High frequency ultrasound as a selective advanced oxidation process to remove penicillinic antibiotics and eliminate its antimicrobial activity from water

Article abstract of DOI:10.1016/j.ultsonch.2016.01.007

This work studies the sonochemical degradation of a penicillinic antibiotic (oxacillin) in simulated pharmaceutical wastewater. High frequency ultrasound was applied to water containing the antibiotic combined with mannitol or calcium carbonate. In the presence of additives, oxacillin was efficiently removed through sonochemical action. For comparative purposes, the photo-Fenton, TiO₂ photocatalysis and electrochemical oxidation processes were also tested. Therefore, the evolution of the antibiotic and its associated antimicrobial activity (AA) were monitored. A high inhibition was found for the other three oxidation processes in the elimination of the antimicrobial activity caused by the additives; while for the ultrasonic treatment, a negligible effect was observed. The sonochemical process was able to completely degrade the antibiotic, generating solutions without AA. In fact, the elimination of antimicrobial activity showed an excellent performance adjusted to exponential kinetic-type decay. The main sonogenerated organic by-products were determined by means of HPLC-MS. Four intermediaries were identified and they have modified the penicillinic structure, which is the moiety responsible for the antimicrobial activity. Additionally, the possible oxacillin sonodegradation mechanism was proposed based on the evolution of the by-products and their chemical structure. Furthermore, the high-frequency ultrasound action over 120 min readily removed oxacillin and eliminated its antimicrobial activity. However, the pollutant was not mineralized even after a long period of ultrasonic irradiation (360 min). Interestingly, the previously sonicated water containing oxacillin and both additives was completely mineralized using non-adapted microorganisms from a municipal wastewater treatment plant. These results show that the sonochemical treatment transformed the initial pollutant into substances that are biotreatable with a typical aerobic biological system.

Full text of DOI:10.1016/j.ultsonch.2016.01.007

Accepted Manuscript
High frequency ultrasound as a selective advanced oxidation process to remove
penicillinic antibiotics and eliminate its antimicrobial activity from water
Efraim A. Serna-Galvis, Javier Silva-Agredo, Ana L. Giraldo-Aguirre, Oscar
A. Flórez-Acosta, Ricardo A. Torres-Palma
PII:
S1350-4177(16)30008-6
DOI:
http://dx.doi.org/10.1016/j.ultsonch.2016.01.007
ULTSON 3083
Reference:
Ultrasonics Sonochemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
12 December 2015
6 January 2016
7 January 2016
Please cite this article as: E.A. Serna-Galvis, J. Silva-Agredo, A.L. Giraldo-Aguirre, O.A. Flórez-Acosta, R.A.
Torres-Palma, High frequency ultrasound as a selective advanced oxidation process to remove penicillinic
antibiotics and eliminate its antimicrobial activity from water, Ultrasonics Sonochemistry (2016), doi: http://
dx.doi.org/10.1016/j.ultsonch.2016.01.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1
2
3
High frequency ultrasound as a selective advanced oxidation process to
remove penicillinic antibiotics and eliminate its antimicrobial activity from
water
4
5
Efraim A. Serna-Galvisa, Javier Silva-Agredoa, Ana L. Giraldo-Aguirreb, Oscar A.
Flórez-Acostab, Ricardo A. Torres-Palmaa*
6
a
7
Grupo de Investigación en Remediación Ambiental y Biocatálisis (GIRAB),
8
Instituto de Química, Facultad de Ciencias Exactas y Naturales, Universidad de
Antioquia UdeA, Calle 70 No. 52-21, Medellín, Colombia.
9
10
11
12
13
14
b
Grupo de Diseño y Formulación de Medicamentos, Cosméticos y Afines
(DYFOMECO), Facultad de Química Farmacéutica, Universidad de Antioquia
UdeA, Calle 70 No. 52-21, Medellín, Colombia.
15
16
17
*Corresponding author: ricardo.torres@udea.edu.co
Abstract
18
19
20
21
22
This work studies the sonochemical degradation of a penicillinic antibiotic
(oxacillin) in simulated pharmaceutical wastewater. High frequency ultrasound was
applied to water containing the antibiotic combined with mannitol or calcium
carbonate. In the presence of additives, oxacillin was efficiently removed through
sonochemical action. For comparative purposes, the photo-Fenton, TiO2
1

23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
photocatalysis and electrochemical oxidation processes were also tested.
Therefore, the evolution of the antibiotic and its associated antimicrobial activity
(AA) were monitored. A high inhibition was found for the other three oxidation
processes in the elimination of the antimicrobial activity caused by the additives;
while for the ultrasonic treatment, a negligible effect was observed. The
sonochemical process was able to completely degrade the antibiotic, generating
solutions without AA. In fact, the elimination of antimicrobial activity showed an
excellent performance adjusted to exponential kinetic-type decay. The main
sonogenerated organic by-products were determined by means of HPLC-MS. Four
intermediaries were identified and they have modified the penicillinic structure,
which is the moiety responsible for the antimicrobial activity. Additionally, the
possible oxacillin sonodegradation mechanism was proposed based on the
evolution of the by-products and their chemical structure. Furthermore, the high-
frequency ultrasound action over 120 min readily removed oxacillin and eliminated
its antimicrobial activity. However, the pollutant was not mineralized even after a
long period of ultrasonic irradiation (360 min). Interestingly, the previously
sonicated water containing oxacillin and both additives, the organic matter was
completely mineralized using non-adapted microorganisms from a municipal
wastewater treatment plant. These results show that the sonochemical treatment
transformed the initial pollutant into substances that are biotreatable with a typical
aerobic biological system.
44
2

45
46
Keywords: Oxacillin; Pharmaceutical wastewater; Sonochemistry; Water
treatment; Selective degradation.
47
48
49
50
51
52
53
54
55
56
1
Introduction
Currently, it is common to find diverse pharmaceutical products in the effluents of
municipal wastewater plants [1,2]. Many of these substances are recalcitrant to
conventional treatment systems and, consequently, they make their way to the
natural environment. Pharmaceuticals in natural water can interact with wild
organisms which could lead to DNA mutations, alterations of the cellular
development and even cause death. Particularly, antibiotic pollution can induce
microorganism population selection and generate microbial resistance [2].
Therefore, resistant genes in environmental bacteria can be transferred to humans
and animals, increasing the risk and diminishing our ability to treat illnesses [1].
57
58
59
60
61
62
63
64
65
66
Penicillinic antibiotics are used against several infections and are recognized
among the most consumed antibiotics in the world [2]. Oxacillin is an example of
such penicillinic antibiotics. It is widely used in clinic treatments against aerobic
gram-positive cocci [3]. Oxacillin has been detected in effluents of wastewater
plants and in natural water bodies [4]. Currently, the treatment of illness associated
to oxacillin-resistant microorganisms represents a serious problem in some
hospitals around the world [5]. One of the main sources of antibiotics in wastewater
is the pharmaceutical industry [2]. Inefficient or inexistent sewage treatments in
plants and factories allow active substances to make their way to municipal
wastewater treatment plants and, subsequently, to natural water bodies. Thus,
3

67
68
studies on how to eliminate this problematic compound by means of effective
processes have become necessary.
69
70
71
72
73
74
75
76
77
78
79
In recent years, the sonochemical treatment (an advanced oxidation process,
AOP) has shown to be efficient to remediate water polluted with pharmaceutical
compounds [6–8]. Sonochemistry is based on a cyclical sequence where micro-
bubbles form and grow until reaching a critical size; then, they collapse violently in
a process called acoustic cavitation, which is induced by the interaction between
ultrasonic waves and dissolved gases in aqueous solutions. The collapse of the
micro-bubbles generates small hot spots with singular conditions of pressure
(~1000 atm) and temperature (~5000 K) [9]. Under such conditions, hydroxyl
radicals are generated by the dissociation of water molecules and oxygen (Eq. 1-
4). By means of the recombination of these radicals, hydrogen peroxide can also
be formed (Eq. 5).
80
81
82
83
84
H2O + ))) → •H + •OH
O2 + ))) → 2 •O
(Eq. 1)
(Eq. 2)
(Eq. 3)
(Eq. 4)
(Eq. 5)
H2O + •O → 2 •OH
O2 + •H → •O + •OH
2 •OH → H2O2
85
86
87
The application of high frequency ultrasound to eliminate the antimicrobial activity
from synthetic pharmaceutical wastewater containing the oxacillin antibiotic (OXA)
was tested in this study. As active ingredients are commonly mixed with other
4

88
89
compounds [10], special attention was paid to other substances that could also be
present in pharmaceutical wastewater. Therefore, this research work was focused
on studying the effects on the efficiency of the sonochemical process of a common
excipient (mannitol, MAN) used in the preparation of antibiotics and other typical
active ingredient (calcium carbonate, CC) [11]. The hydrogen peroxide
accumulation rates during the sonochemical process were determined in order to
understand the action of the additives on the elimination of both OXA and AA. The
antimicrobial activity elimination kinetic constants for different water samples were
obtained and used to quantitatively compare the efficiency of the sonochemical
process with the efficiency of the photo-Fenton, TiO2 photocatalysis and anodic
oxidation processes. Additionally, the main sonogenerated intermediaries were
determined by means of HPLC-MS, and their evolution under ultrasonic treatment
was monitored. Finally, the bio-treatability of previously-sonicated water containing
OXA and additives was assessed by means of the application of an aerobic
biological process using non-adapted microorganisms.
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
2
Materials and methods
105
106
107
108
109
2.1 Reagents
Oxacillin was provided by Sigma-Aldrich. Formic acid, hydrogen peroxide and
calcium carbonate were purchased from Carlo-Erba. Mannitol, sulfuric acid,
methanol, acetonitrile, sodium chloride, sodium phosphate, sodium hydroxide,
sodium meta-bisulfite, potassium iodide, ammonium heptamolybdate, heptahydrate
5

110
111
112
113
114
115
116
117
118
ferrous sulfate and nutrient agar were provided by Merck. Potato dextrose agar
was provided by Oxoid. Titanium dioxide P-25 was supplied by Evonik (Degussa).
Due to the fact that the antibiotic is found in pharmaceutical wastewater in
concentrations of mg L-1 [12,13] and the commercial formulations could contain
additives ten times more concentrated than the active ingredient, the OXA and the
additives (calcium carbonate and mannitol) were used at concentrations of 47.23
µmol L-1 (20 mg L-1) and 472.3 µmol L-1, respectively. All solutions were prepared
with distilled water and the processes were carried out at and initial pH of 5.6
(which is the natural pH of the OXA solution).
119
120
121
122
123
124
125
126
127
128
129
130
2.2 Reaction systems
Sonochemical experiments were carried out in a cylindrical glass reactor
containing 250 mL of the solution, using a homemade ultrasound generator (Figure
SM 1, in Supplementary Material). Ultrasonic waves of 275 kHz (at 60 W) were
emitted from the bottom of the reactor by a piezo-electric disc (with a 4 cm
diameter) fixed on a Pyrex plate (with a 5 cm diameter). It has been reported that a
high production of hydroxyl radical occurs in the range of 200-300 kHz [14].
Therefore, 275 kHz of frequency was used in all of the experimental tests. The
power dissipated by the reactor was 20.7 W (34.5% of the electrical power input)
calculated by the calorimetric method [15]. The reactor was kept at 20 1°C using
a Bioblock Scientific cryothermostat (Huber).
131
132
The biological process was carried out using aerobic microorganisms from the San
Fernando municipal wastewater treatment plant (Medellín, Colombia): 25 mL of
6

133
134
135
136
137
138
inoculum were added to 250 mL of solution in order to biologically treat the water.
This system was aerated with a Resun aquarium pump (AC-9904). It was then
stirred using a VELP BOD shaker and the samples were treated at environmental
temperature (27 0.5 °C). Before applying the biological process, the pH of the
solutions was adjusted to 6.5 with sodium hydroxide, and the residual hydrogen
peroxide was eliminated using sodium meta-bisulfite.
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
Electrochemical oxidation experiments were conducted in an electrolytic cell
(Figure SM 2) containing 150 mL of OXA solution under constant stirring
conditions. Degradation experiments were conducted at constant current density (5
mA cm−2), and using a Ti/IrO2 anode with 4 cm2 of working surface area. The
cathode was a zirconium spiral electrode of 10 cm2. 0.065 mol L-1 NaCl was used
as the supporting electrolyte. For TiO2 photocatalysis and photo-Fenton process
was used a homemade aluminum reflective reactor (Figure SM 3), equipped with
five 30 W Philips lamps (TL-D Actinic BL). 150 W of light power was applied to 100
mL of OXA solution placed in beakers with constant stirring. In the photo-Fenton
process, 1000 µmol L-1 and 90 µmol L-1 of H2O2 and Fe (II) respectively were used.
Before the analyses, the residual H2O2 also was eliminated using sodium meta-
bisulfite. In the heterogeneous photocatalysis, 0.50 g L-1 of TiO2 was used. The
lamps were turned on after the adsorption equilibrium was achieved under
constant stirring (30 min). Before analyzing the treated samples, the catalyst was
separated by centrifugation for 15 min at 3200 rev min-1 in a centrifuge (Centaur 2),
and filtered using a cellulose mesh of 0.45 µm (Advantech).
7

155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
2.3 Analyses
A quantitative analysis of OXA was carried out using a HPLC Waters (486)
instrument with a C-18 column (Merck LiChrospher) and a UV detector set at 225
nm. The injection volume was 40 µL. The mobile phase was a phosphate buffer
(0.02 mmol L-1, pH 5)/acetonitrile/methanol, 64/27/9 (% v/v) in isocratic mode (0.4
mL min-1). The oxidants generated during the process were determined through
iodometry [16].
The antimicrobial activity (AA) was determined by analyzing the inhibition zone in
the agar diffusion test, using Staphylococcus aureus ATCC 6538 as the indicator
microorganism. 30 µL of the sample solution were seeded on Petri dishes
containing 5 mL of potato dextrose agar and 10 mL of nutrient agar inoculated with
10 µL of S. aureus (with an optical density of 0.600 at 580 nm). After 24 h at 37°C
in a Memmert (Schwabach) incubator, confluent bacterial growth was observed,
and the diameter of the inhibitory halo was measured with a vernier. All
antimicrobial activity analyses were conducted at least by duplicate to ensure the
reproducibility of methodology.
The pollutant mineralization degree during the different treatments was determined
by measuring the total organic carbon (TOC). Samples were injected directly by
means of a Shimadzu Total Organic Carbon Analyzer (TOC 5000A). Prior to the
analysis, the analyzer was calibrated with standard potassium hydrogen phthalate
solutions.
The by-products from the initial degradation (formed after 50% of the OXA −203.0
177
µmol L-1− degraded) were extracted from the water and concentrated using Strata
8

178
179
180
181
182
183
184
185
186
X cartridges loaded with 50 mL of the sample. The by-products were desorbed with
2 mL of 2% formic acid, and then analyzed using a Thermo Scientific HPLC
(Ultimate 3000) – MS (Orbitraps) instrument equipped with a C-18 250 x 4.5 mm
column (Merck LiChrosphere). The mobile phase was a mixture of acetonitrile
acidified with 0.1% formic acid (A), and water acidified with 0.1% formic acid (B) at
a flow rate of 0.4 mL min-1. A linear gradient progressed from 10% A (initial
conditions) to 100% A in 50 min, and then it remained stable at 100% A for 5 min.
The injection volume was 20 µL, and the mass spectrometer was operated in the
electrospray positive ion mode.
187
188
3
Results and discussion
189
190
191
192
193
194
195
196
197
198
199
3.1 Sonochemical elimination of the oxacillin antibiotic and its antimicrobial
activity in the presence of additives
Water matrix components can inhibit the sonochemical process and its effect on
the pollutant removal efficiency has been well documented [7,17]. However, there
is scarce information about the impact of other pharmaceutical ingredients on the
sonolytic elimination of antimicrobial activity from water containing antibiotics.
Therefore, this study was focused on evaluating the influence of some additives on
the elimination of oxacillin by means of ultrasonic action. The initial pollutant
degradation rates (rd) in the presence (OXA + MAN and OXA + CC) and absence
(OXA) of additives were determined and are presented in Table 1. The results
show that rd values are considerably similar in all the cases.
9

200
201
202
203
204
205
206
207
208
Although several studies [8,16–18] have demonstrated the efficient action of
sonochemical process in the degradation of antibiotics contained in water, usually
the residual antimicrobial activity (AA) of the treated solutions is not considered.
The AA should be taken into account because it can remain even when the
pollutant has been effectively degraded [21]. Therefore, the evolution of the AA
was assessed during the sonochemical treatment for all the different water
samples (Figure 1A). The AA elimination observed during the treatment of the OXA
exhibited by the different water samples can be adjusted into an exponential form,
and then the AA elimination kinetics can be linearized as follows:
209
exp (AA/AAo) = -kt + b
(Eq. 6)
210
211
212
213
214
215
216
217
218
219
Where AA/AAo is the normalized antimicrobial activity at t time (in min) of the
process; k is the slope of the curve and represents the AA elimination kinetic
constant (in min-1); and b is the y-axis intercept (as the normalized antimicrobial
activity starts at 1 in all cases, this term is equal to exp1). The plot for the linearized
form of the AA evolution of OXA solutions in the presence and absence of the
additives is presented in Figure 1B. The correlation coefficient values show an
adequate fit (R2 > 0.99) between the experimental data and the linearized model in
all the cases. The k value calculated for OXA, OXA + MAN and OXA + CC were
0.0145, 0.0144 and 0.0146 min-1, respectively. These results indicate that additives
did not affect the AA elimination kinetics.
220
221
Beside the AA evolution and the initial oxacillin degradation rates, the H2O2
accumulation rates (ra) were also calculated (Figure 1C). Hydrogen peroxide is
10

222
223
224
225
226
227
generated from the recombination of hydroxyl radicals that do not react with the
matrix components (Eq. 5). The results indicate that the H2O2 accumulation rate in
the presence of the pollutant (OXA) is ~37% lower than in the case of distilled
water, in absence of organic matter (DW). Again, the additives did not show any
interference and the ra values for OXA are similar in both the presence and
absence of other pharmaceutical ingredients.
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
The effect of the additives depends on the relative hydrophobic/hydrophilic
character of both the initial antibiotic and the additive. In fact, the sonochemical
process performs a more effective degradation of the hydrophobic solutes [22].
The octanol/water partition coefficient (log P) is an indicator of the hydrophobicity
of the substances. The compounds with a higher log P have a more hydrophobic
character. The coefficients for OXA, MAN and CC are 2.38, -3.10 and -4.01,
respectively [23]. The log P values for CC and MAN indicate that these additives
are highly hydrophilic substances. Thus, during the ultrasonic action, CC and MAN
are farther from the cavitation bubbles than the antibiotic. Additionally, the similar rd
(Table 1) and ra values for the water samples containing OXA, OXA + MAN and
OXA + CC (Figure 1C) suggest that the hydroxyl radicals that did not react with the
antibiotic and experiments recombination before reacting with the additives despite
their high relative concentration (ten times higher than oxacillin’s concentration).
Therefore, the evaluated additives did not compete with oxacillin for the hydroxyl
radicals. The aforementioned results evidence the selective elimination of the
oxacillin and its antimicrobial activity by means of the sonochemical treatment in
the presence of other pharmaceutical ingredients.
11

245
246
247
248
249
250
251
252
253
254
255
3.2 Comparison between sonochemistry and other oxidation processes
The individual effect of mannitol and calcium carbonate and the combination of
these additives on AA elimination during the sonochemical treatment were
compared with other oxidation processes. For such comparison, the photo-Fenton,
TiO2 photocatalysis and electrochemistry processes were considered (the latter
using Ti/IrO2 and NaCl as anode and supporting electrolyte). Photo-Fenton and
TiO2 photocatalysis are also advanced oxidation processes that generate and use
the hydroxyl radical as a degrading agent [24,25], while the considered
electrochemical system (in presence of chloride ions) produces hypochlorous acid
(HClO), which can easily attack the antibiotic [26,27].
256
257
258
259
260
261
262
For comparative purposes, the AA elimination kinetic constant was determined
(Table SM 1 in the supplementary material) and used as a quantitative indicator in
all conditions and processes. Therefore, the ratio between the kinetic constants (rk
= kOXA+ADDITIVE / kOXA) was calculated and plotted (Figure 2). An rk value higher than
1 indicates that the additive has an accelerating effect on the process, an rk value
equal to 1 indicates that it has no effect, and an rk value lower than 1 indicates it
has a detrimental effect.
263
264
265
266
The individual presence of the additives on the four oxidative systems produced
different results. As Figure 2 shows, in the presence of MAN, rk values lower than 1
were observed in the photo-Fenton (PF, rk ~ 0.40 or 60% inhibition), TiO2 photo-
catalysis (PC, rk ~ 0.40 or 60% inhibition) and electrochemical (EO, rk ~ 0.75 or
12

267
268
269
270
271
272
25% inhibition) processes. The negative effect of MAN can be associated with the
reaction between the additive and the generated oxidants. In fact, mannitol has a
high reaction constant with the hydroxyl radical (1.7 x 109 M-1s-1) [28], and HClO
easily oxidizes primary and secondary alcohols [29], such as those found in the
mannitol structure. In contrast, an rk value equal to 1 (no inhibition) was observed in
the sonochemical system (SC).
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
In the presence of calcium carbonate (such as bicarbonate ions at the
experimental pH) an rk approximately equal to 1 was found in the TiO2
photocatalytic treatment. The formation of bicarbonate radicals by means of hole or
hydroxyl radical action on bicarbonate anions had been reported during the TiO2
photocatalysis process (Eq. 7-8) [30,31]. Due to the fact that carbonate and
bicarbonate radicals are also oxidative species [32], they can degrade oxacillin
and, consequently, the AA elimination was not affected. Additionally, during the
TiO2 photocatalysis (which is a heterogeneous system), carbonate and bicarbonate
radicals could react with OXA in the bulk of the solution, avoiding the diffusional
limitations of the heterogeneous process. Otherwise, in the presence of
bicarbonate ions, the photo-Fenton process has an rk value of 0.57. Bicarbonate
ions are well-known scavengers of the hydroxyl radical (kreaction with •OH = 8.5 x 106 M-
1s-1, [33]). And, in contrast to the TiO2 photocatalysis, in the photo-Fenton process
(which is a system that does not have any diffusional limitations), the change of the
hydroxyl radical by means of the carbonate radical (Eq. 8) negatively affected the
process. Therefore, the antibiotic removal efficiency and, consequently, the AA
elimination are lower.
13

290
291
292
293
294
295
Moreover, the rk value during the electrochemical oxidation was 0.33. Due to the
basic properties of the bicarbonate, it could neutralize the electrogenerated
hypochlorous acid, diminishing the availability of the oxidant agent and inhibiting
the process. Again, the additive did not affect the efficiency of the sonochemical
process (an rk value approximately equal to 1 was obtained for the OXA + CC
combination).
+
-
•
296
297
h + HCO3 → HCO3
(Eq. 7)
(Eq. 8)
-
• -
•OH + HCO3 → CO3 + H2O
298
299
300
301
302
303
304
305
306
307
308
309
310
Frequently, pharmaceutical effluents are mixtures of active compounds, excipients,
stabilizers etc. [10], which can affect the efficiency of the systems . To provide a
better approximation to pharmaceutical wastewater, a mixture of OXA, MAN and
CC was considered. The four oxidation processes were applied and their AA
evolution was monitored. Thus, the rk values were calculated. The treatment of
water with OXA + MAN + CC presented the following rk values for the photo-
Fenton, TiO2 photocatalysis, electrochemical and sonochemical processes,
respectively: 0.09, 0.00, 0.19, and 0.97. Interestingly, even when both MAN and
CC were added to the solution, the sonochemical process remained unaffected
(Figure 2A). On the contrary, the TiO2 photocatalysis, photo-Fenton and
electrochemical treatments showed a high inhibition level (100, 90 and 81%,
respectively). Mannitol combined with calcium carbonate could increase the
consumption of degrading species and, consequently, the removal of oxacillin and
14

311
312
the elimination of its AA are more inhibited than when additives are individually
present.
313
314
315
316
317
Among the photo-Fenton, the TiO2 photocatalysis and the electrochemical
oxidation processes, the latter was the least affected. Although, the hydroxyl
radical and the hypochlorous acid react with MAN and CC, HOCl also has a high
affinity for secondary amines and thioether groups [29], such as those contained in
oxacillin. Therefore, the electrochemical system is inhibited to a lesser extent.
318
319
320
321
322
323
324
325
326
327
328
329
Figure 2A and all the aforementioned results show that the sonochemical process
is the most efficient for treating oxacillin. The cavitation system allows three
differential reaction modes according to the chemical nature of the pollutants: (I)
the volatile substances are pyrolyzed in the hot interior of the bubble, (II) the
hydrophobic compounds are degraded by the hydroxyl radicals available in the
bubble-solution interface, and (III) the hydrophilic pollutants are eliminated in the
bulk of the solution by the radicals ejected during the bubble implosion [9]. As OXA
is non-volatile and hydrophobic but additives have a hydrophilic nature, in the
synthetic pharmaceutical wastewater containing high concentrations of mannitol
and calcium carbonate, oxacillin is selectively degraded in the interfacial zone, as it
is shown in Figure 2B.
330
331
332
3.3 Proposed mechanism for the sonochemical elimination of OXA
To study the structural transformations that remove the antibiotic and eliminate its
antimicrobial activity during the sonochemical treatment, the initial intermediaries
15

333
334
335
336
337
338
were determined by means of HPLC-MS. 3,3-dimethyl-6-{[(5-methyl-3-
phenylisoxazol-4-yl)carbonyl]amino}-7-oxo-4-thia-1-azabicyclo[3.2.0]heptane-2-
carboxylic acid 4-oxide (BP1), two diastereoisomers of 2-(carboxy{[(5-methyl-3-
phenylisoxazol-4-yl)carbonyl]amino}methyl)-5,5-dimethyl-1,3-thiazolidine-4-
carboxylic acid (BP2 and BP3) and 5-methyl-3-phenylisoxazole-4-carboxamide
(BP4) were identified (Table 2).
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
Furthermore, the evolution of these by-products was monitored during the 360 min
of sonication (Figure 3). The by-product formation order indicates that the constant
intermediary production rates (k) are different (k1 > k2 > k4 > k3). As it is indicated in
Figure 3, BP1 is firstly formed. Indeed, the oxidation of the thioether group due to
the attack of •OH is known to be considerably fast [24,34,35]. Secondly, the
diastereoisomer BP2 is produced and, subsequently, BP3 is generated from BP2.
These two last by-products were also found in the degradation of oxacillin by
means of the non-thermal plasma and the photo-Fenton processes [36,37]. Due to
the fact that the central four-member ring of the penicillinic nucleus is significantly
strained and the carbonyl-nitrogen bond is weak [38], such ring is a highly reactive
site. Thus, the β–lactam ring is opened to form BP2, which could evolve into BP3
via enolization, which in turn is less sterically hindranced due to the opposed
stereocenters. Additionally, the action of the sonogenerated •OH caused the
breakdown of the secondary amide group in OXA, BP1, BP2 and BP3, producing
BP4. In fact, during the treatments of OXA by means of the TiO2 photocatalysis
and photo-Fenton processes, the attack of the hydroxyl radical also led to the
formation of BP4 [37,39]. Therefore, based on the intermediaries identified and
16

356
357
358
359
360
361
their evolution during the treatment, a pollutant degradation mechanism was
proposed (Figure 4). Interestingly, as it is shown by the chemical structure of all the
by-products (Table 2 and Figure 4), the central penicillinic nucleus of the oxacillin is
modified and/or removed during the sonochemical process. Such transformations
on the moiety (that is responsible for the biological activity of the pollutant) explain
the effective elimination of OXA and its antibiotic activity.
362
363
364
365
366
367
368
369
370
371
372
373
374
375
In addition to by-product formation, Figure 3 also shows that the prolongation of the
ultrasonic action to 240 min induced the removal of BP1, BP2 and BP3. After 360
min of treatment, the concentration of BP4 decreased ~ 55%. Beside the
elimination of the pollutant and its initial aromatic intermediaries, the by-products
that formed in the process presented the shortest retention time in the
chromatogram (Figure SM 8) and strong UV absorptions at wavelengths of 190-
220 nm (Figure SM 9). And another noteworthy fact is that these by-products are
commonly associated with aliphatic acids. Figure 3 indicates that the concentration
of such by-products increases with the time of treatment. Due to their hydrophilic
nature, these by-products are accumulated in the bulk of the solution, being less
susceptible to a radical attack [40] and, consequently, the system is not able to
completely mineralize them (i.e. transform them into water, carbon dioxide and
inorganic ions). Therefore, as it is indicated in Figure 3, the TOC of the solution
remains unchanged during the sonochemical process.
376
17

377
378
379
380
381
382
383
3.4 Complete mineralization of oxacillin by means of a sonochemical-
biological combination
As it was demonstrated, the sonochemical treatment did not mineralize the OXA
(Figure 3). However, in the presence of highly concentrated additives (Figure 1A),
this process is able to completely eliminate the AA, transforming OXA into highly
oxidized compounds (Figure 4). These results suggest that the solutions generated
during the ultrasonic action could be bio-treated.
384
385
386
387
388
389
390
391
392
393
394
395
396
397
To achieve the mineralization of sonotreated solutions, a biological process was
applied using an inoculum without previous adaptation (as described in section
2.1). Water samples containing OXA, MAN and CC and submitted to different
sonication times (0, 1 and 2 hours) were considered. Additionally, a control
experiment was assessed using a non-sonicated solution of MAN + CC without the
antibiotic. Figure 5 shows that the TOC of the solution containing MAN and CC
was completely removed after 19.5 h of biological treatment, evidencing the highly
biocompatible character of the additives. In a similar period of time, the
microorganisms were also able to remove ~80% of the total organic carbon from
the other solutions. However, the remaining 20% of the TOC was not removed
from the non-sonicated solution and the solution that was sonicated for 1h, not
even after 30 h of bio-treatment (Figure 5). This fact indicates that the biological
step initially removed the biodegradable fraction of the TOC, and the remaining
organic carbon is recalcitrant to microorganism action.
398
399
The residual OXA concentration of the solutions (non–sonicated and sonicated for
1 h) was established after 19.5 h of bio-treatment (Table 3). The results shown in
18

400
401
402
403
404
405
406
Figure 5 and Table 3 suggest that the presence of the antibiotic limits the
microorganism action, hindering the complete mineralization of these solutions.
Contrastingly, in the solution that was sonicated for 2h and from which the oxacillin
had been removed (Table 3), the biological system achieved ~99% of TOC
removal (Figure 5) after 30 h of treatment. Therefore, the sonication of the pollutant
for 120 min transforms it into biodegradable compounds that are mineralized by
means of a conventional biological process.
407
408
409
410
411
412
413
414
415
416
417
Furthermore, the time of residence in an aerobic reactor for practical applications
ranges between 24 h and 48 h [16]. Interestingly, after 30 h, the tested system was
able to achieve a high level of bio-treatability (Figure 5). Additionally, the time for
achieving complete mineralization could be reduced further using previously
adapted microorganisms, optimizing the operational conditions of the biological
reactor, and/or increasing the ultrasonic wave application time. Therefore, the
sonochemical system combined with the biological process is a promising
alternative for achieving the complete mineralization of the oxacillin pollutant in
presence of additives. In contrast to other AOPs, the sonochemical process is able
to selectively eliminate the antibiotic, generating by-products with no AA and
solutions that can be mineralized by means of a subsequent biological treatment.
418
419
Conclusions
420
421
This study shows that the application of ultrasonic waves to water containing
oxacillin is an efficient process to selectively degrade the pollutant and eliminate its
19

422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
antimicrobial activity. During the sonochemical process, the AA was eliminated
after 120 min. Additionally, the AA elimination kinetics followed an adequate
exponential model (R2 > 0.95). The presence of additives (mannitol, calcium
carbonate or their combination) did not affect the process efficiency. In fact, the
ratio between the antimicrobial activity elimination kinetic constants (rk =
kOXA+ADDITIVE / kOXA) for the sonochemical processes was approximately equal to 1
in all the cases. While the photo-Fenton process, the TiO2 photocatalysis and the
electrochemical oxidation mainly presented rk values lower than 1. This fact
highlights the selectivity of sonochemistry in the elimination of the antibiotic activity
from solutions containing the pollutant and other hydrophilic pharmaceutical
ingredients. Moreover, the identification of the main by-products showed that the
attack of the hydroxyl radical modified the penicillinic nucleus, which is the moiety
responsible for the activity of the OXA. Although the ultrasound action also
degraded the initial by-products, the process was unable to achieve the
mineralization of the initial pollutant. However, in the case of the solutions with
oxacillin, mannitol and calcium carbonate that were sonicated for 2h, a
conventional biological system with non-adapted microorganisms removed 99% of
the TOC. This fact indicates that the ultrasonic treatment was able to transform the
initial solution into compounds that are more susceptible to biotreatments.
Therefore, the combination of a biological system with a sonochemical process is a
promising alternative for the remediation of water containing oxacillin and other
pharmaceutical additives.
444
20

445
Acknowledgments
446
447
448
449
450
451
The authors would like thank to Colciencias and to the Swiss National Foundation
for the financial support to this study within the projects: 1115-569-33692,
“Implementación de metodologías eficientes y confiables para degradar residuos
de antimicrobianos en el hogar y en efluentes industriales” and "Treatment of the
hospital wastewaters in Cote d'Ivoire and in Colombia by advanced oxidation
processes", respectively.
452
References
453
454
455
[1]
[2]
[3]
J.L. Martinez, Environmental pollution by antibiotics and by antibiotic
resistance determinants., Environ. Pollut. 157 (2009) 2893–902.
doi:10.1016/j.envpol.2009.05.051.
456
457
458
S.K. Khetan, T.J. Collins, Human Pharmaceuticals in the Aquatic
Environment: A Challenge to Green Chemistry., Chem. Rev. 107 (2007)
2319–2364. doi:10.1021/cr020441w.
459
460
461
462
463
D.W. Hughes, C.R. Frei, P.R. Maxwell, K. Green, J.E. Patterson, G.E.
Crawford, et al., Continuous versus intermittent infusion of oxacillin for
treatment of infective endocarditis caused by methicillin-susceptible
Staphylococcus aureus, Antimicrob. Agents Chemother. 53 (2009) 2014–
2019.
464
465
466
467
[4]
[5]
[6]
J.M. Cha, S. Yang, K.H. Carlson, Trace determination of beta-lactam
antibiotics in surface water and urban wastewater using liquid
chromatography combined with electrospray tandem mass spectrometry., J.
Chromatogr. A. 1115 (2006) 46–57. doi:10.1016/j.chroma.2006.02.086.
468
469
470
471
R.M. Neto, M.A. Ansaldi, M.E.S.M. da Costa, S.O. da Silva, V.H.F. Luz, A
case report of a multi-drug resistant bacterial infection in a diabetic patient
treated in northeast Brazil., Diabet. Foot Ankle. 3 (2012) 1–6.
doi:10.3402/dfa.v3i0.18656.
472
473
474
E. Villaroel, J. Silva-Agredo, C. Petrier, G. Taborda, R. Torres-Palma,
Ultrasonic degradation of acetaminophen in water: effect of sonochemical
parameters and water matrix., Ultrason. Sonochem. 21 (2014) 1763–9.
21

475
476
477
478
479
[7]
P. Villegas-Guzman, J. Silva-Agredo, A.L. Giraldo-Aguirre, O. Flórez-Acosta,
C. Petrier, R. Torres-Palma, Enhancement and inhibition effects of water
matrices during the sonochemical degradation of the antibiotic dicloxacillin,
Ultrason. Sonochem. 22 (2015) 211–219.
doi:10.1016/j.ultsonch.2014.07.006.
480
481
482
[8]
[9]
W. Guo, Y. Shi, H. Wang, H. Yang, G. Zhang, Intensification of sonochemical
degradation of antibiotics levofloxacin using carbon tetrachloride, Ultrason.
Sonochem. 17 (2010) 680–684. doi:10.1016/j.ultsonch.2010.01.004.
483
484
485
Y.G. Adewuyi, Sonochemistryꢀ: Environmental Science and Engineering
Applications, Ind. Eng. Chem. Res. 40 (2001) 4681–4715.
doi:10.1021/ie010096l.
486
487
488
[10] C. Carlsson, A.-K. Johansson, G. Alvan, K. Bergman, T. Kühler, Are
pharmaceuticals potent environmental pollutants?, Sci. Total Environ. 364
(2006) 88–95. doi:10.1016/j.scitotenv.2005.06.036.
489
490
[11] Royal Pharmaceutical Society of Great Britain, Handbook of Pharmaceutical
Excipients, Sixth, Pharmaceutical Press, London, 2009.
491
492
493
[12] D.G.J. Larsson, C. de Pedro, N. Paxeus, Effluent from drug manufactures
contains extremely high levels of pharmaceuticals, J. Hazard. Mater. 148
(2007) 751–755. doi:10.1016/j.jhazmat.2007.07.008.
494
495
496
497
[13] C. Sirtoria, A. Zapata, W. Gernjak, S. Malato, A. Lopez, A. Agüera, Solar
photo-Fenton degradation of nalidixic acid in waters and wastewaters of
different composition. Analytical assessment by LC–TOF-MS, Water Res. 45
(2011) 1736–1744. doi:doi:10.1016/j.watres.2010.11.023.
498
499
500
501
[14] R.A. Torres, C. Pétrier, E. Combet, M. Carrier, C. Pulgarin, Ultrasonic
cavitation applied to the treatment of bisphenol A. Effect of sonochemical
parameters and analysis of BPA by-products., Ultrason. Sonochem. 15
(2008) 605–11.
502
503
[15] N.P. Vichare, P.R. Gogate, V.Y. Dindore, A.B. Pandit, Mixing time analysis of
a sonochemical reactor, Ultrason. Sonochem. 8 (2001) 23–33.
504
505
506
507
508
[16] E.A. Serna-Galvis, J. Silva-Agredo, A.L. Giraldo-Aguirre, R.A. Torres-Palma,
Sonochemical degradation of the pharmaceutical fluoxetine: Effect of
parameters, organic and inorganic additives and combination with a
biological system, Sci. Total Environ. 524-525 (2015) 354–360.
doi:10.1016/j.scitotenv.2015.04.053.
509
510
[17] C. Pétrier, R. Torres-Palma, E. Combet, G. Sarantakos, S. Baup, C.
Pulgarin, Enhanced sonochemical degradation of bisphenol-A by
22

511
512
bicarbonate ions., Ultrason. Sonochem. 17 (2010) 111–5.
doi:10.1016/j.ultsonch.2009.05.01.
513
514
515
516
[18] E. De Bel, J. Dewulf, B.D. Witte, H. Van Langenhove, C. Janssen, Influence
of pH on the sonolysis of ciprofloxacin: Biodegradability, ecotoxicity and
antibiotic activity of its degradation products, Chemosphere. 77 (2009) 291–
295. doi:10.1016/j.chemosphere.2009.07.033.
517
518
519
[19] W. Guo, H. Wang, Y. Shi, G. Zhang, Sonochemical degradation of the
antibiotic cephalexin in aqueous solution, Water SA. 36 (2010) 651–654.
doi:10.4314/wsa.v36i5.61998.
520
521
522
[20] S. Su, W. Guo, C. Yi, Y. Leng, Z. Ma, Degradation of amoxicillin in aqueous
solution using sulphate radicals under ultrasound irradiation, Ultrason.
Sonochem. 19 (2012) 469–474. doi:10.1016/j.ultsonch.2011.10.005.
523
524
525
526
[21] 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. doi:10.1016/j.jenvman.2012.01.010.
527
528
529
[22] Y. Jiang, C. Pétrier, T.D. Waite, Effect of pH on the ultrasonic degradation of
ionic aromatic compounds in aqueous solution., Ultrason. Sonochem. 9
(2002) 163–168. doi:10.1016/S1350-4177(01)00114-6.
530
531
[23] National Library of Medicine, ChemIDplus Advanced, ChemIDplus Adv.
(2013). http://chem.sis.nlm.nih.gov/chemidplus/.
532
533
534
535
[24] J.J. Pignatello, E. Oliveros, A. Mackay, Advanced Oxidation Processes for
Organic Contaminant Destruction Based on the Fenton Reaction and
Related Chemistry, Crit. Rev. Environ. Sci. Technol. 36 (2006) 1–84.
doi:10.1080/10643380500326564.
536
537
538
539
[25] 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.
doi:10.1016/j.jphotochem.2004.11.006.
540
541
542
543
544
[26] A.L. Giraldo, E.D. Erazo-Erazo, O.A. Flórez-Acosta, E.A. Serna-Galvis, R.A.
Torres-Palma, Degradation of the antibiotic oxacillin in water by anodic
oxidation with Ti/IrO2 anodes: Evaluation of degradation routes, organic by-
products and effects of water matrix components, Chem. Eng. J. 279 (2015)
103–114. doi:10.1016/j.cej.2015.04.140.
545
546
[27] I. Sirés, E. Brillas, Remediation of water pollution caused by pharmaceutical
residues based on electrochemical separation and degradation
23

547
548
technologiesꢀ: A review, Environ. Int. 40 (2012) 212–229.
doi:10.1016/j.envint.2011.07.012.
549
550
551
552
[28] G. V Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical-review of
rate constants for reactions of hydrated electrons, hydrogen-atoms and
hydroxyl radicals (.OH/.O-) in aqueous-solution, J. Phys. Chem. Ref. Data.
17 (1988) 513–886. doi:10.1063/1.555805.
553
554
555
[29] M. Deborde, U. von Gunten, Reactions of chlorine with inorganic and organic
compounds during water treatment-Kinetics and mechanisms: A critical
review, Water Res. 42 (2008) 13–51. doi:10.1016/j.watres.2007.07.025.
556
557
558
[30] A. Kumar, N. Mathur, Photocatalytic degradation of aniline at the interface of
TiO2 suspensions containing carbonate ions., J. Colloid Interface Sci. 300
(2006) 244–52. doi:10.1016/j.jcis.2006.03.046.
559
560
561
562
563
[31] D.E. Santiago, J. Araña, O. González-Díaz, M.E. Alemán-Dominguez, A.C.
Acosta-Dacal, C. Fernandez-Rodríguez, et al., Effect of inorganic ions on the
photocatalytic treatment of agro-industrial wastewaters containing imazalil,
Appl. Catal. B Environ. 156-157 (2014) 284–292.
doi:10.1016/j.apcatb.2014.03.022.
564
565
566
[32] P. Mazellier, C. Busset, A. Delmont, J. De Laat, A comparison of fenuron
degradation by hydroxyl and carbonate radicals in aqueous solution, Water
Res. 41 (2007) 4585–4594. doi:10.1016/j.watres.2007.06.066.
567
568
569
570
[33] G. V Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical Review of
Rate Constants for Reactions of Hydrated Electrons, Hydrogen Atoms and
Hydroxyl Radicals (OH/O−) in Aqueous Solution, At. Energy. 17 (1988) 513–
886. doi:10.1063/1.555805.
571
572
573
574
[34] X. He, S.P. Mezyk, I. Michael, D. Fatta-Kassinos, D.D. Dionysiou,
Degradation kinetics and mechanism of β-lactam antibiotics by the activation
of H2O2 and Na2S2O8 under UV-254nm irradiation., J. Hazard. Mater. 279
(2014) 375–383.
575
576
577
578
[35] Y.Y. Gurkan, N. Turkten, A. Hatipoglu, Z. Cinar, Photocatalytic degradation
of cefazolin over N-doped TiO2 under UV and sunlight irradiation: Prediction
of the reaction paths via conceptual DFT, Chem. Eng. J. 184 (2012) 113–
124. doi:10.1016/j.cej.2012.01.011.
579
580
581
[36] M. Magureanu, D. Piroi, N.B. Mandache, V. David, a. Medvedovici, C. Bradu,
et al., Degradation of antibiotics in water by non-thermal plasma treatment,
Water Res. 45 (2011) 3407–3416. doi:10.1016/j.watres.2011.03.057.
24

582
583
584
585
586
[37] E. a. Serna-Galvis, J. Silva-Agredo, A.L. Giraldo, O. a. Flórez, R. a. Torres-
Palma, Comparison of route, mechanism and extent of treatment for the
degradation of a β-lactam antibiotic by TiO2 photocatalysis, sonochemistry,
electrochemistry and the photo-Fenton system, Chem. Eng. J. 284 (2016)
953–962. doi:10.1016/j.cej.2015.08.154.
587
588
589
590
[38] C. Mavronikola, M. Demetriou, E. Hapeshi, D. Partassides, C. Michael, D.
Mantzavinos, et al., Mineralisation of the antibiotic amoxicillin in pure and
surface waters by artificial UVA- and sunlight-induced fenton oxidation, J.
Chem. Technol. Biotechnol. 84 (2009) 1211–1217. doi:10.1002/jctb.2159.
591
592
593
594
595
596
[39] A.L. Giraldo-Aguirre, E.D. Erazo-Erazo, O.A. Flórez-Acosta, E.A. Serna-
Galvis, R.A. Torres-Palma, 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, J. Photochem. Photobiol. A Chem. 311
(2015) 95–103. doi:10.1016/j.jphotochem.2015.06.021.
597
598
599
[40] R. Singla, F. Grieser, M. Ashokkumar, The mechanism of sonochemical
degradation of a cationic surfactant in aqueous solution, Ultrason.
Sonochem. 18 (2011) 484–488. doi:10.1016/j.ultsonch.2010.09.013.
600
601
25

602
603
604
Figure captions
605
606
607
608
Figure 1. Sonochemical elimination of antimicrobial activity (AA) for oxacillin in the
presence of additives. (B) Linearized form of the AA elimination kinetics. (C) H2O2
accumulation rates (ra). [OXA] = 47.23 µmol L-1; [MAN] = [CC] = 472.3 µmol L-1; 60
W (275 kHz); pH = 5.6.
609
1.0
A
0.8
0.6
0.4
AA OXA
AA OXA+MAN
AA OXA+CC
0.2
0.0
0
20
40
60
80
100
120
Time (min)
610
26

3.0
2.8
2.6
2.4
2.2
2.0
1.8
1.6
1.4
1.2
1.0
0.8
B
OXA - R2 = 0.9984
OXA + MAN - R2 = 0.9966
OXA + CC - R2 = 0.9965
0
20
40
60
80
100
120
Time (min)
611
612
3.0
2.5
2.0
1.5
1.0
0.5
0.0
C
DW
OXA
OXA+MAN
OXA+CC
613
614
615
616
27

617
618
619
620
621
622
623
624
Figure 2. (A) Relative effects of the additives on antimicrobial activity elimination by
means of the four oxidation processes. [OXA] = 47.23 µmol L-1; [MAN] = [CC] =
472.3 µmol L-1; pH = 5.6. Photo-Fenton system (PF): [Fe (II)] = 90 µmol L-1; [H2O2]
= 1000 µmol L-1; 150 W. TiO2 photocatalysis (PC): [TiO2] = 0.5 g L-1; 150 W.
Sonochemical treatment (SC): 60 W (275 kHz). Electrochemical oxidation (EO):
[NaCl] = 0.065 mol L-1; Ti/IrO2 anode; J = 5 mA cm−2. (B) Schema of the selective
elimination of OXA by means of the sonochemical process.
625
626
A
1.2
1.0
0.8
OXA + MAN
OXA + CC
OXA + MAN + CC
0.6
0.4
0.2
0.0
PF
PC
SC
EO
627
628
629
630
631
632
633
634
28

635
B
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
29