Science of the Total Environment 557–558 (2016) 257–267

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Degradation of the pharmaceuticals diclofenac and sulfamethoxazole

and their transformation products under controlled
environmental conditions
S. Poirier-Larabie a, P.A. Segura b, C. Gagnon a,⁎
a
Aquatic Contaminants Research Division, Science and Water Technology Directorate, Environment Canada, Montréal, Québec H2Y 2E7, Canada
b
Department of Chemistry, Université de Sherbrooke, Sherbrooke, Québec J1K 2R1, Canada

H I G H L I G H T S G R A P H I C A L A B S T R A C T

• Degradation products vary depending

on the process involved.
• Kinetics of degradation depend on me-
dia physicochemical characteristics.
• Some products of degradation are not
persistent.

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

Article history: Contamination of the aquatic environment by pharmaceuticals via urban effluents is well known. Several classes
Received 15 January 2016 of drugs have been identified in waterways surrounding these effluents in the last 15 years. To better understand
Received in revised form 9 March 2016 the fate of pharmaceuticals in ecosystems, degradation processes need to be investigated and transformation
Accepted 9 March 2016
products must be identified. Thus, this study presents the first comparative study between three different natural
Available online xxxx
environmental conditions: photolysis and biodegradation in aerobic and anaerobic conditions both in the dark of
Editor: D. Barcelo diclofenac and sulfamethoxazole, two common drugs present in significant amounts in impacted surface waters.
Results indicated that degradation kinetics differed depending on the process and the type of drug and the ob-
Keywords: served transformation products also differed among these exposure conditions. Diclofenac was nearly degraded
Diclofenac by photolysis after 4 days, while its concentration only decreased by 42% after 57 days of exposure to bacteria in
Sulfamethoxazole aerobic media and barely 1% in anaerobic media. For sulfamethoxazole, 84% of the initial concentration was still
Photodegradation present after 11 days of exposure to light, while biodegradation decreased its concentration by 33% after 58 days
Biodegradation of exposure under aerobic conditions and 5% after 70 days of anaerobic exposure. In addition, several transforma-
Aquatic environment
tion products were observed and persisted over time while others degraded in turn. For diclofenac, chlorine
atoms were lost primarily in the photolysis, while a redox reaction was promoted by biodegradation under aer-
obic conditions. For sulfamethoxazole, isomerization was favored by photolysis while a redox reaction was also
favored by the biodegradation under aerobic conditions. To summarize this study points out the occurrence of

⁎ Corresponding author.
E-mail address: christian.gagnon@canada.ca (C. Gagnon).

http://dx.doi.org/10.1016/j.scitotenv.2016.03.057
0048-9697/Crown Copyright © 2016 Published by Elsevier B.V. All rights reserved.
258 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267
different transformation products under variable degradation conditions and demonstrates that specific func-
tional groups are involved in the tested natural attenuation processes. Given the complexity of environmental
samples more analytical effort is needed to fully identify new products of potential toxicity.
1. Introduction
The presence of pharmaceuticals in urban discharges is an emerging
concern for the disruption of aquatic ecosystems and human health. Nu-
merous studies have been conducted on the presence of pharmaceuti-
cals in the aquatic environment (Aminot, 2013; Fatta-Kassinos et al.,
2011a; Hernández et al., 2007; Segura et al., 2009; Ziylan and Ince,
2011). These studies show that a vast amount of various drugs from
urban effluents end up in waterways at concentrations of potential con-
cern. The degradation of these pharmaceuticals in the environment also
needs to be considered, thus studies on their environmental fate to un-
derstand degradation mechanisms and transformation products arising
therefrom are required to assess their toxicity. Unfortunately, compared
to the presence of parent compounds, the occurrence of pharmaceutical
transformation products in the aquatic environment is much less
known. Several studies have identified transformation products of key
pharmaceuticals in laboratory and wastewater treatment plants but
few studies have investigated them in the aquatic environment.
The natural degradation of several pharmaceutical classes such as
antibiotics, non-steroidal anti-inflammatory drugs, anti-psychotics, β-
blockers, cholesterol-lowering drugs, hormones and analgesics have
been studied (Challis et al., 2014). Different experimental degradation
conditions were previously investigated, such as photo-transformation
(Boreen et al., 2003; Challis et al., 2014; Yan and Song, 2014), including
the photolysis (Batchu et al., 2014; Gonzalez et al., 2009; Huguet et al.,
2013; Salgado et al., 2013; Schulze et al., 2010) and photocatalysis
(Nasuhoglu et al., 2011), the oxidative decarboxylation (Huguet et al.,
2013), biodegradation (Langenhoff et al., 2013; Zhang et al., 2008), deg-
radation by ultrasound (Hartmann et al., 2008; Ziylan et al., 2014) and
the influence of the degree of sun exposure (Bartels and von Tümpling
Jr, 2007). However, few studies have focused on all aspects of the degra-
dation of pharmaceuticals in environmental media. A variety of degra-
dation processes may occur in the environment: photo-transformation
by sunlight and biodegradation by bacteria under aerobic and anaerobic
conditions. Several factors can also influence environmental degrada-
tion such as physico-chemical water conditions (pH, oxygen, tempera-
ture, turbidity), and the weather (sun brightness).
There are several indications that photochemical degradation is one
of the most significant processes with regard to the environmental fate
of pharmaceuticals. This type of degradation depends on several factors
including physico-chemical characteristics of pharmaceuticals and their
specific structure such as presence of aromatic rings, conjugated π sys-
tems, diverse functional groups and heteroatoms, which facilitate direct
absorption of solar radiation (Challis et al., 2014; Fatta-Kassinos et al.,
2011b). In streams, it can be generally assumed that pharmaceuticals
are exposed to solar radiation composed of photons of UV, infrared
and visible wavelengths, which are mostly above 290 nm, as well as
bacteria in aerobic and anaerobic media. Additionally, photosensitizing
species such as photolytically excited natural organic matter (NOM), ni-
trates, carbonates and iron in the water column may lead to indirect
photolysis at temperatures ranging from 0 to 20 °C and a fairly stable
pH around 8 (Challis et al., 2014; Fatta-Kassinos et al., 2011b). All
these factors play an important role in the photodegradation of pharma-
ceuticals in the environment as shown by a previous study on the
degradation of antibiotics, which demonstrated that the fate of these
compounds was dependent on media pH, organic and chlorine
contents, as well as the irradiation source (Batchu et al., 2014). There-
fore, two pharmaceuticals with different physico-chemical properties
Crown Copyright © 2016 Published by Elsevier B.V. All rights reserved.
were chosen to experiment diverse degradation processes: the anti-
inflammatory drug diclofenac (DCF) and the antibiotic sulfamethoxa-
zole (SMX).
Photo-transformation pathways of DCF in a reconstructed standard
freshwater have been studied in controlled hydrolysis and photolysis
experiments (Agüera et al., 2005). Identification of photo-products gen-
erated by sunlight in aerobic condition was done by gas chromatogra-
phy mass spectrometry and liquid chromatography coupled to time-
of-flight mass spectrometry (LC-TOFMS). This latter technique has
proven to be one of the most powerful approaches currently existing
to investigate suspected or unknown transformation products. Various
degradation products can be proposed following the analysis by GC–
MS spectra (mass and fragments) and LC-TOFMS (exact mass), that sug-
gest loss of chlorine, hydroxylation, cyclisation, loss of CO2, oxidation
and reduction.
Photo-transformation pathways of SMX in water matrices with
enriched environmental components have been studied for a better un-
derstanding of the effects of multi-factors on the drug (Niu et al., 2013).
They concluded that biodegradation and hydrolysis are negligible in
dark. Also, photodegradation is influenced by the concentration of
SMX and the solution pH: the photodegradation decreased with an in-
crease in drug concentration and higher pH as well. Based on LC-MS/
MS triple quadrupole analysis, a photo-transformation mechanism in
pure water was proposed which included three main pathways: hy-
droxylation of the phenyl, fragmentation of the isoxazole ring and
cleavage of the sulfoxide bond. Moreover, increasing concentration
of fulvic acids led to a decrease in the photodegradation of SMX
and the same phenomenon was observed with increase in concen-
tration of suspended sediments.
Analysis of pharmaceuticals and their degradation products in
water samples is typically done by solid phase extraction (SPE)
prior to LC-MS/MS and GC–MS (Baker and Kasprzyk-Hordern,
2011; Fatta et al., 2007; Hollender et al., 2010; Kosjek and Heath,
2008). Accurate mass and high resolution mass spectrometers,
such as time-of-flight (TOF) or hybrid quadrupole-time-of-flight
(QqTOF), are particularly useful to confirm the exact mass of un-
known (Agüera et al., 2005) and associate a chemical formula to a
compound. By performing tandem mass experiments, further molec-
ular fragmentation can be performed and compared to the original
drug's fragmentation pattern to identify similar fragments, find
modifications and suggest probable molecular structures. The devel-
opment of a method to determine the degradation products has,
however, its challenges: the formation of several unknown products
with different physicochemical properties makes difficult the devel-
opment of a single effective extraction method for all compounds
along with a single chromatography method for their separation.
Furthermore, not all degradation products have available commer-
cial analytical standard to confirm the identity of the transformation
products which implies that molecular structures were proposed to
explain potential degradation pathways.
The aim of this study was to determine individually the influence
of controlled natural environmental factors on the degradation of
DCF, which rapidly degrades in the environment (Jiskra, 2008),
and SMX, which is rather likely much more persistent (Niu et al.,
2013) and then compare the degradation products obtained from
different natural degradation processes. These pharmaceuticals
were chosen for the present study because of their ubiquity and
high concentrations in receiving waters of municipal wastewater
 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267
effluents (Gagnon and Lajeunesse, 2012; Guerra et al., 2014; Segura
et al., 2007). Different processes were studied related to environ-
mental conditions in streams downstream of municipal sewage dis-
charges such as exposure period to light and the presence of aerobic
and anaerobic bacteria in water slightly spiked with wastewater
treatment plant effluent (0.1% v/v) at constant water temperature
and pH. To reach these objectives, an analytical method was devel-
oped and validated for the extraction and the analysis of SMX and
DCF by LC-QqTOFMS. Thereafter, the degradation products were
identified with the open platform XCMS online and by analysis of
high resolution tandem mass spectra.
2. Material and method
2.1. Reagents
Standards of diclofenac sodium salt (DCF) and sulfamethoxazole
(SMX) were purchased from Sigma-Aldrich Canada (Oakville, ON).
Isotopically-labeled standards of DCF (acetophenyl ring 13C6) and SMX
(phenyl 13C6) were purchased from Sigma-Aldrich Canada (Oakville,
ON) and used as internal standards. LC-MS grade water (H2O), metha-
nol (MeOH) and acetonitrile (ACN) were purchased from Fisher
Scientific Canada (Ottawa, ON). Sodium Citrate Buffer was purchased
from A&C (Saint-Laurent, Québec). Formic acid (FA) of LC-MS grade
was purchased from Sigma-Aldrich Canada (Oakville, ON). Deionized
water (DI-H2O) passed through a Milli-Q Advantage A10 system
(Millipore, Billerica, MA) was used for method blanks. This system is
equipped with activated carbon, an ion exchange resin and a UV lamp
to reduce total organic carbon (TOC) to ≤5 ppb and increase resistivity
(≥18.0 MΩ cm).
2.2. Analytical method development
A method for the analysis of DCF and its photo-transformation
products in a reconstructed standard freshwater has been developed
by (Agüera et al., 2005). On the other hand, a method for the analysis
of six anti-infectives including SMX in wastewater has been devel-
oped by (Segura et al., 2007). Since we wanted to analyze both DCF
and SMX simultaneously with their potential degradation products
from different degradation mechanisms, optimal recovery was
required for low concentrations levels of degradation products. The
extraction and chromatography methods were therefore optimized
as described below for better recovery, separation and adequate
retention.
2.3. Solid-phase extraction (SPE)
An analytical method based on SPE and LC-QqTOFMS was devel-
oped to identify and analyze simultaneously DCF and SMX and
their potential degradation products. SPE was performed using
Strata-X-CW cartridges manufactured by Phenomenex (Torrance,
CA) and a Vac Master Sample Processing Station from International
Sorbent Technology purchased from Biotage (Charlotte, NC). These
cartridges contain 200 mg of weak cation exchange mixed-mode
polymeric sorbent with a particle size of 33 μm. Cartridges were con-
ditioned first with 5 mL of MeOH and then 5 mL of DI-H2O. Sample
pH was adjusted by adding 10 mL of a solution of 100 mM sodium
citrate buffer at pH 6.5 to 90 mL of water sample. Sample load flow
rate was about 2–4 mL min− 1 . After sample loading, cartridges
were washed with 5 mL of 10 mM citrate buffer solution. Residual
H2O was removed from the cartridges using the manifold at maxi-
mum vacuum of − 0.5 bar. Analytes were eluted with 2 × 2.5 mL of
MeOH. SPE extracts were collected in 10 mL centrifuge tubes and
evaporated to dryness under a gentle stream of nitrogen, in a water
bath set to 40 °C. Evaporated extracts were then reconstituted to
259
1 mL with 0.1% FA in 95% H2O/5% ACN (v/v), vortexed for 10 s and
transferred to 2-mL glass vials for LC-QqTOFMS analysis.
2.4. Liquid chromatography quadrupole-time-of-flight tandem mass
spectrometry (LC-QqTOFMS) analysis
Analysis of samples was done using a 1200 Series liquid chromatog-
rapher coupled to a quadrupole-time-of-flight mass spectrometer Accu-
rate Mass Q-TOF 6530 both manufactured by Agilent (Santa Clara, CA).
Separation of analytes with good retention and separation was efficient-
ly performed using a Waters solid-core Cortecs C18 + column
(150 × 2.1 mm, 2.7 μm, 90 Å) (Fig. 1). This column contains a positively
charged surface that allows for better peak shape for basic compounds.
Solvent A was 0.1% FA in 95% H2O/5% ACN (v/v) and solvent B was 0.1%
FA in 95% ACN/5% H2O (v/v). The injection volume was 10 μL and the
flow rate was 400 μL min−1. The binary LC solvent gradient was: (%B):
0 min (5%), 2 min (5%), 5 min (50%), 10 min (50%), 11 min (5%), and
13 min (5%). Ionization was performed with a heated electrospray ion-
ization (HESI) source in positive mode. Source parameters were the fol-
lowing: gas temperature 300 °C; drying gas flow 7 L min−1, nebulizer
40 psig, sheath gas temperature 350 °C, sheath gas flow 12 L min−1,
fragmentor 160 V; skimmer 60 V and capillary voltage 3000 V. The
QqTOF mass analyzer was operated in the 2GHz high resolution mode
with a low mass range (m/z 1700). In this setting, full width at half max-
imum resolution (RFWHM) was ≈7800 and mass accuracy ≤ 2 ppm for
the reference ion m/z 322. Purine (m/z 121.050873) and Hexakis (m/z
922.009798) were used as internal reference masses to improve mass
accuracy. Two different mass spectrometric modes were used. First, sur-
face water extracts were analyzed in the MS (TOF only) mode to quanti-
fy parent drugs and find transformation products. TOF-only parameters
were chosen as follow: acquisition rage of m/z 80–930; acquisition rate
of 1 spectra s−1; acquisition time of 1000.2 ms spectrum−1; transients
per spectrum of 13,583. Identified compounds were then selected for
MS/MS analysis in the QqTOF mode. QqTOF parameters were: mass
range for MS and MS/MS m/z 100–500; acquisition rate 1 spectra s−1;
acquisition time 1000.2 ms spectrum−1; transients spectrum−1
13,583; quadrupole isolation narrow width 1.3 u; collision energy 10,
20 and 30 eV, and acquisition time 1000 ms spectrum−1.
2.5. Data processing
LC-MS raw data obtained from TOF analysis of degradation samples
were analyzed with XCMS online (https://xcmsonline.scripps.edu/)
open source platform (Segura et al., 2015). This software compares ref-
erence (control) with treatment (degraded samples) acquisition files
and identifies ions that increase or decrease in abundance. Then, the on-
line tool Chemcalc (http://www.chemcalc.org/) was used to determine
the chemical formula of the observed product. Tandem mass spectrom-
etry experiments were performed to obtain product ion spectra and as-
sociate the resulting fragments and their chemical formula to a
molecular structure.
To determine the degradation kinetics of DCF and SMX, results were
processed with the Mass Hunter Qualitative analysis software from
Agilent Technologies. Peaks areas were transferred into a Microsoft
Excel sheet and concentrations of parent drugs (SMX and DCF) in the
samples were determined by internal calibration using least-squares
linear regression. Calibration standards were prepared in water spiked
with 0.1% v/v effluent from a wastewater treatment plant using activat-
ed sludge as secondary treatment. Concentrations of the transformation
products were determined semi-quantitatively using calibration curves
of their respective parent compounds and reported as parent drug
equivalents. This approach assumes that ionization efficiency, precision,
accuracy, recovery and matrix effects of parent drugs are similar to
those of their respective degradation products.
260 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267
Fig. 1. Chromatography and mass spectra; A: diclofenac and B: sulfamethoxazole.
2.6. Degradation experiments
Three degradation experiments were conducted: aerobic
photodegradation, and biodegradation in aerobic and anaerobic con-
ditions, both in the dark. High concentration of DCF and SMX
(9000 ng/mL) were spiked in all degradation experiments in order
ease the characterization of all degradation products, with a single
analytical method. No carry-over effect was observed.
2.7. Photodegradation experiment
The influence of sunlight on DCF and SMX photodegradation was
studied in dechlorinated tap water in order to mimic surface water,
but with negligible levels of bacteria or suspended particles that could
indirectly influence the photolysis results. Tap water samples (a blank
and a spiked sample) were oxygenated in order to maintain aerobic
conditions as in surface waters. Solutions of 9000 ng/mL of DCF or
SMX in 2 L-glass Erlenmeyer flasks were exposed to an artificial sunlight
using an Exo Terra Solar Glo 125 W lamp at a distance of 10 cm. This
lamp emits visual light, UVA, UVB and infrared radiation. Light is emit-
ted at 86,000 lx at the front of the reactors and 12,100 lx which passes
through behind them. The luminous flux was measured by a luxometer
(Traceable Dual-Range Light Meter at a rate of 0.4 s). Knowing that di-
rect sunlight corresponds to 32,000–100,000 lx and a sunny day with-
out direct exposure corresponds to 10,000 to 25,000 lx, one can
conclude that exposure of the sample to artificial sunlight corresponds
to the bright sun illumination. Natural aerobic conditions were kept
using an air bubbling system (Optima Pump 1000 cm3 min−1, 4 psi),
over a period of 12 days. The pH of the solution was 8.0 ± 0.5 which
correspond to the pH of the surface water streams downstream of stud-
ied municipal sewage discharge. Samples were analyzed in triplicate at
the beginning of the experiment and after 1, 4 and 12 days.
2.8. Aerobic biodegradation experiment
The influence of bacteria in an urban effluent dispersion plume was
investigated under aerobic and dark conditions. A solution of 0.1% of
municipal effluent wastewater (activated sludge treatment plant), forti-
fied with 9000 ng/mL of DCF or SMX standard, was exposed to the at-
mospheric oxygen by bubbling. The pH of the solution was 8.0 ± 0.5
which correspond to the pH of the surface water streams downstream
of studied municipal sewage discharge. A blank of 0.1% v/v of municipal
effluent wastewater was also exposed at the same conditions. Analysis
of samples in triplicate was performed at the beginning of the experi-
ment and after 1, 15, 30, 44 and 58 days. Bacterial activity was moni-
tored on an agarose gel.
2.9. Anaerobic biodegradation experiment
The influence of bacteria in an urban effluent dispersion plume
under anaerobic and dark conditions was studied. A solution of 0.1% of
municipal effluent wastewater (activated sludge treatment plant), forti-
fied with 9000 ng/mL of DCF or SMX standard, was bubbled with nitro-
gen gas for 30 min prior being sealed with a Parafilm membrane. The pH
of the solution was 8.0 ± 0.5 which correspond to the pH of the surface
water streams downstream of studied municipal sewage discharge.
Blank of 0.1% of municipal effluent wastewater was also exposed at
the same conditions. Analysis of samples in triplicate was performed
at the beginning of experiment, and at 13, 28, 42, 56 and 69 days.
 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267 261

After each sample collection, the reactor flask was bubbled 15 min with there were only 14% left of the initial concentration (Fig. 2) showing a
nitrogen gas to remove any remaining oxygen. Bacterial activity was calculated half-life of 0.40 days ± 0.05 (Table 4).
monitored on an agarose gel. Two major degradation products were clearly identified (Fig. 3). The
first was TP259 (C14H10ClNO2, [M + H]+: m/z 260.0473) which corre-
3. Results and discussion sponds to 8-chlorocarbazole-1-acetic acid. Its isotopic pattern clearly in-
dicates the presence of only one chlorine atom, which corresponds to a
3.1. Validation loss of chlorine and cyclization forming a five membered ring with ni-
trogen (Agüera et al., 2005; Li, 2012). The second degradation product
All validation results are given in Table 1. Linear range was evaluated was TP241 (C14H11NO3, [M + H]+: m/z 242.0812) which has an isotopic
from 150 to 10,000 ng/mL and calibration curves (concentration range: pattern rather informing of no chlorine, which corresponds to the loss of
150–10,000 ng/L) were linear (R2 N 0.995) for both DCF and SMX. The the second chlorine atom and the addition of a hydroxyl radical instead
lower limit of quantification (LLOQ) was valued at 150 ng/mL and was of chlorine (Table 2) (Agüera et al., 2005; Li, 2012).
reproducible for DCF and SMX with precision of 11% and 0.8% and an ac- Experimental isotopic abundances were also compared to those cal-
curacy of 112% and 70%, respectively. The higher limit of quantification culated from the chemical formula. Their accuracy was variable since
(HLOQ) was valued at 10,000 ng/mL to quantify the concentration of the mass spectra did not come from pure products, but still clearly indicates
spiked samples of degradation, with a precision of 2.9% and 0.8% and an the chlorine contribution. For example for TP259, the experimental iso-
accuracy of 102% and 107% respectively for DCF and SMX. The matrix topic abundance of the M + 1 peak (37.93%) is clearly biased, since the
blanks contained no peak of DCF or SMX. In summary, the method pre- theoretical abundance should be 15.71% and can be explained by co-
cision was b 15% for DCF and b 5% for SMX while accuracy values were elution of a compound of m/z 259 which contributes by a M + 2 to
between 98 and 112% for DCF and between 70 and 97% for SMX. The re- M + 1 of TP259. Likely molecular structures were confirmed by MS/
covery values were between 70 and 93% for DCF and between 83 and MS, which demonstrate that fragmentation products of both contain a
88% for SMX. There was b 3% of matrix effect for DCF and b 8% of matrix carboxylic group and an amino group. A common fragment of m/z 151
effect for SMX. was also observed which corresponds to the loss of the amino group
(Fig. 4).
3.2. Degradation results TP259 and TP241 were then degraded in turn and disappeared
completely in 4 days (Fig. 2). After 4 days of exposure, two new degra-
Photolysis is possible when molecules contain chromophore groups dation products were identified, TP133a and TP133b, C8H7NO
that absorb photons. However, depending on the medium pH and the ([M + H]+: m/z 134.06) (Fig. 3), having the same mass-to-charge
presence of auxochrome groups on the molecule, the photolysis ratio, but two different retention times so these products could be char-
mechanism may vary (Agüera et al., 2005). On the other hand, biodeg- acterized as isomers. The formation of TP133a at 5.33 min was clearly
radation mechanisms do not depend on the presence of specific chro- favored because its concentration increases much faster than his
mophore groups but rather on the presence of functional groups and counterpart at 7.06 min (Fig. 2). The concentration of both degradation
atoms needed for the bacterial metabolism, being either carbon, nitro- products significantly increased between 96 h and 11 days of exposure
gen, oxygen, sulfur, phosphorus, organohalides and iron (Diaz, 2004; (Fig. 2). Those degradation products were found by further interpreta-
O'Connor and Young, 1996). Therefore it can be expected that the tion of data after the experiment was finished, thus no MS/MS experi-
degradation mechanisms under various conditions differ and the degra- ments were done on TP133a and TP133b.
dation products obtained would be also different. Photodegradation has been reported to be the dominant degrada-
tion mechanism for DCF in the environment (Agüera et al., 2005; Yan
3.3. Photodegradation of DCF and SMX and Song, 2014). Effectively, since UV solar spectrum begins at approx-
imately 300 nm, and that maximum UV adsorption of DCF is at 273 nm,
Pharmaceuticals contained in effluents and discharged in receiving and tails up to approximately 320 nm at different pH (Agüera et al.,
waters are likely exposed to sunlight. To determine the influence there- 2005), it is clear that solar radiation absorption by DCF is possible and
of on the degradation of SMX and DCF, light exposure using artificial photodegradation can take place. Among the photodegradation prod-
light matching the solar radiation was reproduced in laboratory. Our ex- ucts of DFC, 8-chlorocarbazole-1-acetic acid was first reported (Yan
periments being done at pH 8, DCF (pKa = 4.15) and SMX (pKa1 = 1.6, and Song, 2014) and observed in this study as well.
pKa2 = 5.7) were initially found in their anionic form, influencing In contrast to DCF, SMX (C10H11N3O3S, [M + H]+: m/z 254.0594)
their degradation based on varying absorbance of chromophores photolysis was much slower. After 11 days of light exposure, there
and auxochromes (Boreen et al., 2004).The photolysis of DCF was still 84% left of the initial concentration of SMX (Fig. 5) showing a
(C14H11Cl2NO2, [M + H]+: m/z 296.0240) was very fast: within 24 h, calculated half-life of 54 ± 41 days (Table 4). Two major degradation
products were observed after 24 h: TP253 (C10H11N3SO3, [M + H]+:
m/z 254.0594) and TP238 (C10H10N2O3S, [M + H]+: m/z 239.0485).
Table 1
Validation tests results. The transformation product TP253 is an SMX isomer where the
isoxazole ring is photoisomerized (Fig. 6) (Periša et al., 2013; Trovó
Validation test Diclofenac Sulfamethoxazole
et al., 2009). The degradation product TP238 indicates a loss of NH,
Linearity (concentration) 150–10,000 (ng/mL) 150–10,000 (ng/mL) named N-(5-methylisozaxol-3-yl)benzenesulfonamide (Bonvin et al.,
Linearity R2 N0.995 N0.995 2013) and was confirmed by MSMS fragmentation which reveals func-
LLOQ (precision) 11% 1%
LLOQ (accuracy) 112% 70%
tional group phenyl (77.0391 u), NO (29.9980 u), CO (27.9949 u) and
HLOQ (precision) 3% 1% NH (15.0109 u). Comparisons of calculated and experimental isotopic
HLOQ (accuracy) 102% 107% abundances were very similar and suggested the presence of sulfur in
Selectivity (% of LLOQ) 0% 0% all degradation products (Table 3). The concentration of TP253 declined
Precision (3 QC low-mid-high) b15% b5%
between its appearance and the 11th day of exposure while that of
Accuracy (3 QC low-mid-high) 98–112% 70–97%
Extraction efficacy (3 QC 76–91% 89–95% TP238 increased between 24 and 48 h and then decreased on day 11,
low-mid-high) suggesting that it began to degrade as well (Fig. 5). After 11 days of ex-
Matrix effect ((3 QC low-mid-high); 97.1% ± 8% 92.5% ± 2% posure, a third degradation product clearly appeared, C10H10N2O4S
moyenne) ([M + H]+: m/z 255.0434), named TP254, which has a mass one unit
Recovery (3 QC low-mid-high) 70–93% 83–88%
higher than SMX and corresponds to a loss of NH and an addition of O.
262 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267

Fig. 2. Degradation kinetics of diclofenac with transformation products (TPs); A: photolysis; B: aerobic biodegradation; C: anaerobic biodegradation. Note: Total length of error bars
represents two standard deviations of triplicates.

Calculated and experimental isotopic abundances of M + 1 has a signif-
icant difference (Δ = 2.11%), but the M + 2 peak of the SMX ion con-
tributed to M + 1 of TP254 as there was co-elution of SMX and TP254.
A photodegradation pathway of SMX has been reported in deionized
water and includes three main pathways: hydroxylation of the aromatic
ring, cleavage of the bond between the sulfur and the nitrogen and a
fragmentation and rearrangement of the isoxazole ring (Niu et al.,
2013). Such degradation products, however, were not observed under
our experimental conditions. Another photolysis degradation pathway
of SMX in distilled water has been proposed (Trovó et al., 2009), and

Fig. 3. Transformation products (TPs) for Diclofenac; A: photolysis; B: aerobic biodegradation.

 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267 263

Table 2
Diclofenac and its transformation/degradation products (TPxxx) properties.

Drug/ m/z RT Exact Mass Chemical Insat Exp. Iso. Exp. Iso. Calc. Iso. Calc. Iso. Difference Degradation process
(TPxxx) [M + H]+ (min) mass accuracy formula Abund. Abund. Abund. Abund.
hν Aero Anae
Δm [M + 1] [M + 2] [M + 1] [M + 2]
biodeg biodeg
(mmu) (%) (%) (%) (%)

DCF 296.0240 9.144 295.0167 −0.0005 C14H11Cl2NO2 9 16.0608 63.0177 15.72 65.56 N/A N/A
TP259 260.0473 7.727 259.0397 −0.0002 C14H10ClNO2 10 37.93 32.9605 15.71 33.56 Cl loss x
TP133b 134.0598 7.064 133.0527 −0.0007 C8H7NO 6 10.5357 negligible 9.15 0.58 Phenyl loss x
TP240 242.0812 6.044 241.0736 −0.0002 C14H11NO3 10 17.1674 4.6753 15.76 1.78 Second Cl loss and x
addition of OH
TP133a 134.0602 5.328 133.0527 −0.0003 C8H7NO 6 9.8536 negligible 9.15 0.58 Phenyl loss x
TP265 266.0134 6.64 265.0059 −0.0003 C13H9Cl2NO 9 16.7638 61.6618 14.58 65.18 CH2O loss x
TP311 312.0189 6.65 311.0108 0.0002 C14H11Cl2NO3 9 18.6916 62.6168 15.76 65.77 O2 addition x
TP294 295.0161 7.1 294.0063 0.0019 C14H10Cl2NO2 9 17.2093 64.1860 15.71 65.55 H loss x x

Note 1: Isotopic abundance M + 1 of m/z 260 is approximative since there is co-elution with m/z 259 (interference with chlorine) which contribute by a M + 2 to M + 1 of 260. Isotopic
abundance of m/z 242 could be distorted by M + 1 of m/z 241. Isotopic ratios are calculated on the average spectra of chromatographic peak.
Note 2: Aerobic biodegradation: Abundances of m/z 266, 295 and 312 are low and isotopic ratios are distorted or there is contamination since products aren’t pure and the MS scan is a
superposition of all elution at the same retention time. Also, m/z 266 and 312 coelute. Isotopic ratios are calculated on the average spectra of chromatographic peak.
Abbreviation:
Insat.: Insaturations
hν: Photolysis
Aero biodeg: Aerobic biodegradation
Anae biodeg : Anaerobic biodegradation
Exp. Iso. Abund. : Experimental isotopic abundance
Calc. Iso. Abund.: Calculated isotopic abundance

only one degradation product of their nine proposed ones was observed
(TP254) which is the result of photo-isomerization reaction of the
isoxazole ring leading to a tautomers of SMX (Periša et al., 2013). It
has been reported that the protonated state of the five membered ring
of SMX is the most photo-reactive (Yan and Song, 2014). Therefore
sulfanilic acid (C6H7NO3S) would be the main photodegradation prod-
uct. Since the five membered ring is not protonated at pH 8, that degra-
dation product was not observed. As a result, SMX was less susceptible
to degradation in its anionic form at natural water pH, and under multi-
ple wavelength photolysis condition. Irradiation with photolysis setup
showed only one major degradation product with m/z 192.1131, as
reported by others (Batchu et al., 2014), in relation with rather a loss
of sulfur dioxide from SMX reported in previous study (Boreen et al.,
2005).
3.4. Aerobic biodegradation of DCF and SMX
Pharmaceuticals contained in effluent receiving waters are also
exposed to bacteria contained in effluents and any other bacteria in
the waterways. In order to reproduce as faithfully as possible biodegra-
dation, DCF and SMX were exposed to 0.1% (v/v) municipal effluents in

Fig. 4. MSMS spectra of TP241 and TP259; notice common fragment at m/z 151.
264 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267
Fig. 5. Degradation kinetics of sulfamethoxazole with transformation products (TPs); A: photolysis; B: aerobic biodegradation; C: anaerobic biodegradation. Note: Total length of error bars
represents two standard deviations of triplicates.
water, either under aerobic and anaerobic conditions, without light
energy.
DCF biodegradation was much slower compared to photolytic deg-
radation and the transformation products were different. In contrast
to photodegradation experiments, no significant biodegradation was
observed before 24 h. But after 14 days, three transformation products
were observed: TP311 (C14H11Cl2NO3, [M + H]+: m/z 312.0186),
TP265 (C13H9Cl2NO, [M + H]+: m/z 266.0138) and m/z 295
(C14H10Cl2NO2, [M + H]+: m/z 295.0141) which is actually a fragment
formed in the electrospray source resulting from the collision-induced
dissociation of a transformation product of mass 324 u (TP324). This
transformation product contains a nitroso group on the carboxylic
acid moiety of DCF, as described previously (Osorio et al., 2014; Pérez
and Barceló, 2008). They persisted up to 57 days of exposure while
the concentration of DCF decreases by 42% only (Fig. 2), showing a
calculated half-life of 70 ± 14 days (Table 4). TP311 could correspond
to the hydroxylation of the phenyl containing no chlorine (Fig. 3), as
described by (Huguet et al., 2013) during a chemical oxidation by
manganese oxide. TP311 corresponds to a previously reported human
metabolite (Wiesenberg-Boettcher et al., 1991), since some bacteria in
aerobic conditions can metabolize the drug by a redox reaction with
their cytochrome P450 (Prior et al., 2010). The mass spectrum of
TP311 demonstrates an isotopic pattern corresponding to two Cl
atoms while a MS/MS fragmentation demonstrated a loss of H2O
(18.0106 u), CO (27.9949 u), Cl (34.9688 u), then a second Cl
(34.9688 u) and a second CO (27.9949 u), suggesting that TP311 is con-
sistent with the proposed molecular formula. TP265 could correspond
to a decarboxylation (Fig. 3) as shown by (Huguet et al., 2013) during
a chemical oxidation by manganese oxide. The mass spectrum demon-
strates an isotopic pattern corresponding to two chlorines and a MS/
MS fragmentation that shows a loss of HCl (35.9767 u), a second chlo-
rine (34.9688 u), a CO (27.9949 u) and a group C2H3 (27.0235 u), sug-
gest that the structure is consistent with the decarboxylation. TP311
and TP265 co-eluted and one would think that TP265 was actually a
fragment of TP311, but their concentrations varied differently so one
can conclude that these are two different molecules. TP324 is a microbi-
al metabolite (Osorio et al., 2014; Pérez and Barceló, 2008). Its mass
spectrum demonstrate that it contains two chlorine and MSMS frag-
mentation reveals loss of H2O (18.0106 u), Cl (34.9688 u), CO
(27.9949 u) and a second Cl (34.9688 u). The isotopic patterns indicate
the presence of two chlorines in all the degradation products observed
(Table 2). The concentration of TP311 gradually increases up to
57 days of exposure. Concentrations of TP265 and TP324, in turn, re-
main rather stable between 14 days and 57 days of exposure (Table 2).
Biodegradation of SMX under aerobic conditions was also different
from degradation by photolysis. SMX concentration decreased by only
33% in 58 days (Fig. 5), showing a calculated half-life of 80 ± 11 days
(Table 4). Two transformation products stood out: the 29th day,
 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267 265

Fig. 6. Transformation products (TPs) for sulfamethoxazole; A: photolysis; B: aerobic biodegradation.

TP269 (C10H11N3SO4, [M + H]+: m/z 270.0537) appeared and remained formula of TP269 contains one more oxygen than SMX and could corre-
stable until the end of the experiment, as well as a product common spond to the hydroxylamine SMX (Fig. 6), a SMX metabolite, as shown
with photodegradation, TP238 (C10H10N2O3S, [M + H]+: m/z by (Niu et al., 2013; Trovó et al., 2009). Under different conditions (ad-
239.0485), which also appeared on the 29th day (Fig. 5). The chemical dition of glucose in a bacterial medium of Rhodococcus equi (R. equi), a

Table 3
SMX and its transformation/degradation products (TPxxx) properties.

Drug/ m/z RT Exact Mass Chemical Insat Exp. Iso. Exp. Iso. Calc Iso. Calc Iso. Difference Degradation process
(TPxxx) [M + H] (min) mass accuracy formula Abund. Abund. Abund. Abund.
hν Aero Anae
Δm [M + 1] [M + 2] [M + 1] [M + 2]
biodeg biodeg
(mmu) (%) (%) (%) (%)

SMX 254.0594 5.026 253.0521 −0.0005 C10H11N3SO3 9 12.1568 5.2157 12.95 5.87 N/A
TP253 254.0594 2.67 253.0521 −0.0005 C10H11N3O3S 9 13.6887 5.7637 12.95 5.87 Structural isomer of 5 x
membered ring
TP238 239.0485 5.887 238.0407 −0.00002 C10H10N2O3S 7 12.2251 5.2356 12.58 5.82 NH loss, 5 membered ring x x x
opening
TP254 255.0434 5.21 254.0368 −0.0012 C10H10N2O4S 7 14.7169 5.6604 12.61 6.03 NH loss, O2 addition, 5 x
membered ring opening
TP269 270.0543 2.99 269.0470 −0.0005 C10H11N3SO4 9 13.2867 4.4755 12.99 6.08 O2 addition x

Note 3: Isotopic abundance M + 1 of m/z 255 is approximative since there is co-elution with SMX which contribute to M + 1 of 255 with its M + 2. Moreover, isotopic abundance M + 1 of
m/z 254 contribute to isotopic abundance of M+ of m/z 255. Isotopic ratios are calculated on the average spectra of chromatographic peak.
Note 4: m/z 270 abundance is low and isotopic ratios are distorted or there is contamination since products aren’t pure and the MS scan is a superposition of all elution at the same re-
tention time. Isotopic ratios are calculated on the average spectra of chromatographic peak.
Abbreviation
Insat.: Insaturations
hν: Photolysis
Aero biodeg: Aerobic biodegradation
Anae biodeg : Anaerobic biodegradation
Exp. Iso. Abund. : Experimental isotopic abundance
Calc. Iso. Abund.: Calculated isotopic abundance
266 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267
Table 4
DCF and SMX half-life.
DCF SMX
k t 1/2 error k t 1/2
Photolysis 1.73 0.40 0.05 0.013 54
Aerobic biodegradation 0.017 70 14 0.014 80
Anaerobic biodegradation ND ND
Note 1: k = constant rate
Note 2: t ½ = half-life = ln(2)/k
Note 3: since the concentration of DCF and SMX didn’t significantly decreased in the ex-
periment of anaerobic biodegradation, the calculation of half-life was not possible.
different degradation metabolite of SMX, an acetylated aromatic
amine of SMX, was found (Larcher and Yargeau, 2011). TP238 [N-
(5-methylisozaxol-3-yl)benzenesulfonamide] (Bonvin et al., 2013)
indicates a loss of NH, and its presence was confirmed by MSMS frag-
mentation which reveals functional group phenyl (77.0391 u), NO
(29.9980 u), CO (27.9949 u) and NH (15.0109 u). Given the low sig-
nal observed for TP269, MS/MS experiments were ineffective. How-
ever, the calculated and experimental isotopic abundances of both
products are consistent and suggest the presence of sulfur and oxy-
gen (Table 3). The experimental isotopic abundance of M + 1 of
TP269 is very low and could be biased by the background noise.
The concentration of TP269 was unfortunately below LOQ. The con-
centration of TP238 was sufficient to be estimated from the 29th
day then decreased significantly until the 57th day (Table 3). The
concentration of TP238 was however slightly lower than in the pho-
tolysis experiment.
3.5. Anaerobic biodegradation of DCF and SMX
Pharmaceuticals in rivers can also be found in an anaerobic environ-
ment, if they are adsorbed on suspended particles which settle thereaf-
ter in sediments. An experiment in an anaerobic environment without
light energy was therefore performed in a matrix containing 0.1% of mu-
nicipal effluent under N2-purged atmosphere. Under these conditions,
only one transformation product was detected for DCF: TP324
(C10H10Cl2NO2, [M + H]+: m/z 295.01601) (Table 2), common with
the aerobic experiment. Its concentration, however, was lower and un-
stable under these conditions. Moreover, its absence was noticed at
55 days of exposure (Fig. 2), suggesting that the concentration was
below detection limit, either by low efficiency and reproducibility of
the extraction step or MS signal suppression in this experiment. DCF
concentration was not significantly reduced (1%) in 70 days of exposure
(Fig. 2). However, biodegradation of DCF in anaerobic sludge digestion
medium has been previously reported (Carballa et al., 2007) with a
rate of 69% degradation.
SMX biodegradation under anaerobic conditions revealed only one
transformation product, TP238 (C10H10N2O3S, [M + H]+: m/z
239.0485) (Table 3), common to the experiments of aerobic biodegra-
dation and photodegradation. Its concentration was, however, lower
than under other degradation conditions, but stable from the 27th day
of exposure until the end, after 70 days (Fig. 5). The concentration of
SMX was slightly reduced (5%) after 70 days of exposure. SMX biodeg-
radation in anaerobic sludge digestion medium in treatment plants was,
on the other hand, reported (Carballa et al., 2007) with a high degrada-
tion rate of 99%.
3.6. Comparisons among degradation processes
Various degradation processes - photodegradation, biodegradation
in aerobic and anaerobic conditions - were investigated with the aim
to evaluate the influence of environmental factors on the fate of two
pharmaceutical of different physicochemical properties. DCF and SMX
were selected as they are found in large amounts in the dispersion
plume of municipal effluents. In analyzing the results of degradation,
it is noted that the degradation products differ depending on the
process.
Kinetics of environmental degradation are clearly dependent on the
error
nature of the drug and the process involved in the degradation. DCF was
much more sensitive to photolysis than SMX with a half-life of 0.40 ±
41
0.05 days compared to 54 ± 41 days for SMX. On the other hand, bio-
11
degradation was similar for DCF and SMX, with half-life of respectively
70 ± 14 days and 80 ± 11 days in aerobic condition (Table 4). However,
the aerobic biodegradation has begun only after 29 days, and this could
be explained by the fact that the bacterial population has exponentially
grown the first 29 days and became significantly active at that moment.
So the calculation of t1/2 was done on the kinetic after 29 days, and on
the value obtained for t1/2, 29 days was added. In anaerobic condition,
the biodegradation was not observed and the concentration did not
decrease significantly over the experiment period to allow half-life
calculation.
Photodegradation of DCF demonstrated the loss of a Cl (TP259)
followed by a second loss of Cl, the addition of a hydroxyl (TP241),
and finally the loss of a phenyl (TP133a and TP133b). Both TP259 and
TP241, resulting from the loss of chlorine, were not persistent as in
turn they completely disappeared after three days of exposure. In con-
trast, the two following degradation products, TP133a and TP133b,
persisted after eleven days of exposure. Biodegradation of DCF under
aerobic conditions showed a very different degradation mechanism
than photodegradation. Indeed, Cl atoms were not involved in the
degradation, but rather O atoms, inducing redox mechanisms such as
hydroxylation (TP311), decarboxylation (TP265) and a nitrosation
(TP324). The DCF-hydroxylated product (TP311) was persistent and
with its concentration increasing beyond 57 days of exposure, it could
be identified as a potential substance to monitor. The other two biodeg-
radation products, TP311 and TP324, were also persistent, but their con-
centration remained rather stable between 14 days and 57 days of
exposure. With regard to biodegradation under anaerobic conditions,
the only observed biodegradation product was common with the aero-
bic environment, TP324, and its concentration decreased with time,
thus no persistent product were identified under these conditions.
Photodegradation of SMX demonstrated isomerization of isoxazole
ring (TP253) as well as the loss of an amine group (TP238), followed
by hydroxylation (TP254). TP253 appeared in very small proportion
compared to the initial concentration of SMX. In contrast, TP238 ap-
peared in much larger concentration reaching a concentration peak
after 3 days of exposure. Given its relative persistence, this degradation
product would be recommended for monitoring. TP254 appeared only
after 11 days of exposure, in relatively large amounts, but its stability
has not been determined. However, no isomerization was observed in
the biodegradation under aerobic conditions, but rather the loss of an
amine group (TP238) and the hydroxylation of SMX (TP269). TP238
concentration peaked after 29 days of exposure, but its concentration
then went down and stayed stable after 57 days of exposure. With re-
spect to biodegradation under anaerobic conditions, the only observed
mechanism is the loss of an amine group (TP238) whose concentration
was the lowest from the three processes. The mechanism leading to the
loss of an amine group could be suggested as common to all SMX degra-
dation processes. TP238 concentration, however, varied widely depend-
ing on the conditions of degradation and its formation was favored
mostly by photodegradation.
4. Conclusion
This study allowed to distinguish different transformation products
from three processes that can be found in the aquatic environment:
photodegradation and biodegradation in aerobic and anaerobic condi-
tions. However, real environmental conditions imply that the various
degradation processes occur simultaneously or in chain, depending on
the weather and patterns of climate variability influencing watersheds
dynamic, on the intensity of solar light and on the composition or
 S. Poirier-Larabie et al. / Science of the Total Environment 557–558 (2016) 257–267 267

properties (e.g., pH, transmittance) of the aquatic ecosystem. The trans- Hartmann, J., Bartels, P., Mau, U., Witter, M., Wv, Tümpling, Hofmann, J., et al., 2008. Deg-
radation of the drug diclofenac in water by sonolysis in presence of catalysts.
formation products can be different from those found in this study, since Chemosphere 70, 453–461.
photolysis and biodegradation can occur simultaneously, and can also Hernández, F., Sancho, J.V., Ibáñez, M., Guerrero, C., 2007. Antibiotic residue determina-
be influenced by photosensitizers that are naturally occurring as report- tion in environmental waters by LC-MS. TrAC Trends Anal. Chem. 26, 466–485.
Hollender, J., Singer, H., Hernando, D., Kosjek, T., Heath, E., 2010. The challenge of the
ed by (Yan and Song, 2014). identification and quantification of transformation products in the aquatic envi-
Further experiments by combining the degradation processes to ronment using high resolution mass spectrometry. In: Fatta-Kassinos, D., Bester,
compare resulting degradation products would be warranted for a K., Kümmerer, K. (Eds.), Xenobiotics in the Urban Water Cycle 16. Springer
Netherlands, pp. 195–211.
better understanding of what is actually found in the environment.
Huguet, M., Deborde, M., Papot, S., Gallard, H., 2013. Oxidative decarboxylation of
However, given the complexity of results interpretation as well as the diclofenac by manganese oxide bed filter. Water Res. 47, 5400–5408.
identification of the degradation products structures, it would also be Jiskra, M., 2008. Fate of the pharmaceutical diclofenac in the aquatic environment. Bio-
geochemistry and Pollutant Dynamics (Term paper 21).
advantageous to combine several analytical methods such as infrared
Kosjek, T., Heath, E., 2008. Applications of mass spectrometry to identifying pharmaceuti-
spectroscopy and nuclear magnetic resonance. These methods, howev- cal transformation products in water treatment. TrAC Trends Anal. Chem. 27,
er, require the use of pure products which is an additional challenge 807–820.
given the complexity of degraded samples. Finally, other pharmaceuti- Langenhoff, A., Inderfurth, N., Veuskens, T., Schraa, G., Blokland, M., Kujawa-Roeleveld, K.,
et al., 2013. Microbial removal of the pharmaceutical compounds ibuprofen and
cal products with different properties should be considered to have a diclofenac from wastewater. BioMed Research International 2013, 9.
more comprehensive perspective of transformation patterns. Larcher, S., Yargeau, V., 2011. Biodegradation of sulfamethoxazole by individual and
mixed bacteria. Appl. Microbiol. Biotechnol. 91-1, 211–218.
Li, M., 2012. Organic chemistry of drug degradation. The Royal Society of Chemistry, Drug
Acknowledgments Discovery Series 29 P001-287.
Nasuhoglu, D., Yargeau, V., Berk, D., 2011. Photo-removal of sulfamethoxazole (SMX) by
We are grateful to Arielle Ariste, Charles Faille and Marc-André photolytic and photocatalytic processes in a batch reactor under UV-C radiation
(λmax = 254 nm). J. Hazard. Mater. 186, 67–75.
Lecours for their help in the laboratory. This work was funded by the Niu, J., Zhang, L., Li, Y., Zhao, J., Lv, S., Xiao, K., 2013. Effects of environmental factors on sul-
St. Lawrence Action Plan. famethoxazole photodegradation under simulated sunlight irradiation: kinetics and
mechanism. J. Environ. Sci. 25, 1098–1106.
O'Connor, O.A., Young, L.Y., 1996. Effects of six different functional groups and their posi-
References
tion on the bacterial metabolism of monosubstituted phenols under anaerobic condi-
tions. Environ. Sci. Technol. 30, 1419–1428.
Agüera, A., Pérez Estrada, L.A., Ferrer, I., Thurman, E.M., Malato, S., Fernández-Alba, A.R.,
Osorio, V., Imbert-Bouchard, M., Zonja, B., Abad, J.-L., Pérez, S., Barceló, D., 2014. Simulta-
2005. Application of time-of-flight mass spectrometry to the analysis of
neous determination of diclofenac, its human metabolites and microbial nitration/
phototransformation products of diclofenac in water under natural sunlight. J. Mass
nitrosation transformation products in wastewaters by liquid chromatography/
Spectrom. 40, 908–915.
quadrupole-linear ion trap mass spectrometry. J. Chromatogr. A 1347, 63–71.
Aminot, Y., 2013. Etude de l'impact des effluents urbains sur la qualité des eaux de la Ga-
Pérez, S., Barceló, D., 2008. First evidence for occurrence of hydroxylated human metabo-
ronne estuarienne: application aux composés pharmaceutiques et aux filtres UV.
lites of diclofenac and aceclofenac in wastewater using QqLIT-MS and QqTOF-MS.
Baker, D.R., Kasprzyk-Hordern, B., 2011. Critical evaluation of methodology commonly
Anal. Chem. 80, 8135–8145.
used in sample collection, storage and preparation for the analysis of pharmaceuticals
Periša, M., Babić, S., Škorić, I., Frömel, T., Knepper, T., 2013. Photodegradation of sulfon-
and illicit drugs in surface water and wastewater by solid phase extraction and liquid
amides and their N 4-acetylated metabolites in water by simulated sunlight irradia-
chromatography–mass spectrometry. J. Chromatogr. A 1218, 8036–8059.
tion: kinetics and identification of photoproducts. Environ. Sci. Pollut. Res. 20,
Bartels, P., von Tümpling Jr, W., 2007. Solar radiation influence on the decomposition pro-
8934–8946.
cess of diclofenac in surface waters. Sci. Total Environ. 374, 143–155.
Prior, J., Shokati, T., Christians, U., Gill, R., 2010. Identification and characterization of a
Batchu, S.R., Panditi, V.R., O'Shea, K.E., Gardinali, P.R., 2014. Photodegradation of antibi-
bacterial cytochrome P450 for the metabolism of diclofenac. Appl. Microbiol.
otics under simulated solar radiation: implications for their environmental fate. Sci.
Biotechnol. 85, 625–633.
Total Environ. 470–471, 299–310.
Salgado, R., Pereira, V.J., Carvalho, G., Soeiro, R., Gaffney, V., Almeida, C., et al., 2013.
Bonvin, F., Omlin, J., Rutler, R., Schweizer, W.B., Alaimo, P.J., Strathmann, T.J., et al., 2013.
Photodegradation kinetics and transformation products of ketoprofen, diclofenac
Direct photolysis of human metabolites of the antibiotic sulfamethoxazole: evidence
and atenolol in pure water and treated wastewater. J. Hazard. Mater. 244-245,
for abiotic back-transformation. Environ. Sci. Technol. 47, 6746–6755.
516–527.
Boreen, A., Arnold, W., McNeill, K., 2003. Photodegradation of pharmaceuticals in the
Schulze, T., Weiss, S., Schymanski, E., von der Ohe, P.C., Schmitt-Jansen, M., Altenburger,
aquatic environment: a review. Aquat. Sci. 65, 320–341.
R., et al., 2010. Identification of a phytotoxic photo-transformation product of
Boreen, A.L., Arnold, W.A., McNeill, K., 2004. Photochemical fate of sulfa drugs in the
diclofenac using effect-directed analysis. Environ. Pollut. 158, 1461–1466.
aquatic environment: sulfa drugs containing five-membered heterocyclic groups. En-
Segura, P.A., Garcia-Ac, A., Lajeunesse, A., Ghosh, D., Gagnon, C., Sauve, S., 2007. Determi-
viron. Sci. Technol. 38, 3933–3940.
nation of six anti-infectives in wastewater using tandem solid-phase extraction and
Boreen, A.L., Arnold, W.A., McNeill, K., 2005. Triplet-sensitized photodegradation of sulfa
liquid chromatography-tandem mass spectrometry. J. Environ. Monit. 9, 307–313.
drugs containing six-membered heterocyclic groups: identification of an SO2 extru-
Segura, P.A., François, M., Gagnon, C., Sauvé, S., 2009. Review of the occurrence of anti-
sion photoproduct. Environ. Sci. Technol. 39, 3630–3638.
infectives in contaminated wastewaters and natural and drinking waters. Environ.
Carballa, M., Omil, F., Ternes, T., Lema, J.M., 2007. Fate of pharmaceutical and personal care
Health Perspect. 117, 675–684.
products (PPCPs) during anaerobic digestion of sewage sludge. Water Res. 41-10,
Segura, P.A., Clair, A., Lecours, M.A., Saadi, K., Yargeau, V., 2015. Application of XCMS On-
2139–2150.
line and toxicity bioassays to the study of transformation products of levofloxacin.
Challis, J.K., Hanson, M.L., Friesen, K.J., Wong, C.S., 2014. A critical assessment of the
Water Sci. Technol. 72, 1578–1587.
photodegradation of pharmaceuticals in aquatic environments: defining our current
Trovó, A.G., Nogueira, R.F.P., Agüera, A., Sirtori, C., Fernández-Alba, A.R., 2009.
understanding and identifying knowledge gaps. Environ. Sci. Processes Impacts 16, 672.
Photodegradation of sulfamethoxazole in various aqueous media: persistence, toxic-
Diaz, E., 2004. Bacterial degradation of aromatic pollutants: a paradigm of metabolic ver-
ity and photoproducts assessment. Chemosphere 77, 1292–1298.
satility. Int. Microbiol. 7, 173–180.
Wiesenberg-Boettcher, I., Pfeilschifter, J., Schweizer, A., Sallmann, A., Wenk, P., 1991.
Fatta, D., Achilleos, A., Nikolaou, A., Meriç, S., 2007. Analytical methods for tracing phar-
Pharmacological properties of five diclofenac metabolites identified in human plas-
maceutical residues in water and wastewater. TrAC Trends Anal. Chem. 26, 515–533.
ma. Agents Actions 34, 135–137.
Fatta-Kassinos, D., Meric, S., Nikolaou, A., 2011a. Pharmaceutical residues in environmen-
Yan, S., Song, W., 2014. Photo-transformation of pharmaceutically active compounds in
tal waters and wastewater: current state of knowledge and future research. Anal.
the aqueous environment: a review. Environmental Science: Processes & Impacts
Bioanal. Chem. 399, 251–275.
16, 697–720.
Fatta-Kassinos, D., Vasquez, M.I., Kümmerer, K., 2011b. Transformation products of phar-
Zhang, Y., Geißen, S.-U., Gal, C., 2008. Carbamazepine and diclofenac: removal in
maceuticals in surface waters and wastewater formed during photolysis and ad-
wastewater treatment plants and occurrence in water bodies. Chemosphere 73,
vanced oxidation processes – degradation, elucidation of byproducts and
1151–1161.
assessment of their biological potency. Chemosphere 85, 693–709.
Ziylan, A., Ince, N.H., 2011. The occurrence and fate of anti-inflammatory and analgesic
Gagnon, C., Lajeunesse, A., 2012. Low removal of acidic and hydrophilic pharmaceutical
pharmaceuticals in sewage and fresh water: treatability by conventional and non-
products by various types of municipal wastewater treatment plants. Journal of Xe-
conventional processes. J. Hazard. Mater. 187, 24–36.
nobiotics 2 (1), 13–17.
Ziylan, A., Dogan, S., Agopcan, S., Kidak, R., Aviyente, V., Ince, N., 2014. Sonochemical deg-
Gonzalez, O., Sans, C., Esplugas, S., Malato, S., 2009. Application of solar advanced oxida-
radation of diclofenac: byproduct assessment, reaction mechanisms and environ-
tion processes to the degradation of the antibiotic sulfamethoxazole. Photochem.
mental considerations. Environ. Sci. Pollut. Res. 21, 5929–5939.
Photobiol. Sci. 8, 1032–1039.
Guerra, P., Kim, M., Shah, A., Alaee, M., Smyth, S.A., 2014. Occurrence and fate of antibiotic,
analgesic/anti-inflammatory, and antifungal compounds in five wastewater treat-
ment processes. Sci. Total Environ. 473–474, 235–243.