Standards and reagents
As EDCs, two steroids 17β-estradiol (E2) and 17α-ethinylestradiol (EE2) were chosen as target compounds, because they are frequently identified in the environment as a consequence of their high public consumption (Vajda et al. 2008; Fent et al. 2006). E2 is a natural hormone, relatively bioaccumulative and persistent in the environment while EE2 is a synthetic hormone from cholesterol and the active ingredient in birth control pills (Maniero et al. 2008).
Standards for E2 (98 %) and EE2 (98 %) were acquired in powder form from Sigma–Aldrich and Fluka (USA). As reference compound in competitive kinetics, sodium phenolate 99 % was used; potassium dihydrogen phosphate (KH2PO4) and tert-butyl alcohol were acquired from Sigma–Aldrich (USA). Tert-butyl alcohol was used as scavenger of OH radical in the experiments (reagent grade) and without further purification. All solvents were HPLC grade; acetonitrile was acquired from Tedia (USA), methanol from JT Baker (USA) and ethyl acetate from Burdick and Jackson (USA). Derivatization was performed directly using N,O-bis (trimethylsilyl) trifluoroacetamide (BSTFA) + 1 % trimethylchlorosilane (TMCS), from Sigma-Aldrich. Stock solutions of the selected compounds were prepared with deionized water (Millipore). Ozone gas was generated from a Pacific Ozone G11 equipment operated at 20 °C from pure O2, and then MilliQ water was saturated with ozone gas to obtain the standard solution of ozone (0.25 mmol L−1).
Analytical methods
Steroids concentration was determined by solid phase extraction; sodium phenolate concentration in the extracted samples was determined by high performance liquid chromatography (HPLC) using a Varian ProStar 7725 equipment with a Varian ProStar 230 diode array detector (DAD) (Walnut Creek, CA) using maximal absorption wavelengths (λmax) of 197 nm following the methodology from Vallejo-Rodríguez et al. (2011). Residual steroid was determined by gas chromatography (GC) using an Agilent Technologies chromatograph model 6890 coupled with a mass spectrometer (MS) 5975 and with a quadruple mass filter with an autosampler model 7683, prior the sample was derivatized. The GC was equipped with a HP5MS 30 m × 0.25 mm capillary column (Agilent, USA), 0.25 mm internal diameter (i.d.) with a stationary phase of 5 % phenyl and 95 % dimethyl polysiloxane, and a 0.25 μm film thickness. Oven temperature was programmed at 120 °C for 20 min, ramped at 15 °C min−1 to 250 °C and finally increased 5 °C min−1 up to 300 °C and held for 5 min. The injector temperature was 300 °C in splitless mode using an injection volume of 1.0 μL. Helium (99.999 %, INFRA) was used as the carrier gas at a constant flow rate of 1.0 mL min−1. Mass spectra was obtained by electron impact (EI) at 70 eV using ionization source at 200 °C. Mass scanning was used in SCAN mode for optimizing the separation and identification of compounds and selected ion monitoring (SIM) for quantification (Bowden et al. 2009). Ozone concentration in water was performed by indigo colorimetric method 4500 (Eaton 2005).
Determination of rate constants
Competitive kinetics method
Competitive kinetics was performed by duplicate in batch reactors using 50 mL volumetric flasks at T = 20 ± 1 °C and pH = 6 to ensure the molecular action of ozone and prevent fast decomposition into hydroxyl radicals. Steroid/sodium phenolate ratio was equimolar; E2/ozone and EE2/ozone stoichiometric ratios were 5:1 and 1:5 from standard solutions of 3.6 and 11.8 μmol L−1 for E2 and EE2, respectively (Huber et al. 2003). In general form, the model for the competitive kinetics method is expressed by Eq. (1) (Benitez et al. 2009; Huber et al. 2003).
$$ { \ln }\left( {{{\left[ {\text{M}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{M}} \right]_{ 0} } {\left[ {\text{M}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{M}} \right]_{\text{t}} }}} \right)\,\, = \,\,\frac{{{\text{n}}_{\text{M}} }}{{{\text{n}}_{\text{Phen}} }}\left( {\frac{{{\text{k}}_{{{\text{O}}_{ 3} , {\text{M}}}} }}{{{\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} }}} \right)\,{ \ln }\,\left( {{{\left[ {\text{Phen}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{Phen}} \right]_{ 0} } {\left[ {\text{Phen}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{Phen}} \right]_{\text{t}} }}} \right) $$
(1)
Here [M]0 and [M]t are the concentrations of the steroid at t = 0 and at a time t, respectively, nM is the stoichiometric coefficient and \( {\text{k}}_{{{\text{O}}_{ 3} , {\text{M}}}} \) is the second order rate constant. Sodium phenolate was used as the reference compound, thus \( \left[ {\text{Phen}} \right] \), \( {\text{n}}_{\text{Phen}} \) and \( {\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} \) represent the concentration, the stoichiometric coefficient and the rate constant of sodium phenolate, respectively. Stoichiometric coefficients for E2 and EE2 used in this work was taken from Vallejo-Rodríguez et al. (2014), where 1 mol of ozone is necessary to oxidize 1 mol of E2 or EE2.
From Eq. (1), the plot \( { \ln }\left( {{{\left[ {\text{M}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{M}} \right]_{ 0} } {\left[ {\text{M}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{M}} \right]_{\text{t}} }}} \right) \) versus \( { \ln }\left( {{{\left[ {\text{Phen}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{Phen}} \right]_{ 0} } {\left[ {\text{Phen}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{Phen}} \right]_{\text{t}} }}} \right) \), gave the slope \( {{{\text{k}}_{{{\text{O}}_{ 3} , {\text{M}}}} } \mathord{\left/ {\vphantom {{{\text{k}}_{{{\text{O}}_{ 3} , {\text{M}}}} } {{\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} }}} \right. \kern-0pt} {{\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} }} \), thus \( {\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} \) is known, and \( {\text{k}}_{{{\text{O}}_{ 3} , {\text{M}}}} \) can be determined, here \( {\text{k}}_{{{\text{O}}_{ 3} , {\text{Phen}}}} \) = 2.4 × 105 L mol−1 s−1 at pH 6 (Hoigné and Bader 1983).
Stopped flow method
The acquisition of kinetic data and photometric measurements by stopped flow were performed on a BioLogic SFM-3000/S equipment. Stoichiometric ratios from 1:1 to 1:20 for E2/ozone and EE2/ozone were established injecting with the syringe system of the equipment (corresponding to 1:1 to 1:5 for E2-EE2/ozone volume ratios), different volumes of ozone at pH = 6 to ensure the molecular action of ozone and prevent fast decomposition of ozone into hydroxyl radicals. For both steroids, the detections were performed at 197 nm. In order to obtain the second order kinetic constant, the first step is the determination of the absolute rate constant under pseudo-first-order conditions (Eq. (2)):
$$ { \ln }\left( {{{\left[ {\text{M}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{M}} \right]_{ 0} } {\left[ {\text{M}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{M}} \right]_{\text{t}} }}} \right) = {\text{k}}_{\text{obs}} \,{\text{t}} $$
(2)
For each experiment performed in the stopped flow, a value of kobs is obtained by plotting \( { \ln }\left( {{{\left[ {\text{M}} \right]_{ 0} } \mathord{\left/ {\vphantom {{\left[ {\text{M}} \right]_{ 0} } {\left[ {\text{M}} \right]_{\text{t}} }}} \right. \kern-0pt} {\left[ {\text{M}} \right]_{\text{t}} }}} \right) \) versus time, where the slope is kobs. Here kobs is the pseudo-first-order rate constant and t is time.
From Eq. (3), \( {\text{k}}_{{{\text{O}}_{ 3} }} \) values were obtained from the slope by plotting values of k1 versus \( [ {\text{O}}_{ 3} ]_{ 0} \) (Hoigné et al. 1985):
$$ {\text{k}}_{ 1} = {\text{k}}_{{{\text{O}}_{ 3} ,{\text{M}}}} [ {\text{O}}_{ 3} ]_{ 0} $$
(3)
Here \( {\text{k}}_{{{\text{O}}_{ 3} }} \) is the second order rate constant for M and \( [ {\text{O}}_{ 3} ]_{ 0} \) is the initial concentration of ozone for each experiment.