAR) models were used for in-silico analysis in order to provide preliminary indication for fate and toxicity of new and unknown TPs of TCS.
2. Material and methods 2.1. Chemicals and reagents All chemicals used were analytical or HPLC grade: 5-Chloro-2-(2,4dichlorophenoxy)phenol (TCS, N 97%) was purchased from Sigma (Steinheim, Germany); methanol (N99.8) and hydrochloric acid (32%) from Bio-Lab ltd. (Jerusalem, Israel); acetone (N98%) and sulfuric acid (N98%) from Gadot ltd. (Netanya, Israel); N2 gas (N99.999%) and air from Maxima (Israel). Deionized (DI) water 18.2 MΩ was obtained by MilliPore system (Direct Q3, Merck, Germany).
2.2. UV-C and V-UV light source Low-pressure mercury lamp (λ = 254 nm or λ = 254 + 185 nm) were used (Jelight; 7″, double-bore, PN: 78–2046-7 or 81–3306-7, respectively). According to the manufacturer, both lamps have an input power of 6 W and output intensity (on lamp's surface) of ~ 30 mW cm− 2 at 254 nm and (for the combined lamp) 1.3 mW cm−2 at 185 nm. UV-C irradiation intensity was measured using spectroradiometer (USB2000 + UV-VIS-ES, Ocean Optics, USA); whereas photon flux at 185 nm was measured in batch experiments using methanol photolysis as an actinometer (Oppenländer and Schwarzwälder, 2002). Samples of 80 mM of methanol were placed in the temperaturecontrolled wells (as above), and exposed to 254/185 nm UV-C/V-UV. The photon flux was calculated from the methanol photodegradation rate.
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
2.3. Experimental procedures Triclosan (TCS) degradation under UV-C (254 nm) and V-UV (185 nm) radiation was investigated in batch settings. In order to isolate the various parameters that affect its degradation (i.e., direct vs. indirect photolysis) the batch experiments were conducted both with TCS dissolved in aqueous solution and adsorbed on glass slides as a dry thin film. 2.3.1. Photodegradation of TCS thin films Dry thin films of TCS were obtained by applying 20 μL of 100 mg L−1 TCS solution in acetone on clean glass slides (Duran, Germany) and letting the solvent evaporate. For each experiment six replicates were prepared; five film-coated slides were placed in round sealed glass chamber (60 mm inner diameter; 200 mm length), the sixth slide was immediately extracted in order to evaluate initial TCS concentration. The UV lamp (254 nm or 254/185 nm, see details above) was placed 1 cm above the samples. By altering the inflow gas mixture and radiation wavelength different photo-oxidative environments were obtained, as summarized in Table 1. Ozone was generated via O2 photolysis under 185 nm light (in situ or upstream of the glass chamber). Addition of humidity to the irradiated air resulted in formation of •OH radicals (Petrick et al., 2013). The chamber was purged (200 mL min−1) by either N2 or air. When needed, inflow gas was humidified by bubbling it through DI water upstream of the reactor. In all experiments, the UV lamp was allowed to warm up for 15 min prior to film exposure. Based on the geometry of the films and the amount applied, surface residues generated in our experiments are likely to be multilayer films. In order to avoid changes in reactivity due to changes in film thickness, much attention was paid to the uniformity of film generation procedure. The impact of areal concentration on reactivity of TCS films was tested in a separate set of experiments, in which TCS was deposited in shallow glass caps of two diameters: 1.3 and 0.9 mm. The caps were filled with TCS dissolved in acetone and let to dry. In each run two small diameter caps and two large diameter ones were exposed to 254/185 nm radiation. This was performed in a wooden chamber (details below), in which the samples were placed 7 cm below the lamp. 2.3.2. Photodegradation of TCS dissolved in deionized water These experiments were performed in a wooden chamber reactor (width: 150 mm; length: 200 mm; Height: 180 mm), equipped with a metal shatter under which the samples (0.8 mL of 2.5 mg L−1 TCS in DI water at pH of 5.6–5.7) were placed in four wells (15 mm diameter × 3 mm deep) in a temperature controlled aluminum block. TCS was dissolved directly in DI water, without a co-solvent. In one set of experiments methanol (150 mM) was used as •OH scavenger. The UV lamp (either 254 nm or 254/185 nm) was placed 7 cm above the block. Exposure durations to UV light were controlled using the shatter. A thermocouple located on the metal block provided real time temperature monitoring. 2.4. Analytical analysis TCS was extracted from glass slides in 1–3 mL acetone for 10 min at 30 °C by sonication (MRC LTD, Israel). Extracts were analyzed by High Performance-Liquid Chromatography (HPLC) with Photo-Diode Array
399
(PDA) detection (Agilent 230 solvent delivery, 410 autosampler, 335 PDA detector). 50 μL of samples were injected onto a C18 column (Halo®, 4.6 × 150 mm, 2.7 μm particles). A bisolvent system (A and B) was used at 0.4 mL min−1 flow rate. For initial 14 min solvent mixture was 80% of A (10 mM ammonium acetate in DIW) and 20% of B (acetonitrile), then varied to 3% A: 97% B within 8 min, and back to 80:20 ratio over additional 9 min. Molar absorption coefficient (ε) for TCS was calculated based on absorbance spectra (200–400 nm wavelength) of seven different concentrations of TCS (0–2.6·10− 5 M) measured with spectrophotometer (Genesys 10uv scanning, Thermo Fisher Scientific, USA). Methanol was measured using gas chromatography coupled with flame ionization detector (Agilent 3800). Transformation products were detected using a Dionex Ultimate 3000 UHPLC system (Thermo Scientific, Bremen, Germany) equipped with a solvent rack (SRD-3600), a binary pump (HPG-3400-RS), an autosampler (WPS-3000-TRS), column oven (TCC-3000), UV detector (VWD-3400RS) coupled to a high-resolution mass spectrometer (HRMS; LTQ Orbitrap XL, Thermo Scientific, Dreieich, Germany). The chromatographic separation was carried out on a Nucleoshell RP 18 column (150/3 mm, 2.7 μm, Macherey Nagel, Düren, Germany). The mobile phase consisted of 0.1% formic acid and acetonitrile at a flow rate of 0.3 mL min−1. The gradient for solvent B (acetonitrile) was as follows: 10% B (up to 2 min), increasing to 95% B within 8 min, held at 95% B for 10 min, decreasing back to initial concentration within 2 min, reconditioning up to the end for 8 min. The oven temperature was 20 °C and the detector wavelength were 254 nm and 280 nm for TCS. The MS ionization was done with atmospheric pressure chemical ionization (APCI) in negative ion mode. The parameters for the APCI source for sheath and auxiliary gas flow rates were 40.5 and 5.0 arbitrary units, respectively. The source voltage was set to 6 kV. The vaporizer temperature and the capillary temperature were set at 225 °C and 275 °C, respectively. 3. Results and discussion In order to distinguish between direct photodegradation of TCS and its oxidation by radicals, TCS degradation was investigated both in aqueous solution and in thin dry films. The dry films enable to test separately direct photolysis and photo-induced degradation by radicals. 3.1. Photodegradation of TCS dissolved in DI water Fig. 1 shows the degradation of TCS dissolved in DI water upon exposure to 254 nm radiation (UV-C) and to 254/185 nm radiation (V-UV/ UV-C). TCS degradation under the combined lamp was much faster than under the UV-C lamp, despite the low intensity of 185 nm radiation in this lamp (only ~4% of the energy emitted at 254 nm). This may be the result of (1) faster direct photolysis of TCS under V-UV and (2) oxidation of TCS by •OH generated via water photolysis at 185 nm (i.e., indirect photolysis). The enhancement in TCS degradation rate under combined radiation vanished when methanol was added to solution as an •OH scavenger (Fig. 1), suggesting it was mainly due to TCS oxidation.
Table 1 Photodegradation of TCS on dry thin film, using different experimental conditions. Exp. number
Experiment name
Carrier gas
Radiation (λ) (nm)
Degradation process
1 2 3 4 5 6
254_N2 254/185_N2 254/185 + O3 254/185 + •OH 254/185 + O3 + •OH O3
Dry N2 Dry N2 Dry air Humidified N2 Humidified air Dry air
254 254 254 254 254 −
Direct photolysis Direct photolysis Direct photolysis + heterogeneous oxidation by O3 Direct photolysis + heterogeneous oxidation by •OH Direct photolysis + heterogeneous oxidation by O3 and •OH Heterogeneous oxidation by O3
+ + + +
185 185 185 185
400
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
under the two types of lamps. In addition, its reactivity towards the relevant oxidants (ozone and •OH) was also studied.
3.2. Photodegradation of TCS dry thin films
Fig. 1. TCS (dissolved in water) relative concentration (C/C0) as function of UV dose for the two types of lamps: UV-C emitting at 254 nm (triangles); combined V-UV/UV-C lamp emitting at 254 nm and 185 nm (circles). For the latter, experiments were repeated also in the presence of methanol which acts as •OH scavenger (open green circles). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Quantum yield (Φ(λ)) for TCS photolysis under the two different lamps was calculated using Eq. (1): ΦðλÞ ¼
k C0 V I0 ðλÞ 1−10−εðλÞC0 l A
ð1Þ
where: k - is observed reaction rate constant for TCS degradation (s−1), C0 - initial TCS concentration (M), V - volume of irradiated solution (L), A - irradiated surface area of the solution (cm2) I0 - incident photon flux (Einstein cm−2 s−1), ε - TCS molar absorption coefficient (M−1 cm−1), and l - optical path length (cm). For 254 nm irradiation a quantum yield of 0.29 was obtained, which is in very good agreement with the value 0.3 reported previously (Carlson et al., 2015). However, for the combined lamp the calculated Φ was 1.27; N 4 times higher (Table 2). This extremely high yield supports the assumption that TCS degradation under V-UV/UV-C irradiation involves also indirect photolysis, such as chain reactions with photo-induced radicals. It should be noted that for the combined lamp (254/185 nm) the 185 nm photon flux was neglected in the quantum yield calculation because it contributes only 4% of the radiation intensity and the absorption coefficient of TCS at this wavelength is still unknown. In order to verify that the enhanced TCS degradation observed under combined V-UV/UV-C radiation is not due to faster direct photolysis under 185 nm, TCS photodegradation was also studied as thin dry film
In contrast with the results of the aqueous solution (in the absence of methanol), TCS in dry films did not show significant difference in degradation kinetics under the 254 nm lamp (254_N2) and the combined 254/185 nm lamp (254/185_N2) (Table 2; Fig. S1). This indicates that indeed the small amount of 185 nm radiation from the combined lamp is not sufficient to affect TCS direct photolysis, and the enhanced TCS photodegradation in aqueous solution under the combined radiation is due to its oxidation by photo-induced •OH (i.e. indirect photolysis). Interestingly, the presence of •OH and/or ozone in the gas phase did not increase TCS degradation rates in the dry films (Table 2). While ozone reactivity is not expected to be high towards saturated compounds as TCS, •OH should react with it very quickly (Finlayson-Pitts and Pitts, 2000). Probable explanation for the observed negligible effect of gaseous •OH is that these radicals were formed near the lamp surface, 1 cm away from the TCS dry films. Due to the very short life-time of •OH, only very low •OH flux reached the films' surface to react with TCS. Presence of high ozone levels (treatments III and V, Table 2) resulted in reduced TCS degradation rate, most likely due to absorption of the 254 nm radiation by ozone and its low reactivity towards TCS. In the same line, the slower TCS degradation observed when •OH were generated by irradiating humidified air (treatment V, Table 2) rather than humidified N2 (treatment IV, Table 2), results from ozone absorption of 254 nm radiation and conversion of •OH to less reactive •HO2 (Finlayson-Pitts and Pitts, 2000). The possible positive impact of oxidants presence on TCS degradation was observed at the lower irradiation doses (b 8 J/cm2), i.e. shorter exposure times (Fig. S1). Based on the data obtained for the thin films, it seems that under all treatments involving the combined lamp the reaction rate constants decrease above UV dose of 8 J/cm2. This rate reduction at high doses may suggest the formation of a “passivation layer” of TCS's TPs at the surface, protecting underlying TCS molecules from interacting with gaseous oxidants and maybe even from the UV radiation. The degradation of TCS in dry films was tested under different areal concentrations (μg cm−2) in order to check its impact of film reactivity. Varying surface concentrations was obtained in two ways: (1) applying the same amount of TCS (10 μg) on different slide areas; (2) applying different amount of TCS on the same slide area. Fig. 2a depicts the observed degradation of dry TCS films applied on two slide areas (0.69 and 1.36 cm2) corresponding to two areal concentrations (14.5 and 7.35 μg·cm−2, respectively) upon exposure to 254/ 185 nm light under ambient air. Indeed, faster degradation rate was observed when TCS was spread on larger area (i.e., lower surface concentration); k = (1.81 ± 0.11) × 10−4 and (2.61 ± 0.19) × 10−4 s−1, for 0.69 and 1.36 cm2 slide area, respectively. When varying masses of TCS (23 vs. 31 μg) were applied on the same area (0.69 cm2; 33.3 and 44.9 μg·cm−2, respectively), TCS degradation rate was faster in the films with the lower mass (Fig. 2b). Worth noting that the maximum removal of TCS in both cases did not exceed 60%,
Table 2 Summary of reaction rate constants (k), time required for 50% TCS removal and quantum yields in the different photodegradation experiments. #
Treatment
Lamp type
TCS (dry/dissolved in DI water)
k (s−1)
Time for C/C0 = 50% (s)
Quantum yield
I II III IV V VI VII
254_N2 254/185_N2 254/185 + O3 254/185 + •OH 254/185 + O3 + •OH 254 254/185
254 254 254 254 254 254 254
Dry film Dry film Dry film Dry film Dry film Aqueous solution Aqueous solution
1.57·10−3 (±7·10−5) 1.61·10−3 (±7·10−5) 1.15·10−3 (±5·10−5) 1.66·10−3 (±1·10−4) 1.22·10−3 (±8·10−5) 3.70·10−3 (±3·10−4) 1.61·10−2 (±8·10−4)
442 430 605 419 567 187 43
0.29 1.27⁎
⁎ Effective quantum yield (see text).
nm + 185 nm + 185 nm + 185 nm + 185 nm nm + 185 nm
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
401
Fig. 2. TCS (dry film) relative concentration (C/C0) as function of UV dose using 245/185 lamp: a) Different areal concentration (different slide areas) and constant TCS mass (10 μg); b) Different TCS mass on the same slide area (0.69 cm2).
suggesting that high areal TCS concentration may have limited the maximum TCS removal. Fig. 3 summarizes the observed photo-oxidation rate constants for TCS film under combined 254/185 nm radiation as function of areal concentration of the film, showing a negative non-linear correlation (i.e., larger areal concentration induces slower reaction rate). As the processes involved occur at the solid-gas interface, there is high importance to the film's exposed-surface to volume ratio. Higher areal concentration (i.e. lower surface to volume ratio) implies that lower proportion of the film was exposed to radiation and gaseous oxidants. Furthermore, as the photo-oxidation processes continue, a “passivation layer” of TPs may form on the surface of the film, protecting its deeper parts from oxidation (supporting the hypothesis portrayed above). 3.3. Formation of photo transformation products (TPs) Three photo-TPs were detected and identified following TCS photolysis in water. The structure prediction of the formed TPs was based on the LC-HRMS analysis. The sum formula and the ring double bond
equivalent (RDB) were predicted by Xcalibur Qual browser (Xcalibur 2.2, Thermo Fisher Scientific Inc.) from the acquired accurate masses of the TPs. The mass accuracy of the used mass spectrometer was found to be better than 1 mmu and therefore allows the determination of the exact mass of TPs. Furthermore, the mass from the predicted sum formula and the mass error to the acquired mass was calculated. Results are shown in Table 3.
Table 3 Retention time, acquired mass, predicted formula, ring double bond (RDB) equivalent, calculated mass and mass error of the photo transformation products. ID
Retention time [min]
Acquired mass [m/z]
Predicted formula
RDB
Calculated mass
Mass error [mmu]
TCS TP-1 TP-2 TP-3
9.1 8 8.6 6.3
286.94384 252.98303 252.98214 235.01674
C12H7O2Cl3 C12H6O2Cl2 C12H7O2Cl2 C12H8O3Cl
8.5 8.5 8.5 8.5
286.94389 252.98286 252.98286 235.01675
−0.05 0.17 −0.72 −0.01
Table 4 Predicted structure, ion, logP, in source fragments and MS2 fragments of the TPs.
Fig. 3. Observed reaction rate constant of TCS photo-oxidation under 254/185 nm light as function of the areal concentration of the TCS film. Error bars represent standard error (R2 = 0.745; n = 96).
402
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
From the predicted sum formula, the RDB and the MS2, fragments structure prediction was done, and the results are listed in Table 4. Log P was additionally calculated to confirm the order of the retention times of the predicted products. TP-1 was found to be 2,8-dichlorodibenzo-p-dioxin (2,8-DCDD), which was reported as photoproduct of TCS under UV-C irradiation in several other studies (e.g. Gao et al., 2014; Kliegman et al., 2013; Latch et al., 2003; Qiao et al., 2014). TP-2 is a dechlorinated derivate of TCS. The possible products can be (TP-2a) 5-chloro-2-(4-chlorophenoxy)phenol and (TP-2b) 5-chloro-2(2-chlorophenoxy)phenol due to the MS2 fragmentation to 141.984 m/z. The fragments 141,983 m/z can be formed only when having the chlorhydroxyphenyl substructure in the molecule (Fig. S2). Therefore, dechlorination only happened on the dichlorophenoxy part of the molecule. TP-3 is 2-hydroxy-8-chlorodibenzodioxine and is likely a dechlorination product of TP-1. The temporal change in concentrations of the TPs and parent compound (TCS) during irradiation is shown in Fig. 4. 92% and 99% removal of TCS was observed upon exposure to 1.1 J cm−2 dose of 254 nm and 254/185 nm light, respectively. In line with the faster degradation of TCS, the formation kinetics of the TPs was faster under the combined radiation. Moreover, all TPs also degraded faster under this radiation (via
reaction with •OH and/or direct photolysis by 185 nm radiation), leading to overall lower concentrations in solution after exposure to 1.1 J cm−2 irradiation dose. During TCS photolysis, TP-3 reaches its maximum concentration after TP-1 and TP-2, (Fig. 4) supporting the assumption it is a secondary TP arising from degradation of TP-1. Due to the predicted sum formula a substitution of one Cl-atom of 2,8-DCDD with a hydroxyl group can be assumed. A proposed reaction pathway for the formation of the three detected TPs is given in Fig. 5. Conversion of TCS to the more toxic compound 2,8- dichlorodibenzop-dioxins (2,8- DCDD) has focused much of the concern regarding the environmental fate of TCS (e.g., (Anger et al., 2013; Latch et al., 2003; Mezcua et al., 2004). Toxicological data for TP-2 (5-chloro-2-(4chlorophenoxy)phenol/5-chloro-2-(2-chlorophenoxy)phenol) is scarce. Yet, its reported EC50 (460 μg L−1, Melnikov et al., 2016) is similar to that of TCS (0.96–390 μg L−1, Orvos et al., 2002) and both are classified as very toxic to aquatic life with long lasting effects (ECHA, 2017; MERCK, 2017). To the best of our knowledge, no toxicological information is available for TP-3, 2-hydroxy-8-chlorodibenzodioxin toxicity. Hence a first estimation of its biodegradability and toxicity as compared with that of TCS and the other two TPs was obtained by QSAR analysis using CASE Ultra (v. 1.5.2.0, MultiCASE Inc.) and Oasis Catalogic software (V.5.11.6 TB) (further details in Text S1.2). The prediction of readily/
Fig. 4. Temporal changes in TCS and TP's concentrations (aqueous) as function of UV dose. For each TP, concentrations were normalized to its highest detected concentration (i.e., peak area ratios).
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
403
Fig. 5. Proposed reaction pathway of TCS photodegradation.
non-readily biodegradable molecules under different conditions was made based on data from the OECD 301C MITI-I test (MITI, 2006). The statistical Case Ultra AU6 model and the Catalogic OECD 301C model for readily biodegradability predicted that TCS and its TPs should be not readily biodegradable (Table S1) and, thus, could persist as a dead-end products in the environment. TCS, TP-2b and TP-3 showed positive alert in the microtox toxicity to environmental bacteria model (Case Ultra) (Table S2). The three acute Vibrio fischeri and Daphnia magna models (Oasis Catalogic; Table S2) predicted that TP-1 (2,8DCDD) is more toxic than TCS and the other TPs as reflected by its lower IC50 and EC50 values. These predictions fall in line with literature data (see above). TP-2 was predicted to be less toxic than TCS. However, its toxicity was reported as similar to that of TCS and both are classified as very toxic to aquatic life with long lasting effects (see above). The acute toxicity of TP-3 towards Vibrio fischeri and Daphnia magna was predicted as similar to that of TCS, yet it was predicted as less biodegradable than TCS. The in-silico predicted toxicity and low biodegradability of the TPs suggest potential increase in the aquatic toxicity.
4. Conclusions The goal of this study was to investigate the potential of unmasking the V-UV radiation from low pressure Hg lamps (emitting at 185 and 254 nm), commonly used in decentralized treatment systems, for enhancing MPs removal efficiency. TCS served as a model compound for MPs that are not very biodegradable. In-order to characterize TCS photo-oxidation process different experiments were conducted, using different types of lamps, experimental conditions (gases, humidity
etc.) and even the matrix in which the TCS was present (dry film vs. dissolve in DI water). In aqueous solution TCS photolysis was much faster when combined 254/185 nm radiation was applied rather than 254 nm alone, even though the added V-UV radiation (i.e., 185 nm) accounted for only 4% of the lamp UV-C emission. This significant impact of 185 nm radiation was not observed during photolysis of TCS in solution containing methanol nor in dry films, suggesting that hydroxyl radicals, which are produced via water photolysis under V-UV radiation, play a major role in TCS degradation. Photodegradation of TCS in dry films was slower than in aqueous solution, and was shown to be reversibly correlated to the surface concentration of the film (indicating the importance of surface to volume ratio in such conditions). Under both irradiation conditions (i.e., 254 nm and 254/185 nm) the same three TPs were identified in aqueous solutions: TP-1: 2,8dichlorodibenzo-p-dioxin (DCDD), TP-2a/b: 5-chloro-2-(4chlorophenoxy)phenol, 5-chloro-2-(2-chlorophenoxy)phenol and TP3: 2-hydroxy-8-chlorodibenzodioxin. The degradation kinetics suggests that TP-1 and TP-2 are first formed, while TP-3 is formed by photodegradation of TP-1. It should be noted that under the combined irradiation (254/185 nm) both the maximum concentrations and the persistence against photodegradation of these intermediates was lower than under 254 nm radiation alone. Overall, using combined UV-C and V-UV radiation results in faster photo-oxidation of both TCS and its TPs. In silico QSAR toxicity assessment performed in this study provides a first indication that, additionally to the concerns about 2,8dichlorodibenzo-p-dioxin, photo-degradation of TCS might lead to a not readily biodegradable and similar ecotoxic 2-hydroxy-8chlorodibenzodioxin. The fate and effects of these byproducts in aquatic
404
A. Yuval et al. / Science of the Total Environment 601–602 (2017) 397–404
environments cannot be ignored and deserves further investigations. This is especially true for TP-2 and TP-3 for which hardly any empirical data exist. UV-based AOP of MPs is likely to exhibit significant disinfection efficiency, as the latter often requires lower UV doses (Zoschke et al., 2014). It should be noted however that treated wastewater or greywater contain substances that absorb UV irradiation or scavenge radicals, which would reduce the process efficiency compared with the current findings. Hence, in order to evaluate the required UV dose (controlled by radiation flux and retention time) the process should be further studied in flow-through reactors treating wastewater or greywater effluent. Considering the low cost and small size of such low-pressure mercury lamps, the process can be implemented in the future in small on-site systems, such as decentralized wastewater and greywater treatment units, for chemical-free post treatment of persistent pollutants. Acknowledgement Financial support from the State of Lower Saxony, Ministry of Science and Culture, and by the Volkswagen Foundation (Grant No. VWZN2830) is acknowledged. The authors would like to thank Multicase Inc. for kindly providing the CASE Ultra software. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2017.05.172. References Anger, C.T., Sueper, C., Blumentritt, D.J., McNeill, K., Engstrom, D.R., Arnold, W.A., 2013. Quantification of triclosan, chlorinated triclosan derivatives, and their dioxin photoproducts in lacustrine sediment cores. Environ. Sci. Technol. 47:1833–1843. http:// dx.doi.org/10.1021/es3045289. Bock, M., Lyndall, J., Barber, T., Fuchsman, P., Perruchon, E., Capdevielle, M., 2010. Probabilistic application of a fugacity model to predict triclosan fate during wastewater treatment. Integr. Environ. Assess. Manag. 6:393–404. http://dx.doi.org/10.1002/ ieam.134. Cantwell, M.G., Wilson, B.A., Zhu, J., Wallace, G.T., King, J.W., Olsen, C.R., Burgess, R.M., Smith, J.P., 2010. Temporal trends of triclosan contamination in dated sediment cores from four urbanized estuaries: evidence of preservation and accumulation. Chemosphere 78:347–352. http://dx.doi.org/10.1016/j.chemosphere.2009.11.021. Carlson, J.C., Stefan, M.I., Parnis, J.M., Metcalfe, C.D., 2015. Direct UV photolysis of selected pharmaceuticals, personal care products and endocrine disruptors in aqueous solution. Water Res. 84:350–361. http://dx.doi.org/10.1016/j.watres.2015.04.013. De la Cruz, N., Giménez, J., Esplugas, S., Grandjean, D., De Alencastro, L.F., Pulgarín, C., 2012. Degradation of 32 emergent contaminants by UV and neutral photo-fenton in domestic wastewater effluent previously treated by activated sludge. Water Res. 46:1947–1957. http://dx.doi.org/10.1016/j.watres.2012.01.014. ECHA, 2017. European chemical agency [WWW document]. URL https://echa.europa.eu/ home Fatta-Kassinos, D., Hapeshi, E., Achilleos, A., Meric, S., Gros, M., Petrovic, M., Barcelo, D., 2011. Existence of pharmaceutical compounds in tertiary treated urban wastewater that is utilized for reuse applications. Water Resour. Manag. 25:1183–1193. http:// dx.doi.org/10.1007/s11269-010-9646-4. Finlayson-Pitts, B.J., Pitts, J.N., 2000. Chemistry of the upper and lower atmosphere: theory, experiments, and applications. Chemistry of the Upper and Lower Atmosphere. Academic Press http://dx.doi.org/10.1016/B978-0-12-257060-5.50027-7. Gao, Y., Ji, Y., Li, G., An, T., 2014. Mechanism, kinetics and toxicity assessment of •OHinitiated transformation of triclosan in aquatic environments. Water Res. 49: 360–370. http://dx.doi.org/10.1016/j.watres.2013.10.027.
Gonzalez, M.G., Oliveros, E., Wörner, M., Braun, A.M., 2004. Vacuum-ultraviolet photolysis of aqueous reaction systems. J Photochem Photobiol C: Photochem Rev 5:225–246. http://dx.doi.org/10.1016/j.jphotochemrev.2004.10.002. Herrmann, M., Menz, J., Olsson, O., Kümmerer, K., 2015. Identification of phototransformation products of the antiepileptic drug gabapentin: biodegradability and initial assessment of toxicity. Water Res. 85:11–21. http://dx.doi.org/10.1016/j. watres.2015.08.004. Khaleel, N.D.H., Mahmoud, W.M.M., Olsson, O., Kümmerer, K., 2017. Initial fate assessment of teratogenic drug trimipramine and its photo-transformation products – role of pH, concentration and temperature. Water Res. 108:197–211. http://dx.doi. org/10.1016/j.watres.2016.10.078. Kim, I., Yamashita, N., Tanaka, H., 2009. Photodegradation of pharmaceuticals and personal care products during UV and UV/H2O2 treatments. Chemosphere 77:518–525. http://dx.doi.org/10.1016/j.chemosphere.2009.07.041. Kliegman, S., Eustis, S.N., Arnold, W.A., McNeill, K., 2013. Experimental and theoretical insights into the involvement of radicals in triclosan phototransformation. Environ. Sci. Technol. 47:6756–6763. http://dx.doi.org/10.1021/es3041797. Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., Buxton, H.T., 2002. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999–2000: a national reconnaissance. Environ. Sci. Technol. 36: 1202–1211. http://dx.doi.org/10.1021/es011055j. Latch, D.E., Packer, J.L., Arnold, W.A., McNeill, K., 2003. Photochemical conversion of triclosan to 2,8-dichlorodibenzo-p-dioxin in aqueous solution. J. Photochem. Photobiol. A Chem. 158:63–66. http://dx.doi.org/10.1016/S1010-6030(03)00103-5. Lindström, A., Buerge, I.J., Poiger, T., Bergqvist, P.A., Müller, M.D., Buser, H.R., 2002. Occurrence and environmental behavior of the bactericide triclosan and its methyl derivative in surface waters and in wastewater. Environ. Sci. Technol. 36:2322–2329. http:// dx.doi.org/10.1021/es0114254. Maimon, A., Tal, A., Friedler, E., Gross, A., 2010. Safe on-site reuse of greywater for irrigation - a critical review of current guidelines. Environ. Sci. Technol. 44, 3213–3220. Melnikov, F., Kostal, J., Voutchkova-Kostal, A., Zimmerman, J.B., Anastas, Paul T., 2016. Assessment of predictive models for estimating the acute aquatic toxicity of organic chemicals. Green Chem. 18:4432–4445. http://dx.doi.org/10.1039/C6GC00720A. MERCK, 2017. The Merck index online - chemicals, drugs and biologicals [WWW document]. URL https://www.rsc.org/Merck-Index/ (accessed 2.9.17). Mezcua, M., Gómez, M.J., Ferrer, I., Aguera, A., Hernando, M.D., Fernández-Alba, A.R., 2004. Evidence of 2,7/2,8-dibenzodichloro-p-dioxin as a photodegradation product of triclosan in water and wastewater samples. Anal. Chim. Acta 524:241–247. http://dx. doi.org/10.1016/j.aca.2004.05.050. MITI, 2006. OECD 2006. Ready Biodegradability: Modified Miti Test (I). Oppenländer, T., Schwarzwälder, R., 2002. Vacuum-UV oxidation (H2O-VUV) with a xenon excimer flow-trough lamp at 172 nm: use of methanol as Actinometer for VUV intensity measurement and as reference compound for •OH-radical competition kinetics in aqueous systems. J. Adv. Oxid. Technol. 5, 155–163. Orvos, D.R., Versteeg, D.J., Inauen, J., Capdevielle, M., Rothenstein, A., Cunningham, V., 2002. Aquatic toxicity of triclosan. Environ. Toxicol. Chem. 21:1338–1349. http://dx. doi.org/10.1002/etc.5620210703. Petrick, L.M., Sabach, S., Dubowski, Y., 2013. Degradation of VX surrogate profenofos on surfaces via in situ photo-oxidation. Environ. Sci. Technol. 47:8751–8758. http://dx. doi.org/10.1021/es4016537. Qiao, X., Zheng, X., Xie, Q., Yang, X., Xiao, J., Xue, W., Chen, J., 2014. Faster photodegradation rate and higher dioxin yield of triclosan induced by cationic surfactant CTAB. J. Hazard. Mater. 275:210–214. http://dx.doi.org/10.1016/j.jhazmat.2014. 05.012. Singer, H., Müller, S., Tixier, C., Pillonel, L., 2002. Triclosan: occurrence and fate of a widely used biocide in the aquatic environment: field measurements in wastewater treatment plants, surface waters, and lake sediments. Environ. Sci. Technol. 36: 4998–5004. http://dx.doi.org/10.1021/es025750i. Sirés, I., Oturan, N., Oturan, M.A., Rodríguez, R.M., Garrido, J.A., Brillas, E., 2007. ElectroFenton degradation of antimicrobials triclosan and triclocarban. Electrochim. Acta 52:5493–5503. http://dx.doi.org/10.1016/j.electacta.2007.03.011. Wong-Wah-Chung, P., Rafqah, S., Voyard, G., Sarakha, M., 2007. Photochemical behaviour of triclosan in aqueous solutions: kinetic and analytical studies. J. Photochem. Photobiol. A Chem. 191:201–208. http://dx.doi.org/10.1016/j.jphotochem.2007.04. 024. Zoschke, K., Börnick, H., Worch, E., 2014. Vacuum-UV radiation at 185 nm in water treatment–a review. Water Res. 52:131–145. http://dx.doi.org/10.1016/j.watres. 2013.12.034.