Water Research 104 (2016) 441e448

Contents lists available at ScienceDirect

Water Research
journal homepage: www.elsevier.com/locate/watres

Degradation of benzotriazole and benzothiazole in treatment

wetlands and by artificial sunlight
Ewa Felis a, b, Adam Sochacki a, b, *, Sylwia Magiera a, c
a
Centre for Biotechnology, Silesian University of Technology, ul. B. Krzywoustego 8, PL-44-100, Gliwice, Poland
b
Environmental Biotechnology Department, Faculty of Power and Environmental Engineering, Silesian University of Technology, ul. Akademicka 2, PL-44-
100, Gliwice, Poland
c
Department of Inorganic, Analytical Chemistry and Electrochemistry, Faculty of Chemistry, Silesian University of Technology, ul. M. Strzody 7, PL-44-100,
Gliwice, Poland

a r t i c l e i n f o a b s t r a c t

Article history: Laboratory-scale experiments were performed using unsaturated subsurface-flow treatment wetlands
Received 13 May 2016 and artificial sunlight (with and without TiO2) to study the efficiency of benzotriazole and benzothiazole
Received in revised form removal and possible integration of these treatment methods. Transformation products in the effluent
17 August 2016
from the treatment wetlands and the artificial sunlight reactor were identified by high performance
Accepted 19 August 2016
Available online 21 August 2016
liquid chromatography coupled with tandem mass spectrometry. The removal of benzothiazole in the
vegetated treatment wetlands was 99.7%, whereas the removal of benzotriazole was 82.8%. The vege-
tation positively affected only the removal of benzothiazole. The major transformation products in the
Keywords:
Constructed wetlands
effluents from the treatment wetlands were methylated and hydroxylated derivatives of benzotriazole,
Transformation products and hydroxylated derivatives of benzothiazole. Hydroxylation was found to be the main process gov-
Titanium dioxide erning the transformation pathway for both compounds in the artificial sunlight experiment (with and
Photodegradation without TiO2). Benzotriazole was not found to be susceptible to photodegradation in the absence of TiO2.
Micropollutants The integration of the sunlight-induced processes (with TiO2) with subsurface-flow treatment wetlands
Corrosion inhibitors caused further elimination of the compounds (42% for benzotriazole and 58% for benzothiazole). This
was especially significant for the elimination of benzotriazole, because the removal of this compound
was 96% in the coupled processes.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction are used as components of dishwasher detergents (Giger, 2006;

Benzotriazoles (BTAs) and benzothiazoles (BTs) are high pro-
duction volume substances that are widely used in various indus-
trial processes and in households (Weiss and Reemtsma, 2005;
Jover et al., 2009; Matamoros et al., 2010; Herrero et al., 2014).
BTAs are heterocyclic compounds based on the structure of ben-
zotriazole (BTA), which contains two fused rings (benzene and
triazole) and three nitrogen atoms. The most common BTAs are the
parent compound and tolyltriazole, which is a mixture of the iso-
mers 4- and 5-methyl-BTA (Herrero et al., 2014). BTAs due to their
complexing properties are used as corrosion inhibitors (Giger,
2006; Ding et al., 2010; Herrero et al., 2014). In households, BTAs
* Corresponding author. Centre for Biotechnology, Silesian University of Tech-
nology, ul. B. Krzywoustego 8, PL-44-100, Gliwice, Poland.
E-mail address: adam.sochacki@polsl.pl (A. Sochacki).
Ding et al., 2010). BTAs have been classified as being toxic to
aquatic organisms, potentially carcinogenic and showing estrogenic
activities (Giger, 2006; Matamoros et al., 2010; Asimakopoulos
et al., 2013).
The structure of benzothiazole (BT) and its derivatives (BTs) are
based on two fused rings e benzene and thiazole. BTs are very often
used as vulcanization accelerators in rubber production, and as
antifungal agents in paper and leather manufacturing (Matamoros
et al., 2010). The commonly used BTs include the parent compound,
2-morpholin-4-yl-BT, 2-hydroxy-BT and 2-methylthio-BT
(Reemtsma et al., 1995; Herrero et al., 2014). Asimakopoulos et al.
(2013), additionally mentioned 2-(mercapto)-BT as one of the
most widely used BT derivatives, which is used as a biocorrosion
inhibitor in various industries.
Only in the US the annual production of BTAs exceeds 9000 tons
and the production of BT was reported to be in the range of
4.5e450 tons (Asimakopoulos et al., 2013). Because of their

http://dx.doi.org/10.1016/j.watres.2016.08.037
0043-1354/© 2016 Elsevier Ltd. All rights reserved.
442 E. Felis et al. / Water Research 104 (2016) 441e448
widespread application, both BTAs and BTs have been identified in
different environmental compartments, especially in the aquatic
environment. The major sources of BTAs and BTs in the aquatic
environment are wastewater treatment plants' (WWTPs) effluents
and runoff from urbanized areas (Ni et al., 2008; Janna et al., 2011).
Municipal and domestic WWTPs are not optimized for the removal
of organic micropollutants (Huntscha et al., 2014). In the case of
BTAs and BTs, their removal efficiency in conventional activated
sludge system ranged from 0 to 74% for BTAs and 5e28% for BTs
(Kloepfer et al., 2005; Voutsa et al., 2006). BTA has been proposed
as a proxy compound for assessing compliance to removal targets
in Swiss WWTPs (Eggen et al., 2014). Both BTAs and BTs are rela-
tively high water-soluble and do not partition significantly into
organic phases or the atmosphere, and are found in average con-
centrations of up to 10 mg/L in WWTP effluents (Stasinakis et al.,
2013; Huntscha et al., 2014) and at 0.1e6.0 mg/L in surface waters
(Matamoros et al., 2010). The highest concentrations of BTAs (5-
methyl-BTA) were found in the runoff water from several airports
in the US and were in the range of 1.1e90 mg/L (Sulej et al., 2011).
The conventional activated sludge WWTPs are not efficient in the
removal of BTs and BTAs, but it was found that treatment wetlands
(TWs) can be more suitable for the removal of these and other
organic pollutants as they support co-existence of more diverse
degradation processes: photodegradation, microbial degradation,
sorption and plant uptake (Matamoros et al., 2010). Indeed, Miksch
et al. (2015) indicated the elucidation of the removal and fate of
corrosion inhibitors in TWs as one of the research needs.
Only several articles address the removal of BTs and BTAs in TWs
but their transformation products (TPs) were not identified. Castro
et al. (2005) stated that TWs have the potential for the removal of
BTAs especially by the vegetation-induced processes. High removal
(98%) of 2-methylthio-BT in an activated soil filter planted with
common reed operated as a continuous down-flow (DF) system
was attributed to sorption by peat (Bester and Scha €fer, 2009). The
removal of BT and BTA in a surface flow TW was approximately 80%
and 50% during warm season, but negative or equal to 0% during
cold season, respectively (Matamoros et al., 2010). On the other
hand, in a waste stabilization pond even during cold season the
removal of BT and BTA was 72% and 45%, respectively, and was
markedly less affected by the seasonality (Matamoros et al., 2016).
Pilot-scale vertical DF bed ensured removal efficiency within
80e90% for both compounds (measured only during the warm
season) (Matamoros et al., 2010). The elimination of BTA in hori-
zontal flow TWs was found to be very low (<5%) during both cold
and warm season, and the removal of BT was moderate and was
affected by the seasonality, this is, 63% for the warm season and 39%
for the cold season (Matamoros et al., 2016). Aerobic conditions
prevailing in unsaturated vertical flow TWs and photodegradation
in open water surface systems seems to be conducive to BT and BTA
removal. In the case of subsurface-flow TWs, which were applied in
the present study, the photodegradation is not a major factor
(Matamoros et al., 2010), but this type of TW can be coupled with a
system with open water surface systems to promote sunlight-
induced processes (Vymazal, 2013).
Indeed, photochemical processes induced by sunlight are
mentioned as an effective tool for the removal of persistent
micropollutants from biologically treated wastewater (Lin and
Reinhard, 2005; Klavarioti et al., 2009; Oller et al., 2011; Gan
et al., 2014). In most cases, the sunlight-induced processes were
accelerated by the addition of a semiconductor (predominantly
TiO2) to a reaction solution during the irradiation process (Lin and
Reinhard, 2005; Pereira et al., 2011; De la Cruz et al., 2013). As a
result of an excitation of TiO2 molecule by means of the sunlight (in
an aqueous environment and in the presence of oxygen), hydroxyl
radicals are generated, which react indiscriminately with most of
organic contaminants. There are only several studies, in which the
photochemical degradation of BTA and BT is discussed (Andreozzi
et al., 1998; Benitez et al., 2013; Bahnmüller et al., 2015), but
none of them described application of both suspended TiO2 and UV
radiation in the solar range for BTA and BT decay. Importantly, TiO2-
based photocatalysis process was suggested as a post-treatment
method for the TW-treated effluents, in which the turbidity
would not reduce radiation penetration (Liu et al., 2015).
The main objective of the present study was to determine the
effect of the processes occurring in TWs and the processes induced
by solar light on the degradation and transformation of BT and BTA
in the wastewater.
2. Materials and methods
2.1. Chemicals
Analytical standards of 1H-benzotriazole (BTA) and benzothia-
zole (BT) were purchased in highest available purity from Sigma
Aldrich (Poland). Acetonitrile Chromoscan® for HPLC analysis was
purchased from POCH S.A. (Poland). Acetonitrile Ultra CHROMA-
SOLV® from Sigma Aldrich (Poland) was used for LC/MS-MS ana-
lyses. Titanium (IV) oxide (CAS number 1317-70-0) was purchased
from Sigma-Aldrich. In the experiments, TiO2 was used as
powdered anatase (particle size < 5 mm; purity ¼ 99.8%).
2.2. Laboratory-scale TWs
The experimental system consisted of twelve PVC columns filled
up to 70 cm with three layers of bed media: 4e8 mm gravel
(0e5 cm from the bottom), 0.5e1 mm sand (5e65 cm) and sand
mixed with organic soil (65e70 cm). Polyester mesh was used to
separate the layers and to protect the outlet from clogging. The
dimensions of the columns are shown in Fig. S1 (Supplementary
data). The experimental system was operated under laboratory
conditions. In order to enable plant growth, the experimental sys-
tem was illuminated using high pressure sodium lighting system
(400 W; photosynthetic photon flux 750 mmoL/s) with various
light/dark conditions to mimic seasonal changes: for summer 16 h
of light and 8 h of dark (16/8), for fall 9/15, and winter 8/16. The
duration of the experiment was 23 weeks (162 days; 10 days of
summer, 68 days of fall, 64 days of winter), but this period was
preceded by a 3-month adaptation period when the columns were
fed with the influent without BT and BTA. The mean min. and max.
temperature during the experiment was 23.0  C and 28.7  C, and
the humidity was 50.3e76.4%. Selected columns were planted with
the seedlings of reed canary grass ‘Picta’ (Phalaris arundinacea var.
picta L.) 3 months before the experiment. The seedling plants were
rooted in organic soil, therefore the top most layers of all the col-
umns contained a mixture of organic soil and sand. This mixed
layer was also present in the unplanted columns.
The types of columns (in duplicate) used in the experiment were
selected based on the presence or absence of vegetation in the
columns and the presence or absence of BT or BTA in the influent.
This combination yielded six types of columns: columns fed with
simulated wastewater (control columns, CTRL), respectively plan-
ted (CTRL-P) and unplanted (CTRL-U); columns fed with waste-
water spiked with BT, respectively planted (BT-P) and unplanted
(BT-U) and columns fed with wastewater spiked with BTA,
respectively planted (BTA-P) and unplanted (BTA-U).
The columns were operated in an intermittent DF unsaturated
mode and fed by simulated municipal wastewater prepared based
on the modified protocol by Nowrotek et al. (2016). The detailed
information on the composition of the wastewater used in the
study is given in Table S1. The influent for the BTA-U and BTA-P
 E. Felis et al. / Water Research 104 (2016) 441e448
columns was spiked with 10 mg/L of BTA (1H-BTA), and the influent
for the BT-U and BT-P columns was spiked with 10 mg/L of BT. Each
column was fed with 1 L of the wastewater (single pulse per day;
hydraulic loading rate 32 mm/pulse) in the loading and resting
regime of 5 d and 2 d (Monday-Friday loading; Saturday-Sunday
resting).
2.3. Sampling
The effluent from the TW system was collected in buckets
placed underneath the columns. The effluent samples were taken
fortnightly after third feeding in the feeding cycle. The exception to
this routine was on day 2 of the experiment when additional
samples were taken after fourth feeding in the first week of the
experiment. The influent samples were withdrawn from plastic
barrels in which the three types of influents were prepared. All the
samples were filtered using qualitative filter paper (MN 615 filter
for medium fast filtration, Macherey-Nagel GmbH & Co. KG, Ger-
many) prior to analysis.
2.4. Artificial sunlight system
The studies on the degradation of investigated substances by
means of an artificial sunlight were performed in the Solarbox
1500e System (Co.fo.me.gra, Italy), which was equipped with a
polychromatic Xenon lamp (1500 W). The radiation generated by
the Xenon lamp was filtered by an extra-long life soda lime glass
filter, which provided the natural outdoor exposure conditions. The
emission spectrum of the Xenon lamp, depending on the irradiance
intensity, is shown in Fig. S2. The Xenon lamp and the reaction
chamber of the system were cooled by means of an air generated by
two impeller fans installed inside the reactor. The temperature
inside the system was controlled and it was 32 ± 2  C.
During the experiments with the artificial sunlight three
different aspects were tested: (i) the behavior of the investigated
substances in distilled water at different irradiances emitted by the
Xenon lamp; (ii) the influence of TiO2 doses on BTA and BT pho-
todegradation efficiency; (iii) the decay of investigated compounds
in effluents from the planted TW columns by means of artificial
sunlight in the presence of TiO2. The aspect (i) was carried out at
four different lamp irradiances: 250 W/m2; 500 W/m2; 750 W/m2
and 1000 W/m2. In the case of aspect (ii), all the experiments were
performed at lamp irradiance equal to 500 W/m2, but at three
different TiO2 doses: 100 mg/L; 300 mg/L and 500 mg/L. The ex-
periments, which were carried out within aspect (iii), were carried
out at the lamp irradiance equal to 500 W/m2 (specified as an
average insolation for the 52nd parallel north, which crosses
Poland, Germany, the Netherlands, England, Canada, the US
(Aleutian Islands) at noon in September (Burgess, 2009) and the
concentration of TiO2 equal to 500 mg/L.
In the case of aspects (i) and (ii), 350 mL of an aqueous solution
respectively of BTA or BT was filled into a glass beaker and then
placed on a magnetic stirrer inside the Solarbox system. In the case
of aspect (iii), the experiments were conducted in grab samples
from the effluents of the planted TW columns, however the pro-
cedure was the same. All the experiments were performed in
duplicate.
2.5. Sorption studies
In order to determine the adsorption isotherms of the investi-
gated substances on TiO2 surface, the aqueous stock solutions of
respectively BT and BTA at a concentration of 500 mg/L were pre-
pared. The stock solutions were diluted with distilled water to
obtain the following concentrations of the investigated substances:
443
1.0 mg/L, 5.0 mg/L, 10 mg/L, 20 mg/L, 50 mg/L, 75 mg/L, 100 mg/L,
200 mg/L, 300 mg/L and 500 mg/L 40 mL of each solutions was
transferred into 50 mL polypropylene centrifuge tubes and 10 mg of
TiO2 was added. The tubes were capped and placed on a shaker. The
samples were shaken for 18 h. Next, the samples were filtered and
the concentrations of the investigated compounds in the liquid
phase were determined by means of HPLC-UV. Further calculations
were performed using MATLAB 10 software (MATLAB 10). The test
was performed in quadruplicate.
2.6. Standard wastewater analysis
Chemical oxygen demand (COD), total organic carbon (TOC),
ammonium nitrogen (N-NH4), orthophosphate phosphorus (P-
PO4), and water volume were measured according to the APHA
Standard Methods (Eaton et al., 2005) and the protocol of Nowrotek
et al. (2016).
2.7. HPLC-UV wastewater analysis
High-performance liquid chromatography coupled with vari-
able wavelength detector (UltiMate 3000 system; Dionex Corpo-
ration, Sunnyvale, CA, USA) was used for quantification of BTA and
BT during all the processes. The chromatographic separation was
performed using C18 Hypersil™ Gold column (250 mm  4.6 mm;
pore size: 5 mm) (Thermo Scientific, Polygen, Poland). The mobile
phase was a mixture of acetonitrile and acetate buffer (pH ¼ 5.7), in
a volumetric ratio 40:60 (v/v). During the analysis an isocratic flow
rate was used and it was 1.0 mL/min. In the above-described con-
ditions, the retention times (RT) of the investigated compounds
were equal to 6.3 ± 0.1 min and 10.5 ± 0.1 min for BTA and BT,
respectively. The limit of quantification (LOQ) of BTA and BT was
equal to 0.1 mg/L. It was established as the first lowest calibration
point of their calibration curves (linear regression, R2 > 0.99). The
calculated value of “signal to noise” ratio (S/N) in the case of both
investigated compounds was greater than 10. The limits of detec-
tion (LOD) of BTA and BT were defined for S/N at the level of 3 and
they were equal to 0.02 mg/L. The method was validated according
to PN-EN ISO/IEC 17025. The analyses were performed and
confirmed at four different wavelengths, i.e., 210 nm; 218 nm;
245 nm and 262 nm. The data was evaluated by means of Dionex
Chromeleon™ 6.8 software.
2.8. HPLCeMS/MS wastewater analysis
The identification of benzotriazole transformation products
(TPs-BTA) and benzothiazole transformation products TPs-BT was
performed using Dionex UPLC system (UltiMate 3000 RS; Dionex
Corporation, Sunnyvale, CA, USA) coupled with a triple quadrupole
mass spectrometer (AB SCIEX 4000 Q TRAP; Applied Biosystems/
MDS SCIEX, Foster City, CA, USA) equipped with an electrospray
ionization (ESI) source. The analytical method of BTA and its TPs
identification was adapted from Huntscha et al. (2014) and the
analytical method of BT and its TPs identification were adapted
from Haroune et al. (2002), Kloepfer et al. (2004) and Wick et al.
(2010). All the analyses were performed using the multiple reac-
tion monitoring (MRM) mode. MRM transitions and crucial
compound-dependent parameters for BTA and its TPs and BT and
its TPs are listed in Tables S2 and S3, respectively with their full
names given in the footnotes. The chromatographic separation was
performed using C18 Hypersil™ Gold column (250 mm  4.6 mm;
pore size: 5 mm) (Thermo Scientific, Polygen, Poland). The mobile
phase was a mixture of 0.1% formic acid in water and acetonitrile
(60:40; v/v) using an isocratic elution with a flow rate set at 1.0 mL/
min. Mass spectrometry with a TurboIonSpray source was
444 E. Felis et al. / Water Research 104 (2016) 441e448
performed in positive and negative ion mode. In the case of BTA and
its TPs, electrospray ionization was used only in positive mode. The
method for identification of BT and its TPs required two different
injections of the same sample, one with positive ionization and one
with negative ionization. Crucial detector parameters such as
ionization voltage (IS), collision assisted dissociation (CAD) gas,
source temperature (TEM), sheath gas (GS1), desolvation gas (GS2),
and curtain gas (CUR) were optimized by flow injection analysis
(FIA) to obtain better ionization. The mass spectrometric parame-
ters were as follows: TEM ¼ 500  C, IS ¼ 4000 V (or 4000 V),
CAD ¼ high, CUR ¼ 15 psi, GS1 ¼ 35 psi and GS2 ¼ 35 psi. Tuning
and optimization of the compound-dependent parameters
declustering potential (DP), collision energy (CE), entrance poten-
tial (EP), collision cell exit potential (CXP)) were performed for BTA
and BT by the direct infusion of a 1 mg/mL standard solution into the
ion source. Quadruples Q1 and Q3 were maintained at unit reso-
lution with a dwell time of 150 ms per channel. Analyst 1.5.1 soft-
ware (Applied Biosystems, Foster City, CA, USA) was applied for
instrumental control, data acquisition and quantitative analysis.
2.9. Statistical analysis
Median and the median absolute deviation (MAD) were used as
descriptors of the central tendency and dispersion of data distri-
butions for non-normally distributed data. Arithmetic mean and
standard deviation (SD) were used for normally distributed data.
The Shapiro-Wilk W test was employed to test for normality of the
data. Dependent groups of non-normally distributed data were
analyzed using the Wilcoxon matched pair test. Differences were
considered statistically significant when p < 0.05. Statistical testing
was performed using the STATISTICA 10 software (StatSoft, Inc.,
2011).
3. Results and discussion
3.1. Removal of BT and BTA in TWs
The influent and effluent concentrations of BT and BTA in the
TW columns are shown in Fig. 1.
Fig. 1. Influent (BT feed and BTA feed) and effluent (BT-U, BT-P, BTA-U, BTA-P) concentrations of BT and BTA in the TW system (central point indicates median; box shows 10 and 90
percentile range; whiskers show min and max values), n ¼ 16e18.
The effluent concentrations of BT and BTA were significantly
different from the BT and BTA influent concentration, respectively.
The median concentration (±MAD) of BT and BTA in the feed was
10.57 ± 0.30 and 9.93 ± 0.55 mg/L, respectively. The effluent con-
centration of BT in the unplanted columns BT-U was
0.60 ± 0.31 mg/L and was higher than for the planted columns BT-P,
which was 0.04 ± 0.03 mg/L, which corresponded to the removal
efficiency of 93.6% and 99.7%, respectively (Table S4). This differ-
ence was statistically significant. This allows for conclusion that the
presence of plants enhanced the removal of BT. For the columns fed
with the influent containing BTA the outflow concentration for the
unplanted columns BTA-U was 1.97 ± 0.51 mg/L, and for the planted
columns BTA-P it was 1.66 ± 0.74 mg/L, which corresponded to the
removal efficiency of 81.2% and 82.8%, respectively (Table S4). This
difference was not statistically significant. In conclusion, DF-TWs
can efficiently remove BT and BTA from synthetic municipal
wastewater. The removal of BT in the vegetated TWs was almost
complete, whereas the removal of BTA was 17% lower. The presence
of vegetation positively affects only the removal of BT.
The additional performance data corresponding to the results
presented in Fig. 1 is shown in Table S4 and is accompanied with the
light intensity and aboveground biomass in the TW system
(Table S5). Matamoros et al. (2010) reported similar removal effi-
ciencies (range 80e90%) for BT and BTA in the same type of a TW
system treating real wastewater containing these compounds at
the level of several mg/L. It can be concluded that DF-TWs ensure
high removal efficiency of BT and BTAs and taking into account the
results of Matamoros et al. (2010) this observation is valid for a
wide range of influent concentrations of these compounds: from
mg/L to mg/L levels.
3.2. Effect of BT and BTA on the removal of COD, N-NH4 and P-PO4
in TWs
The main objective of treating municipal wastewater is the
removal of nutrients, i.e. carbon, nitrogen and phosphorus. The
main focus of the present study was the removal of BT and BTA, but
the removal of nutrients was also monitored throughout the
experiment. Significant differences between feed and effluent
 E. Felis et al. / Water Research 104 (2016) 441e448
concentrations were observed for COD, BT and BTA, N-NH4, but not
for P-PO4. The effect of BT and BTA on the effluent concentration of
COD, N-NH4 and P-PO4 is shown in Table 1.
The effluent concentrations of COD, N-NH4 and P-PO4 were not
affected (no statistically significant differences) by the presence of
the studied micropollutants. This allows conclusion that the rela-
tively high concentrations of BT or BTA can be treated in DF-TWs
without compromising the removal efficiency of COD, N-NH4 and
P-PO4, which are the routinely measured parameters in the TW
effluents.
3.3. Products of BT and BTA transformation in TWs
The treatment of BT and BTA in the TW system resulted in the
formation of their TPs. The TPs of the investigated compounds were
tentatively identified by means of LC-MS/MS analysis in MRM
mode, according to the mass spectra (or MRM transitions pairs)
which were available in the references (Kloepfer et al., 2004; Wick
et al., 2010; Huntscha et al., 2014).
In the case of BTA, seven TPs-BTA were identified (Fig. 2). Six of
them were detected both in the effluents from the planted and
unplanted columns, and only 5-COOH-BTA was detected in the
effluent from the unplanted DF-TWs. The results may indicate a
similar mechanism of BTA transformation in TWs, regardless of the
presence of plants. Most of the identified TPs-BTA in DF-TWs were
mentioned in the literature as potential TPs of BTA, in batch ex-
periments with activated sludge (Liu et al., 2011; Huntscha et al.,
2014). Such an example is 1-CH3-BTA, which may be formed in
activated sludge systems both under aerobic and anaerobic condi-
tions (Liu et al., 2011; Huntscha et al., 2014). This substance was also
reported as an anaerobic biotransformation product in aquifer
materials (Liu et al., 2013). Another issue is the possibility of BTA
transformation to 4-CH3-BTA and 5-CH3-BTA. Admittedly,
Huntscha et al. (2014) stated the ability of BTA to transformation
under aerobic conditions to its derivatives containing a methyl
group attached to the benzene ring, but the methyl group was al-
ways accompanied by a hydroxyl group substituted in the other
part of benzene ring. A methylation of the aromatic ring is dis-
cussed as likely reaction of the microbial transformation of benzene
derivatives under anaerobic or anoxic conditions (Vogt et al., 2011).
According to Weelink et al. (2010), the methylation of a benzene
ring has been found in the presence of sulfate-reducing microor-
ganisms and the methyl group possibly might be generated from
bicarbonate via a reverse CO-dehydrogenase pathway. The ob-
tained results may suggest that both aerobic and anoxic zones
occurred in the investigated wetland systems. An aerobic decom-
position pathway of BTA was confirmed by the presence of other
TPs, namely: 1-OH-BTA, 4-OH-BTA and TP1-BTA (as shown in
Fig. 2). Additionally, the presence of the above-mentioned TPs-BTA
in the effluents from DF-TWs indicates that the BTA hydroxylation
may occur in both the benzene ring and triazole ring. In contrast,
Huntscha et al. (2014) in batch activated sludge experiments,
observed only the formation of 4-OH-BTA, while the formation of
1-OH-BTA was not observed. In the case of the wetland system,
Table 1
The effect of BT or BTA on the effluent concentrations of COD, N-NH4 and P-PO4 in the planted and unplanted columns (Median ± MAD, mg/L; n ¼ 12e13 for COD; n ¼ 13e14 for
N-NH4, n ¼ 7 for P-PO4).
Parameter Influent Planted columns (effluent)
CTRL-P BT-P
COD 822 ± 72 44 ± 8 50 ± 14
N-NH4 54.3 ± 25.7 5.6 ± 1.7 6.5 ± 2.3
P-PO4 34.0 ± 5.4 21.1 ± 7.4 21.1 ± 6.6
445
both 4-OH-BTA and 1-OH-BTA were present in the treated waste-
water and the signals corresponding to these two TPs were the
most intense compared to the other identified TPs-BTA.
During the experiments performed in the DF-TWs, it was found
that BT also undergoes transformation into several TPs (Fig. 3). After
10 weeks of the process, in the effluents of both planted and
unplanted DF-TWs, two TPs-BT were identified, namely: 2-OH-BT
and 2-SCH3-BT. Additionally, only in the effluents from the planted
DF-TWs two other TPs-BT were detected, but their structures were
not confirmed. The MS full scan of the TP2-BT (Fig. 3) is presented
in Fig. S3. 2-OH-BT was mentioned to be one of the main TPs of BT,
both in the natural aquatic environment (De Wever et al., 2001) and
during the laboratory tests (Haroune et al., 2002). The presence of
2-OH-BT indicates an aerobic degradation pathway of BT. However,
the presence of 2-SCH3-BT in the effluents from DF-TWs may
suggest the anoxic (or anaerobic) pathway of BT degradation. It is
supposed that the transformation of BT into 2-SCH3-BT takes
several steps. Firstly, the selected microorganisms are able to
transform BT into 2-SH-BT, then 2-SH-BT is converted into more
stable methylated product, i.e. 2-SCH3-BT. De Wever et al. (2001)
described the BT-methylating activity in crude extracts of variety
of soil and water bacterial isolates, e.g. Pseudomonas sp. As it was
mentioned previously, in the wetland system, the aerobic and
anoxic (or anaerobic) niches inhabited by different kind of micro-
organisms may occur simultaneously and it means that the aerobic
and anaerobic processes of BT decay can run in parallel. After 20
weeks of the experiment, both in the effluents from planted and
unplanted DF-TWs, a dicarboxylic acid intermediate derived from
BT (diacid-BT) was detected. As it was reported by Haroune et al.
(2002) the presence of this substance demonstrates that the ben-
zene ring cleavage (which is still attached to the thiazole ring)
occurs as a result of the activity of catechol 1,2-dioxygenase
(intradiol mechanism). Admittedly, the supposed catechol inter-
mediate of BT was not detected in the effluents samples, but the
detection of this substance may be difficult due to its immediate
conversion to diacid-BT by the action of the aforementioned
enzyme. In addition, the presence of 2-OH-BT and 2,6-diOH-BT in
the effluents from the unplanted DF-TWs, may support the hy-
pothesis that the microbial decay of BT occurs in several stages.
Firstly, the hydroxylation steps from BT to 2-OH-BT and then from
2-OH-BT to 2,6-diOH-BT are observed. Subsequently, 2,6-diOH-BT
is converted to the catechol intermediate of BT and then to diacid-
BT (Haroune et al., 2002). The above-described degradation
mechanism is mentioned to be a common decay pathway observed
in Gram-positive bacteria, particularly widespread in the soil,
namely Rhodococcus spp. (Haroune et al., 2002). It may suggest, that
the various Rhodococcus species can be involved in the trans-
formation process of BT in TW systems.
3.4. Removal and transformation of BT and BTA in the artificial
sunlight system
The decay efficiencies of the investigated substances in the
artificial sunlight system were dependent on the exposure time and
Unplanted columns (effluent)
BTA-P CTRL-U BT-U BTA-U
49 ± 8 44 ± 8 52 ± 14 55 ± 12
5.4 ± 1.1 13.0 ± 3.3 12.5 ± 2.8 11.3 ± 3.1
26.8 ± 3.0 25.6 ± 3.5 27.0 ± 2.4 23.6 ± 5.1
446 E. Felis et al. / Water Research 104 (2016) 441e448

Fig. 2. TPs of BTA formed during its degradation in the DF-TWs.

Fig. 3. TPs of BT degradation in DF-TWs: (a) after 10 weeks of the experiment, (b) after 20 weeks of the experiment.

intensity of irradiance. The initial concentrations of BT and BTA in and 9.88 ± 0.19 mg/L, respectively. The studies have shown that BT
aqueous solutions (distilled water) were equal 10.77 ± 0.11 mg/L was relatively high susceptible to degradation by the artificial
 E. Felis et al. / Water Research 104 (2016) 441e448 447

sunlight (Fig. S4.A). For example, after 480 min of the process obtained during direct photolysis processes. In the case of BTA, the
performed at an irradiance of 1000 W/m2, BT concentration in the kinetic constants calculated based on the results obtained by UV/
reaction solution was equal to 0.29 mg/L. It corresponded to 97% TiO2 process are at least by two orders of magnitude higher than
removal. The second investigated substance, BTA, was not signifi- the values obtained during the direct photolysis of BTA induced by
cantly degraded by the sunlight (Fig. S4.B). After 660 min of the the artificial sunlight. The main transformation products of the
irradiation (irradiance ¼ 1000 W/m2), only 26% BTA removal was investigated substances in these processes were 4-OH-BTA and 2-
observed. It was a significantly lower removal of BTA, in compari- OH-BT, for BTA and BT, respectively.
son to the results obtained by the other authors (Andreozzi et al., As it has been proved, the solar photocatalysis assisted by the
1998; Benitez et al., 2013; Bahnmüller et al., 2015). However, suspended TiO2 was an effective process for the elimination of both
those literature results were obtained in the processes induced by BT and BTA from the aqueous solution. Based on these results,
low-pressure mercury lamps, which are characterized by a different subsequent study on the photocatalytic post-treatment of the ef-
emission spectrum than the Xenon lamp used in the study. As it is fluents from the planted DF-TWs was carried out at a concentration
known, the emission spectrum of the light source has an important of the suspended TiO2 equal to 500 mg/L and the irradiance equal to
impact on the efficiency of the photodegradation process. 500 W/m2. The median concentrations of BT and BTA in the efflu-
As the low removal of BTA was observed, additional photo- ents from the planted DF-TWs, were equal to 0.64 ± 0.05 mg/L and
chemical experiments in the presence of TiO2 as photocatalyst were 0.94 ± 0.01, respectively. However, other organic pollutants
carried out. At the beginning of the experiment, the kinetic study generated during the wastewater treatment in the DF-TWs were
on the sorption of BT and BTA onto the TiO2 surface was performed. present in these effluents. The average TOC concentrations in the
It was estimated that for both compounds, the best fit of the treated wastewater were equal to 5.29 mg/L and 4.34 mg/L,
mathematical model to the experimental data was found for the respectively for BT and BTA. In such conditions, after 3 h of the
BrunauereEmmetteTeller (BET) isotherm (Table S6) (Ebadi et al., photocatalytic post-treatment, the removal of BT was at the level of
2009). It means that multilayer adsorption occurred, that is, not 58%, and for BTA the removal efficiency was equal to 42%. This was
all the adsorbed molecules of BT and BTA were in contact with the especially significant in the case of the elimination of BTA, because
surface layer of TiO2. the removal of this compound was 96% in the coupled processes,
The photocatalytic studies on the BT and BTA decomposition this is, incorporation of TWs with the TiO2-photocatalysis induced
were performed at the irradiance equal to 500 W/m2 and the by the sunlight. In both cases, the TOC removal after 3 h of irradi-
following doses of TiO2: 100, 300, 500 mg/L. The addition of TiO2, ation was higher than 25%. These results indicated, that photo-
improved considerably the degradation efficiencies of the investi- catalysis may be a viable technique for the removal of the
gated compounds (Figs. S4.C and S4.D). After 60 min of the pro- investigated compounds from the wastewater treated in TWs,
cesses performed at the dose of TiO2 ¼ 500 mg/L, the however its efficiency depends on the presence of other (organic)
concentrations of both BT and BTA in the reaction solutions were substances in the wastewater matrices. The other substances may
below their LODs. The increase of the reaction rate of BT and BTA compete with target compounds for an access to the surface of TiO2,
photodegradation in the presence of TiO2 in comparison to the the OH generated during the process and the solar radiation. The
reaction rate of both compounds obtained in the direct photolysis efficiency of the photocatalysis as a tertiary treatment can be
was confirmed by the estimation of the kinetic parameters of these improved by applying higher concentrations of TiO2, however
reactions. The direct photolysis is very often described by a first below certain turbidity threshold above which penetration of the
order equation (Miller and Olejnik, 2001) In the case of the sunlight into the wastewater is impaired. Such conditions may be
photochemical processes performed in the presence of TiO2, the ensured by the application of immobilized TiO2, which is planned
Langmuir-Hinshelwood kinetic model can be used (Herrmann, in the future studies.
2005). According to this theory, at low initial concentrations of
the substrate (C0 < 103 M), its decomposition process occurs ac-
cording to the model of pseudo first order reaction and a major 4. Conclusions
impact on the reaction rate has the capacity of the substance to
adsorb onto the surface of TiO2. The calculated kinetic parameters  The relatively high concentrations of BT or BTA (at the level of
of the above-mentioned processes indicated that the addition of 10 mg/L) in a municipal wastewater can be reduced in DF-TWs
TiO2 caused a significant increase of the reaction rates (Table 2). The without compromising the removal efficiency of COD, N-NH4
kinetic constants of the reaction of BT degradation in the UV/TiO2 and P-PO4.
process are by an order of magnitude higher than the values  DF-TWs ensure high removal efficiency of BT both in planted
(99.7%) and unplanted (93.6%) systems. Thus, it can be
concluded that the presence of plants enhanced the removal of
this compound.
Table 2
 DF-TWs ensure also relatively high removal efficiency of BTA
Kinetic parameters of BT and BTA photochemical degradation.
both in planted (82.8%) and unplanted (81.2%) systems, how-
Process BT BTA ever, the plants did not affect the efficiency of the process.
ka, h1 R2 ½ tb, h k, h1 R2 ½ t, h  It was proved that BTA is poorly removed by using the artificial
UV 250c W/m2 0.152 0.998 4.56 0.020 0.830 35.19
sunlight; however, the addition of TiO2 significantly affected the
500c W/m2 0.244 0.999 2.84 0.020 0.944 35.19 efficiency and rate of its removal.
750c W/m2 0.309 0.997 2.24 0.022 0.981 32.09  The combination of DF-TWs with TiO2-photocatalysis under
1000c W/m2 0.450 0.998 1.54 0.026 0.985 26.26 artificial sunlight irradiation contributed to an increase in the
UV/TiO2 100d mg/L 2.845 0.973 0.24 7.283 0.991 0.10
efficiency of the decay of the investigated substances. It is
300d mg/L 3.062 0.950 0.23 8.935 0.987 0.08
500d mg/L 3.338 0.980 0.21 10.028 0.991 0.07 especially significant for BTA, because the abatement of this
a
compound in the coupled processes was 96%.
Pseudo first order reaction constant.
b
Half-life time.
 During the treatment in DF-TWs, mainly hydroxylated de-
c
Lamp irradiance. rivatives of BT were formed, whereas for BTA the formation of
d
TiO2 concentration. both methylated and hydroxylated derivatives was observed.
448 E. Felis et al. / Water Research 104 (2016) 441e448
 The main transformation products of BT and BTA in the presence
of TiO2 under artificial sunlight were their hydroxylated
derivatives.
 The coupling of DF-TWs with TiO2-photocatalysis induced by
the sunlight seems to be a promising method for elimination of
selected corrosion inhibitors from the aquatic environment.
Acknowledgement
This work was supported by the National Science Centre
(Poland) Project UMO-2011/03/D/ST8/04595.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.watres.2016.08.037.
References
Andreozzi, R., Caprio, V., Insola, A., Longo, G., 1998. Photochemical degradation of
benzotriazole in aqueous solution. J. Chem. Technol. Biotechnol. 73, 93e98.
Asimakopoulos, A.G., Wang, L., Thomaidis, N.S., Kannan, K., 2013. Benzotriazoles
and benzothiazoles in human urine from several countries: a perspective on
occurrence, biotransformation, and human exposure. Environ. Int. 59, 274e281.
Bahnmüller, S., Loi, C.H., Linge, K.L., von Gunten, U., Canonica, S., 2015. Degradation
rates of benzotriazoles and benzothiazoles under UV-C irradiation and the
advanced oxidation process UV/H2O2. Water Res. 74, 143e154.
Benitez, F.J., Acero, J.L., Real, F.J., Roldan, G., Rodriguez, E., 2013. Photolysis of model
emerging contaminants in ultra-pure water: kinetics, by-products formation
and degradation pathways. Water Res. 47 (2), 870e880.
Bester, K., Scha€fer, D., 2009. Activated soil filters (bio filters) for the elimination of
xenobiotics (micro-pollutants) from storm- and waste waters. Water Res. 43
(10), 2639e2646.
Burgess, P., 2009. Variation in Light Intensity at Different Latitudes and Seasons,
Effects of Cloud Cover, and the Amounts of Direct and Diffused Light. Contin-
uous Cover Forestry Group Scientific Meeting. Westonbirt Arboretum, Glou-
cestershire, UK. September 29, 2009. http://www.ccfg.org.uk/conferences/
downloads/P_Burgess.pdf (accessed February, 22 2016).
Castro, S., Davis, L.C., Erickson, L.E., 2005. Natural, cost-effective, and sustainable
alternatives for treatment of aircraft deicing fluid waste. Environ. Prog. 24 (1),
26e33.
De la Cruz, N., Dantas, R.F., Gime nez, J., Esplugas, S., 2013. Photolysis and TiO2
photocatalysis of the pharmaceutical propranolol: solar and artificial light.
Appl. Catal. B Environ. 130e131, 249e256.
De Wever, H., Besse, P., Verachtert, H., 2001. Microbial transformations of 2-
substituted benzothiazoles. Appl. Microbiol. Biotechnol. 57, 620e625.
Ding, Y., Yang, C., Zhu, L., Zhang, J., 2010. Photoelectrochemical activity of liquid
phase deposited TiO2 film for degradation of benzotriazole. J. Hazard Mater. 175
(1e3), 96e103.
Eaton, A.D., Clesceri, L.S., Rice, E.W., Greenberg, A.E., 2005. Standard Methods for the
Examination of Water and Wastewater, twenty-first ed. APHA, AWWA, WEF,
Washington, USA.
Ebadi, A., Mohammadzadeh, J.S.S., Khudiev, A., 2009. What is the correct form of
BET isotherm for modeling liquid phase adsorption? Adsorption 15 (1), 65e73.
Eggen, R.I., Hollender, J., Joss, A., Schaerer, M., Stamm, C., 2014. Reducing the
discharge of micropollutants in the aquatic environment: the benefits of
upgrading wastewater treatment plants. Environ. Sci. Technol. 48 (14),
7683e7689.
Gan, Z., Sun, H., Wang, R., Hu, H., Zhang, P., Ren, X., 2014. Transformation of ace-
sulfame in water under natural sunlight: joint effect of photolysis and
biodegradation. Water Res. 64, 113e122.
Giger, W., 2006. Benzotriazoles anticorrosives in municipal wastewaters and in the
aquatic environment. In: NORMAN Workshop, Stresa, Italy, June 19, 2006.
http://www.norman-network.net/sites/default/files/files/Events/2006-2008/
2006Jun19-20-Stresa-EmEnvPollutants-KeyIssuesChallenges/08-giger_nor-
man_stresa_06_session_i.pdf (accessed December, 10 2015).
Haroune, N., Combourieu, B., Besse, P., Sancelme, M., Reemtsma, T., Kloepfer, A.,
Diab, A., Knapp, J.S., Baumberg, S., Delort, A.M., 2002. Benzothiazole degrada-
tion by Rhodococcus pyridinovorans strain PA: evidence of a catechol 1,2-diox-
ygenase activity. Appl. Environ. Microbiol. 68 (12), 6114e6120.
Herrero, P., Borrull, F., Pocurull, E., Marce , R.M., 2014. An overview of analytical
methods and occurrence of benzotriazoles, benzothiazoles and benzenesulfo-
namides in the environment. TrAC Trend. Anal. Chem. 62, 46e55.
Herrmann, J.M., 2005. Heterogeneous photocatalysis: state of the art and present
applications. Top. Catal. 34 (1e4), 49e65.
Huntscha, S., Hofstetter, T.B., Schymanski, E.L., Spahr, S., Hollender, J., 2014.
Biotransformation of benzotriazoles: insights from transformation product
identification and compound-specific isotope analysis. Environ. Sci. Technol. 48
(8), 4435e4443.
Janna, H., Scrimshaw, M.D., Williams, R.J., Churchley, J., Sumpter, J.P., 2011. From
dishwasher to tap? Xenobiotic substances benzotriazole and tolyltriazole in the
environment. Environ. Sci. Technol. 45 (9), 3858e3864.
Jover, E., Matamoros, V., Bayona, J.M., 2009. Characterization of benzothiazoles,
benzotriazoles and benzosulfonamides in aqueous matrixes by solid-phase
extraction followed by comprehensive two-dimensional gas chromatography
coupled to time-of-flight mass spectrometry. J. Chromatogr. A 1216, 4013e4019.
Klavarioti, M., Mantzavinos, D., Kassinos, D., 2009. Removal of residual pharma-
ceuticals from aqueous systems by advanced oxidation processes. Environ. Int.
35 (2), 402e417.
Kloepfer, A., Jekel, M., Reemtsma, T., 2004. Determination of benzothiazoles from
complex aqueous samples by liquid chromatography-mass spectrometry
following solid-phase extraction. J. Chromatogr. A 1058, 81e88.
Kloepfer, A., Jekel, M., Reemtsma, T., 2005. Occurrence, sources, and fate of ben-
zothiazoles in municipal wastewater treatment plants. Environ. Sci. Technol. 39,
3792e3798.
Lin, A.Y., Reinhard, M., 2005. Photodegradation of common environmental phar-
maceuticals and estrogens in river water. Environ. Toxicol. Chem. 24 (6),
1303e1309.
Liu, R., Zhao, Y., Doherty, L., Hu, Y., Hao, X., 2015. A review of incorporation of
constructed wetland with other treatment processes. Chem. Eng. J. 279,
220e230.
Liu, Y.S., Ying, G.G., Shareef, A., Kookana, R.S., 2013. Biodegradation of three selected
benzotriazoles in aquifer materials under aerobic and anaerobic conditions.
J. Contam. Hydrol. 151, 131e139.
Liu, Y.S., Ying, G.G., Shareef, A., Kookana, R.S., 2011. Biodegradation of three selected
benzotriazoles under aerobic and anaerobic conditions. Water Res. 45 (16),
5005e5014.
Matamoros, V., Rodríguez, Y., Albaige s, J., 2016. A comparative assessment of
intensive and extensive wastewater treatment technologies for removing
emerging contaminants in small communities. Water Res. 88, 777e785.
Matamoros, V., Jover, E., Bayona, J.M., 2010. Occurrence and fate of benzothiazoles
and benzotriazoles in constructed wetlands. Water Sci. Technol. 61 (1),
1991e197.
MATLAB 10, The MathWorks, Inc., Natick, Massachusetts, United States.
Miksch, K., Cema, G., Corvini, P.F.X., Felis, E., Sochacki, A., Surmacz-Go  rska, J.,
_
Wiszniowski, J., Zabczynski, S., 2015. R&D priorities in the field of sustainable
remediation and purification of agro-industrial and municipal wastewater.
N. Biotechnol. 32 (1), 128e132.
Miller, J.S., Olejnik, D., 2001. Photolysis of polycyclic aromatic hydrocarbons in
water. Water Res. 35, 233e243.
Ni, H.-G., Lu, F.-H., Luo, X.-L., Tian, H.-Y., Zeng, E.Y., 2008. Occurrence, phase dis-
tribution, and mass loadings of benzothiazoles in riverine runoff of the Pearl
River Delta, China. Environ. Sci. Technol. 42 (6), 1892e1897.
Nowrotek, M., Sochacki, A., Felis, E., Miksch, K., 2016. Removal of diclofenac and
sulfamethoxazole from synthetic municipal waste water in microcosm down-
flow constructed wetlands: start-up results. Int. J. Phytoremediat. 18 (2),
157e163.
Oller, I., Malato, S., Sanchez-Perez, J.A., 2011. Combination of advanced oxidation
processes and biological treatments for wastewater decontamination e A re-
view. Sci. Total Environ. 409, 4141e4166.
Pereira, J., Vilar, V., Borges, M., Gonzalez, O., Esplugas, S., Boaventura, R., 2011.
Photocatalytic degradation of oxytetracycline using TiO2 under natural and
simulated solar radiation. Sol. Energy 85, 2732e2740.
Reemtsma, T., Fiehn, O., Kalnowski, G., Jekel, M., 1995. Microbial transformation and
biological effect of fungicide derived benzothiazoles determined in industrial
wastewater. Environ. Sci. Technol. 29, 478e485.
Stasinakis, A.S., Thomaidis, N.S., Arvaniti, O.S., Asimakopoulos, A.G., Samaras, V.G.,
Ajibola, A., Mamais, D., Lekkas, T.D., 2013. Contribution of primary and sec-
ondary treatment on the removal of benzothiazoles, benzotriazoles, endocrine
disruptors, pharmaceuticals and perfluorinated compounds in a sewage treat-
ment plant. Sci. Total Environ. 463, 1067e1075.
StatSoft, Inc, 2011. STATISTICA (Data Analysis Software System), Version 10. www.
statsoft.com.
Sulej, A.M., Polkowska, Z.,_ Namiesnik, J., 2011. Analysis of airport runoff waters. Crit.
Rev. Anal. Chem. 41 (3), 190e213.
Vogt, C., Kleinsteuber, S., Richnow, H.H., 2011. Anaerobic benzene degradation by
bacteria. Microb. Biotechnol. 4 (6), 710e724.
Voutsa, D., Hartmann, P., Schaffner, C., Giger, W., 2006. Benzotriazoles, alkylphenols
and bisphenol A in municipal wastewaters and in the Glatt River, Switzerland.
Environ. Sci. Pollut. Res. Int. 13 (5), 333e341.
Vymazal, J., 2013. The use of hybrid constructed wetlands for wastewater treatment
with special attention to nitrogen removal: a review of a recent development.
Water Res. 47 (14), 4795e4811.
Weelink, S.A.B., van Eekert, M.H.A., Stams, A.J.M., 2010. Degradation of BTEX by
anaerobic bacteria: physiology and application. Rev. Environ. Sci. Biotechnol. 9,
359e385.
Weiss, S., Reemtsma, T., 2005. Determination of benzotriazole corrosion inhibitors
from aqueous environmental samples by liquid chromatography-electrospray
ionization-tandem mass spectrometry. Anal. Chem. 77, 7415e7420.
Wick, A., Fink, G., Ternes, T.A., 2010. Comparison of electrospray ionization and
atmospheric pressure chemical ionization for multi-residue analysis of biocides,
UV-filters and benzothiazoles in aqueous matrices and activated sludge by
liquid chromatographyetandem mass spectrometry. J. Chromatogr. A 1217 (14),
2088e2103.