Chemosphere 111 (2014) 13–17
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
The impact of chromophoric dissolved organic matter on the photodegradation of 17a-ethinylestradiol (EE2) in natural waters Waldemar Grzybowski ⇑, Jerzy Szydłowski Gdansk University, Institute of Oceanography, Al. Pilsudskiego 46, 81-378 Gdynia, Poland
h i g h l i g h t s Photodegradation of 17a-ethinylestradiol (EE2) was studied natural water samples. Photodegradation was enhanced in presence of chromophoric dissolved organic matter. Half-life times in river and sea water were about one and two days, respectively.
a r t i c l e
i n f o
Article history: Received 25 November 2013 Received in revised form 10 March 2014 Accepted 15 March 2014
Handling Editor: Klaus Kümmerer Keywords: 17a-Ethinylestradiol Photodegradation Sunlight Chromophoric dissolved organic matter
a b s t r a c t 17a-Ethinylestradiol (EE2), the potent estrogen which forms the basic constituent of the contraceptive pill, can undergo degradation in natural waters by sunlight and via secondary reactions initiated by photo-excited dissolved organic matter. The current paper presents the findings of an investigation into the irradiation process of EE2 when dissolved in natural waters. This investigation was carried out under simulated sunlight in samples of sea, river and distilled water at a 17a-ethinylestradiol concentration of 300 ng L1. Several notes of significance may be made on the basis of these results. Firstly, an enhancement of the degradation, observed in the presence of co-absorbing dissolved organic matter, was shown to be proportional to the absorbance of the sample. Secondly, the kinetics of the process obtained during this investigation were within the range of previously reported findings, despite the fact that significantly higher concentrations of EE2 were used in earlier studies. Finally, the environmental half-life times for 17a-ethynyloestradiol, calculated from the results of the experiments, were found to be one and two days in the top layer of river and sea water respectively. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction 17a-Ethinylestradiol (EE2), the basic constituent of the contraceptive pill, is released into surface waters from communal sewage treatment plants and, despite occurring at very low concentrations (in the range of ng L1), can exert noticeable effects on aquatic biota (Sumpter and Johnson, 2005; Kidd et al., 2007). It is not only more stable in the natural environment than natural estrogens (Ying and Kookana, 2003; Robinson and Hellou, 2009) but also possesses the greatest estrogenic potency (Thorpe et al., 2003; van den Belt et al., 2004). The light absorption band of EE2 partly overlaps with the solar spectrum and this xenobiotic may therefore be subject to ⇑ Corresponding author. Tel.: +48 585236842; fax: +48 585236678. E-mail address: [email protected] (W. Grzybowski). http://dx.doi.org/10.1016/j.chemosphere.2014.03.062 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
degradation via direct/primary photoreactions. Indeed, there are studies showing that radiation representative to that of sunlight (wavelength > 280 nm) decreases the concentration of 17a-ethynyloestradiol in solutions (Table 2). Aside from direct photolysis, a significant role in the removal of EE2 is played by indirect (sensitized) reactions initiated by other light absorbers. Reports of experiments with natural water samples (Table 2) clearly show enhancement of EE2 photodegradation as a result of secondary reactions initiated by photo-excited chromophoric dissolved organic matter (CDOM), the main component in the pool of dissolved natural organic substances. This material, referred to in the past as humic substances, is the most important sunlight absorber in natural waters (Lean, 1998). Its significance in aquatic photochemistry has been proved in numerous studies (see Mopper and Kieber, 2002 and references therein).
14
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
Table 1 Physico-chemical characteristics of irradiated waters. pH Baltic Sea Vistula River a b
8.1 7.9
Salinity (practical salinity units) 6.9 0.2
DOC (mg L1)
NO 3 + NO2 (lM)
Fetot (lM) a
4.9 ± 0.2 10.8 ± 0.3
b.db 6.2 ± 1.2
b.d 0.18 ± 0.05
Below detection limit (0.02 lM). Below detection limit (0.8 lM).
Table 2 Degradation rate constant (k) and half-life for EE2 photodegradation under different experimental conditions. EE2 concentration (solute)
k (h1)/half-life time
Irradiation device; optical path length (l)
Source
0.04 ± 0.02/17 h 0.06 ± 0.02/12 h 0.11 ± 0.03/6 h
See text
This study
0.03/23 h
Sunlight, summer sunny day 41°N; l = 1.4 cm
Zuo et al. (2013)
0.08 ± 0.02/7 h 0.13 ± 0.03/5 h
Solar simulator k > 300 nm; l = 1 cm
Whidbey et al. (2012)
0.013 ± 0.006/53 h 0.001 ± 0.004/693 h 0.021 ± 0.014/33 h 0.010 ± 0.011/69 h
Fluorescent lamp UVB = 133 lW cm2
Atkinson et al. (2011)
0.09/7 h
Fluorescent lamp UVA = 115.6 Wm2; k > 300 nm; l = 25.5 cm
Puma et al. (2010)
0.61/1 h 0.69/1 h 0.73/1 h 0.62/1 h 0.007/99 h
Sunlight simulator (300–800 nm); l 2 cm
Matamoros et al. (2009)
0.12/6 h
Fluorescent lamp (k > 290 nm); l = 6.75 cm
0.05/14 h
Sunlight 41°N; l = 1.4 cm
1
300 ng L (Distilled water) (Baltic Sea) (Vistula River) 0.6 mg L1 (Lake water) 10 lg L1 (Distilled water) (10 mg L1 Solution of Swuanee River fulvic acid) 500 ng L1 (Distilled water) (Ottawa River) (Lake Cromwell) (Raisin River) 1 mg L1 (Distilled water) 10–40 mg L1 (Distilled water) (Ebre river) (Besos River) (Mediterranean Sea) (Besos River) 0.150 mg L1 (Distilled water) 8 mg L1 (Estuarine seawater) 1–2 lg L1 (Milli-Q water) (Santa Ana River) 100 lg L1 (Thames River)
Sunlight, May, 41°N; l 2 cm Mazellier et al. (2008) Zuo et al. (2006) 2
0.02 ± 0.002/28 h 0.30 ± 0.015/2 h
Sunlight simulator (290–700 nm) 765 W m
0.0055/126 h
Solar light simulator
In the majority of laboratory experiments EE2 concentrations exceeded ‘‘natural’’ levels by three to six orders of magnitude. Those results can still be reasonably extrapolated to natural aquatic systems, however, provided that the irradiated water sample contains low amounts of other light absorbing species (as is the case, for example, with ocean water), thus showing EE2 photodegradation to be driven mainly by the primary reactions. However, in samples of high concentration of natural organic matter the reactions sensitized by CDOM have to be taken into account. As already mentioned, much of the research has been conducted with EE2 concentrations comparably high to that of natural organic matter. The question arises therefore whether the relative share of indirect reactions could have been underestimated in those studies. This issue was addressed within the current study which purpose was to examine the photodegradation of EE2 in solutions where its concentration was several orders of magnitude lower than that of natural organic matter.
2. Materials and methods River and sea water samples were taken from the Vistula River at Tczew (Poland) and from Gdansk Bay (Baltic Sea) respectively. The samples were aged at room temperature for several months. The physico-chemical parameters were determined for water
; l = 1.5 cm
Lin and Reinhard (2005)
Jürgens et al. (2002)
samples filtered through 0.8 lm Whatman GFF filters. The concentration of dissolved organic carbon was measured using a Shimadzu TOC-5000 analyser. The total iron concentration was determined by a colorimetric method (Collins et al., 1959) using 100 mm long glass cuvette. Prior to performing the analysis the samples were UV irradiated in order to release organically bound iron (for details see Kononets et al., 2002). The sum of nitrate and nitrite was measured colorimetrically as described by Grashoff et al. (1983). The changes in oxygen concentration in water were monitored using a dissolved oxygen meter. Directly before photochemical experiment, the filtered samples of river, sea and Millli-Q water were shaken vigorously for 5 min in glass jars then enriched with 17a-ethinylestradiol (Merck) to concentration of 300 ng mg L1 and poured into quartz glass tubes (1 cm inner diameter). The tubes contained small air-filled headspaces. The samples were irradiated in triplicates, with the tubes tightly sealed with ground glass stoppers to prevent evaporation and dark control samples additionally covered with aluminum foil. The tubes were placed horizontally in a shallow water bath, the temperature of which was kept below 30 °C using tap water cooling, and irradiation was then performed using a solar simulator (SOL 500, Dr. Hönle, Germany) equipped with a H1 filter. Sub-samples of 0.5 ml were withdrawn during the irradiation process. Radiation intensity was measured with UV-B SKU 430 (280–315 nm) and UV-A SKU 420 (315–380 nm) broadband
15
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
300 ng/l EE2 river water
irradiance [W m-2]
absorbance
sea water irradiance
wavelength [nm] Fig. 1. Emission spectrum of the light source and absorbance of 0.01 m deep layers of Vistula River and Baltic Sea water (absorbance of EE2 was multiplied by 1000 to fit the scale).
sensors produced by Skye Instruments. The readouts of the UVB and UVA sensors at the tubes surface were 2.0 W m2 and 31.2 W m2 respectively. The sensors, when exposed to sunlight radiation on a sunny afternoon in late June in Gdynia (54 °N), showed corresponding values of 2.3 W m2 and 26.8 W m2. The absorption spectra of the natural water samples were measured with a Perkin Elmer Lambda 3b dual-beam spectrophotometer, using 100 mm long quartz cuvettes with Milli-Q water as reference. The absorbance values of EE2 solution (Fig. 1) were calculated using data from the measurements of 3 mg L1 EE2 solution. 10% methanol solution served as a reference in this case. Analysis of 17a-ethinylestradiol concentration changes was carried out using commercial EE2 enzyme-linked immunosorbent assay (ELISA) kits in accordance with the instructions of the manufacturer (Tokiwa Chemical Industries, Japan). This method is based on the competition between a labelled standard EE2 and the EE2 in the sample, in terms of binding to a specific monoclonal antibody. The amount of labelled EE2 bound to the antibody (determined by the addition of a colouring reagent) is inversely proportional to the amount of EE2 in the sample. The assay is calibrated using a standard solution of EE2 supplied with the kit. The method has been verified against instrumental techniques by Huang and Sedlak (2001) and has been applied in other environmental research (Hintemann et al., 2006; Shved et al., 2008; Ferguson et al., 2013).
3. Results and discussion Selected physico-chemical characteristics of water serving as EE2 solute were determined in the aged samples of Baltic Sea and Vistula River water after filtering through Whatman GFF filters (Table 1). In addition, the filtered samples have been checked for bacterial activity by measuring biochemical oxygen demand after 24 h long incubation in the dark at room temperature. There were no measurable changes in dissolved oxygen concentration during incubation period (0.3 mg L1: the highest standard deviation of dissolved oxygen measurement was used as a conservative measure of precision). Fig. 1 shows the optical properties of the irradiated samples and the emission spectrum of the sunlight simulator.
As seen in Fig. 1, the absorbance of the samples at wavelengths longer than 290 nm was below 0.1. This means that irradiated samples were optically thin and that absorbing species (i.e. EE2 and CDOM) did not overshadow each other. Irradiation resulted in a decrease in CDOM absorption along the whole spectrum. Absorbance values at 300 nm decreased 20% in Baltic Sea water and 25% in Vistula River water. The EE2 concentration in the non-enriched water samples was not statistically different from the zero standard provided by the ELISA kit manufacturer. An initial EE2 concentration of 300 ng L1 therefore permitted monitoring without the need for pre-concentration and, in the enriched samples kept in darkness during irradiation, no measurable decrease was observed. However, in contrast to the distilled water solutions, initial analysis of the enriched natural water samples showed lower than nominal concentrations of EE2. The apparent decrease ranged up to 30 ng L1 (see Fig. 2). In spite of the very low absorbance of the distilled water solution, a significant decrease in EE2 concentration was noted after 8 h of irradiation. The direct photodegradation of EE2 by UV representative of sunlight (wavelength longer than 280 nm) had already been observed in previous studies (Table 2) but at significantly higher EE2 concentrations. The EE2 concentrations decreased at the highest rate in the Vistula River water solution. The photodegradation rates were in agreement with the optical properties and iron content of the irradiated solutions in that the higher the absorbance/iron concentration, the greater the effects of irradiation. The enhanced photodegradation may be attributed to secondary reactions initiated (sensitized) by CDOM present in the natural water samples (Mopper and Kieber, 2002). Since most of the iron in natural waters is organically bound (Morel and Hering, 1993), CDOM photochemistry cannot be straightforwardly separated from iron-induced photochemistry (Boule et al., 1999). The efficiency of CDOM in 17a-ethynyloestradiol degradation has been assessed by dividing the concentration change by the absorbed energy dose (the latter was calculated using the absorbance spectra of the samples and the emission spectrum of the radiation source). The calculations made for an irradiation period of 8 h show that sea water CDOM is approximately two times more efficient. Since all of the irradiated solutions were optically thin, it was assumed that photoreactions during irradiation had adhered to
16
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
350 0h
8h
16 h
EE2 concentration [ng L-1]
300 250 200 150 100 50 0 distilled water + EE2 (300 ng L-1) sea water + EE2 (300 ng L-1) river water + EE2 (300 ng L-1) Fig. 2. Photodegradation of 17a-ethynyloestradiol (EE2) dissolved in sea, river and distilled water; mean ± mean standard deviation (n = 3).
the law of first order kinetics (Leifer, 1988). For each sample, degradation rate constant (k) and half-life (t½) were calculated using the following equations:
cðtÞ ¼ cðt 0 Þekt
ð1Þ
t 1=2 ¼ ln 2=k
ð2Þ
Note: C(t0) = EE2 concentration at time zero, C(t) = EE2 concentration at time of irradiation. To compare the obtained results with those presented in literature is difficult due to the fact that different light sources and irradiation conditions were used in different studies. In addition to this, a lack of detailed optical characteristics for light sources and irradiated samples does not allow for normalization of the relationship between degradation and the amount of light absorbed. Table 2 presents the photodegradation characteristics obtained during this study, together with previously published findings (some values have been calculated from the available figures). Only data from experiments in which sunlight or a source of sun-like radiation was used has been included here. Despite different concentrations, the degradation kinetics obtained for pure EE2 solutions are generally comparable within the precision of the methods. The only exception (an order of magnitude higher reaction rate constant) was reported for an experiment with very high EE2 concentration. In this case, however, calculation of solution absorbance (e280 nm 2000 M1 cm1, c 0.1 mM, l = 2 cm) suggests that the irradiated sample was not optically thin and that an assumption of first order kinetics cannot be applied (Leifer, 1988). The rates for degradation in solutions containing CDOM obtained in this study do not differ significantly from those acquired in previous studies for solutions of significantly higher EE2 concentrations. Moreover, the range of reported photodegradation rates is relatively narrow considering the enormous span of EE2 concentrations and CDOM levels supposedly tested during these experiments. This suggests high and ubiquitous potential of natural organics to assist in the sunlight-induced removal of EE2 from natural waters although such a conclusion may only be given tentatively due to lack of data on amount/absorbance of co-irradiated natural organic matter. For the results obtained in this study to be extrapolated to actual conditions in the aquatic environment, adjustment to the shape of irradiation vessel is required. A sample in a tube receives radiation from practically all directions due to radiation reflection (e.g. off walls) and, as a result, direct photoreactions in tubes are faster than in a water column, where radiation is only received from above. The so-called geometry factor, defined as a ratio of rate
constant in a tube to that in a ‘‘flat’’ layer, was assumed to be 2 after Leifer (1988). Thus, the corrected half-life times for Baltic Sea and Vistula River water are 24 and 12 h respectively. It should be stressed that the presented data enable predictions to be made in relation to 17a-ethynyloestradiol degradation only at the very top of the water column. Photodegradation in the thicker, optically obscured layer will be dominated by CDOM-induced, secondary reactions and will be difficult to monitor on account of light attenuation and mixing in the water column. Moreover, a comprehensive study of this phenomenon would not only require the basic optical characteristics of a sample and an irradiation source, but also discriminating the impact of other non-CDOM photoreactive components present in the water. Of these, nitrate, iron (Chowdhury et al., 2011) and halide ions (Grebel et al., 2012) have already been proved to be significant factors in the photodegradation of 17b-estradiol, the natural counterpart of EE2.
4. Conclusions Since the radiation intensity used during the experiment was constant (in contrast to natural conditions), irradiation time was converted into a number of ‘‘real’’ days using data published by Diffey (2002) on the daily dose of UVA radiation at a similar latitude (55 °N). The daily dose measured in June was up to 150 J cm2, equivalent to 13 h of irradiation time under the lamp used during the current study. Thus, 17a-ethinylestradiol concentration in the top layer of the Vistula River and the Baltic Sea can be halved in the course of one and two sunny days respectively. In comparison to biodegradation, in which EE2 half-life time may range from 5 (Robinson and Hellou, 2009) to 35 d (Ying and Kookana, 2003) and up to even 80 d (Zuo et al., 2013), sunlight-driven degradation seems to be the most effective natural mode of removing EE2 from the aquatic environment. Acknowledgment This study was funded by the Polish Ministry of Science and Higher Education through Grant No. N N306 254939 for the research project entitled ‘‘Phototransformation of the synthetic steroid estrogen in Gdansk Bay water’’. References Atkinson, S.K., Marlatt, V.L., Kimpe, L.E., Lean, D.R., Trudeau, V.L., Blais, J.M., 2011. Environmental factors affecting ultraviolet photodegradation rates and estrogenicity of estrone and ethinylestradiol in natural waters. Arch. Environ. Contam. Toxicol. 60, 1–7.
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17 Boule, P., Bolte, N., Richard, C., 1999. Phototransformations induced in aquatic media by NO 3 /NO2 , Fe and humic substances. In: Boule, P. (Ed.), Environmental Photochemistry. Springer-Verlag, Berlin, Heidelberg, pp. 181–215. Chowdhury, R.R., Charpentier, P.A., Ray, M.B., 2011. Photodegradation of 17bestradiol in aquatic solution under solar irradiation: kinetics and influencing water parameters. J. Photochem. Photobiol., A 219, 67–75. Collins, P.F., Diehl, H., Smith, G.F., 1959. 2,4,6-Tripyridyls-triazine as a reagent for iron—determination of iron in limestone, silicates, and refractories. Anal. Chem. 31, 1862–1867. Diffey, B.L., 2002. Sources and measurement of ultraviolet radiation. Methods 28, 4– 13. Ferguson, E.M., Allinson, M., Allinson, G., Swearer, S.S., Hassell, K.L., 2013. Fluctuations in natural and synthetic estrogen concentrations in a tidal estuary in south-eastern Australia. Water Res. 47, 1604–1615. Grashoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. Verlag Chemie, Weinheim. Grebel, J.E., Pignatello, J.J., Mitch, W.A., 2012. Impact of halide ions on natural organic matter-sensitized photolysis of 17b-estradiol in saline waters. Environ. Sci. Technol. 46, 7128–7134. Hintemann, T., Schneider, Ch., Schöler, H.F., Schneider, R.J., 2006. Field study using two immunoassays for the determination of estradiol and ethinylestradiol in the aquatic environment. Water Res. 40, 2287–2294. Huang, C.H., Sedlak, D.L., 2001. Analysis of estrogenic hormones in municipal wastewater effluent and surface water using enzyme-linked immunosorbent assay and gas chromatography/tandem mass spectrometry. Environ. Toxicol. Chem. 20, 133–139. Jürgens, M.D., Holthaus, K.I.E., Johnson, A.C., Smith, J.J.L., Hetheridge, M., Williams, R.J., 2002. The potential for estradioland ethinylestradiol degradation in English Rivers. Environ. Toxicol. Chem. 21, 480–488. Kidd, K.A., Blanchfield, P.J., Mills, K.H., Palace, V.P., Evans, R.E., Lazorchak, J.M., Flick, R.W., 2007. Collapse of a fish population after exposure to a synthetic estrogen. Proc. Nat. Acad. Sci. 104, 8897–8901. Kononets, M.Y., Pakhomova, S.V., Rozanov, A.G., Proskurnin, M.A., 2002. Determination of soluble iron species in seawater using ferrozine. J. Anal. Chem. 57, 586–589. Lean, D., 1998. Attenuation of solar radiation in humic waters. In: Hessen, D.O., Tranvik, L. (Eds.), Aquatic Humic Substances, Ecology and Biogeochemistry. Springer, Berlin, pp. 109–113. Leifer, A., 1988. The Kinetics of Environmental Aquatic Photochemistry. American Chemical Society, Washington, DC. Lin, A.Y., Reinhard, M., 2005. Photodegradation of common environmental pharmaceuticals and estrogens in river water. Environ. Toxicol. Chem. 24, 1303–1309.
17
Matamoros, V., Duhec, A., Albaigés, J., Bayona, J.M., 2009. Photodegradation of carbamazepine, ibuprofen, ketoprofen and 17a-ethinylestradiol in fresh and seawater. Water Air Soil Pollut. 196, 161–168. Mazellier, P., Meéité, L., Laat, J.D., 2008. Photodegradation of thesteroid hormones 17b-estradiol (E2) and 17a-ethinylestradiol (EE2) in dilute aqueous solution. Chemosphere 73, 1216–1223. Mopper, K., Kieber, D.J., 2002. Photochemistry and the cycling of carbon, sulfur, nitrogen and phosphorus. In: Hansell, D.A., Carlsson, C.A. (Eds.), Biogeochemistry of Marine Dissolved Organic Matter. Academic Press, Amsterdam, pp. 153–247. Morel, F.M., Hering, J., 1993. Principles and Applications of Aquatic Chemistry. John Wiley & Sons, New York. Puma, G.L., Puddu, V., Tsang, H.K., Gora, A., Toepfer, B., 2010. Photocatalytic oxidation of multicomponent mixtures of estrogens (estrone (E1), 17b-estradiol (E2), 17a-ethynylestradiol (EE2) and estriol (E3)) under UVA and UVC radiation: photon absorption, quantum yields and rate constants independent of photon absorption. Appl. Catal. B 99, 388–397. Robinson, B.J., Hellou, J., 2009. Biodegradation of endocrine disrupting compounds in harbour seawater and sediments. Sci. Total Environ. 407, 5713–5718. Shved, N., Berishvili, G., Baroiller, J., Segner, H., Reinecke, M., 2008. Environmentally relevant concentrations of 17a-ethinylestradiol (EE2) interfere with the growth hormone (GH)/insulin-like growth factor (IGF)-I system in developing bony fish. Toxicol. Sci. 106, 93–102. Sumpter, J.P., Johnson, A.C., 2005. Lessons from endocrine disruption and their application to other issues concerning trace organics in the aquatic environment. Environ. Sci. Technol. 39, 4321–4332. Thorpe, K.L., Cummings, R.I., Hutchinson, T.H., Scholze, M., Brighty, G., Sumpter, J.P., Tyler, C.R., 2003. Relative potencies and combination effects of steroidal estrogens in fish. Environ. Sci. Technol. 37, 1142–1149. van den Belt, K., Berckmans, P., Vangenechten, C., Verheyen, R., Witters, H., 2004. Comparative study on the in vitro/in vivo estrogenic potencies of 17b-estradiol, estrone, 17a-ethynylestradiol and nonylphenol. Aquat. Toxicol. 66, 183–195. Whidbey, Ch.M., Daumit, K.E., Nguyen, T., Ashworth, D.D., Davis, J.C.C., Latch, D.E., 2012. Photochemical induced changes of in vitro estrogenic activity of steroid hormones. Water Res. 46, 5287–5296. Ying, G., Kookana, R., 2003. Degradation of five selected endocrine-disrupting chemicals in seawater and marine sediment. Environ. Sci. Technol. 37, 1256– 1260. Zuo, Y., Zhang, K., Deng, Y., 2006. Occurrence and photochemical degradation of 17a-ethinylestradiol in Acushnet river estuary. Chemosphere 63, 1583–1590. Zuo, Y., Zhang, K., Zhou, S., 2013. Determination of estrogenic steroids and microbial and photochemical degradation of 17a-ethinylestradiol (EE2) in lake surface water, a case study. Environ. Sci.: Proc. Impacts 15, 1529–1535.
Contents lists available at ScienceDirect
Chemosphere journal homepage: www.elsevier.com/locate/chemosphere
The impact of chromophoric dissolved organic matter on the photodegradation of 17a-ethinylestradiol (EE2) in natural waters Waldemar Grzybowski ⇑, Jerzy Szydłowski Gdansk University, Institute of Oceanography, Al. Pilsudskiego 46, 81-378 Gdynia, Poland
h i g h l i g h t s Photodegradation of 17a-ethinylestradiol (EE2) was studied natural water samples. Photodegradation was enhanced in presence of chromophoric dissolved organic matter. Half-life times in river and sea water were about one and two days, respectively.
a r t i c l e
i n f o
Article history: Received 25 November 2013 Received in revised form 10 March 2014 Accepted 15 March 2014
Handling Editor: Klaus Kümmerer Keywords: 17a-Ethinylestradiol Photodegradation Sunlight Chromophoric dissolved organic matter
a b s t r a c t 17a-Ethinylestradiol (EE2), the potent estrogen which forms the basic constituent of the contraceptive pill, can undergo degradation in natural waters by sunlight and via secondary reactions initiated by photo-excited dissolved organic matter. The current paper presents the findings of an investigation into the irradiation process of EE2 when dissolved in natural waters. This investigation was carried out under simulated sunlight in samples of sea, river and distilled water at a 17a-ethinylestradiol concentration of 300 ng L1. Several notes of significance may be made on the basis of these results. Firstly, an enhancement of the degradation, observed in the presence of co-absorbing dissolved organic matter, was shown to be proportional to the absorbance of the sample. Secondly, the kinetics of the process obtained during this investigation were within the range of previously reported findings, despite the fact that significantly higher concentrations of EE2 were used in earlier studies. Finally, the environmental half-life times for 17a-ethynyloestradiol, calculated from the results of the experiments, were found to be one and two days in the top layer of river and sea water respectively. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction 17a-Ethinylestradiol (EE2), the basic constituent of the contraceptive pill, is released into surface waters from communal sewage treatment plants and, despite occurring at very low concentrations (in the range of ng L1), can exert noticeable effects on aquatic biota (Sumpter and Johnson, 2005; Kidd et al., 2007). It is not only more stable in the natural environment than natural estrogens (Ying and Kookana, 2003; Robinson and Hellou, 2009) but also possesses the greatest estrogenic potency (Thorpe et al., 2003; van den Belt et al., 2004). The light absorption band of EE2 partly overlaps with the solar spectrum and this xenobiotic may therefore be subject to ⇑ Corresponding author. Tel.: +48 585236842; fax: +48 585236678. E-mail address: [email protected] (W. Grzybowski). http://dx.doi.org/10.1016/j.chemosphere.2014.03.062 0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
degradation via direct/primary photoreactions. Indeed, there are studies showing that radiation representative to that of sunlight (wavelength > 280 nm) decreases the concentration of 17a-ethynyloestradiol in solutions (Table 2). Aside from direct photolysis, a significant role in the removal of EE2 is played by indirect (sensitized) reactions initiated by other light absorbers. Reports of experiments with natural water samples (Table 2) clearly show enhancement of EE2 photodegradation as a result of secondary reactions initiated by photo-excited chromophoric dissolved organic matter (CDOM), the main component in the pool of dissolved natural organic substances. This material, referred to in the past as humic substances, is the most important sunlight absorber in natural waters (Lean, 1998). Its significance in aquatic photochemistry has been proved in numerous studies (see Mopper and Kieber, 2002 and references therein).
14
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
Table 1 Physico-chemical characteristics of irradiated waters. pH Baltic Sea Vistula River a b
8.1 7.9
Salinity (practical salinity units) 6.9 0.2
DOC (mg L1)
NO 3 + NO2 (lM)
Fetot (lM) a
4.9 ± 0.2 10.8 ± 0.3
b.db 6.2 ± 1.2
b.d 0.18 ± 0.05
Below detection limit (0.02 lM). Below detection limit (0.8 lM).
Table 2 Degradation rate constant (k) and half-life for EE2 photodegradation under different experimental conditions. EE2 concentration (solute)
k (h1)/half-life time
Irradiation device; optical path length (l)
Source
0.04 ± 0.02/17 h 0.06 ± 0.02/12 h 0.11 ± 0.03/6 h
See text
This study
0.03/23 h
Sunlight, summer sunny day 41°N; l = 1.4 cm
Zuo et al. (2013)
0.08 ± 0.02/7 h 0.13 ± 0.03/5 h
Solar simulator k > 300 nm; l = 1 cm
Whidbey et al. (2012)
0.013 ± 0.006/53 h 0.001 ± 0.004/693 h 0.021 ± 0.014/33 h 0.010 ± 0.011/69 h
Fluorescent lamp UVB = 133 lW cm2
Atkinson et al. (2011)
0.09/7 h
Fluorescent lamp UVA = 115.6 Wm2; k > 300 nm; l = 25.5 cm
Puma et al. (2010)
0.61/1 h 0.69/1 h 0.73/1 h 0.62/1 h 0.007/99 h
Sunlight simulator (300–800 nm); l 2 cm
Matamoros et al. (2009)
0.12/6 h
Fluorescent lamp (k > 290 nm); l = 6.75 cm
0.05/14 h
Sunlight 41°N; l = 1.4 cm
1
300 ng L (Distilled water) (Baltic Sea) (Vistula River) 0.6 mg L1 (Lake water) 10 lg L1 (Distilled water) (10 mg L1 Solution of Swuanee River fulvic acid) 500 ng L1 (Distilled water) (Ottawa River) (Lake Cromwell) (Raisin River) 1 mg L1 (Distilled water) 10–40 mg L1 (Distilled water) (Ebre river) (Besos River) (Mediterranean Sea) (Besos River) 0.150 mg L1 (Distilled water) 8 mg L1 (Estuarine seawater) 1–2 lg L1 (Milli-Q water) (Santa Ana River) 100 lg L1 (Thames River)
Sunlight, May, 41°N; l 2 cm Mazellier et al. (2008) Zuo et al. (2006) 2
0.02 ± 0.002/28 h 0.30 ± 0.015/2 h
Sunlight simulator (290–700 nm) 765 W m
0.0055/126 h
Solar light simulator
In the majority of laboratory experiments EE2 concentrations exceeded ‘‘natural’’ levels by three to six orders of magnitude. Those results can still be reasonably extrapolated to natural aquatic systems, however, provided that the irradiated water sample contains low amounts of other light absorbing species (as is the case, for example, with ocean water), thus showing EE2 photodegradation to be driven mainly by the primary reactions. However, in samples of high concentration of natural organic matter the reactions sensitized by CDOM have to be taken into account. As already mentioned, much of the research has been conducted with EE2 concentrations comparably high to that of natural organic matter. The question arises therefore whether the relative share of indirect reactions could have been underestimated in those studies. This issue was addressed within the current study which purpose was to examine the photodegradation of EE2 in solutions where its concentration was several orders of magnitude lower than that of natural organic matter.
2. Materials and methods River and sea water samples were taken from the Vistula River at Tczew (Poland) and from Gdansk Bay (Baltic Sea) respectively. The samples were aged at room temperature for several months. The physico-chemical parameters were determined for water
; l = 1.5 cm
Lin and Reinhard (2005)
Jürgens et al. (2002)
samples filtered through 0.8 lm Whatman GFF filters. The concentration of dissolved organic carbon was measured using a Shimadzu TOC-5000 analyser. The total iron concentration was determined by a colorimetric method (Collins et al., 1959) using 100 mm long glass cuvette. Prior to performing the analysis the samples were UV irradiated in order to release organically bound iron (for details see Kononets et al., 2002). The sum of nitrate and nitrite was measured colorimetrically as described by Grashoff et al. (1983). The changes in oxygen concentration in water were monitored using a dissolved oxygen meter. Directly before photochemical experiment, the filtered samples of river, sea and Millli-Q water were shaken vigorously for 5 min in glass jars then enriched with 17a-ethinylestradiol (Merck) to concentration of 300 ng mg L1 and poured into quartz glass tubes (1 cm inner diameter). The tubes contained small air-filled headspaces. The samples were irradiated in triplicates, with the tubes tightly sealed with ground glass stoppers to prevent evaporation and dark control samples additionally covered with aluminum foil. The tubes were placed horizontally in a shallow water bath, the temperature of which was kept below 30 °C using tap water cooling, and irradiation was then performed using a solar simulator (SOL 500, Dr. Hönle, Germany) equipped with a H1 filter. Sub-samples of 0.5 ml were withdrawn during the irradiation process. Radiation intensity was measured with UV-B SKU 430 (280–315 nm) and UV-A SKU 420 (315–380 nm) broadband
15
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
300 ng/l EE2 river water
irradiance [W m-2]
absorbance
sea water irradiance
wavelength [nm] Fig. 1. Emission spectrum of the light source and absorbance of 0.01 m deep layers of Vistula River and Baltic Sea water (absorbance of EE2 was multiplied by 1000 to fit the scale).
sensors produced by Skye Instruments. The readouts of the UVB and UVA sensors at the tubes surface were 2.0 W m2 and 31.2 W m2 respectively. The sensors, when exposed to sunlight radiation on a sunny afternoon in late June in Gdynia (54 °N), showed corresponding values of 2.3 W m2 and 26.8 W m2. The absorption spectra of the natural water samples were measured with a Perkin Elmer Lambda 3b dual-beam spectrophotometer, using 100 mm long quartz cuvettes with Milli-Q water as reference. The absorbance values of EE2 solution (Fig. 1) were calculated using data from the measurements of 3 mg L1 EE2 solution. 10% methanol solution served as a reference in this case. Analysis of 17a-ethinylestradiol concentration changes was carried out using commercial EE2 enzyme-linked immunosorbent assay (ELISA) kits in accordance with the instructions of the manufacturer (Tokiwa Chemical Industries, Japan). This method is based on the competition between a labelled standard EE2 and the EE2 in the sample, in terms of binding to a specific monoclonal antibody. The amount of labelled EE2 bound to the antibody (determined by the addition of a colouring reagent) is inversely proportional to the amount of EE2 in the sample. The assay is calibrated using a standard solution of EE2 supplied with the kit. The method has been verified against instrumental techniques by Huang and Sedlak (2001) and has been applied in other environmental research (Hintemann et al., 2006; Shved et al., 2008; Ferguson et al., 2013).
3. Results and discussion Selected physico-chemical characteristics of water serving as EE2 solute were determined in the aged samples of Baltic Sea and Vistula River water after filtering through Whatman GFF filters (Table 1). In addition, the filtered samples have been checked for bacterial activity by measuring biochemical oxygen demand after 24 h long incubation in the dark at room temperature. There were no measurable changes in dissolved oxygen concentration during incubation period (0.3 mg L1: the highest standard deviation of dissolved oxygen measurement was used as a conservative measure of precision). Fig. 1 shows the optical properties of the irradiated samples and the emission spectrum of the sunlight simulator.
As seen in Fig. 1, the absorbance of the samples at wavelengths longer than 290 nm was below 0.1. This means that irradiated samples were optically thin and that absorbing species (i.e. EE2 and CDOM) did not overshadow each other. Irradiation resulted in a decrease in CDOM absorption along the whole spectrum. Absorbance values at 300 nm decreased 20% in Baltic Sea water and 25% in Vistula River water. The EE2 concentration in the non-enriched water samples was not statistically different from the zero standard provided by the ELISA kit manufacturer. An initial EE2 concentration of 300 ng L1 therefore permitted monitoring without the need for pre-concentration and, in the enriched samples kept in darkness during irradiation, no measurable decrease was observed. However, in contrast to the distilled water solutions, initial analysis of the enriched natural water samples showed lower than nominal concentrations of EE2. The apparent decrease ranged up to 30 ng L1 (see Fig. 2). In spite of the very low absorbance of the distilled water solution, a significant decrease in EE2 concentration was noted after 8 h of irradiation. The direct photodegradation of EE2 by UV representative of sunlight (wavelength longer than 280 nm) had already been observed in previous studies (Table 2) but at significantly higher EE2 concentrations. The EE2 concentrations decreased at the highest rate in the Vistula River water solution. The photodegradation rates were in agreement with the optical properties and iron content of the irradiated solutions in that the higher the absorbance/iron concentration, the greater the effects of irradiation. The enhanced photodegradation may be attributed to secondary reactions initiated (sensitized) by CDOM present in the natural water samples (Mopper and Kieber, 2002). Since most of the iron in natural waters is organically bound (Morel and Hering, 1993), CDOM photochemistry cannot be straightforwardly separated from iron-induced photochemistry (Boule et al., 1999). The efficiency of CDOM in 17a-ethynyloestradiol degradation has been assessed by dividing the concentration change by the absorbed energy dose (the latter was calculated using the absorbance spectra of the samples and the emission spectrum of the radiation source). The calculations made for an irradiation period of 8 h show that sea water CDOM is approximately two times more efficient. Since all of the irradiated solutions were optically thin, it was assumed that photoreactions during irradiation had adhered to
16
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17
350 0h
8h
16 h
EE2 concentration [ng L-1]
300 250 200 150 100 50 0 distilled water + EE2 (300 ng L-1) sea water + EE2 (300 ng L-1) river water + EE2 (300 ng L-1) Fig. 2. Photodegradation of 17a-ethynyloestradiol (EE2) dissolved in sea, river and distilled water; mean ± mean standard deviation (n = 3).
the law of first order kinetics (Leifer, 1988). For each sample, degradation rate constant (k) and half-life (t½) were calculated using the following equations:
cðtÞ ¼ cðt 0 Þekt
ð1Þ
t 1=2 ¼ ln 2=k
ð2Þ
Note: C(t0) = EE2 concentration at time zero, C(t) = EE2 concentration at time of irradiation. To compare the obtained results with those presented in literature is difficult due to the fact that different light sources and irradiation conditions were used in different studies. In addition to this, a lack of detailed optical characteristics for light sources and irradiated samples does not allow for normalization of the relationship between degradation and the amount of light absorbed. Table 2 presents the photodegradation characteristics obtained during this study, together with previously published findings (some values have been calculated from the available figures). Only data from experiments in which sunlight or a source of sun-like radiation was used has been included here. Despite different concentrations, the degradation kinetics obtained for pure EE2 solutions are generally comparable within the precision of the methods. The only exception (an order of magnitude higher reaction rate constant) was reported for an experiment with very high EE2 concentration. In this case, however, calculation of solution absorbance (e280 nm 2000 M1 cm1, c 0.1 mM, l = 2 cm) suggests that the irradiated sample was not optically thin and that an assumption of first order kinetics cannot be applied (Leifer, 1988). The rates for degradation in solutions containing CDOM obtained in this study do not differ significantly from those acquired in previous studies for solutions of significantly higher EE2 concentrations. Moreover, the range of reported photodegradation rates is relatively narrow considering the enormous span of EE2 concentrations and CDOM levels supposedly tested during these experiments. This suggests high and ubiquitous potential of natural organics to assist in the sunlight-induced removal of EE2 from natural waters although such a conclusion may only be given tentatively due to lack of data on amount/absorbance of co-irradiated natural organic matter. For the results obtained in this study to be extrapolated to actual conditions in the aquatic environment, adjustment to the shape of irradiation vessel is required. A sample in a tube receives radiation from practically all directions due to radiation reflection (e.g. off walls) and, as a result, direct photoreactions in tubes are faster than in a water column, where radiation is only received from above. The so-called geometry factor, defined as a ratio of rate
constant in a tube to that in a ‘‘flat’’ layer, was assumed to be 2 after Leifer (1988). Thus, the corrected half-life times for Baltic Sea and Vistula River water are 24 and 12 h respectively. It should be stressed that the presented data enable predictions to be made in relation to 17a-ethynyloestradiol degradation only at the very top of the water column. Photodegradation in the thicker, optically obscured layer will be dominated by CDOM-induced, secondary reactions and will be difficult to monitor on account of light attenuation and mixing in the water column. Moreover, a comprehensive study of this phenomenon would not only require the basic optical characteristics of a sample and an irradiation source, but also discriminating the impact of other non-CDOM photoreactive components present in the water. Of these, nitrate, iron (Chowdhury et al., 2011) and halide ions (Grebel et al., 2012) have already been proved to be significant factors in the photodegradation of 17b-estradiol, the natural counterpart of EE2.
4. Conclusions Since the radiation intensity used during the experiment was constant (in contrast to natural conditions), irradiation time was converted into a number of ‘‘real’’ days using data published by Diffey (2002) on the daily dose of UVA radiation at a similar latitude (55 °N). The daily dose measured in June was up to 150 J cm2, equivalent to 13 h of irradiation time under the lamp used during the current study. Thus, 17a-ethinylestradiol concentration in the top layer of the Vistula River and the Baltic Sea can be halved in the course of one and two sunny days respectively. In comparison to biodegradation, in which EE2 half-life time may range from 5 (Robinson and Hellou, 2009) to 35 d (Ying and Kookana, 2003) and up to even 80 d (Zuo et al., 2013), sunlight-driven degradation seems to be the most effective natural mode of removing EE2 from the aquatic environment. Acknowledgment This study was funded by the Polish Ministry of Science and Higher Education through Grant No. N N306 254939 for the research project entitled ‘‘Phototransformation of the synthetic steroid estrogen in Gdansk Bay water’’. References Atkinson, S.K., Marlatt, V.L., Kimpe, L.E., Lean, D.R., Trudeau, V.L., Blais, J.M., 2011. Environmental factors affecting ultraviolet photodegradation rates and estrogenicity of estrone and ethinylestradiol in natural waters. Arch. Environ. Contam. Toxicol. 60, 1–7.
W. Grzybowski, J. Szydłowski / Chemosphere 111 (2014) 13–17 Boule, P., Bolte, N., Richard, C., 1999. Phototransformations induced in aquatic media by NO 3 /NO2 , Fe and humic substances. In: Boule, P. (Ed.), Environmental Photochemistry. Springer-Verlag, Berlin, Heidelberg, pp. 181–215. Chowdhury, R.R., Charpentier, P.A., Ray, M.B., 2011. Photodegradation of 17bestradiol in aquatic solution under solar irradiation: kinetics and influencing water parameters. J. Photochem. Photobiol., A 219, 67–75. Collins, P.F., Diehl, H., Smith, G.F., 1959. 2,4,6-Tripyridyls-triazine as a reagent for iron—determination of iron in limestone, silicates, and refractories. Anal. Chem. 31, 1862–1867. Diffey, B.L., 2002. Sources and measurement of ultraviolet radiation. Methods 28, 4– 13. Ferguson, E.M., Allinson, M., Allinson, G., Swearer, S.S., Hassell, K.L., 2013. Fluctuations in natural and synthetic estrogen concentrations in a tidal estuary in south-eastern Australia. Water Res. 47, 1604–1615. Grashoff, K., Ehrhardt, M., Kremling, K., 1983. Methods of Seawater Analysis. Verlag Chemie, Weinheim. Grebel, J.E., Pignatello, J.J., Mitch, W.A., 2012. Impact of halide ions on natural organic matter-sensitized photolysis of 17b-estradiol in saline waters. Environ. Sci. Technol. 46, 7128–7134. Hintemann, T., Schneider, Ch., Schöler, H.F., Schneider, R.J., 2006. Field study using two immunoassays for the determination of estradiol and ethinylestradiol in the aquatic environment. Water Res. 40, 2287–2294. Huang, C.H., Sedlak, D.L., 2001. Analysis of estrogenic hormones in municipal wastewater effluent and surface water using enzyme-linked immunosorbent assay and gas chromatography/tandem mass spectrometry. Environ. Toxicol. Chem. 20, 133–139. Jürgens, M.D., Holthaus, K.I.E., Johnson, A.C., Smith, J.J.L., Hetheridge, M., Williams, R.J., 2002. The potential for estradioland ethinylestradiol degradation in English Rivers. Environ. Toxicol. Chem. 21, 480–488. Kidd, K.A., Blanchfield, P.J., Mills, K.H., Palace, V.P., Evans, R.E., Lazorchak, J.M., Flick, R.W., 2007. Collapse of a fish population after exposure to a synthetic estrogen. Proc. Nat. Acad. Sci. 104, 8897–8901. Kononets, M.Y., Pakhomova, S.V., Rozanov, A.G., Proskurnin, M.A., 2002. Determination of soluble iron species in seawater using ferrozine. J. Anal. Chem. 57, 586–589. Lean, D., 1998. Attenuation of solar radiation in humic waters. In: Hessen, D.O., Tranvik, L. (Eds.), Aquatic Humic Substances, Ecology and Biogeochemistry. Springer, Berlin, pp. 109–113. Leifer, A., 1988. The Kinetics of Environmental Aquatic Photochemistry. American Chemical Society, Washington, DC. Lin, A.Y., Reinhard, M., 2005. Photodegradation of common environmental pharmaceuticals and estrogens in river water. Environ. Toxicol. Chem. 24, 1303–1309.
17
Matamoros, V., Duhec, A., Albaigés, J., Bayona, J.M., 2009. Photodegradation of carbamazepine, ibuprofen, ketoprofen and 17a-ethinylestradiol in fresh and seawater. Water Air Soil Pollut. 196, 161–168. Mazellier, P., Meéité, L., Laat, J.D., 2008. Photodegradation of thesteroid hormones 17b-estradiol (E2) and 17a-ethinylestradiol (EE2) in dilute aqueous solution. Chemosphere 73, 1216–1223. Mopper, K., Kieber, D.J., 2002. Photochemistry and the cycling of carbon, sulfur, nitrogen and phosphorus. In: Hansell, D.A., Carlsson, C.A. (Eds.), Biogeochemistry of Marine Dissolved Organic Matter. Academic Press, Amsterdam, pp. 153–247. Morel, F.M., Hering, J., 1993. Principles and Applications of Aquatic Chemistry. John Wiley & Sons, New York. Puma, G.L., Puddu, V., Tsang, H.K., Gora, A., Toepfer, B., 2010. Photocatalytic oxidation of multicomponent mixtures of estrogens (estrone (E1), 17b-estradiol (E2), 17a-ethynylestradiol (EE2) and estriol (E3)) under UVA and UVC radiation: photon absorption, quantum yields and rate constants independent of photon absorption. Appl. Catal. B 99, 388–397. Robinson, B.J., Hellou, J., 2009. Biodegradation of endocrine disrupting compounds in harbour seawater and sediments. Sci. Total Environ. 407, 5713–5718. Shved, N., Berishvili, G., Baroiller, J., Segner, H., Reinecke, M., 2008. Environmentally relevant concentrations of 17a-ethinylestradiol (EE2) interfere with the growth hormone (GH)/insulin-like growth factor (IGF)-I system in developing bony fish. Toxicol. Sci. 106, 93–102. Sumpter, J.P., Johnson, A.C., 2005. Lessons from endocrine disruption and their application to other issues concerning trace organics in the aquatic environment. Environ. Sci. Technol. 39, 4321–4332. Thorpe, K.L., Cummings, R.I., Hutchinson, T.H., Scholze, M., Brighty, G., Sumpter, J.P., Tyler, C.R., 2003. Relative potencies and combination effects of steroidal estrogens in fish. Environ. Sci. Technol. 37, 1142–1149. van den Belt, K., Berckmans, P., Vangenechten, C., Verheyen, R., Witters, H., 2004. Comparative study on the in vitro/in vivo estrogenic potencies of 17b-estradiol, estrone, 17a-ethynylestradiol and nonylphenol. Aquat. Toxicol. 66, 183–195. Whidbey, Ch.M., Daumit, K.E., Nguyen, T., Ashworth, D.D., Davis, J.C.C., Latch, D.E., 2012. Photochemical induced changes of in vitro estrogenic activity of steroid hormones. Water Res. 46, 5287–5296. Ying, G., Kookana, R., 2003. Degradation of five selected endocrine-disrupting chemicals in seawater and marine sediment. Environ. Sci. Technol. 37, 1256– 1260. Zuo, Y., Zhang, K., Deng, Y., 2006. Occurrence and photochemical degradation of 17a-ethinylestradiol in Acushnet river estuary. Chemosphere 63, 1583–1590. Zuo, Y., Zhang, K., Zhou, S., 2013. Determination of estrogenic steroids and microbial and photochemical degradation of 17a-ethinylestradiol (EE2) in lake surface water, a case study. Environ. Sci.: Proc. Impacts 15, 1529–1535.
