TiO2 photocatalysis applied to the degradation and antimicrobial activity removal of oxacillin: Evaluation of matrix components, experimental parameters, degradation pathways and identification of organics by-products

Article abstract of DOI:10.1016/j.jphotochem.2015.06.021

The TiO₂ photocatalytic degradation of oxacillin (OXA) in synthetic and natural waters was studied. The matrix effects, in terms of antibiotic and antimicrobial activity removal, were evaluated in the presence of iron ions, natural mineral water and additives contained in commercial formulations of the antibiotic. A slight improvement in degradation was observed in the presence of iron ions. On the other hand, the presence of excipients in a commercial formulation or inorganic ions in natural mineral water slightly inhibited the efficiency of the system. An experimental design using pH, catalyst load and light intensity as variables was also evaluated. The best performances were achieved at natural pH (～6.0) using 2.0 g L^-1 of TiO₂ with 150 W of applied power. The evaluation of OXA concentration indicated that the photodegradation process showed a Langmiur-Hinshewoold kinetic model. The extent of the process was evaluated following the evolution of chemical oxygen demand (COD) and dissolved organic carbon (DOC). Total removal of both, the antibiotic and its antimicrobial activity, was achieved after 120 min; while 100% of mineralization was observed within 480 min of treatment. Finally, five by-products were identified, the degradation routes were elucidated and a schema of the antibiotic degradation was proposed.

Full text of DOI:10.1016/j.jphotochem.2015.06.021

Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
Contents lists available at ScienceDirect
Journal of Photochemistry and Photobiology A:
Chemistry
journal homepage: www.elsevier.com/locate/jphotochem
TiO2 photocatalysis applied to the degradation and antimicrobial
activity removal of oxacillin: Evaluation of matrix components,
experimental parameters, degradation pathways and identification of
organics by-products
Ana L. Giraldo-Aguirrea, Edgar D. Erazo-Erazoa, Oscar A. Flórez-Acostaa,
Efraim A. Serna-Galvisb, Ricardo A. Torres-Palmab,
*
a
Grupo de Diseño y Formulación de Medicamentos, Cosméticos y Afines (DYFOMECO), Facultad de Ciencias Farmacéuticas y Alimentarias, Universidad de
Antioquia UdeA, Calle 70 No. 52-21, Medellín, Colombia
b
Grupo de Remediación Ambiental y Biocatálisis (GIRAB), Instituto de química, Facultad de Ciencias Exactas y Naturales, Universidad de Antioquia UdeA,
Calle 70 No. 52-21, Medellín, Colombia
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 3 March 2015
Received in revised form 16 June 2015
Accepted 22 June 2015
Available online 23 June 2015
The TiO2 photocatalytic degradation of oxacillin (OXA) in synthetic and natural waters was studied. The
matrix effects, in terms of antibiotic and antimicrobial activity removal, were evaluated in the presence of
iron ions, natural mineral water and additives contained in commercial formulations of the antibiotic. A
slight improvement in degradation was observed in the presence of iron ions. On the other hand, the
presence of excipients in a commercial formulation or inorganic ions in natural mineral water slightly
inhibited the efficiency of the system. An experimental design using pH, catalyst load and light intensity
as variables was also evaluated. The best performances were achieved at natural pH (ꢀ6.0) using 2.0 g Lꢁ1
of TiO2 with 150 W of applied power. The evaluation of OXA concentration indicated that the
photodegradation process showed a Langmiur–Hinshewoold kinetic model. The extent of the process
was evaluated following the evolution of chemical oxygen demand (COD) and dissolved organic carbon
(DOC). Total removal of both, the antibiotic and its antimicrobial activity, was achieved after 120 min;
while 100% of mineralization was observed within 480 min of treatment. Finally, five by-products were
identified, the degradation routes were elucidated and a schema of the antibiotic degradation was
proposed.
Keywords:
Oxacillin
Advanced oxidation process
Tio2 photocatalysis
Water treatment
Antibiotics
Antimicrobial activity
ã 2015 Elsevier B.V. All rights reserved.
1. Introduction
Antibiotics are one of the groups of pharmacologically active
substances, which have caused adverse environmental effects.
Active pharmaceutical compounds are designed to intervene in
the health–disease process and fulfill a biological function in living
organisms. Therefore, they are considered essential to preserve
public health and quality of life. However, like many foods and
supplements that are consumed by humans and animals,
pharmaceuticals are not completely absorbed or metabolized in
the body and are eventually excreted into the environment [1–3].
Although little is known of the potential effects of pharmaceuticals
in the environment, the problems of microbial resistance,
allergenicity and endocrine disruption are of greatest concern to
the scientific community [4–7].
Among them, resistance to antibiotics by microorganisms is a
serious and growing problem in contemporary medicine, to the
point of being considered one of the most eminent risks to public
health of the 21st century.
penicillin derivatives, cephalosporins, carbapenems, monobac-
tams, carbacephems, and -lactamase inhibitors, are examples of
b-lactam antibiotics, which include
b
antimicrobials that generate widespread bacteria resistance.
Oxacillin (OXA) an isoxazolyl penicillin, is a semisynthetic
pharmaceutic member of the group of b-lactam antibiotics, which
are resistant to acidic conditions and penicillinase. OXA is widely
used in clinic for the treatment of infections caused by aerobic
gram-positive cocci [8–10].
A high percentage of the antibiotic OXA is then excreted
unchanged into domestic wastewater. Eventually, OXA ends up in
wastewater treatment plants (WWTPs); and therefore, it is
* Corresponding author. Fax: +57 4 2195666.
E-mail address: ricardo.torres@udea.edu.co (R.A. Torres-Palma).
http://dx.doi.org/10.1016/j.jphotochem.2015.06.021
1010-6030/ã 2015 Elsevier B.V. All rights reserved.

96
A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
released into the environment. Indeed, OXA was found at the
concentration of 10 ng Lꢁ1 in surface waters [11]. Once in the
environment OXA spreads its bacterial resistance, which leads to
problems in the therapeutic effectiveness of this antibiotic. In fact,
the proliferation of methicillin-resistant Staphylococcus aureus
(MRSA) and borderline oxacillin-resistant S. aureus (BORSA) has
been reported [12]. As a consequence, efficient treatment systems
that eliminate the antibiotic and the antimicrobial activity of the
treated water need to be developed urgently.
hþVB þ OHꢁðadsorbedÞ ! ꢄHO
(3)
The holes can directly oxidize the adsorbed organic matter
(Eq. (4)).
þ
hVB þ organic matter ðOMÞ ! OMþ ! oxidized products
(4)
ꢄ
The treatment of water contaminated by antibiotics using
conventional technologies is not easy. The usual complex mixture
of chemicals in the effluents impedes the recovery of the organics.
Furthermore, the low calorific power of the waste preclude the use
of incineration, and the non-biodegradable character of antibiotics
exclude the use of classical biological treatments [13]. Therefore,
advanced oxidation processes (AOPs) would appear to be a very
promising solution for the environmental problems generated by
the discharge of these effluents [13].
Heterogeneous TiO2 photocatalysis is an AOP that has earned an
important place among the new emerging technologies for the
degradation of compounds such as antibiotics. The photocatalytic
process is initiated when a semiconductor absorb photons with an
Given its high efficiency, TiO2 photocatalysis has been amply
described as one of the most promising advanced oxidation
technologies for eliminating contaminants in low concentration
using artificial or natural light. In spite of this, to our knowledge-
ment, the application of the TiO2 photocatalysis for the removal of
oxacillin has not been reported.
In this study, the application of TiO2 photocatalysis for the
treatment and removal of OXA in waters is investigated. Special
focus is given to evaluate the effect of the water matrix
composition: ions in natural mineral water and excipients in a
commercial product of the antibiotic. Additionally, the influence of
the initial pH and antibiotic concentration are evaluated. The
effects of the most relevant operational parameters (TiO2 loading
and light intensity), the extend of degradation in terms of the
chemical oxygen demand (COD) and dissolved organic carbon
(DOC), and the changes in antimicrobial activity are also
investigated. Finally, the degradation routes are elucidated, the
primary organic by-products are identified and a degradation
schema for this antibiotic is depicted.
energy equal or superior to their band gap (
electrons from the valence band to the conduction band (Eq. (1)),
l
ꢂ 380 nm), promoting
and thus producing an electron–hole pair [14–16].
Semiconductor þ h ! eꢁCB þ hVþB
n
Pollutant elimination during TiO2 photocatalytic oxidation,
2. Material and methods
mainly ascribed to strong oxidative reactions at the hole ðhþVBÞ, is
due to the semiconductor high oxidative power (Eꢃ = +2.7 V vs NHE)
[17]. Thus, the vacancies generated in the valence band can be
trapped on the catalyst surface, which promote the splitting of
adsorbed water molecules or hydroxide anions to produce
hydroxyl radicals (ꢄOH) (Eqs. (2) and (3)). These hydroxyl radicals
are considered responsible for the oxidation of organic matter due
to their high oxidation capability (Eꢃ = +2.8 V vs NHE) [15].
2.1. Reagents
Oxacillin sodium purchased from Sigma–Aldrich had a 95% of
purity and was used without further purification. OXA sterile
powder for injection, from laboratory Blaskov Ltd., was used to
evaluate the effect of excipients in a real commercial product of the
antibiotic. Titanium dioxide Evonik P-25 was used for all experi-
ments. Sulfuric acid and sodium hydroxide were provided by
Merck and used for pH adjustment. The sample solution was
prepared using deionized water, which was obtained by passing
water through a Pureline deionization system. Milli-Q water,
hþVB þ H2O
! Hþ þ ꢄHO
(2)
ðadsorbedÞ
Table 1
Experimental results of the applied factorial design for oxacillin degradation by TiO2 photocatalysis.
Block
[TiO2]
Light intensity (W)
pH
Degradation percentage observed
Degradation percentage estimated
(g Lꢁ1
)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
2
150
90
30
30
150
30
150
150
30
90
150
90
9
6
9
9
3
3
9
3
3
6
9
6
9
9
3
3
9
3
3
6
61.52
53.8
33.24
14.68
20.38
7.18
47.71
64.82
37.31
52.71
62.83
56.33
34.26
16.48
18.42
9.64
64.99
39.33
41.70
13.67
36.96
13.67
36.96
64.99
41.70
39.33
65.04
39.38
41.75
13.72
37.02
13.72
37.02
65.04
41.75
39.38
1.025
2
0.05
0.05
0.05
0.05
2
2
1.025
2
1.025
2
0.05
0.05
0.05
0.05
2
30
30
150
30
150
150
30
36.99
64.09
37.62
57.22
2
1.025
90

A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
97
acetonitrile and methanol (Merck, HPLC grade) were used for the
preparation of the HPLC mobile phase. 2-Propanol provided by
Merck was used to evaluate the effect of the presence of organics in
the degradation efficiency. Characterized natural mineral water
was employed to investigate the antibiotic elimination in natural
waters.
factor was the degradation percentage of the antibiotic after
20 min of irradiation. The test concentration of OXA was 203 M.
m
The statistical analysis of the data was performed using the
software STATGRAPHICS CENTURION XVI v16.0.07. The most
favorable conditions, in the range of the evaluated variables,
was used to evaluate the reaction kinetic, the matrix effects, the
degradation pathway, the reduction or loss of antimicrobial
activity and the organic by-products. All experiments were carried
out at least in duplicated to confirm reproducibility and average
results were reported.
2.2. Photocatalytic reactor
Photocatalytic oxidation experiments were carried out using
100 mL of tested solution placed in beakers with constant stirring.
An aluminum closed reactor, containing 5 Phillips cylindrical black
lamps (maximum emission at 365 nm), which supplied between
30 and 150 W of power, was used as a reactor. The reactor
temperature was controlled using mechanical ventilation. Sol-
utions containing the pollutant were placed into the vessel reactor
and the system was conditioned in darkness until the adsorption-
desorption equilibrium was established. During degradation,
samples were taken at regular time intervals and filtered through
2.4. Chemicals analysis
Sampling was carried out at specific time intervals using 1.0 mL
of sample to perform the chromatographic readings. 1.0 mL of
sample was also used to take COD, antimicrobial activity, and DOC
readings.
The evolution of OXA concentration was monitored using a
Waters1 liquid chromatograph equipped with a 486 absorbance
(UV–vis) detector set at 225 nm. Isocratic separation was
cellulose nitrate filters of 0.45
HPLC–UV–VIS.
mm in diameter and analyzed by
performed at 25 ꢃC in a 250 ꢅ 4.6 mm ID RP18 5
mm reverse phase
HPLC column (Merck LiChrospher1). Optimum separation oc-
curred using a mixture of a phosphate buffer: acetonitrile:
methanol (64:27:9 v/v), operating at a flow-rate of 0.6 mL minꢁ1
2.3. Experimental design
A two-level full factorial design (23) with two replicate center
points and multivariate analysis (response surface methodology)
was used in order to find the optimal conditions of the process. The
replicate center points serve to evaluate the experimental error
and the curvature of the evolution of a response factor, that is
whether or not the evolution of the response factor is linear within
the experimental range studied. Using the experimental design the
effects of several parameters were evaluated. Three variables
were analyzed: pH in the range of 3–9, TiO2 concentration
(0.05–2.0 g Lꢁ1) and light intensity (30–150 W). The response
and using 20 mL as the injection volume. Under work conditions,
the accuracy of the method, in terms of the recovery (percent), was
always higher than 93%, while uncertainties were found to be
lower than 3.5%.
The chemical oxygen demand (COD) was determined using the
methodology established in the Standard Methods for the
Examination of Water and Wastewater (Standard Methods
5220), in accordance with the closed reflux titrimetric method.
To quantify the COD, the optical density required to measure the
change of color of the dichromate solution was determined at
Fig. 1. Optimization of TiO2 load, power light and pH for photocatalytic degradation of oxacillin (203 m
mol Lꢁ1) at natural pH after 20 min of irradiation. (a) Response surface
for TiO2 load vs power light. (b) Response surface for power light vs. pH. (c) Pareto diagram plot. All figures were obtained from the software STATGRAPHICS CENTURION XVI
v16.0.07 according to the experimental data presented in Table 1.

98
A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
445 nm with
a
Spectronic Genesys 2.0 spectrophotometer.
Meanwhile, the dissolved organic carbon (DOC) was analyzed
using a Shimadzu TOC 5000A analyzer in accordance with the
methodology reported previously [18]. Uncertainties were found
to be lower than 5 and 7% for figures reporting DOC and COD
evolution, respectively.
2.5. Biological analysis
To determine the antibacterial activity of treated OXA samples
the disk diffusion method was used. In this method Mueller–Hilton
agar (DIFCO1) plates were seeded with a 105 CFUmLꢁ1 suspension
of S. aureus (ATCC 6538). The plates were loaded with 100 mL of
irradiated samples and incubated for 24 h at 37 ꢃC. The inhibition
halo formed around the holes was measured (in mm) to determine
the decrease in potency of the antibiotic during the treatment.
2.6. Identification of degradation by-products
Degradation intermediates for oxacillin were extracted from
treated solutions by solid-phase extraction, using Strata X1
cartridges to improve intermediate detectability. Cartridges were
conditioned with 2 mL of methanol and 2 mL of water, and loaded
with a 50 mL aliquot of the sample. The sorbent was washed with
2 mL of Milli-Q1 water and then eluted with two aliquots of 2 mL of
2% formic acid in water. The by-products generated were
monitored by an ORBITRAPS mass spectrometer (Thermo Scien-
tific1) connected to a HPLC Ultimate 3000 system (Thermo
Dionex1) equipped with a 250-mm reverse-phase C18 analytical
column that had a 5-mm particle size (LiChrospher, RP-18,
Merck1). The mobile phase was a mixture of acetonitrile acidified
with 0.1% formic acid (A/F) and water acidified with 0.1% formic
acid (W/F) at a flow rate of 0.4 mL minꢁ1. A linear gradient
progressed from 10% A/F (initial conditions) to 100% A/F in 50 min,
and then remained stable at 100% A/F for 5 min. The injection
volume was 20 mL. This HPLC system was connected to an
ORBITRAPS mass spectrometer (Thermo Scientific1) equipped
with an electrospray interface, which was operated under the
following conditions: spray voltage 3.5 KV, sheath gas flow 40,
auxiliary gas flow 10, capillary temperature 320 ꢃC, heater
temperature 300 ꢃC, S-lens RF level 55 and a positive polarity.
Data were processed with the Thermo Xcalibur Instrument
software.
Fig. 2. (a) Effect of the antibiotic concentration during oxacillin degradation by UV/
TiO2. 2.0 g Lꢁ1 of TiO2, 150 W and natural pH. (b) Initial oxacillin degradation rate (r)
as a function of their initial concentrations. Inset: linear regression 1/C vs 1/r.
Experiments were carried out at least in duplicated.
irradiation intensity, the number of catalyst particles and pH.
Typically, the rate of the photocatalytic degradation increases with
an increase in light intensity and catalyst concentration until it
reaches a limiting value [19], which is particular to each organic
compound and photocatalytical system. Regarding the pH, it has
been generally observed that best degradation occurs at pH values
where there is a higher adsorption of the pollutant onto the
catalyst surface [20,21]. However, in some cases, the pollutant
adsorption can also be detrimental. Consequently, to maximize the
efficiency of the process an evaluation and optimization of the
experimental parameters should be always carried out.
Table 1 shows the factorial array used to analyze the influence
of TiO2 concentration, pH and light intensity when subjected to
photocatalytical degradation of OXA. The response variable was
the degradation percentage after 20 min of photocatalytical action.
As can be seen in Table 1, the catalyst load and the light intensity
are the most important factors influencing the process. For a better
interpretation of the results, Fig. 1a and b shows the response
surfaces for OXA degradation; while Fig. 1c depicts the Pareto
chart. The response surface figure (Fig. 1a and b) shows a
significant improvement in the degradation percentage as the
applied power increased from 30 to 150 W. This is because an
increase in light intensity generates more emitted photons,
thereby forming more electron–holes. These can generate a higher
number of hydroxyl radicals or electronic vacancies where direct
3. Results and discussion
3.1. OXA adsorption onto the photocatalyst
The adsorption of the pollutant onto the catalyst surface can be
a determining step in the heterogeneous photocatalytic process.
Therefore, the adsorption behavior of OXA onto the catalyst in dark
conditions was evaluated. The results indicated that the
adsorption–desorption equilibrium was reached after 30 min
(Fig. S1, see Supplementary data). Tests conducted at different
pH values indicate that at pH 3.0, natural pH (6.0) and pH 9.0
adsorptions were approximately 5%, 30% and 1% respectively
(Fig. S1, see Supplementary data). The implications and reasons for
these different adsorptions will be discussed later.
3.2. Evaluation and optimization of operative degradation parameters
during the TiO2 photocatalytic degradation of oxacillin
3.2.1. Evaluation and optimization of TiO2 loading, light intensity and
initial pH
During the photocatalytic degradation process, the efficiency of
the system is related to experimental parameters such as light

A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
99
oxidation of the pollutant can occur [22]. Higher catalyst
concentrations also improved the efficiency of the process (Table 1
and Fig. 1a), which can be explained by a greater availability of
active sites on the catalyst surface. However, TiO2 concentrations
higher than 2 g Lꢁ1 were not taken into account in this study to
avoid having an excessive catalyst load that would have to be
recovered at the end of the treatment.
As can be seen in the Pareto diagram (Fig. 1c), the light intensity
and the catalyst concentration significantly influenced the
efficiency of OXA degradation. On the contrary, according to
Fig. 1b and the Pareto diagram (Fig. 1c), the initial pH of the
solution (3–9) did not significantly affect the removal percentage
of the antibiotic.
To interpret the effect of initial pH on the process, the stability
of OXA at the different tested pH values was investigated. The
results (Fig. S2, see Supplementary data) indicated that in the
working conditions, no change on the initial antibiotic concentra-
tion was observed. Therefore, changes to both, TiO2 (Point of zero
charge, pzc = 6.30 for P25 Evonik) [20] and the pollutant structure
(pKa 2.8 OXA) [23], must be considered. At natural and basic pH, the
OXA carboxylic group is deprotonated, giving to the antibiotic a
negative charge. Meanwhile in acidic media (pH 3.0), according to
the pKa of the substrate, a significant fraction of OXA is in its neutral
and uncharged form. On the other hand, in acidic media (pH 3.0)
the surface TiO2 is positively charged, while in basic media (pH 9.0)
it is negatively charged. At natural pH the TiO2 is very close to its
pzc value; and thus it has practically the same number of negative
and positive sites on its surface [20,24]. Therefore, when using the
basic pH, repulsion occurs between the OXA structure and
the catalyst, and consequently a negligible adsorption (ꢀ1%) of
the pollutant is observed (Fig. S1, see Supplementary data). At the
more acidic pH (3.0), electrostatic attraction forces between the
antibiotic and the catalyst is not predominant, because at this pH
value the substance is close to their pKa value. Under such
conditions, only the ionized form of such molecule interacts with
the positively charged catalyst, which may explain the low
adsorption observed (ꢀ5%) (Fig. S1, see Supplementary data).
Furthermore, at natural pH (ꢀ6.0) electrostatic attractive forces
between the antibiotic and the catalyst can be important. In fact,
even if under such pH values the catalyst is close to its isoelectric
point and the pollutant is negatively charged, the high
concentration of catalyst molecules in relation to the pollutant
concentration makes the adsorption of a considerable fraction of
this compound possible. This explains why better adsorption of
OXA (ꢀ30% of the antibiotic) is observed at this pH value (Fig. S1,
see Supplementary data). The higher adsorption at natural pH
suggests that at this pH pollutant adsorption can also contribute to
the destruction of the antibiotic. On the contrary, at basic or acid
pH, where an insignificant absorption was observed, free OH
radicals seem to be the main factor responsible for OXA
degradation. However, whether or not the adsorption of the
antibiotic on the catalyst surface takes place, the degradation rate
of the substrate is similar and as a result, the effect of pH on the
photocatalytical system was not significant (Fig. 1b and c).
Therefore, from a practical point of view, natural pH is the most
suitable pH condition to perform the OXA degradation by TiO2
photocatalysis.
According to the response surface diagrams (Fig. 1a and b), the
most favorable conditions for the degradation of OXA in aqueous
TiO2 suspensions are: natural pH (6.0), 2.0 g Lꢁ1 of TiO2 and a light
power of 150 W.
These conditions can also be presented in a polynomial
expression (Eq. (5)); where Y (%), represents the degradation
percentage of the antibiotics:
Y ð%Þ ¼ 7:15944 þ 14:3724 ꢅ ½TiO2ꢆ þ 0:194115 ꢅ Power
(5)
The expression shows the coefficients of the reduced model
calculated by multiple regression analysis, which directly relates
the response factor with the influential variables. Table 1 presents
both the experimental values and the results calculated with the
mathematical model. As seen in Table 1, the estimated values are
similar to the experimental values in most of the experiments.
Therefore, as observed by other treatment systems and pollutants
[25,26], the reduced model can satisfactorily describe the TiO2
photocatalytical process for the degradation of the antibiotic OXA.
3.2.2. Effect of the initial concentration
To investigate the influence of the initial concentration of the
studied penicillin during the photocatalytical treatment, different
solutions of OXA, in a concentration range of two orders of
magnitude (from 25 m m
mol Lꢁ1 to 438 mol Lꢁ1) at a fixed TiO2
concentration (2.0 g Lꢁ1) and light intensity (150 W) were tested
(Fig. 2a). The evolution of the curves seems to indicate that the
Fig. 3. Relative oxacillin degradation percent by TiO2 photocatalysis in presence of
different matrices after 60 min of irradiation and 2.0 g Lꢁ1 of TiO2,150 W and natural
pH. %DW stands for the degradation percent in deionized water and %Matrix
represent the degradation percent in presence of: mineral natural water (MW),
mineral natural water containing a commercial product of the antibiotic (FP in
Fig. 4. TiO2 photocatalytic degradation of oxacillin (OXA) and antimicrobial activity
removal (AA), in deionized water (DW) and in presence of mineral natural water
containing a commercial product of the antibiotic (FP in MW). Test conditions were
the same as those in Fig. 3. Experiments were carried out at least in duplicated.
MW), Fe2+ 5075
m m
mol Lꢁ1 (Fe(II) 5075), Fe2+ 90 mol Lꢁ1 (Fe(II) 90) and Fe3+
90
m
mol Lꢁ1 (Fe(III) 90). Experiments were carried out at least in duplicated.

100
A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
reaction fits a pseudo-first order kinetics law. However, Fig. 2b
clearly shows that the degradation rate does not increase linearly
with the concentration of the pollutant, as seen with first order
kinetics. Instead, particularly at relatively high pollutant concen-
tration, the slopes of the curves gradually decrease. This behavior
can be better represented in terms of a Langmuir–Hinshelwood
(LH) kinetic model. To obtain the kinetic parameters from the LH
model, linearization can be performed (Eq. (6)).
results shown that in the case of mineral water, the degradation is
slightly affected. Fig. 3 shows that the percentage of degradation
decreases by approximately 6% after 60 min of reaction compared
to degradation in deionized water. Interestingly, a reduction in the
adsorption was also observed in natural water. In fact during the
equilibration time, in the natural water the adsorption decreases
from ꢀ30% to 1%, which probably leads to the slight reduction
observed in the percentage of degradation. The reduction in the
percentage of degradation may be due to the presence of inorganic
anions in water, such as phosphate, sulfate, nitrate, bicarbonate
and chloride, which are known scavengers of hydroxyl radicals.
Additionally, bicarbonate, the ion present in the highest concen-
tration, can easily adsorb onto the catalyst and avoid the pollutant-
catalyst interaction, thereby negatively affecting the process. The
lower adsorption of the pollutants in the presence of this water
matrix supports the previous hypothesis [28,29].
1
r
1 1
KkC
1
k
¼
þ
(6)
In Eq. (6), r represents the initial degradation rate for OXA
mol Lꢁ1 minꢁ1), C is the initial concentration of the pollutant
mol Lꢁ1), k is the apparent rate constant and K is the adsorption–
(
(
m
m
desorption equilibrium constant. The calculated parameters are:
k = 5.62
m
mol Lꢁ1 minꢁ1 and K = 3.78 ꢅ 10ꢁ3
m
molꢁ1 L; with a small
When iron is present in the reaction medium, interesting
phenomena occur during the degradation of the substance. When
the concentrations of Fe2+ and Fe3+ are below the concentrations of
the substance used, the degradation of the antibiotic is slightly
enhanced (Fig. 3). In water solutions and due to the presence of
dissolved oxygen, Fe2+ ions easily undergoes into Fe3+ ions. In
presence of UV light, the generated ferric ion is able to produce
extra hydroxyl radicals [18,30], which improves OXA degradation.
The higher improve in the OXA degradation due to Fe3+, is in
agreement with this hypothesis. Additionally, the electron
scavenging effect of iron ions may prevent the recombination of
electrons and holes, which favors the formation of ꢄOH and O2ꢄꢁ on
the TiO2 surface [28]. However, when the concentration of Fe2+ is
average percentage error (APE = 5.6%). It should be noted, that the
values of the constants obtained in the development of this work
are very similar to those obtained by Dimitrakopoulou et al. when
working with amoxicillin, another antibiotic of the penicillin class
[27]. As can also be seen in the inset of Fig. 2b, the experimental
data shows a good fit to the LH model (R2 is 0.9979).
3.3. Study of OXA degradation in presence of different matrix
composition
The presence of ions, organic matter, biological materials and
xenobiotics can affect the processes of degradation, reducing or
increasing the speed of degradation, either by interaction or by
competition with the substances of interest. To enhance our
understanding of these phenomena, it is important to test the
performance of the aforementioned technology in samples of
natural waters containing the pollutant. It is also important to
evaluate the effect of different species that may be present
during OXA degradation by TiO2 photo-catalysis. Thereforꢁe,
25 times higher (5075 m
mol Lꢁ1) than the concentration of OXA,
the percentage of degradation is significantly reduced (Fig. 3). This
can be attributed to the suppression of ꢄOH radicals by Fe2+, as the
conduction band electrons are trapped by the adsorbed metal ions
[28]. Such behavior is in accordance with the reduction in OXA
adsorption during equilibration in the dark in the presence of Fe2+
(Fig. S3, see Supplementary data).
On the other hand, in commercial formulations OXA is found
mixed with a number of excipients or inert substances so that it
can be supplied as a medicine. To further prove the potential of the
technology, a commercial finished product (FP), OXA sterile
powder for injection, was treated by TiO2 photocatalysis. The
results shown in Fig. 3 indicate that the excipients in the finished
mineral spring water containing 4.930ꢁmmol Lꢁ1 HCO3
,
1.160 mmol Lꢁ1 Mg2+
,
0.137 mmol Lꢁ1 NO3
,
0.043 mmol Lꢁ1
SO42ꢁ, 0.020 mmol Lꢁ1 K+, 1.220 mmol Lꢁ1 Ca2+, 0.252 mmol Lꢁ1
Na+ and 0.068 mmol Lꢁ1 Clꢁ was used as an example of a natural
water sample. Moreover, the effect of the presence of Fe2+ ions and
the degradation of the pharmaceutical, in a commercial formula-
tion, were investigated. The Fe2+ ion was chosen to be studied given
its widespread presence in a lot variety of water sources. The
product only cause
degradation.
a
slight inhibition (ꢀ5%) in the OXA
The aforementioned results highlight the ability of the
technology to remove the antibiotic even in the presence of
excipients in commercial products or ions in natural waters.
3.4. Evaluation of the loss of antimicrobial activity
One of the most important aspects to be taken into account in
the treatment of antibiotics is the residual antimicrobial activity
(potency) of the substance and the by-products produced in the
course of treatment. Thus, the susceptibility of S. aureus (ATCC
6538) strains was used to evaluate the antimicrobial activity of the
treated solutions during the application of the photocatalytical
process. S. aureus was chosen because of its good response at low
antibiotic concentrations using the inhibition halo methodology
[31]. In this analysis, the smaller inhibition halo around the
microdrop spread on the agar plate and inoculated with the
bacteria suggests the antimicrobial activity is less important.
Fig. 4 indicates that the inhibition halo of the treated solution is
completely eliminated by photocatalytical action. Interestingly,
the elimination of antimicrobial activity is consistent with the
elimination of the antibiotic. The concomitant elimination of the
antimicrobial activity with the pollutant suggests that the initial
Fig. 5. TiO2 photocatalytic degradation of oxacillin in presence of some additives: 2-
propanol at 5075
5075
mol Lꢁ1. Test conditions were the same as those in Fig. 3. Experiments were
carried out at least in duplicated.
m m m
mol Lꢁ1 and 130800 mol Lꢁ1, and KI at 203 mol Lꢁ1 and
m

A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
101
Fig. 6. Main reaction pathways during the TiO2 photocatalytical degradation of the antibiotic oxacillin.
breaking up of the compound takes place in the
b
-lactam ring.
also retard the loss of antimicrobial activity. In fact, 180 min of
photocatalytic treatment, instead of 120 min under experimental
conditions, is required to eliminate the pollutant and its
antimicrobial activity. However, even in as such complex matrix,
the TiO2 photocatalytical system resulted effective to eliminate the
antibiotic and its antimicrobial activity acting against S. aureus.
Finally, the action of light alone is not able to reduce either the
pollutant concentration or the antimicrobial activity (data not
shown). For a better interpretation, the UV emission spectrum of
the lamp and the OXA UV absorption spectrum were obtained
(Fig. S4, see Supplementary data). As seen, the fact that there were
no band matches indicates that the OXA cannot absorb light
radiation from the lamp. Accordingly, direct photolysis of the
antibiotic is not plausible.
3.5. Determination of OXA photocatalytic degradation route
On the other hand, Fig. 4 also shows the joining effects of
excipients in a commercial formulation of the antibiotic with the
effects of matrix components in natural water on the removal of
antimicrobial activity. From the figure it can be noted that ions in
the mineral water and excipients of the commercial product
increase and delay the degradation time of the OXA and therefore
The application of AOPs for organic pollutant degradation can
be carried out by different degradation pathways, which involve
variations on both the degradation rate and the by-products
generated. Thus, organic chemicals in TiO2 photocatalysis can be
oxidized by three different mechanisms: direct photolysis,
oxidation at the photo-generated holes or hydroxyl radical attack.

102
A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
phenylisoxazole-4-carboxamide (compound E). The chromato-
gram and mass spectrum of the identified compounds is shown in
Fig. S5 (see Supplementary data). Most of the by-products were
produced half-way through the irradiation and, as time pro-
gressed, they gradually disappeared. The results indicated that the
photocatalytical degradation of OXA began with the hydroxylation
on the aromatic ring to produce compound A and with the
breaking up of the b-lactam ring to produce compound B. Such as
suggested in the previous section, both of initial routes are carried
at the catalyst surface by attack of the adsorbed hydroxyl radicals.
The signal at m/z 160, in the mass spectra of compounds A and B
(m/z 160), support the presence of the penicillin nucleus in both
intermediates [33]. Decarboxylation of oxacillin penicilloic acid
(compound B), via photo-Kolbe reaction, produces compound C.
The reduced sulfur moiety of compound B can also be oxidized by
the hydroxyl radical to produce a sulfoxide group (Compound D).
Secondary amine group are also susceptible to hydroxyl radical
attack. Therefore, compound E can be formed from hydroxyl
radical attack on oxacillin and compounds B–D. Finally, successive
attack of hydroxyl radicals would conduct to the formation of
carboxylic acids and then to the complete mineralization of the
initial pollutant into CO2, water and inorganic ions. A schema
depicting the main degradation pathways of OXA is shown in Fig. 6.
Fig. 7. Evolution of oxacillin, COD, TOC and pH during OXA degradation by TiO2
photocatalysis and photolysis (UV). Test conditions were the same as those in Fig. 3.
Experiments were carried out at least in duplicated.
In the previous section it was shown that, under the working
conditions, direct photolysis is not plausible.
To investigate whether the other two pathways are followed in
the degradation of OXA by TiO2 photocatalysis, ꢄOH radical
scavengers and substances that act on the valence band holes of
the catalyst can be tested. Alcohols, such as 2-propanol, have
commonly been used to estimate the oxidation mechanism in
photocatalytic processes due to their powerful ꢄOH radical
scavenger effect at the solution bulk [31,32]. In it turn, the iodide
ion is an excellent probe specie in TiO2 photocatalysis, as it reacts
3.7. Evaluation of the extend of oxidation and mineralization during
the photocatalytical treatment
The evolution of the OXA concentration, COD, and DOC in
individual solutions containing the antibiotic was determined to
comparatively evaluate pollutant elimination, oxidation and
mineralization of the antibiotic, respectively. The experiments
were performed at natural pH (ꢀ6.0), 2.0 g Lꢁ1 of TiO2 and 150 W of
light intensity. Fig. 7 reveals that the initial OXA concentration
decrease. Total removal of the antibiotic is reached after 120 min of
reaction. In turn, COD evolution indicates that the antibiotic is
transformed into more oxidized compounds. This is consistent
with the by-products identified previously. Interestingly, the pH
decrease, to ꢀ3.5–4.0 after two hours of treatment (Fig. 7),
suggests that these by-products are subsequently transformed into
carboxylic acids. Additionally, as shown by the DOC evolution,
successive photocatalytical action is able to further degrade the
initial pollutant into CO2, water and inorganic ions. In fact, after
480 min of treatment both COD and DOC are practically eliminated.
As can also be observed, with the elimination of these by-products
the solution pH progressively increases.
Recently Magureanu et al. [34] evaluated the use of non-
thermal plasma to remove OXA in water, which is up to now, the
only AOPs reported to deal with this antibiotic. TiO2 photocatalysis
have the advantage over non-thermal plasma by using ambient
sunlight as energy source and being environmental friendly
technology. Additionally, as indicated by Malato et al. [14], TiO2
is a cheap photo-stable catalyst, and the process may run at
ambient temperature and pressure. In fact, in principle, the process
involves a mild catalyst working under mild conditions with mild
oxidants. The aforementioned characteristics of the TiO2 photo-
catalytic system could promote its use for real applications and the
feasibility of the process has been already demonstrated at large
scale [14–15,35].
ꢄ
with valence band holes and adsorbed OH radicals [32].
The results presented in Fig. 5 show that a concentration of 2-
propanol 25 times higher than the antibiotic produces a slight
reduction (ꢀ3%) in the removal rate of the pollutant. On the other
hand, a concentration 645 times higher than the pollutant causes
ꢀ30% of inhibition. This indicates that hydroxyl radicals at the
solution bulk may contribute to the degradation of these
molecules. However, it is not the main mechanism involved given
the moderate reduction in the extent of OXA removal. Assays in the
presence of KI were also conducted to study the involvement of the
catalyst surface in the pollutant degradation. When
a KI
concentration 25 times higher than the concentration of the
substance was employed, ꢀ75% inhibition occurred (Fig. 5). When
an equimolar KI concentration was employed, a 13% reduction was
observed in the degradation rate of OXA. Interestingly, in all of
aforementioned cases, the reduction in the degradation rate is
associated with a reduction in the adsorption of the pollutant to
the catalyst surface. Consequently, the degradation of OXA by
heterogeneous photocatalysis seems to occur mostly at the catalyst
surface via two routes: radical attack and photo-Kolbe mechanism.
3.6. Identification of OXA by-products
During the photocatalytic degradation of OXA, using TiO2 as the
photocatalyst, various organic intermediates were produced. Five
OXA by-products were identified by HPLC-MS: 6-({[3-(4-hydrox-
yphenyl)-5-methylisoxazol-4-yl]carbonyl}amino)-3,3-dimethyl-
7-oxo-4-thia-1-azabicyclo[3.2.0]heptane-2-carboxylic acid (com-
pound A), 2-(carboxy{[(5-methyl-3-phenylisoxazol-4-yl) carbon-
yl]amino}methyl)-5,5-dimethyl-4,5-dihydro-1,3-thiazole-4-car-
boxylic acid (compound B), (5,5-dimethyl-4,5-dihydro-1,3-thiazol-
2-yl) {[(5-methyl-3-phenyl-1,2-oxazol-4-yl) carbonyl]amino}ace-
tic acid (compound C), 2-(carboxy{[(5-methyl-3-phenylisoxazol-
4-yl) carbonyl]amino}methyl)-5,5-dimethyl-4,5-dihydro-1,3-thia-
zole-4-carboxylic acid 1-oxide (compound D) and 5-methyl-3-
4. Conclusions
The results of this work show that the TiO2 photocatalysis
system has the potential to eliminate the environmentally
damaging effects of
b-lactam antibiotics such as OXA. High light
intensity and catalyst loading increased the efficiency of the
process. Experiments at natural pH (6.0) showed that the system

A.L. Giraldo-Aguirre et al. / Journal of Photochemistry and Photobiology A: Chemistry 311 (2015) 95–103
103
followed Langmuir–Hinshelwood kinetics. The identification of
OXA by-products and assays in the presence of probe substances
[13] M. Faouzi, P. Cañizares, A. Gadri, J. Lobato, B. Nasr, R. Paz, M.A. Rodrigo, C. Saez,
Advanced oxidation processes for the treatment of wastes polluted with azoic
dyes, Electrochim. Acta 52 (2006) 325–331.
[14] S. Malato, P. Fernández-Ibáñez, M.I. Maldonado, J. Blanco, W. Gernjak,
Decontamination and disinfection of water by solar photocatalysis: recent
overview and trends, Catal. Today 147 (2009) 1–59.
[15] S. Malato, J. Blanco, A. Vidal, C. Richter, Photocatalysis with solar energy at a
pilot-plant scale: an overview, Appl. Catal. B: Environ. 37 (2002) 1–15.
[16] S. Malato, J. Blanco, A. Vidal, D. Alarcón, M.I. Maldonado, J. Cáceres, W. Gernjak,
Applied studies in solar photocatalytic detoxification: an overview, Solar
Energy 75 (2003) 329–336.
[17] T. Xu, Y. Cai, K.E. O’Shea, Adsorption and photocatalyzed oxidation of
methylated arsenic species in TiO2 suspensions, Environ. Sci. Technol. 41
(2007) 5471–5477.
showed that OXA degradation occurred via
a photo-Kolbe
mechanism with some participation of hydroxyl radical attack.
Under the working conditions, the system showed to be able to
remove the pollutant and the antimicrobial activity acting against
S. aureus, transforming OXA into CO2, water and inorganic ions. It
has the additional advantage that excipients in a commercial
preparation of the antibiotic, as well as inorganic species present in
natural waters, do not significantly affect the efficiency of the
system.
[18] R.A. Torres-Palma, J.I. Nieto, E. Combet, C. Pétrier, C. Pulgarin, An innovative
ultrasound, Fe2+ and TiO2 photoassisted process for bisphenol
a
mineralization, Water Res. 44 (2010) 2245–2252.
Acknowledgements
[19] C. Martínez, M. Canle L, M.I. Fernández, J.A. Santaballa, J. Faria, Aqueous
degradation of diclofenac by heterogeneous photocatalysis using
nanostructured materials, Appl Catal B: Environ. 107 (2011) 110–118.
[20] X. Van Doorslaer, K. Demeestere, P.M. Heynderickx, H. Van Langenhove, J.
Dewulf, UV-A and UV-C induced photolytic and photocatalytic degradation of
aqueous ciprofloxacin and moxifloxacin: reaction kinetics and role of
adsorption, Appl. Catal.B: Environ. 101 (2011) 540–547.
[21] K.-I. Ishibashi, A. Fujishima, T. Watanabe, K. Hashimoto, Quantum yields of
active oxidative species formed on TiO2 photocatalyst, J. Photochem.
Photobiol. A: Chem. 134 (2000) 139–142.
[22] A.R. Khataee, M. Fathinia, S.W. Joo, Simultaneous monitoring of photocatalysis
of three pharmaceuticals by immobilized TiO2 nanoparticles: chemometric
assessment, intermediates identification and ecotoxicological evaluation,
Spectrochim. Acta Part A: Mol. Biomol. Spectrosc. 112 (2013) 33–45.
[23] G. Yang, S. Feng, H. Liu, J. Yin, L. Zhang, L. Cai, On-line clean-up and screening of
oxacillin and cloxacillin in human urine and plasma with a weak ion exchange
monolithic column, J. Chromatogr. B 854 (2007) 85–90.
The authors would like to thank the Administrative Department
of Science, Technology and Innovation (Colciencias, Colombia) for
its financial support throughout the project: “Implementación de
metodologías eficientes y confiables para degradar residuos de
antimicrobianos en el hogar y en efluentes industriales”.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the
online
version,
at
http://dx.doi.org/10.1016/j.
jphotochem.2015.06.021.
[24] M. Sleiman, D. Vildozo, C. Ferronato, J.-M. Chovelon, Photocatalytic
degradation of azo dye metanil yellow: optimization and kinetic modeling
using a chemometric approach, Appl. Catal. B: Environ. 77 (2007) 1–11.
[25] P. Durango-Usuga, F. Guzmán-Duque, R. Mosteo, M.V. Vazquez, G. Peñuela, R.A.
Torres-Palma, Experimental design approach applied to the elimination of
crystal violet in water by electrocoagulation with Fe or Al electrodes, J. Hazard.
Mater. 179 (2010) 120–126.
References
[1] K. Kümmerer, The presence of pharmaceuticals in the environment due to
human use—present knowledge and future challenges, J. Environ. Manage. 90
(2009) 2354–2366.
[26] R.A. Torres, R. Mosteo, C. Pétrier, C. Pulgarin, Experimental design approach to
the optimization of ultrasonic degradation of alachlor and enhancement of
treated water biodegradability, Ultrason. Sonochem. 16 (2009) 425–430.
[27] D. Dimitrakopoulou, I. Rethemiotaki, Z. Frontistis, N.P. Xekoukoulotakis, D.
Venieri, D. Mantzavinos, Degradation, mineralization and antibiotic
inactivation of amoxicillin by UV-A/TiO2 photocatalysis, J. Environ. Manage.
98 (2012) 168–174.
[28] S. Ahmed, M.G. Rasul, W.N. Martens, R. Brown, M.A. Hashib, Heterogeneous
photocatalytic degradation of phenols in wastewater: a review on current
status and developments, Desalination 261 (2010) 3–18.
[29] X. Zhu, M.A. Nanny, E.C. Butler, Effect of inorganic anions on the titanium
dioxide-based photocatalytic oxidation of aqueous ammonia and nitrite, J.
Photochem. Photobiol. A: Chem. 185 (2007) 289–294.
[30] H.E. Kim, J. Lee, H. Lee, C. Lee, Synergistic effects of TiO2 photocatalysis in
combination with Fenton-like reactions on oxidation of organic compounds at
circumneutral pH, Appl. Catal. B: Environ. 115 (2012) 219–224.
[2] N. Kemper, Review veterinary antibiotics in the aquatic and terrestrial
environment, Ecol. Indic. 8 (2008) 1–13.
[3] J.L. Martinez, Review environmental pollution by antibiotics and by antibiotic
resistance determinants, Environ. Pollut. 157 (2009) 2893–2902.
[4] F. Gagné, C. André, P. Cejka, R. Hausler, M. Fournier, Evidence of
neuroendocrine disruption in freshwater mussels exposed to municipal
wastewaters, Sci. Total Environ. 409 (2011) 3711–3718.
ꢀ
[5] D. Błe˛dzka, J. Gromadzinska, W. Wa˛sowicz, Parabens. From environmental
studies to human health, Environ. Int. 67 (2014) 27–42.
[6] N. Gundogan, S. Citak, N. Yucel, A. Devren, A note on the incidence and
antibiotic resistance of Staphylococcus aureus isolated from meat and chicken
samples, Meat Sci. 69 (2005) 807–810.
[7] J.H. Savage, E.C. Matsui, R.A. Wood, C.A. Keet, Urinary levels of triclosan and
parabens are associated with aeroallergen and food sensitization, J. Allergy
Clin. Immunol. 130 (2012) 453–460.
[8] R. Sutherland, E.A. Croydon, G.N. Rolinson, Flucloxacillin, a new isoxazolyl
penicillin, compared with oxacillin, cloxacillin, and dicloxacillin, Br. Med. J. 4
(1970) 455–460.
[31] R.A. Palominos, M.A. Mondaca, A. Giraldo, G. Peñuela, M. Pérez-Moya, H.D.
Mansilla, Photocatalytic oxidation of the antibiotic tetracycline on TiO2 and
ZnO suspensions, Catal. Today 144 (2009) 100–105.
[9] J.P. Bru, R. Garraffo, Role of intravenous cloxacillin for inpatient infections,
[32] Y. Chen, S. Yang, K. Wang, L. Lou, Role of primary active species and TiO2 surface
characteristic in UV-illuminated photodegradation of acid orange 7, J.
Photochem. Photobiol. A: Chem. 172 (2005) 47–54.
[33] L.A. Mitscher, H.D. Hollis Showalter, K. Shirahata, Chemical-ionization mass
spectrometry of beta-lactam antibiotics, J. Antibiot. (Tokyo) 28 (1975)
668–675.
Méd. Mal. Infect. 42 (2012) 241–246.
[10] D.W. Hughes, C.R. Frei, P.R. Maxwell, K. Green, J.E. Patterson, G.E. Crawford, J.S.
Lewis, Continuous versus intermittent infusion of oxacillin for treatment of
infective endocarditis caused by methicillin-susceptible Staphylococcus
aureus, Antimicrob. Agents Chemother. 53 (2009) 2014–2019.
[11] J.M. Cha, S. Yang, K.H. Carlson, Trace determination of
b-lactam antibiotics in
[34] M. Magureanu, D. Piroi, N.B. Mandache, V. David, A. Medvedovici, C. Bradu, V.I.
Parvulescu, Degradation of antibiotics in water by non-thermal plasma
treatment, Water Res. 45 (2011) 3407–3416.
surface water and urban wastewater using liquid chromatography combined
with electrospray tandem mass spectrometry, J. Chromatogr. A 1115 (2006)
46–57.
[35] B. Lopez-Alvarez, R.A. Torres-Palma, G. Peñuela, Solar photocatalitycal
treatment of carbofuran at lab and pilot scale: Effect of classical
parameters, evaluation of the toxicity and analysis of organic by-products,
J. Hazard. Mater. 191 (2011) 196–203.
[12] S. Croes, P.S. Beisser, P.H. Terporten, C. Neef, R.H. Deurenberg, E.E. Stobberingh,
Diminished in vitro antibacterial activity of oxacillin against clinical isolates of
borderline oxacillin-resistant Staphylococcus aureus, Clin. Microbiol. Infect. 16
(2010) 979–985.