Heterogeneous catalytic ozonation of ciprofloxacin in aqueous solution using a manganese-modified silicate ore

Article abstract of DOI:10.1039/C8RA06880A

Manganese modified silicate ore (MnSO) prepared using an impregnation method was used as a heterogeneous ozonation catalyst, and the catalytic activity was evaluated by the degradation of ciprofloxacin (CIP). The results showed that the manganese oxide was successfully loaded onto natural silicate ore (SO). The degradation and mineralization efficiencies of CIP were considerably improved in the presence of MnSO. Under optimal conditions, the CIP removal process followed the pseudo-first-order reaction model well. The degradation rate constant of MnSO/O₃ was 1.7 times and 3.3 times higher than those of SO/O₃ and only O₃, respectively. During the ozonation of the CIP aqueous solution in the presence of MnSO, the TOC removal rate reached 61.2% at 60?min, but was only 30.8% using ozonation alone. The addition of tert-butanol (TBA) significantly inhibited the degradation efficiency of CIP, which indicated that catalytic ozonation of MnSO followed a hydroxyl radical (·OH) reaction mechanism. Furthermore, MnSO showed great stability and durability over several reaction cycles.

Full text of DOI:10.1039/C8RA06880A

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Heterogeneous catalytic ozonation of
ciprofloxacin in aqueous solution using
a manganese-modified silicate ore†
Cite this: RSC Adv., 2018, 8, 33534
ab
c
b
Donglei Zou,a Dongwei Lu,
Bingjing Xin,
Ming Zhou,b
*
Lisha Luo,
c
c
*
Xuedong Zhai and Jun Ma
Manganese modified silicate ore (MnSO) prepared using an impregnation method was used as
a heterogeneous ozonation catalyst, and the catalytic activity was evaluated by the degradation of
ciprofloxacin (CIP). The results showed that the manganese oxide was successfully loaded onto natural
silicate ore (SO). The degradation and mineralization efficiencies of CIP were considerably improved in
the presence of MnSO. Under optimal conditions, the CIP removal process followed the pseudo-first-
order reaction model well. The degradation rate constant of MnSO/O3 was 1.7 times and 3.3 times
higher than those of SO/O3 and only O3, respectively. During the ozonation of the CIP aqueous solution
in the presence of MnSO, the TOC removal rate reached 61.2% at 60 min, but was only 30.8% using
ozonation alone. The addition of tert-butanol (TBA) significantly inhibited the degradation efficiency of
CIP, which indicated that catalytic ozonation of MnSO followed a hydroxyl radical ($OH) reaction
mechanism. Furthermore, MnSO showed great stability and durability over several reaction cycles.
Received 17th August 2018
Accepted 11th September 2018
DOI: 10.1039/c8ra06880a
rsc.li/rsc-advances
on its surface to accelerate subsequent reactions between the
$OH and pollutant.8 Manganese oxide is among the most widely
1. Introduction
studied metal oxides as an ozonation catalyst because of its
unique characteristics, such as environmental friendliness,
diverse crystallographic structure, facile fabrication, and
abundance in soil.12–14 As demonstrated in this study, manga-
nese oxide supported by various porous materials (carbon-
based multi-pore materials,15,16 honeycomb ceramics,17 and
mesoporous molecular sieve18) can effectively enhance
mechanical performance and provide more catalytic active
sites. In addition, porous materials supported by manganese
oxide can be easily separated, avoiding the waste of catalyst and
secondary pollution. However, the carbon-based material is
easily oxidized by ozone and can generate a large quantity of
residual sludge that needs to be processed.19–21 In addition, the
production of synthetic materials may cause certain pollution
problems to the environment and increase costs.22 These
disadvantages limit the application of carbon-based materials
and synthetic catalysts in full-scale systems.
Natural mineral materials have been reported as heteroge-
neous ozone catalysts or supports because they are environ-
mentally friendly and of low cost. Previous studies have
indicated that natural mineral materials and their modied
forms have been investigated for their potential to catalyze the
ozonation of various organic contaminants in water and
wastewater including bauxite,21 magnetite ore,22 brucite,23
goethite,24 and perovskite.25 However, the specic surface area
of some natural mineral materials is small, and the adsorption
performance of the target materials is poor;21–25 thus, they
In recent years, antibiotics have been widely developed and
increasingly used in the aquaculture, animal farming, and
human pharmaceutical industries.1,2 Most antibiotics enter
aquatic systems and lead to chemical pollution and resistant
gene pollution.3 Ciprooxacin (CIP), as the second generation of
uoroquinolones, has been extensively used because of its
capacity for efficient disease control and enhancing animal
growth.4,5 The level of CIP concentration detected in the envi-
ronment has increased from ng Lꢀ1 to mg Lꢀ1.6 Conventional
biological wastewater treatment processes used in sewage
treatment plants (STPs) are not effective for the removal of CIP
because of its poor biodegradability.7 Therefore, it is urgent and
necessary to develop an effective technology to remove CIP.
Heterogeneous catalytic ozonation, which combines ozone
with a solid catalyst, is an effective advanced oxidation tech-
nology for the degradation and mineralization of CIP.8–11 The
solid catalyst can accelerate the decomposition of ozone to
hydroxyl radical ($OH) and can also adsorb organic pollutants
aKey Laboratory of Ministry of Education for Groundwater Resources and Environment,
College of New Energy and Environment, Jilin University, Changchun 130000, P. R.
China
bJilin Institute of Chemical Technology, Jilin 130022, P. R. China
cState Key Laboratory of Urban Water Resource and Environment, Harbin Institute of
Technology, Harbin 150090, P. R. China. E-mail: lvdongwei@hit.edu.cn
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c8ra06880a
33534 | RSC Adv., 2018, 8, 33534–33541
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
provide limited catalytic active sites. Therefore, it is necessary to a Fourier transform infrared (FT-IR) spectrum instrument
seek a natural mineral material that simultaneously possesses (Spectrum 2000, Perkin Elmer, USA). X-ray photoelectron spec-
a large surface area and a strong adsorption capacity for cata- troscopy (XPS, Kratos-AXIS ULTRA DLD) using Al Ka as the
lytic ozonation of pollutants.
source was used to determine the surface components of the
In this study, we used a type of natural silicate ore as the composite. The characterization results of SEM, XRD, BET and
support to fabricate a composite catalyst. MnOx was selected as FT-IR are provided in ESI (Fig. S1–S4 and Table S1 †).
the active component of the catalyst. A composite catalyst was
prepared using the impregnation method, and the MnSO ob-
tained was characterized using different techniques. The
inuence of catalytic dose, solution pH, CIP concentration, and
the reaction temperature on the degradation and mineraliza-
tion of CIP were studied. Furthermore, a possible reaction
mechanism for ozonation of CIP with MnSO is discussed.
2.4 Catalytic ozonation procedure
Batch catalytic ozonation degradation experiments of CIP were
conducted in a 500 mL at bottom beaker. A microporous
titanium diffuser was connected at the bottom of the reactor to
diffuse the ozone gas entering the reactor. Ozone was produced
via a DHX-I ozone generator (Harbin Jiujiu Electrochemistry
Technology Co., Ltd., China). The inlet ozone gas concentration
was approximately 0.4 mg minꢀ1 and the ozone ow rate was
300 mL minꢀ1. During each semi-intermittent experiment,
400 mL of CIP aqueous solution (20 mg Lꢀ1) and a certain
quality catalyst were added to the beaker. During the whole
process, the slurry was continuously stirred to ensure that the
catalyst was completely uidized in the reaction solution. The
experiment was conducted at 20 ꢁC using a cryostat. The water
samples were collected at regular intervals and ltered through
cellulose acetate lters (0.22 mm) before analysis of CIP and total
organic carbon (TOC) concentrations. A small amount of
sodium thiosulfate solution was added as the termination
agent. To test the stability of the MnSO, the catalyst was
collected by centrifugal separation aer the catalytic ozonation
reaction, washed several times with deionized water, and dried
at 70 ꢁC for 24 h before reuse. The pH value of the solution was
regulated using normal HNO3 and NaOH in all reactions for
a comparison. To obtain effective experimental data, the
experiment was repeated three times, and the results were
averaged.
2. Materials and methods
2.1 Chemicals
Silicate ore was obtained from advanced oxidation technology
research center of the Harbin Institute of Technology. CIP was
purchased from Sigma-Aldrich ($98%). Manganese nitrate was
purchased from the Tianjin Baishi Chemical Reagent Co., Ltd,
tert-butyl alcohol (TBA) was purchased from the Tianjin Ruijinte
Chemical Reagent Co., Ltd. Sodium thiosulphate was supplied
by the Tianjin Reagent Chemical Co., Ltd. All chemicals used in
this study were analytic grade and used without further puri-
cation. Deionized water was used during the experiment. The
pH of the solution was adjusted with nitric acid and sodium
hydroxide.
2.2 Preparation of MnSO catalysts
Modied natural silicate ore was prepared using an impregna-
tion method. Detailed preparation information is as follows.
First, the natural silicate ore particles were washed with
deionized water several times to remove surface dust, and then
grinded to 60–80 mesh through a high-speed universal mill.
Second, 5 g of crushed natural silicate ore was soaked in
Mn(NO3)2 (0.5 mol Lꢀ1, 100 mL) solution, and oscillated at
a constant speed (120 rpm) for 24 h at 30 ꢁC. Following
2.5 Analysis methods
CIP concentration was analyzed using high-performance liquid
chromatography (HPLC; Agilent 1200 Series, DAD detector and
XDB-C18 column) with a UV detector at 276 nm. The mobile
phase was acetonitrile and 0.1% formic acid (88 : 12, v/v) with
a 1.0 mL minꢀ1 ow rate.
The rst-order model was used to simulate the degradation
kinetics of CIP, Ct/C0 ¼ exp(ꢀkt). Here, k (minꢀ1) is the pseudo-
rst-order rate constant of CIP; t (min) is the reaction time; and
C0 and Ct denote the CIP concentrations in the solution at the
initial concentration and a specic time point, respectively.
Total organic carbon (TOC) was measured using a total
organic carbon analyzer. (TOC-VCPH, Shimadzu Japan). The
concentration of leaching manganese was measured using
inductively coupled plasma (ICP) (Perkin-Elmer optima
5300DV).
ꢁ
precipitation and ltration, the silicate ore was dried at 70 C
ꢁ
for 36 h before calcination treatment at 500 C for 5 h in air.
Then, the MnSO samples were cooled to room temperature and
stored in a desiccator before use.
2.3 Catalyst characterization
Scanning electron microscopy (SEM, Hitachi SU8000, Japan)
with an acceleration voltage of 15 kV was used for analyzing the
morphology of the samples, and the catalyst surface was sputter
coated with gold. The crystalline shape of the catalyst was
analyzed using X-ray diffraction (XRD, D/max-RA, Rigaku),
operating with a Cu anode at 40 kV and 100 mA. The 2 theta data
from 10ꢁ to 80ꢁ were obtained at a scanning speed of 0.03ꢁ sꢀ1
.
The specic surface area, total pore volume, and average pore
size of the catalysts were measured using a nitrogen gas
adsorption analyzer (ASAP-2020, Micromeritics, USA) based on
3. Results and discussion
3.1 Characterization of catalysts
the Brunauer–Emmett–Teller (BET) method. Infrared spectra of The morphologies of the SO and MnSO were observed by
the catalysts were measured under ambient conditions using scanning electron microscope (Fig. S1†). As shown in Fig. S1a,†
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 33534–33541 | 33535

View Article Online
RSC Advances
Paper
Fig. 1 EDS spectrum of (a) SO, (b) MnSO, and (c)–(f) EDS mapping images of the MnSO.
the surface of the SO was rough and uneven. Many particles and Mn2O3 (642.2 eV),30,31 which were consistent with the
were piled atop one another and interrelated. Pores were aforementioned XRD results. This indicated that the MnO2 and
abundant. As shown in Fig. S1b,† the surface of the MnSO Mn2O3 were successfully doped into the SO. As shown in Fig. 2d,
gradually smoothed with the MnOx introduction, probably the O 1s peaks at 530.7 eV, 531.3 eV, 532.7 eV, and 531.9 eV were
owing to the MnOx on the surface of the SO.
assigned to the Mn–O, Al–O, Si–O, and OH(ads), respectively.
Fig. 1a and b show the EDS images of the SO and MnSO. The The OH(ads) is benecial for the generation of $OH in catalytic
results conrmed the presence of the Mn element in the MnSO. ozonation.32
The mass ratio of the surface Mn element for MnSO was
As shown in Fig. S3,† the N2 adsorption–desorption
13.35%. Based on the EDS mapping image as shown in Fig. 2c isotherms of the SO and MnSO were both identied as type IV
the manganese was uniformly distributed on the MnSO surface. according to the International Union of Pure and Applied
The formation of metal oxides and their effective distribution Chemistry classication. This means that the SO and MnSO
play an important role in MnSO degrading refractory organic
pollutants.26 The results indicated that the Mn element was
successfully doped in the SO catalyst. XRD of the SO and MnSO
sample is shown in Fig. S2.† In the pattern of the SO, the
diffraction peaks matched with the published XRD pattern of
the cristobalite (JCPDS no. 89-3607) and a-quartz (JCPDS no. 85-
1054). Compared to the SO, there were some unique diffraction
peaks (2q ¼ 28.8ꢁ, 37.5ꢁ, 56.8ꢁ, 32.9ꢁ and 55.1ꢁ) which corre-
sponded to the (110), (101), (221), (222), and (440) faces,
respectively, on the MnSO sample. These diffraction peaks
belong to MnO2 and Mn2O3.27,28 Therefore, it can be deduced
that the MnOx were effectively loaded on the SO surface.
The surface elemental composition and chemical states of
the SO and MnSO were further investigated using XPS. As
shown in Fig. 2a, compared to the characteristic peaks of the
SO, there not only were characteristic peaks corresponding to Si
2p, C 1s, O 1s and Al 2p, but also to Mn 2p in the XPS spectrum
of the MnSO composites, which conrms the presence of Mn in
the MnSO composites. Fig. 2b shows the peak at 102.9 eV cor-
Fig. 2 XPS spectra analysis of the SO and MnSO catalyst. (a) Wide-
responding to Si 2p.29 Fig. 2c shows the spectrum of Mn 2p3/2
range scan of the SO and MnSO; (b) Si 2p, (c) Mn 2p, and (d) O 1s of the
MnSO composites.
shows three peaks, corresponding to MnO2 (641.1 and 643.4 eV)
33536 | RSC Adv., 2018, 8, 33534–33541
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
catalyst surface were mesoporous structures.33 The BET of both was relatively high efficient, widespread, easily prepared and
the SO and MnSO are listed in Table S1.† The specic surface low cost, which has a better application prospect.
areas of the SO and MnSO composite were determined to be
In addition, the TOC removal efficiency obtained over 60 min
75.56 and 57.82 m2 gꢀ1, respectively. The results showed that during the MnSO catalytic ozonation process (61.2%) was
the loading of MnOx resulted in a decrease in the average pore signicantly higher than that during the SO catalytic ozonation
volume, specic surface area, and pore size of the SO. This (53.6%) and single ozonation (30.8%) processes. The adsorp-
illustrated that the manganese oxide dispersed into the pores of tion process still had a great effect on the removal of TOC. The
the SO and blocked them.
improvement of CIP degradation and TOC removal may have
The FT-IR spectra were measured to investigate the surface been a result of the following. First, the MnSO catalyst had
function groups of the SO and MnSO. As shown in Fig. S4,† the a good adsorption capacity which could accelerate subsequent
absorption peak at 1105 cmꢀ1 originated from Si–O.21 The reactions between the ozone and pollutant.37 Second, the pres-
bands at 478 cmꢀ1 were attributed to the bending and stretch- ence of the MnSO catalyst could promote decomposition of
ing modes of AlO6.34 The presence of water molecules was ozone to a hydroxyl radical ($OH).8 The $OH was non-selective
supported by the appearance peaks of the bending vibration and could enhance the CIP and TOC removal. However, the
absorption of H–O–H at 1635 cmꢀ1 and the stretching mode removal rate of TOC was signicantly lower than that of the
stretching vibration of O–H at 3419 cmꢀ1
.
degradation of CIP. This result was because a large number of
intermediates produced during the CIP degradation process
were difficult to degrade via ozone. Although CIP can be largely
removed, the intermediate products cannot be completely
converted into CO2 and H2O, thus reducing the removal effect of
the TOC.
35,36
3.2 Adsorption and catalytic ozonation of CIP by MnSO
To evaluate the catalytic activity of the catalyst, the degradation
and mineralization of CIP under different experimental condi-
tions including adsorption, single ozonation, and catalytic
ozonation experiments were determined as shown in Fig. 3.
Aer 30 min, the adsorption efficiencies in the presence of SO
and MnSO were approximately 25.3% and 17.5%, respectively.
This type of natural SO and MnSO had good adsorption
performance for the CIP, which beneted further ozonation
reactions. The degradation of CIP reached 53.% over 30 min
using a single ozonation process, whereas it increased to 74.9%
and 92.1% by adding SO and MnSO during the catalytic ozon-
ation process, respectively. The results showed that MnSO was
an effective heterogeneous ozonation catalyst, which increased
the degradation efficiency of CIP by approximately 39.1% as
compared to the single ozonation process. As shown in Table
S2,† the CIP degradation rates by adsorption, single ozonation,
and catalytic ozonation followed the pseudo-rst-order kinetic
model. The CIP degradation rate constant with MnSO/O3 was
1.7 times higher than that of SO/O3, and 3.3 times higher than
that of single ozonation. The results show that MnSO enhanced
the catalytic activity in the ozonation of CIP. In comparison with
the catalyst of MnOx/MNCNT16 and MnOx/MCM-41 (ref. 18) for
refractory pollutant oxidative degradation, MnSO in this work
3.3 Inuence of the catalyst dosage on the catalytic
ozonation process
The catalyst dosage relates to the reaction rate of the ozone with
an organic reaction, and also relates to the water treatment cost
in practical application. Therefore, it is necessary to discuss the
optimum amount of catalyst. As shown in Fig. 4, the degrada-
tion and mineralization efficiency of CIP signicantly increase
with an increase in catalyst dose. The CIP removal efficiencies
improved from 82% to 92.5% and the removal efficiency of the
TOC increased by 20%.
The CIP degradation kinetic constants during the MnSO/O3
process were enhanced from 0.058 minꢀ1 to 0.094 minꢀ1 as the
catalyst dosage increased from 0.2 g Lꢀ1 to 1.0 g Lꢀ1. Simulta-
neously, the TOC removal rate constants by MnSO/O3 also
increased from 0.007 minꢀ1 to 0.014 minꢀ1 with the catalyst
dosage increasing from 0.2 g Lꢀ1 to 1.0 g Lꢀ1. These results
indicated that increasing the amount of catalyst in the solution
was conducive to improving the specic surface area of the
catalyst in the solution and facilitating decomposition of ozone
Fig. 3 Degradation and mineralization of CIP by ozonation alone and Fig. 4 Effect of catalyst dosage on the removal of CIP and TOC.
with MnSO. Experimental conditions: pH ¼ 7.0; C0 ¼ 20 mg Lꢀ1; ozone (Experimental conditions: pH ¼ 7.0; C0 ¼ 20 mg Lꢀ1; ozone gas flow
gas flow rate ¼ 0.3 L minꢀ1; ozone gas concentration ¼ 0.4 mg minꢀ1
;
rate ¼ 0.3 L minꢀ1; ozone gas concentration ¼ 0.4 mg minꢀ1; T ¼
catalyst dose ¼ 0.5 g Lꢀ1; T ¼ 20 ꢁC.
20 ꢁC).
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 33534–33541 | 33537

View Article Online
RSC Advances
Paper
molecules into more $OH.38 However, the removal efficiency
and mineralization efficiency of CIP were not signicantly
improved when the addition of the catalyst was 1.0 g Lꢀ1
because of the limited amount of ozone in the system. In
addition, when the catalytic dose continued to increase, the
particles collided with each other, thus affecting the effective
contact between the catalyst and ozone.
3.4 Inuence of CIP initial concentration on catalytic
ozonation process
Fig. 6 Effect of reaction temperature on the removal of CIP and TOC.
(Experimental conditions: pH ¼ 7.0; C0 ¼ 20 mg Lꢀ1; ozone gas flow
rate ¼ 0.3 L minꢀ1; ozone gas concentration ¼ 0.4 mg minꢀ1; catalyst
dose ¼ 0.5 g Lꢀ1).
Fig. 5 shows the effect of the initial concentration on the CIP
and TOC removal during the MnSO catalytic ozonation process.
It can be seen that the CIP or TOC removal decreased with
increasing initial concentration. The CIP degradation kinetic
constant was 0.104 minꢀ1 with the initial concentration of
10 mg Lꢀ1, and it was approximately twice that with an initial
concentration of 40 mg Lꢀ1. This result indicates that when the
initial concentration of CIP was low, the amount of dissolved
ozone in the solution and the amount of $OH from the
decomposition of ozone in the solution were high. They can
rapidly degrade CIP and also result in a higher mineralization
effect. When the initial concentration of CIP increases, the
ozone is rapidly utilized by the high concentration of CIP. At the
same time, the intermediate product competes with ozone for
CIP, thus the degradation effect of CIP was signicantly
reduced.39
temperature was 20 ꢁC, the degradation rate was the fastest.
However, when the temperature continued to increase to 35 ꢁC,
the removal efficiency was suppressed. Generally, a higher
reaction temperature would lead to a faster reaction rate for
a chemical reaction and mass transfer. In addition, a tempera-
ture increase can accelerate the decomposition of ozone and
produce more hydroxyl radicals, which is benecial to
improving the catalytic ozonation effect. Nonetheless, the
solubility of the ozone in water could decrease at higher
temperatures. During the process of catalytic ozonation, the
increase in reaction temperature results in these two opposing
effects.40,41 Therefore, the reaction temperature of 20 ꢁC has
a positive effect on CIP degradation and mineralization during
the process of ozonation of MnSO.
The removal rate of the TOC corresponded to the trend in the
catalytic ozonation degradation rate of CIP, but it was still much
lower than that of the CIP degradation. It can be inferred that
the speculation produced a large number of intermediates that
were more resistant to ozonation than CIP.
3.6 Inuence of initial pH on the catalytic ozonation process
In the catalytic ozonation system, the initial pH value was the
key factor that affected the ozone decomposition rate and
hydroxyl formation.8,41 In addition, the charge of the organic
compounds in the solution varied under different pH condi-
tions.40,41 Fig. 7 shows the inuence of the initial pH value on
the CIP degradation and TOC removal efficiencies in catalytic
ozonation. As shown in Fig. 7, the CIP removal efficiency clearly
improved with an increase in the initial pH from 3.7 (79.5%) to
10.3 (95.8%). This was because in the alkaline solution more
$OH were yielded which was benecial to organic decomposi-
tion.8,42 A similar trend was found in the TOC removal. At initial
pH values of 3.7, 5.2, 7, 8.5, and 11.3, the TOC removals were
44.5%, 52.7%, 61.5%, 65.4%, and 73.2% aer 60 min. The CIP
3.5 Inuence of reaction temperature on the catalytic
ozonation process
Fig. 6 shows the inuence of reaction temperature (10, 20, and
35 ꢁC) on the catalytic activity of the MnSO for ozonation of CIP.
As shown in Fig. 6, when the reaction temperature increased
from 10 ꢁC to 20 ꢁC, the degradation efficiencies of CIP and TOC
greatly increased from 82.3% to 92.1% and 33.1 to 61.3%,
respectively. The kinetic information for CIP degradation and
mineralization is shown in Fig. 6. It shows that when the
Fig. 5 Effect of initial concentration of the target on removal of CIP Fig. 7 Effect of pH on the removal of CIP and TOC. (Experimental
and TOC. (Experimental conditions: pH ¼ 7.0; ozone gas flow rate ¼ conditions: C0 ¼ 20 mg Lꢀ1; ozone gas flow rate ¼ 0.3 L minꢀ1; ozone
0.3 L minꢀ1; ozone gas concentration ¼ 0.4 mg minꢀ1; catalyst dose ¼ gas concentration ¼ 0.4 mg minꢀ1; catalyst dose ¼ 0.5 g Lꢀ1; T ¼
0.5 g Lꢀ1; T ¼ 20 ꢁC).
20 ꢁC).
33538 | RSC Adv., 2018, 8, 33534–33541
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
degradation kinetic constants of the MnSO/O3 process were produced during the process of catalytic ozonation decompo-
enhanced from 0.0043 minꢀ1 to 0.100 minꢀ1 as the pH sition. The reaction rate constant of TBA and hydroxyl radicals
increased from 3.7 to 10.3. Simultaneously, the TOC removal was 6.0 ꢂ 108 Mꢀ1 Sꢀ1, which was much higher than the reac-
rate constants by MnSO/O3 also increased from 0.010 minꢀ1 to tion rate constant of 3.0 ꢂ 10ꢀ3 Mꢀ1 Sꢀ1 for TBA and ozone.12 As
0.018 minꢀ1 as the pH increased from 3.7 to 10.3.
shown in Fig. S6,† the effects of TBA on the degradation of CIP
Therefore, even though the intermediates of CIP were resis- were investigated. Aer the addition of TBA (100 mg Lꢀ1), the
tant to ozonation, they could be further oxidized or even CIP degradation was signicantly inhibited. When 100 mg Lꢀ1
mineralized by $OH generated from the O3 decomposition. The of TBA was added into the aqueous solution, reductions of
increased production of $OH during ozonation was an effective 54.5% and 28.9% for the degradation efficiency during the
method to improve the TOC removal.
processes of catalytic ozonation and ozonation alone at 30 min
were noted, respectively. It was shown that during the process of
degradation of CIP, TBA and CIP competition for $OH produced
intermediates of high selectivity and inertia, ending the chain
reaction of free radicals.41 This hindered the reaction of $OH
and CIP, resulting in a reduction in the CIP degradation rate in
the system. At the same time, it was also indirectly proven that
the catalytic ozonation degradation of the CIP system by MnSO
was the mechanism of hydroxyl radical action.
To further explore the reaction mechanism, O3 was replaced
by the N2 in same experimental setup. As shown in Fig. S7,†
under the MnSO/N2 system, the removal of CIP was 18.2%.
When 100 mg Lꢀ1 TBA was added to the MnSO/N2 system, the
removal of CIP remained unchanged. It indicated that there was
no $OH in the water, and the removal of CIP was probably due to
the adsorption of catalyst. In comparison, MnCO/O3 system had
signicantly higher CIP removal. However, the addition of TBA
greatly inhibited the degradation of CIP in MnCO/O3 system.
Base on above experiment results, it can be concluded that O3
decomposition was the sole source of $OH under MnSO cata-
lytic ozonization.
3.8.2 Mechanisms of catalytic ozonation by MnSO. From
the aforementioned discussion, it was quite obvious that MnSO
showed superior adsorption performance and perfect catalytic
activity in heterogeneous catalytic ozonation to remove CIP. The
oxidation reaction on the surface of the catalyst was the most
important step in catalytic ozonation. When ozone or organic
molecules were adsorbed on the catalyst surface, the oxidation
reaction was signicantly accelerated. To illustrate the possible
interaction relation of ozone and the MnSO catalyst in the
ozonation of CIP and its intermediates, a simplied mechanism
showing the catalytic ozonation of CIP with the MnSO catalyst is
shown in Fig. 8. Ozone can directly attack CIP in aqueous
solution to generate CO2 and H2O, but the rate of direct reaction
is much less than that of the oxidation of the hydroxyl radicals.
3.7 Catalyst stability and reusability
During the application process of the catalyst, the catalyst may
have some problems such as the dissolution of the active
component and damage to the active sites, which can lead to
a signicant decrease in its activity. Therefore, in the hetero-
geneous catalytic ozonation system, among the important
measures to evaluate the effect of the catalyst was its reuse rate.
Recycling experiments were conducted to evaluate the activity
and stability of MnSO catalyst.
As shown in Table 1, the CIP and TOC removal efficiency
slightly decreased with the increase in the catalyst reuse times.
CIP degradation was reduced by ca. 3.0% and the TOC removal
by ca. 2.9%. It was found that aer four times of reuse, the
MnSO catalyst still had good catalytic activity on CIP degrada-
tion and mineralization. The concentration of the released
manganese ions in solution repeated over four times were
detected to be 0.097 mg Lꢀ1, 0.088 mg Lꢀ1, 0.076 mg Lꢀ1, and
0.069 mg Lꢀ1, respectively. Although slight leaching of manga-
nese ions was found, the catalytic capacity of the MnSO was
hardly affected. The dissolution concentration of manganese
ions was less than the limit in the drinking water of China
(0.1 mg Lꢀ1).43 Thus, it had less effect on the safety of drinking
water. We have characterized the catalyst by XRD aer reactions
(Fig. S5†). The results showed that the diffraction peaks of the
catalyst remained unchanged before and aer the reaction. This
further demonstrated the good stability of the catalyst. There-
fore, it could be concluded that the MnSO had great catalytic
activity and stability, and is a promising catalyst for the removal
of antibiotic wastewater.
3.8 Possible reaction mechanism during the MnSO process
3.8.1. Radical scavenging experiment. TBA is a well-known
hydroxyl radical scavenger that can capture hydroxyl radicals
Table 1 Stability and activity of the MnSO in the catalytic ozonation of
CIP
CIP
TOC
Reuse time
degradationa (%)
removalb (%)
c (Mn)a/mg Lꢀ1
0
1
2
3
92.1
90.4
89.7
89.1
60.2
58.9
57.8
57.3
0.097
0.088
0.076
0.069
a
b
Reaction time is 30 min. Reaction time is 60 min.
Fig. 8 Reaction mechanism of the MnSO process.
RSC Adv., 2018, 8, 33534–33541 | 33539
This journal is © The Royal Society of Chemistry 2018

View Article Online
RSC Advances
Paper
The existence of $OH resulted in successive oxidation reactions
of CIP and its intermediates and promoted TOC removal.
The chain reactions of ozone decomposition were as
followed:19
4. Conclusions
Manganese modied natural SO catalysts were successfully
prepared using the impregnation method and showed
remarkable catalytic activity for the degradation and minerali-
zation of CIP. The characterizations of the MnSO catalyst
showed that MnOx was distributed on the surface of the SO. The
CIP removal rate with an MnSO catalyst at 30 min could reach
92.1%, but just 53% using O3 alone. The removal rate of TOC
was 30.4% higher during the catalytic ozonation than with O3
alone. During MnSO catalytic ozonation, with the increase in
catalyst dosage and the initial solution pH, the CIP and TOC
removal efficiency in aqueous solution both increased.
However, the capability of the catalyst for CIP degradation and
TOC removal decreased with an increase in the initial concen-
tration of CIP. The MnSO catalytic ozonation followed the $OH
production mechanism. The CIP removal was attributed to
a synergistic effect of the adsorption, ozone oxidation, and $OH
oxidation. The activity of MnSO on CIP degradation and
mineralization slightly decreased with repeated use several
times. This was because of the leaching of the manganese ion
from MnSO into solution, while the MnSO catalyst maintained
excellent activity. This study may provide information on an
inexpensive, viable, and promising catalyst towards effective
catalytic ozonation of antibiotics from wastewater for practical
application.
O3 + H2O / 2$OH + O2, k2 ¼ 1.1 ꢂ 10ꢀ4 Mꢀ1 sꢀ1
O3 + OHꢀ / O2cꢀ + HO2c, k2 ¼ 70 Mꢀ1 sꢀ1
O3 + $OH / O2 + HO2c 4 O2cꢀ + H+
O3 + HO2c / 2O2 + $OH, k2 ¼ 1.6 ꢂ 109 Mꢀ1 sꢀ1
2HO2c / O2 + H2O2
(1)
(2)
(3)
(4)
(5)
Furthermore, the adsorption of CIP and the intermediates
on the MnSO catalyst contributed to the increase in the CIP and
TOC removal, which is also benecial to the subsequent
oxidation process.
Consequently, in this study, there were three pathways for
heterogeneous catalytic ozonation as follows:41 (1) chemisorp-
tion of ozone on the catalyst surface was decomposed by the
formation of surface oxygenated radical species, which
promoted the formation of hydroxyl radicals. The hydroxyl
radicals rapidly removed CIP and the intermediates. (2) Organic
molecules (CIP or intermediates) adsorbed on the surface of the
catalyst were oxidized by gaseous or aqueous ozone. (3)
Adsorption of both ozone and organic molecules (CIP or inter-
mediates) on the surface of the MnSO catalyst and their direct
and indirect reactions occurred between ozone and CIP, leading
to a rapid degradation of CIP.
Conflicts of interest
There are no conicts to declare.
The reactions of CIP degradation can be simplied as:44
(I). Indirect (radical) reactions.
(a). Catalyst surface reaction:
Acknowledgements
This work was nancially supported by the National Science and
Technology Major Projects for Water Pollution Control and
Treatment (Grant No. 2017ZX07201003); Natural Science
Foundation and Technology Department of Jilin Province,
China (20150101093C); and Technology Department Key
Research of Jilin Province, China (20140204038SF).
MnSOꢀ$OH + CIP / MnSO + CO2+H2O + intermediates (6)
MnSOꢀ$CIP + $OH / MnSO + CO2+H2O + intermediates (7)
(b). Bulk solution reaction:
CIP + $OH / intermediates + CO2 + H2O
(8) References
1 D. Li, X. Guo, H. Song, T. Sun and J. Wan, J. Hazard. Mater.,
2018, 180, 957.
(II). Direct (molecular O3) reactions.
(a). Catalyst surface reaction:
2 T. Paul, P. L. Miller and T. J. Strathmann, Environ. Sci.
Technol., 2007, 41, 4720.
MnSOꢀ$O + CIP / MnSO + CO2+H2O + intermediates (9)
3
3 H. Xiong, S. Dong, J. Zhang, D. Zhou and B. E. Rittmann,
Water Res., 2018, 136, 75.
MnSOꢀ$CIP + O3 / MnSO + CO2+H2O + intermediates (10)
4 H. Nakata, K. Kanaan, P. D. Jones and J. P. Giesy,
Chemosphere, 2005, 58, 759.
(b). Bulk solution reaction:
5 D. W. Kolpin, E. T. Furlong, M. T. Meyer, E. M. Thurman,
S. D. Zaugg, L. B. Barber and H. T. Buxton, Environ. Sci.
Technol., 2002, 36, 1202.
6 D. G. Larsson, P. C. De and N. Pedro, J. Hazard. Mater., 2007,
148, 751.
7 P. Zheng, B. Bai, W. S. Guan, H. L. Wang and Y. R. Suo, RSC
Adv., 2016, 6, 4101.
CIP + O3 / intermediates + CO2+H2O
(11)
In addition, according to the catalyst characterization results
shown in Fig. 2 and S2,† different states of manganese on the
surface of MnSO led to more electron transfer opportunities and
promoted its high catalytic activity.45
33540 | RSC Adv., 2018, 8, 33534–33541
This journal is © The Royal Society of Chemistry 2018

View Article Online
Paper
RSC Advances
´
8 B. Kasprzyk-Hordern, M. Ziołek and J. Nawrocki, Appl. Catal., 27 F. Arena, G. Truno, J. Negro, B. Fazio and L. Spadaro,
B, 2003, 46, 639. ChemInform, 2007, 38, 2269.
9 J. Nawrockia and B. Kasprzyk-Hordern, Appl. Catal., B, 2010, 28 C. K. Sang and G. S. Wang, Appl. Catal., B, 2010, 98, 180.
99, 27.
29 B. M. Reddy, B. Chowdhury and P. G. Smirniotis, Appl.
Catal., A, 2001, 211, 19.
30 W. Wang, J. Yu, J. Zou and X. Yu, Electrochim. Acta, 2016, 191,
426.
31 D. Li, T. Sun, L. Wang and N. Wang, Electrochim. Acta, 2018,
282, 416.
32 X. Tan, Y. Wan, Y. Huang, C. He, Z. Zhang, Z. He, L. Hu,
J. Zeng and D. Shu, J. Hazard. Mater., 2017, 321, 162.
33 H. Yan, P. Lu, Z. Pan, X. Wang, Q. Zhang and L. Li, J. Mol.
Catal. A: Chem., 2013, 377, 57.
34 A. D. Kiadehi, A. Ebadi and A. Aghaeinejad-Meybodi, Sep.
Purif. Technol., 2017, 182, 238.
35 Y. Xu, H. Bai, G. Lu, C. Li and G. Shi, J. Am. Chem. Soc., 2008,
130, 5856.
36 Z. Ji, X. Shen, J. Yang, G. Zhu and K. Chen, Appl. Catal., B,
2014, 144, 454.
10 L. Yang, C. Hu, Y. Nie and J. Qu, Appl. Catal., B, 2010, 97, 340.
11 H. M. Jalali, Mater. Sci. Eng., C, 2016, 59, 924.
12 B. Roshani, I. Mcmaster, E. Rezaei and J. Soltan, Sep. Purif.
Technol., 2014, 135, 158.
13 F. Nawaz, H. Cao, Y. Xie, J. Xiao, Y. Chen and Z. A. Ghazi,
Chemosphere, 2017, 168, 1457.
14 R. Andreozzi, A. Insola, V. Caprio, R. Marotta and V. Tufano,
Appl. Catal., A, 1996, 138, 75.
15 J. Ma, M. Sui, T. Zhang and C. Guan, Water Res., 2005, 39,
779.
16 M. Sui, S. Xing, L. Sheng, S. Huang and H. Guo, J. Hazard.
Mater., 2012, 227–228, 227.
17 W. Chen, X. Li, Z. Pan, S. Ma and L. Li, Sep. Purif. Technol.,
2017, 173, 99.
18 M. Sui, J. Liu and L. Sheng, Appl. Catal., B, 2011, 106, 195.
19 R. Huang, H. Yan, L. Li, D. Deng, Y. Shu and Q. Zhang, Appl. 37 S. P. Tong, W. P. Liu, W. H. Leng and Q. Q. Zhang,
Catal., B, 2011, 106, 264. Chemosphere, 2003, 50, 1359.
20 X. Lu, Q. Zhang, W. Yang, X. Li, L. Zeng and L. Li, RSC Adv., 38 S. P. Ghuge and A. K. Saroha, J. Environ. Manage., 2018, 211,
2015, 14, 10537. 83.
21 F. Qi, B. Xu, L. Zhao, Z. Chen, L. Zhang, D. Sun and J. Ma, 39 Z. Liu, J. Tu, Q. Wang, Y. Cui, L. Zhang, X. Wu, B. Zhang and
Appl. Catal., B, 2012, 121–122, 171. J. Ma, Sep. Purif. Technol., 2018, 200, 51.
22 G. Moussavi, R. Khosravi and N. R. Omran, Appl. Catal., A, 40 C. A. Orge, J. J. M. Orfao, M. F. R. Pereira, A. M. D. D. Farias,
2012, 445, 42. R. C. R. Neto and M. A. Fraga, Appl. Catal., B, 2011, 103, 190.
23 K. He, Y. Dong, L. Yin, A. Zhang and Z. Li, Acta Geochim., 41 L. Zhao, J. Ma, Z. Sun and H. Liu, Appl. Catal., B, 2009, 89,
2006, 25, 587.
326.
24 J. S. Park, H. Choi and J. Cho, Water Res., 2004, 38, 2284.
42 T. Zhang, J. Mol. Catal. A: Chem., 2008, 279, 82.
25 N. Sano, T. Yamamoto, D. Yamamoto, S. I. Kim, A. Eiad-Ua, 43 Ministry of Health of the People's Republic of China,
H. Shinomiya and M. Nakaiwa, Chem. Eng. Process., 2007, 46,
513.
26 H. Zhang, F. Ji, Y. Zhang, Z. Pan and B. Lai, Sep. Purif.
Technol., 2017, 193, 368.
Standards for drinking water quality (GB5749-2006), 2006.
44 G. Moussavi, A. A. Aghapour and K. Yaghmaeian, Chem. Eng.
J., 2014, 249, 302.
45 Y. Huang, Y. Sun, Z. Xu, M. Luo, C. Zhu and L. Li, Sci. Total
Environ., 2017, 575, 50.
This journal is © The Royal Society of Chemistry 2018
RSC Adv., 2018, 8, 33534–33541 | 33541