Degradation and estrogenic activity removal of 17β-estradiol and 17α-ethinylestradiol by ozonation

Degradation and estrogenic activity removal of 17β-estradiol and 17α-ethinylestradiol by ozonation and O 3/H 2O 2

Milena Guedes Maniero a ,Daniele Maia Bila b ,Márcia Dezotti a,?

a Chemical Engineering Program —COPPE,Federal University of Rio de Janeiro,P.O.Box 68502,21941-972,Rio de Janeiro,RJ,Brazil b

Department of Environmental and Sanitary Engineering,State University of Rio de Janeiro,Rio de Janeiro,RJ,Brazil

A R T I C L E D A T A

A B S T R A C T

Article history:

Received 13February 2008

Received in revised form 21July 2008Accepted 6August 2008

Available online 20September 2008This work investigated the degradation of a natural (17β-estradiol)and a synthetic (17α-ethinylestradiol)estrogens (pure or in the mixture)and the removal of estrogenic activity by the ozonation and O 3/H 2O 2process in three different pHs (3,7and 11).The effect of oxidation via OH radical was evaluated adding a radical scavenger (t-butanol)in the medium.Estrogenic activity was performed using the YES assay.17β-estradiol and 17α-ethinylestradiol presented similar estrogenic potential and the association of these estrogens resulted in an addictive effect for estrogenic activity.Ozonation and O 3/H 2O 2processes were effective in removing the estrogens in aqueous solution.In the mixture at pH 11,removals were higher than 98%and 96%for 17β-estradiol and 17α-ethinylestradiol,respectively.In pH 3,17β-estradiol and 17α-ethinylestradiol removals were 100%and 99.7%,respectively.When estrogens were treated separately,the removals in pH 11were superior to 99.7and 98.8%,while in pH 3were 100%and 99.5%for 17β-estradiol and 17α-ethinylestradiol,respectively.17α-ethinylestradiol has been always removed at lower rates (pure or in the mixture)for all applied conditions.Estrogenic activity was completely removed in pH 3for ozonation or O 3/H 2O 2.The samples oxidized in pH 11presented higher estrogenic activity than those in pH 7.Estrogens removal was lower at pHs 7and 11,when the scavenger was added to the media.The higher estrogen residual concentrations found in ozonation in presence of tert-butanol are contributing for higher estrogenic activity observed in pHs 7and 11.By-products with estrogenic activity were formed by oxidation via OH radical.Only a few compounds could be identified in pHs 7and 11and they have a phenolic ring,which,probably is contributing to the estrogenic activity observed.

?2008Elsevier B.V.All rights reserved.

Keywords:Ozonation O 3/H 2O 217β-estradiol 17α-ethinylestradiol EDC YES assay

1.Introduction

There is a growing concern about water quality all over the world.Recently,many studies reported alterations on the reproduction of animals and humans.They are possibly associated to the presence of some micropollutants,known as “Endocrine Disrupters Chemicals ”(EDCs),found in super-ficial and underground waters.Works published in the literature showed that endocrine disrupters can increase the incidence of cancer of the testicle,ovary and breast as well as to reduce fertility and spermatozoids number and promote fish feminization (Coleman et al.,2005;Harrison et al.,1997).

Natural and synthetic estrogens are the main substances responsible for estrogenic activity found in domestic sewage.Estrogens 17β-estradiol and 17α-ethinylestradiol are excreted daily by humans in domestic sewage and are only partially removed in domestic wastewater treatment plants (DWTP).Consequently,they are continuously discharged into receiv-ing bodies (Ternes et al.,1999).17β-estradiol is the main natural estrogen responsible by feminine characteristics

S C I E N C E O F T H E T O T A L E N V I R O N M E N T 407(2008)105–115

?Corresponding author.Tel.:+552125628347;fax:+552125628300.E-mail address:mdezotti@peq.coppe.ufrj.br (M.

Dezotti).

0048-9697/$–see front matter ?2008Elsevier B.V.All rights reserved.doi:

10.1016/j.scitotenv.2008.08.011

ava i l a b l e a t w w w.s c i e n c e d i r e c t.c o m

w w w.e l s ev i e r.c o m /l o c a t e /s c i t o t e n v

formation.17α-ethinylestradiol is a synthetic estrogen found in contraceptive pills and applied in hormonal reposition therapies.

Natural and synthetic estrogens are effective endocrine disrupters at ng l?1levels(Nogueira,2003;Routledge et al., 1998).They are among the substances which cause endocrine alterations on aquatic organisms(Gomes et al.,2004;Lai et al., 2002).In vivo potency of17α-ethinylestradiol is10–50fold higher than that of17β-estradiol,probably due its lower metabolization (Brian et al.,2005).Already0.1ng l?1of17α-ethinylestradiol induces the expression of vitellogenin in fish,0.1–15ng l?1can affect sex differentiation and2–10ng l?1may affect fecundity. Life-long exposure to5ng l?1leads to significant reduction on fish fecundity(Fent et al.,2006;Nash et al.,2004).Thus,given that its concentration in the environment frequently is between 0.5and7ng l?1(Bila and Dezotti,2003),17α-ethinylestradiol may be a significant contributor to reproductive disfunction in wild fish(Fent et al.,2006).

Estrogenic activity of a substance is defined as its capacity to bind to the estrogen receptor and elucidate an estrogenic response.In this context,the molecular structure of the pollutant is very relevant.The presence of a polar group capable to form hydrogen bridge(hydroxyl,for example)with an aromatic ring seems to be of paramount importance to an estrogenic response(Hamblen et al.,2003).Studies suggest that there is only one perfect bind of17β-estradiol molecule with estrogen receptor;i.e.,through the phenolic ring of the steroid.

There are a variety of in vivo and in vitro assays to evaluate estrogenic activity.In vivo assays use parameters as sexual organs weight,cell differentiation,expression of proteins and enzymatic activity(Gray Jr.et al.,1997;Baker,2001).In vivo experiments for investigating estrogenic effects are,in gen-eral,time-consuming and expensive.However,several in vitro assays have been established to identify estrogenic potentials in environmental samples.Among in vitro assays there are those that evaluate the interaction with hormonal receptors and cell proliferation,such as YES and MCF-7assays, respectively.

The presence of endocrine disruptors in sewage treatment plants and sources of potable water indicates the need for evaluating the treatment processes and their efficiency in removing such substances.Ozonation,photocatalysis,O3/ H2O2and O3/UV,showed to be promising techniques for removing these micropollutants(Bila et al.,2007;Ternes et al., 2003;Huber et al.,2003;Zwiener and Frimmel,2000;Irmak et al.,2005).According to Bila et al.(2007)ozonation was very efficient for removal of17β-estradiol in aqueous solutions (N99%),using low ozone dosages.According to Ternes et al. (2003),for ozone doses of15mg l?1,the natural estrogen estrone was reduced below the limit of quantification.Irmak et al.(2005)showed that the time needed for complete conversion of0.1mmol of17β-estradiol was55min for the applied O3dose of15.78·10?3mmol min?1.According to Huber et al.(2003)17-estradiol and17α-ethinylestradiol exhibited high rate constants with ozone,showing that ozonation and advanced oxidative processes are promising processes for efficient removal of pharmaceuticals in waters.However,in addition to the removal of the endocrine disrupter it should be assured that the biological activity is also removed,since some reactions products may be active.

The objective of this study was to evaluate the effect of molecular ozone and hydroxyl radical for removing estrogenic activity during degradation of17β-estradiol,17α-ethinylestradiol and their mixture by ozonation and O3/H2O2process.The influence of pH and ozone dosage in the removal of these substances and in the formation of by-products was investigated.

2.Materials and methods

2.1.Chemicals

17β-Estradiol(98%purity),17α-ethinylestradiol(98%purity), BSTFA(bis(trimethylsilyl)tri-fluoroacetamide),KH2PO4, (NH4)2SO4,MgSO4,Fe2(SO4)3,L-leucine,L-histidine,adenine,L-arginine-HCl,L-methionine,L-tyrosine,L-isoleucine,L-lysine-HCl,L-phenylalanine,L-glutamic acid,L-valine,L-serine,thia-mine,pyridoxine,calcium pantetonate,inositol,D-glucose, aspartic acid,L-threonine,copper sulfate(II)and KOH pellets were supplied by Sigma-Aldrich.Biotin and absolute ethanol were supplied by Merck.Hexane,methanol,acetone and tert-butanol were supplied by Tedia Brazil.Chlorophenol red-β-D-galactopyranoside(CPRG)was supplied by Roche Diagnostics GmbH.

2.2.Stock solutions

Stock solutions of17β-estradiol and17α-ethinylestradiol were prepared at100mg l?1in acetone and stored at4°C.Samples for ozonation and O3/H2O2were prepared by spiking Milli-Q Biocell water with the stock solution in order to achieve initial concentrations of10μg l?1and50μg l?1for each estrogen.

2.3.Experimental set-up

Experiments were carried out in an ozonation unit.This unit is constituted of an ozone generator(Unitek—model UTK),a glass contact column(500mm height×70mm diameter?1.0l volume)and an ozone analyzer for the gas phase(IN USA, ASX-Mod H1).The ozone generator is able to produce up to5g O3h?1,using a blend of pure oxygen and pure nitrogen as the feed gases at a flow rate of93.6g h?1(1.56g min?1).This mixture was used instead of air in order to allow operation with low ozone concentrations.

Ozonation experiments were performed in aqueous solu-tion at different initial pH values:3,7and11.Sulfuric acid or sodium hydroxide was used to adjust the solution pH.Tests to evaluate degradation and estrogenic activity were performed using initial concentrations of10μg l?1for each estrogen separately and20μg l?1for the mixture(10μg l?1of each estrogen).For UV spectrophotometry,an initial concentration of50μg l?1for each estrogen was used.The ozone consumed concentration ranged from1.0to25mg O3l?1,corresponding to ozonation times of10s to7min.The ozone consumed concentration corresponds to the amount of ozone absorbed by the aqueous solution volume and was calculated using the ozone gas concentration at the inlet and the outlet of the bubble column,the liquid volume and the ozonation time.

To identify the by-products formed during ozonation, aqueous solutions containing1mg l?1of17β-estradiol and

106S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

17α-ethinylestradiol were used.Experiments were conducted at three pH values(3,7and11).A higher estrogen concentra-tion was used in these experiments to render feasible the analytical determination of by-products.

In O3/H2O2experiments,the hydrogen peroxide was added to the sample immediately before ozone bubbling.Molar ratio of O3/H2O2was2:1.

Tests with radical scavenger were carried out and tert-butanol(4mM)was added to the sample immediately before of the assay start.

2.4.Analytical methods

The method for17β-estradiol and17α-ethinylestradiol deter-mination used solid phase extraction(SPE),followed by derivatization with BSTFA(bis(trimethylsilyl)tri-fluoroaceta-mide)and analysis by gas chromatography(GC)or gas chromatography coupled with mass spectrometer(GC-MS).

Extraction was carried out using a solid phase cartridge C18 (Varian0.5g),previously conditioned by flushing with hexane (3×2ml),followed by acetone(1×2ml)and methanol(3×2ml). The cartridges were,then,washed with water(5×2ml,with pH adjusted to3)and then one liter of sample(pH3)was percolated through the cartridges at a flow rate of10ml min?1 with the aid of vacuum.Subsequently,the solid phase was completely dried and the analytes were eluted with acetone (4×1ml).Acetone extracts were completely evaporated under a gentle nitrogen stream,derivatized by adding50μl of BSTFA and heated at60°C for30min.

The GC analyses were performed on a GC-17A Shimadzu gas chromatograph equipped with flame ionization detector (FID)and a fused-silica capillary column(Varian,Factor Four 5MS—30m×0.25mm×0.25μm).The carrier gas was hydrogen (2.0ml min?1),and the temperature program was as follows: initial temperature130°C,at10°C min?1to230°C(2min),at 2°C min?1to260°C(2min),at10°C min?1to300°C(2min). The injector and the detector temperatures were set to280and 320°C,respectively.The injection volume was0.6μl in splitless mode.

The GC-MS analyses were carried out on a Hewlett-Packard Model5972MSD coupled to an HP5890GC(Agilent Technologies, Avondale,USA).Helium was used as carrier gas at a flow rate of1ml min?1.The mass spectrometer was operated on electron impact mode at70eV and scan mode of analysis was used in the mass range of45–600Da.The same capillary column was used with the following temperature program:initial temperature 50°C,at30°C min?1to150°C and at10°C min?1to300°C(5min). The column head pressure was kept at65kPa.The derivatized extract was injected in splitless mode,1.0μl injection volume. The injector and the FID temperatures were set to290and300°C, respectively.The ozonation process by-products were analyzed by GC/MS.The compounds were identified by comparison of the mass spectra with those from the literature data,NIST and Wiley275mass spectra electronic library.

2.5.The yeast estrogen screen(YES)

The estrogenic activity of the ozonated samples was deter-mined using a recombinant receptor gene assay in yeast cells, known as yeast estrogen screen(YES).2.5.1.Solid phase extraction of aqueous samples for the YES assay

The ozonated samples were extracted following the same methodology already described in Section 2.4.The only difference was that the extracts were reconstituted in2ml of ethanol and stored at4°C until use in the YES assay.

2.5.2.Assay procedure

Recombinant yeast(Saccharomyces cerevisae)cells were kindly provided by Professor J.P.Sumpter from Brunel University, Uxbridge,UK.Yeast cells culture and YES assay was performed according to the method of Routledge and Sumpter (1996)with some modifications.

The samples were serially diluted in absolute ethanol.A total of10μl aliquots of each concentration was then transferred to a96-well optically flat microtitre plate and tested in duplicate.200μl of seeded yeast medium containing chlorophenol red-β-D-galactopyranoside(CPRG)were,then, added to the microtitre plate,after which the plates were sealed,shaken vigorously for5min and incubated for72h at 30°C.Each plate contained duplicate rows containing a serial dilution of sample,duplicate of ethanol(solvent control)and a row containing a serial dilution of the appropriate17β-estradiol(positive control).After incubation,the absorbance was read at540nm(for color)and at620nm(for turbidity) using a plate read(BIO-TEK EL808).

The concentrations that elucidate50%of estrogenic activity(EC50)were obtained from dose–response curves. Relative potency(RP)was calculated taking EC50of17β-estradiol as reference.EC50values were very close for both estrogens,individually or in the mixture(Eq.(1)).

RP substance or mixture?

EC50E2

EC50substance or mixture

e1T2.6.UV spectrophotometry

UV scanning was performed on a Shimadzu spectrophot-ometer,model UV Mini-1240.Ozonized samples of17β-estradiol and17α-ethinylestradiol in Milli-Q water(initial concentrations of50μg l?1)were analyzed.The samples were scanned from190to300nm.

The analytical methodology employed in the preparation of the samples was based on SPE,in which cartridges were conditioned by washing with methanol(4×2ml),followed by washing with water(5×2ml,pH3).Afterwards,1l of sample with pH3was percolated through the cartridge with a10ml min?1flow rate.Subsequently,the analytes were eluted with methanol(4×1ml).This solution was analyzed at the UV–Vis spectrophotometer.

3.Results and discussion

Two main oxidants(molecular O3and OH radical)may be acting in the ozonation reaction.Lower pH(pH b4)favors the oxidation via molecular ozone that is more selective and reacts with specific groups.Increase of pH or addiction of H2O2 favors O3decomposition into hydroxyl radicals.In pH N10,O3 is instantaneously decomposed into hydroxyl radicals.In pH7 both oxidants can be acting.

107

S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

According to the literature,the oxidation of organic micropollutants by O3is an efficient process for compounds with functional groups such as amino groups,activated aromatic systems(i.e.,phenolics)or double bonds(von Gunten,2003). Ozone reacts rapidly with phenolic compounds in aqueous solution(Huang and Shu,1995;Mvula et al.,2001).This fact indicates that ozonation can be efficient for removal of estro-genic activity related to17β-estradiol and17α-ethinylestradiol, since phenolic group has affinity to bond with estrogen receptor and elucidate an estrogenic response(Birkett and Lester,2003). In general,ozonation increases the number of functional groups and the molecule’s polarity(Ternes et al.,2003).

3.1.Removal of17β-estradiol and17α-ethinylestradiol

3.1.1.Ozonation process

Ozonation was effective to remove17β-estradiol and17α-ethinylestradiol and their mixture from aqueous solution.In both cases the extent of17β-estradiol removal(pure or in the mixture)was slightly higher than that of17α-ethinylestradiol for the same ozone dose consumed and pH value.Fig.1shows the removal of estrogens in the mixture by ozonation and O3/ H2O2at different pH values.pH variation during ozonation was not observed.Estrogens were not oxidized by pure oxygen and not stripped by gas bubbling,as confirmed by blank experi-ments results.

In the mixture at pH11,removals were higher than98% and96%for17β-estradiol and17α-ethinylestradiol,respec-tively.In pH3,17β-estradiol and17α-ethinylestradiol removals were100%and99.7%,respectively,as can be observed in Fig.1A.When estrogens were treated separately, the removals in pH11were superior to99.7and98.8%,while in pH3were100%and99.5%for17β-estradiol and17α-ethinylestradiol,respectively.Thus,for all ozone doses and pH tested,while in mixture,17β-estradiol and17α-ethinyles-tradiol residual concentrations were lower than200ng l?1

and

Fig.1–Removal of the estrogens in the mixture17β-estradiol/17α-ethinylestradiol by(A)ozonation in pHs3,7and11.(B)O3/H2O2 at pHs3and7.

108S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

400ng l?1,respectively.However,when these substances were treated separately,residual concentrations were lower than 30ng l?1and120ng l?1,respectively.17α-ethinylestradiol was always removed at slower rates(pure or in the mixture)for all applied conditions.The detection limit of the analytical method was 5.0±0.1ng l?1.However,for concentrations below0.1μg l?1the variability was higher.Thus,due to analytical challenges,estrogens removal in low concentra-tions should be investigated in more detail to clearly establish the influence of pH on the oxidation reaction rate.

The estrogens removal increased with ozone dose.How-ever,this increment was small when the estrogens residual concentrations were very low.Thus,indicating that these substances are hardly degraded in these very low concentra-tions.This is not the case,when they are present in higher concentrations,as experimentally observed.

According to Huber et al.(2003)the reactive group of17β-estradiol is the phenolic group.These authors have deter-mined the reaction rate constants for17α-ethinylestradiol,a synthetic estrogen,with O3(k O

3

=~7×109M?1s?1)and with OH radical(k OH=9.8×109M?1s?1)in pHs6and7.They have demonstrated that near neutral pH the reaction rate constants for O3and OH radical have practically the same value.

Different forms of phenolic group can be found in different pH values.In acidic pH(pH b4),phenolate concentration is negligible.In alcaline pH(pH N10),phenolic ring is present as phenolate,which is more reactive than phenol(Huber et al., 2003).According to Hoignéand Bader(1983)the reaction rate constant for the reaction of ozone with phenolate(1.4±0.4×109M?1s?1)is much higher than that with phenol(1.3±0.2×103M?1s?1)in the pH range of2–6.Thus,one should expect to find slower oxidation rates at pH3for17β-estradiol and17α-ethinylestradiol removal.However,a significant difference was not verified for estrogens residual concentra-tions at all pHs evaluated,mainly at the higher ozone dosages, when estrogens removals were very similar.

Although extensive removals of estrogens were achieved, residual concentrations in the range of nanograms per liter were always found.This is a matter of concern because these concentrations are sufficiently high to cause effects in fishes (Routledge et al.,1998).

3.1.2.O3/H2O2process

O3/H2O2process was effective for removal of17β-estradiol and 17α-ethinylestradiol(individually or mixed together)from aqueous solutions(Fig.1B).However,the addition of H2O2to the ozonation process did not improve estrogens removal.The hydrogen peroxide dose applied(O3/H2O2:2/1,molar basis) was considered the most adequate by other authors.It avoids the peroxide accumulation in the medium.Peroxide can act as hydroxyl radical scavenger,hindering the oxidation process (Paillard et al.,1988;Balcioglu and Otker,2003).H2O2was added only in pHs3and7in order to promote the formation of OH radicals.This was not necessary in pH11because the oxidation already occurs via OH radical.

Ozone decomposition into OH radicals can be accelerated by H2O2addition,which constitutes a beneficial factor. However,H2O2should be present in a dissociated form to assure its reaction with ozone,forming hydroxyl radicals.As peroxide dissociation is favored when pH is increased,we can expect that at pH3,the formation of hydroxyl radicals is small (Staehelin and Hoigné,1982);hence,at this pH the mechanism governing the oxidation process(O3/H2O2)is via molecular ozone.

3.2.Estrogenic activity

The dose–response curves of17β-estradiol were obtained for a concentration range of2.72μg l?1to1.33ng l?1,corresponding to the concentrations inside of each well in the plate.The standard curve of17β-estradiol obtained showed a sigmoidal shape,which was adjusted by a non linear regression method (Routledge and Sumpter,1996).A different dose–response curve was established for each test.The maximal induction ofβ-galactosidase was2.8781±0.02and EC50was4.53·10?8±2.15·10?9g l?1.

3.2.1.Estrogenic potential of17β-estradiol and

17α-ethinylestradiol

The estrogenic activity determination by the YES assay is subjected to some effects(synergism,antagonism and addi-tive effect)when more than one estrogen is present.When the individual activities are known,it is possible to confirm what effect is prominent.The YES assay is adequate for use,since it is rapid and highly reproducible and sensitive(Payne et al., 2000).

The YES assay was used to compare the estrogenic potential of the two estrogens.The dose–response curves were obtained using samples consisting of estrogen in

ethanol

Fig.2–Dose–response curves of(A)17β-estradiol and

17α-ethinylestradiol and(B)17β-estradiol and mixture of 17β-estradiol/17α-ethinylestradiol.

109

S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

solution,corresponding to the following range of concentra-tion:2.72μg l?1to1.33ng l?1(Fig.2A).A relatively lower estrogenic potential for17α-ethinylestradiol was obtained as revealed by the YES assay.This behavior was also reported by Folmar et al.(2002),whereas equipotent responses of17α-ethinylestradiol and17β-estradiol were observed by Van den Belt et al.(2004))in the YES screen,evaluating EC50values.

In order to verify the association effect of the two estrogens a dose–response of their mixture(50%of17α-ethinylestradiol and50%of17β-estradiol)was obtained.The mixture pre-sented the similar estrogenic potential,which was revealed for17β-estradiol(Fig.2B).Therefore,the association of the two estrogens revealed an additive effect(estrogenic activity).A similar result was reported by Chen et al.(2007)who observed additive effect in mixtures of17α-estradiol and17α-ethiny-lestradiol,relatively to the responses of single components solutions,determined by the YES assay.

The concentrations that elucidate50%of estrogenic activity(EC50)were obtained from the dose–response curves and are shown in Table1.Relative potency(RP)was calculated taking EC50of17β-estradiol as reference.EC50values were very close for both estrogens,individually or in the mixture.

3.2.2.Estrogenic activity reduction by ozonation(mixture of estrogens)

Fig.3shows dose–response curves of17β-estradiol and of extracts of ozonated samples of the mixture(17β-estradiol/ 17α-ethinylestradiol)at three different pH values.

Significant levels of residual estrogenic activity were observed in experiments conducted at pH values of7and11, even when the ozone dose was increased.However,the activity was not completely removed even when high ozone doses were applied.Auspiciously,complete removal of activity was achieved in pH3,even applying low ozone doses(1mg l?1),as illustrated in Fig.3.The same trend was observed when the estrogens were ozonated separately(not shown results).

The ozone doses usually applied in water treatment plants are close to1.0mg l?1(Harrison,2000).Our results show that this dose was not sufficient to completely remove the estrogenic activity of solutions containing17α-ethinylestra-diol and17β-estradiol ozonated at pH values of7and11.

Probably,the by-products formed during ozonation in those pH values present estrogenic activities similar to those of17β-estradiol and17α-ethinylestradiol,once these sub-stances were rapidly removed at the beginning of ozonation (N96%).Some by-products formed during ozonation in pHs7 and11were identified to verify if they are responsible for the residual estrogenic activity(presented in Section3.4).

Our results indicate that molecular ozone,the predomi-nant form present at pH3,was able to destroy the phenolic ring,to which is attributed the estrogenicity of the tested estrogens.On the other hand,at pH value of11,the non-specific hydroxyl radical is the predominant oxidant found in the reaction medium.Its action on the phenolic ring was not so effective as that of molecular ozone.

The results obtained in our work are in accordance to those of Huber et al.(2004)and Kim et al.(2004).Huber et al.(2004) observed that ozone doses typically applied for the disinfec-tion of drinking waters were sufficient to reduce estrogenicity by a factor of more than200.However,it was impossible to complete by remove the estrogenic activity of17α-ethinyles-tradiol in pH8.

Results from the assays of17β-estradiol ozonation per-formed by Kim et al.(2004)have shown that estrogenicity was not reduced significantly with the increase of ozone dosage (pH6.0).This suggests the formation of by-products that may present estrogenicity similar to that of17β-estradiol,once this substance was rapidly removed at the beginning of the ozonation experiment.

Table1–CE50of estrogens and their mixture

Estrogen CE50(g l?1)RP

17β-estradiol 4.53·10?8±2.15·10?91 17α-ethinylestradiol 6.34·10?8±4.11·10?90.71 17β-estradiol/17α-ethinylestradiol 5.42·10?8±2.16·10?90.84Fig.3–Dose–response curves of17β-estradiol and of the extracts of ozonated samples of mixture of17β-estradiol/ 17α-ethinylestradiol in(A)pH3,(B)pH7and(C)pH11.

110S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

https://www.360docs.net/doc/c29228032.html,parison of O 3and O 3/H 2O 2processes for estrogenic activity reduction

Fig.4shows the estrogenic activity of aqueous samples treated by ozone and O 3/H 2O 2at pH 7.Higher estrogenic activity was observed in samples treated by O 3/H 2O 2.The lower efficiency of the combined process to remove estrogenic activity is related to its capacity to produce OH radicals.As already commented the oxidation pathways involving OH radicals produce by-products with higher estrogenic activity.This was also observed for samples ozonized at pHs 11and 7.Under alkaline conditions the oxidation by OH radicals is preponderant and the treated sample presents a higher estrogenicity level (Fig.3B,C).

3.2.

4.OH radical and estrogenic activity removal

Tert-butanol,a known OH radical scavenger,was added so as to verify the action of hydroxyl radicals at different pH values in both oxidation processes (O 3and O 3/H 2O 2).The concentra-tion of tert-butanol was 4·10?3mol l ?1in all assays.According to López et al.(2004)a concentration of 10?3mol l ?1is enough for scavenger action.

The results showed that the addition of tert-butanol did not change estrogens removal at pH 3for ozonation.It is known that at acidic pH the pathway of OH radical production is suppressed,since the conjugate base of hydrogen peroxide (HO 2?),which is favored at alkaline pH,is the species that reacts with ozone to produce OH radicals (Staehelin and Hoigné,1982).Thus,in pH 3estrogens oxidation occurs mainly via molecular ozone.

However,removal of estrogens was lower at pH 7and 11,when the scavenger was added to the media,as presented in Fig.5.17β-estradiol and 17α-ethinylestradiol residual concen-trations found in the ozonation experiments were 90ng l ?1and 230ng l ?1for ozone dosage of 4.0mg l ?1in pH 7and 90ng l ?1and 300ng l ?1for ozone dosage of 4.5mg l ?1in pH 11,respectively.On the other hand the residual concentrations of 17β-estradiol and 17α-ethinylestradiol for ozonation experi-ments without tert-butanol were 25ng l ?1and 40ng l ?1for ozone dosage of 4.0mg l ?1in pH 7and 70ng l ?1and 165ng l ?1for ozone dosage of 4.5mg l ?1in pH 11,respectively.Therefore,the higher estrogen residual concentrations found in ozona-tion in presence of tert-butanol are contributing for higher estrogenic activity observed.

Indeed,higher estrogenic activities were observed in pH 7and 11for ozonation and H 2O 2/O 3process,as showed in Fig.6

.

Fig.4–Dose –response curves of mixture extracts containing 17β-estradiol (10μg l ?1)and 17α-ethinylestradiol (10μg l ?1)treated with O 3and O 3/H 2O 2at pH 7and O 3/H 2O 2molar ratio of

2:1.

Fig.5–Removal of the estrogens in the mixture 17β-estradiol/17α-ethinylestradiol by ozone in pHs 3,7and 11added of TBA.

111

S C I E N C E O F T H E T O T A L E N V I R O N M E N T 407(2008)105–115

This higher activity was probably caused by the estrogens (not degraded)remaining in solution.

Thus,our results indicated that the radicalar oxidation mechanism,which is preponderant at pH 11and significant at pH 7,formed oxidation by-products with higher estrogenic activity.This was somewhat an unexpected result and a question arises calling for an appropriated answer:why oxidation by OH radicals was not so effective to remove estrogenic activity?To have a correct answer more work is necessary,focusing the relationship between chemical struc-ture of by-products and estrogenicity.

3.3.UV spectrophotometry

The relative position of the hydroxyl phenolic group in the ring is considered critical to its affinity with the estrogen receptor (RE),which is hence responsible for its estrogenic potential.

According to the literature,absorbance of a substance in the wavelength range of 190–300nm is representative of conjugated double and triple C –C bonds and the UV absorption band at 288nm is associated with the phenolic ring (Liu and Liu,2004).The UV spectra of 17β-estradiol and 17α-ethinyles-tradiol in methanol indicate that these molecules present a characteristic peak of absorption at λ=204nm and a smaller absorption at λ=280nm.

These UV spectra were obtained using methanol,which proved to be the best solvent for identification of absorbance maxima (peaks).This procedure could be employed because samples had been extracted and concentrated in C 18and eluted in organic https://www.360docs.net/doc/c29228032.html,ually,it is not possible to obtain these spectra from samples containing estrogens at very low concentrations.

Fig.7presents the UV spectra for the ozonated samples of 17α-ethinylestradiol at different pH values.These spectra are similar to those of ozonated samples of 17β-estradiol obtained by Bila et al.(2007)

.

Fig.6–Dose –response curves of 17β-estradiol and extracts of ozonated samples of the mixture containing 17β-estradiol (10μg l ?1)and 17α-ethinylestradiol (10μg l ?1)at pHs 3,7and 11with and without

TBA.

Fig.7–UV spectra of aqueous samples of 17α-ethinylestradiol ozonated at pHs 3,7and 11with different ozone doses.

112

S C I E N C E O F T H E T O T A L E N V I R O N M E N T 407(2008)105–115

For the three pH values tested attenuation and disappear-ance of absorption bands were observed as time of ozonation and,consequently,ozone dosage increased (Fig.7).All absorbance bands were significantly reduced after ozonation,indicating that the phenolic ring was disrupted.In pH 3,a larger reduction of the absorbance peaks was observed,while in pH 11such effect was less pronounced.This feature is in agreement with estrogenic activity results,once at pH 3the activity was completely reduced,whereas at pH 11it was higher,seeming to indicate the existence of a relationship between estrogenic activity and the phenolic ring configura-tion.If the rupture of the phenolic ring allows estrogenic activity removal,a practical consequence for water and waste-water treatment is non-necessity of performing oxidation until complete mineralization of the estrogen molecule.

3.4.Identification of by-products formed during 17β-estradiol and 17α-ethinylestradiol ozonation

Published works have proposed or identified by-products formed during the oxidation of 17β-estradiol and 17α-ethinylestradiol (Huber et al.,2004;Ohko et al.,2002;Irmak et al.,2005;Bila et al.,2007).

Huber et al.(2004)identified some by-products formed during the ozonation of 17α-ethinylestradiol.Those authors assigned that two groups were firstly oxidized:the phenolic ring,which is highly reactive (k O 3=3×106M ?1s ?1in pH 7)and the ethinyl group,which is less reactive (k O 3=200M ?1s ?1).

The proposed ozonation by-products,which were deter-mined in our experiments with 17α-ethinylestradiol are shown in Table 2.The structure of the by-products seems to corroborate the reactive pattern of different chemical groups,as proposed by Huber et al.(2004).The similarities obtained for

substances estra-1,3,5(10)-trien-17-one,2,3-bis[(trimethyl)oxy]and estra-1,3,5(10)-trien-17-one,3,4-bis[(trimethyl)oxy]were 90%and 70%,respectively.These proposed by-products are isomers,they were identified by mass spectrometry,but unfortunately confirmation with a standard was not possible.

The identification of ozonation by-products formed in pH 3was not possible,neither for 17β-estradiol nor 17α-ethinyles-tradiol.However,it was clearly observed that in pH 3the formed by-products were different from those formed at pH 7and 11.This fact highlights the relevance of the oxidation mechanism (molecular ozone and OH radical)on by-products formation and characteristics.

The by-products identified in our ozonation experiments with 17β-estradiol (2-hydroxyestradiol and testosterone)were already described in the works of Ohko et al.(2002)and Bila et al.(2007).Irmak et al.(2005)proposed mechanisms for the direct reaction of ozone with the phenolic ring of 17β-estradiol and identified some by-products formed during ozonation,among them 2-hydroxyestradiol.The similarities found for 2-hydroxyestradiol and testosterone in our work were 99and 62.5%,respectively.

According to Bila et al.(2007),2-hydroxyestradiol and testos-terone present different responses in the YES assay.The former presents estrogenic activity and the latter not.The estrogenicity of by-products estra-1,3,5(10)-trien-17-one,2,3-bis[(trimethyl)oxy]and estra-1,3,5(10)-trien-17-one,3,4-bis[(trimethyl)oxy]is prob-ably associated to the phenolic rings of these molecules.

4.Conclusion

Ozonation and O 3/H 2O 2processes were effective for 17β-estradiol and 17α-ethinylestradiol removal in aqueous solu-tion (pure or in the mixture).Estrogens removals were higher than 96%for all ozone doses and pH values investigated.The addition of H 2O 2to the ozonation process did not improve estrogens removal.

Estrogenic activity removal increased with ozone dosage;however it was not completely removed in pHs 7and 11.In pH 3,it was totally removed,even applying low ozone dosages,in both ozonation and O 3/H 2O 2processes.Therefore,oxidation via molecular ozone was more effective for estrogenicity removal.

Residual estrogenic activity remained after oxidation at pHs 7and 11,probably due to by-products formed via OH radical.When the O 3/H 2O 2process was applied using the same conditions of the ozonation process (pH and ozone dosage),the estrogenic activity was slightly higher.The results suggest that more by-products were formed,presenting estrogenic activity,in the O 3/H 2O 2process,due the more intense production of OH radicals.

Only few by-products have been proposed.However,the by-products formed present a phenolic ring that is,probably,responsible for the estrogenic activity.

It was observed that in spite of high removals of estrogens obtained,the estrogenic activity was not completely removed by ozonation and O 3/H 2O 2processes,mainly in experiments conducted at pHs 7and 11.Therefore,combined analyses,chemical and biological,should be performed to evaluate the efficiency of water or domestic wastewater treatment.

Table 2–By-products proposed for 17α-ethinylestradiol oxidation

ND –not detected.

113

S C I E N C E O F T H E T O T A L E N V I R O N M E N T 407(2008)105–115

This work reveals the challenge of removing these sub-stances,mainly when they are present in very low concentra-tions(range ofμg l?1and ng l?1).It is important to highlight that these micropollutants,found in the environment in such concentration range,cause adverse effects in humans and others animals.

Acknowledgements

The authors thank Professor John P.Sumpter for kindly providing the recombinant Saccharomyces cerevisiae(yeast estrogen screen).The authors thank Professor Débora de Azevedo for GC-MS analyses.We also want to thank FUJB, FAPERJ,CAPES and CNPq for research funding.

R E F E R E N C E S

Baker VA.Endocrine disrupters—testing strategies to assess human hazard.Toxicol in Vitro2001;15:413–9.

Balcioglu IA,?tker M.Treatment of pharmaceutical wastewater containing antibiotics by O3and O3/H2O2processes.

Chemosphere2003;50:85–95.

Bila DM,Dezotti M.Fármacos no Meio Ambiente.Quim Nova 2003;26:1–8.

Bila D,Montalv?o AF,Azevedo DA,Dezotti M.Estrogenic activity removal of17β-estradiol by ozonation and identification of by-products.Chemosphere2007;69:736–46.

Birkett JW,Lester JN.Endocrine disrupters in wastewater and sludge treatment processes.Boca Raton:CRC Press Company;2003. Brian JV,Harries CA,Scholze M,Backhaus T,Booy P,Lamoree M, Pojana G,Jonkers N,Runnalls T,BonfàA,Marcomini A,

Sumpter JP.Accurate prediction of the response of freshwater fish to a mixture of estrogenic chemicals.Environ Health

Perspect2005;113:722–8.

Chen PJ,Rosenfeldt EJ,Kullman SW,Hinton DE,Linden KG.

Biological assessments of a mixture of endocrine disruptors at environmentally relevant concentrations in water following UV/H2O2oxidation.Sci Total Environ2007;376:18–26. Coleman HM,Abdullah MI,Eggins BR,Palmer FL.Photocatalytic degradation of17β-oestradiol,oestriol and17α-ethynyloestradiol in water monitored using fluorescence spectroscopy.Appl Cat B:Environ2005;55:23–30.

Fent K,Escher C,Caminada D.Estrogenic activity of pharmaceuticals and pharmaceutical mixtures in a yeast reporter gene system.

Reprod Toxicol2006;22:175–85.

Folmar LC,Hemmer MJ,Denslow ND,Kroll K,Chen J,Cheek A, Richman H,Meredith H,Grau G.A comparison of the estrogenic potencies of estradiol,ethynylestradiol,diethylstilbestrol,

nonylphenol and methoxychlor in vivo and in vitro.Aquat

Toxicol2002;60:101–10.

Gray Jr LE,Kelce WR,Wiese T,Tyl R,Gaido K,Cook J,et al.

Endocrine screening methods workshop report:detection of estrogenic and androgenic hormonal and antihormonal

activity for chemicals that act via receptor or steroidogenic enzyme mechanisms.Reprod Toxicol1997;11(5):719–50. Gomes RL,Avcioglu E,Scrimshaw MD,Lester JN.Steroid estrogen determination in sediment and sewage sludge:a critique of sample preparation and chromatographic/mass spectrometry considerations,incorporating a case study in method

development.Trends Anal Chem2004;23(10–11):737–44. Hamblen EL,Cronin MTD,Schultz TW.Estrogenicity and acute toxicity of selected anilines using a recombinant yeast assay.

Chemosphere2003;52:1173–81.Harrison JF.Ozone for point-of use,point-of entry,and small water system water treatment aplications—a reference manual.

Lisle:The Water Quality Association;2000.

Harrison PTC,Holmes P,Humfrey CDN.Reproductive health in humans and wildlife:are adverse trends associated with

environmental chemical exposure?Sci Total Environ

1997;205:97–106.

HoignéJ,Bader H.Rate constants of reactions of ozone with organic and inorganic compounds in water—I non-dissociating organic compounds.Water Res1983;17:173–83.

Huang CR,Shu HY.The reaction kinetics,decomposition pathways and intermediate formations of phenol in ozonation, UV/O3and UV/H2O2processes.J Hazard Mat1995;41:47–64. Huber MM,Canonica S,Park GY,von Gunten U.Oxidation of pharmaceuticals during ozonation and advanced oxidation processes.Environ Sci Technol2003;37:1016–24.

Huber M,Ternes TA,von Gunten U.Removal of estrogenic activity and formation of oxidation products during ozonation of

17α-ethinylestradiol.Environ Sci Technol2004;38(19):5177–86. Irmak S,Erbatur O,Akgerman A.Degradation of17β-estradiol and bisphenol A in aqueous medium by using ozone and ozone/UV techniques.J Hazard Mat B2005;126:54–62.

Kim SE,Yamada H,Tsuno H.Evaluation of estrogenicity for17β-estradiol decomposition during ozonation.Ozone Sci Eng

2004;26(6):563–71.

Lai KM,Scrimshaw MD,Lester JN.The effects of natural and synthetic steroid estrogens in relation to their environmental occurrence.Crit Rev Toxicol2002;32(2):113–32.

Liu B,Liu XL.Direct photolysis of estrogens in aqueous solutions.

Sci Total Environ2004;320:269–74.

López A,Benbelkacem H,Pic JS,Debellefontaine H.Influence of t-butanol and of pH on hydrodynamic and mass transfer

parameters in an ozone bubble column.International

Conference Barcelona,March10–12;2004.

Mvula E,Schuchmann MN,von Sonntag C.Reactions of phenol-OH-adduct radicals.Phenoxyl radical formation by

water elimination vs.oxidation by dioxygen.J Chem Soc,

Perkin Trans2001;2:264–8.

Nash JP,Kime DE,Van der Ven LT,Wester PW,Brion F,Maack G, Stahlschmidt-Allner P,Tyler CR.Long-term exposure to

environmental concentrations of the pharmaceutical

ethynylestradiol causes reproductive failure in fish.Environ Health Perspect2004;112:1725–33.

Nogueira JMF.Desreguladores endócrinos:efeitos adversos e estratégias para monitora??o dos sistemas aquáticos.Química 2003;88:65–71.

Ohko Y,Iuchi K,Niwa C,Tatsuma T,Nakashima T,Iguchi T, Kubota Y,Fujishima A.17β-estradiol degradation by TiO2

photocatalysis as a means of reducing estrogenic activity.

Environ Sci Technol2002;36:4175–81.

Paillard H,Brunet R,Dore M.Optimal conditions for applying an ozone–hydrogen peroxide oxidizing system.Water Res1988;22(1): 91–103.

Payne JP,Rajapakse NA,Wilkins M,Kortenkamp A.Prediction and assessment of the effects of mixtures of four xenoestrogens.

Environ Health Perspect2000;108:983–7.

Routledge EJ,Sheahan D,Desbrow C,Brighty GC,Waldock M, Sumpter JP.Identification of estrogenic chemicals in STW

effluent.2.In vivo responses in trout and roach.Environ Sci Technol1998;32:1559–65.

Routledge EJ,Sumpter JP.Estrogenic activity of surfactants and some of their degradation products assessed using a

recombinant yeast screen.Environ Toxicol Chem1996;15(3): 241–8.

Staehelin J,HoignéJ.Decomposition of ozone in water:rate of initiation by hydroxide ion and hydrogen peroxide.Environ Sci Technol1982;16:676–81.

Ternes TA,Stüber J,Hermann N,McDowell D,Ried A,Kampmann M,Teiser B.Ozonation:a tool for removal of pharmaceuticals,

114S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

contrast media and musk fragrances from wastewater?Water Res2003;37:1976–82.

Ternes TA,Stumph M,Mueller J,Haberer H,Wilken RD,Servos M.

Behavior and occurrence of estrogens in municipal sewage treatment plants—I.Investigations in Germany,Canada and Brazil.Sci Total Environ1999;225:81–90.

Van den Belt K,Berckmans P,Vangenechten C,Verheyen R, Witters https://www.360docs.net/doc/c29228032.html,parative study on the in vitro/in vivo estrogenic

potencies of17β-estradiol,estrone,17α-ethynylestradiol and nonylphenol.Aquat Toxicol2004;66:183–95.

von Gunten U.Ozonation of drinking water:part I.Oxidation kinetics and product formation.Water Res2003;37:1443–67. Zwiener C,Frimmel FH.Oxidative treatment of pharmaceuticals in water.Water Res2000;34(6):1881–5.

115

S C I E N C E O F T H E T O T A L E N V I R O N M E N T407(2008)105–115

相关文档
最新文档