Accepted Manuscript

Degradation of UV-filter Benzophenone-3 in Aqueous Solution Using Persul-

fate catalyzed by Cobalt Ferrite

Xiaoxue Pan, Liqing Yan, Chenguang Li, Ruijuan Qu, Zunyao Wang

PII: S1385-8947(17)31019-7
DOI: http://dx.doi.org/10.1016/j.cej.2017.06.068
Reference: CEJ 17149

To appear in: Chemical Engineering Journal

Received Date: 6 April 2017

Revised Date: 13 June 2017
Accepted Date: 14 June 2017

Please cite this article as: X. Pan, L. Yan, C. Li, R. Qu, Z. Wang, Degradation of UV-filter Benzophenone-3 in
Aqueous Solution Using Persulfate catalyzed by Cobalt Ferrite, Chemical Engineering Journal (2017), doi: http://
dx.doi.org/10.1016/j.cej.2017.06.068

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.
Degradation of UV-filter Benzophenone-3 in Aqueous Solution
Using Persulfate catalyzed by Cobalt Ferrite

Xiaoxue Pana, Liqing Yanb, Chenguang Lia, Ruijuan Qua*, Zunyao Wanga

a
State Key Laboratory of Pollution Control and Resources Reuse, School of the Environment,

Nanjing University, Jiangsu Nanjing 210023, P.R. China

b
Environmental Engineering School of Civil and Environmental Engineering, Georgia Institute of

Technology, Atlanta, Georgia 30332-0373, USA

*
Corresponding author: Ruijuan Qu. Tel: +86-25-89680358; Fax: +86-25-89680358. E-mail:
quruijuan0404@nju.edu.cn.

1
Abstract

The occurrence of benzophenone-3 (BP-3) in aquatic environments constitutes a

potential risk to the environment and human health due to its phytotoxicity,
carcinogenicity and endocrine disrupting effects. In this work, cobalt ferrite (CoFe2O4)
prepared by chemical co-precipitation method was used as a catalyst for the
degradation of BP-3. The catalyst was characterized by SEM, TEM, XRD, XPS, BET
and FT-IR spectroscopy, and its catalytic activity on BP-3 removal by persulfate (PS)
was then evaluated at different operating parameters (catalyst loading, reaction
temperature, solution pH and PS dose). The removal of BP-3 (1.31 µM) in 6 h
reached 91% under the condition of [BP-3]0: [PS]0 = 1 : 1000, initial pH = 7.0, T = 25
o
C and catalyst load = 500 mg L-1. In addition to the satisfactory catalytic
performance, the CoFe2O4 also showed high stability and excellent recyclability.
Electron paramagnetic resonance and radical quenching tests showed that sulfate
radicals predominated in the decomposition of BP-3 by PS/CoFe2O4, while hydroxyl
radicals also contributed to the catalytic oxidation process. Fifteen degradation
products were identified by liquid chromatography-mass spectrometry (LC-MS), and
two reaction pathways involving hydroxylation, demethylation, direct oxidation and
benzene ring opening were proposed. This study could provide useful information for
the potential application of CoFe2O4 activated PS technology to treat water and
wastewaters containing BP-3.

Keywords: Benzophenone-3; Cobalt Ferrite; Persulfate; Catalytic degradation;

Reaction mechanisms

2
1. Introduction
Benzophenone-3 (BP-3), a typical hydroxylated derivative of benzophenone, can
absorb sunlight in the UVA and UVB regions with limited phototransformation. This
property leads to its widely application as sunlight filter in personal care products
(PCPs) and as photostabiliser in polymeric materials to protect humans and materials
against harmful exposure to UV irradiation. Due to the use in recreational water
activities or the incomplete elimination in wastewater treatment plants (WWTP) [1],
BP-3 has been widely detected in WWTP effluents [2], lakes [3], rivers [4], tap waters
[5], and sediments [6] at the concentrations ranging from several ng L-1 to µg L-1. As a
persistent and bio-accumulative compound (LogKow = 3.79) [3], BP-3 constitutes a
potential risk to the environment and human health. Considering that BP-3 has
potential endocrine disrupting effect [7-9] and high phytotoxicity and carcinogenicity
[10], it is necessary to effectively remove BP-3 in aquatic environment.
In the past few years, sulfate radicals-based advanced oxidation processes
(SR-AOPs) have received wide attention in water treatment. Sulfate radicals (SO4-•),
which are strong oxidants with a redox potential of 2.5 - 3.1 V, could achieve a higher
mineralization degree of organic pollutants in the oxidation process [11,12]. Because
they are more selective and stable than hydroxyl radicals, SO4-• are promising
alternatives to •OH for advanced oxidation treatment of compounds in aqueous
solutions. Recent studies have demonstrated the efficiency of SR-AOPs in removing a
broad spectrum of organic pollutants, such as cylindrospermopsin [13], flumequine
[14], tetrabromobisphenol A [15] and PCBs [16]. However, the information about the
SO4-•-based degradation process of BP-3 is still limited.
Since it is stable, soluble, cheap (< $2.65/kg for Na2S2O8 [17]) and owns high
redox potential (2.01 V), persulfate (PS, S2O82-) is widely used in laboratory
researches as a source of SO4-• [18]. Various methods such as heat [19], UV [20,21],
transition metal ions (Fe2+, Co2+, Cu2+) [22] or metal oxides [23], and quinones [16]
can activate PS to produce SO4-•. Currently, the key to improve the efficiency of

3
SR-AOPs is to develop highly effective activation methods for PS. Metaloxide
nanoparticles have been widely used in AOPs for decades [24,25]. The fine
physicochemical characteristics of cobalt ferrite (CoFe2O4) nanoparticles make it
suitable to be used as magnetic material in electric films and as photocatalyst in
pollutants removal [26,27]. However, at present, there is little information on cobalt
ferrite activated PS oxidation, which limits the application of this technology to
degrade persistent organic pollutants.
In this study, we investigated the degradation kinetics and reaction mechanisms
of BP-3 by PS activated with CoFe2O4 catalyst in aqueous solution. The catalyst was
synthesized and characterized, and its catalytic performance was then assessed under
different conditions. Two different radical scavengers were adopted to clarify the
contributions of radical species in the reaction system. Based on the identified
degradation products of BP-3, the reaction pathways were proposed. Additionally, the
mineralization degree of reaction solutions was evaluated by total organic carbon
(TOC) measurement, and the ecotoxicity of BP-3 and its degradation products was
estimated using Ecological Structure Activity Relationships (ECOSAR) program. To
the best of our knowledge, this is the first attempt to employ CoFe2O4 as a catalyst of
PS for the degradation of BP-3.
2. Materials and methods
2.1 Chemicals
BP-3 (CAS NO. 131-57-7, purity 98%) and the spin-trapping agent
5,5-dimethyl-pyrroline-N-oxide (DMPO) was obtained from J&K and Aladdin
Chemistry, respectively. Potassium persulfate (K2S2O8, purity ≥ 99.5%) was provided
by Nanjing Chemical Reagent Co., Ltd. (Nanjing, China). The chromatographic grade
formic acid and methanol were acquired from Merck (Darmstadt, Germany). The rest
of reagents used in the present work were of analytical grade or higher. Ultrapure
water (18.2 MΩ cm) was prepared with a Milli-Q water purification system (Millipore,
Bedford, USA).
2.2 Synthesis and characterization of CoFe2O4
The CoFe2O4 catalyst was synthesized using the chemical co-precipitation
4
method, according to Gao et al. [28]. First, 3.00 g Co(NO3)2•6H2O and 5.50 g
FeCl3•6H2O were dissolved into 100 mL ultrapure water (Co2+ : Fe3+ = 1 : 2 at a
molar concentration ratio) with continuous mechanical and ultrasonic stirring at 62 oC.
Second, 8 M NaOH solution was added dropwise into the mixture solution until the
pH value of the solution was approximately 10. Then, stop stirring and keep the
mixture solution at 75 oC for 30 min. After cooling down to room temperature, the
suspension was filtered and washed with ultrapure water at least three times. Finally,
the material was calcined at different temperatures (100, 200, 300, 400, 500 and 600
o
C) for 4 h in the muffle furnace and then ground to powders for further use.
The morphology and internal structure of CoFe2O4 was observed by scanning
electron microscope (SEM, Quanta FEG 250, FEI, the Netherlands) and transmission
electron microscopy (TEM, JEM-200CX, JEOL, Japan), respectively. Meanwhile,
energy dispersive spectroscopy (EDS) analysis was carried out during SEM
characterization to identify the chemical component. X-ray powder diffraction (XRD)
spectrum was analyzed by an X’TRA X-ray diffractometer (ARL, Switzerland). The
metal valance state was identified by X-ray photo-electron spectroscopy (XPS, Al
K-Alpha, Thermo Fisher Scientific, USA). The specific surface area and pore size of
CoFe2O4 were measured by a Brunauer-Emmett-Teller (BET) analyzer (ASAP 2020,
USA). The Fourier-transform infrared (FT-IR) spectrum was recorded on a Bruker
Vertex 70 FT-IR spectrometer (Bruker Optics Inc., Billerica, MA, USA). The leaching
of Co 2+ and Fe3+ were determined by an atomic absorption spectrophotometer (AAS,
Solar M6, Thermo Fisher Scientific, USA).
2.3 Oxidation procedure
In a typical run, conical flasks containing 50 mL reaction solution were kept at a
controlled temperature in a water bath shaker (SHA-B, Jintan, China) to conduct the
experiments. The initial concentration of BP-3 was 1.31 µM. After 30 min of
adsorption equilibrium between BP-3 and CoFe2O4, the oxidation reaction was
initiated by adding a desired amount of PS. The pH in the solution was adjusted by
H2SO4 or NaOH solution using a Mettler-Toledo S20 SevenEasy pH Meter. The
reason for selecting H2SO4 was that SO42- can also be generated in PS oxidation
5
process so that no other anions such as Cl-, NO3-, H2PO4- and HPO4- were introduced
into the reaction. After 0, 30, 60, 90, 120, 180, 240 and 360 min, 800 µL of reaction
solution was withdrawn and filtered through 0.22 µm organic filters into 1.5 mL
brown sample vials, and 200 µL of 100 mM NaNO2 solution was immediately added
to terminate the reaction. To evaluate the stability and reusability, the catalyst after
reaction was collected by a magnet, washed and dried, and then reused five times
under the same reaction conditions.
To ensure the accuracy of data in this study, the tests were performed in triplicate
and the average values were presented in the figures.
2.4 Analytical methods
The concentration of BP-3 was analyzed at 239 nm by an Agilent 1200 high
performance liquid chromatography (HPLC) system (Agilent Technologies, Palo Alto,
CA, USA) equipped with a quaternary pump and a UV detector. A Zorbax
300SB-C18 column (4.6 × 150 mm, 5 µm) was used for the separation of analytes,
and the column temperature was maintained at 30 oC. The mobile phase was 0.3%
formic acid in water (30%) and methanol (70%) at a flow rate of 1 mL min-1. The
injection volume was 20 µL.
To detect the degradation intermediates, the reaction solution was purified by an
SPE workstation (Supleco, USA) using an Oasis HLB cartridge (Waters, USA)
according to the following procedure [29-31]: The SPE columns were activated with 5
mL methanol and 5 mL Milli-Q water in sequence, loaded with 2 mL samples, washed
with 5 mL Milli-Q water and then dried under vacuum. After eluting with 2 mL
methanol, the extract was collected for product analysis. The LC-MS analysis was
carried out on an Agilent 1260 infinity high performance liquid chromatography
system coupled with a high-resolution hybrid quadrupole time-of-flight mass
spectrometer (LCMS-Triple TOF 5600, AB Sciex, Foster City, CA). The electrospray
ionization (ESI) source was operated in both positive and negative ion mode. The
chromatographic separation was conducted on a Thermo BDS Hypersil C18 column
(100 × 2.1 mm, 2.4 µm). The mobile phase consisting of 0.1% formic acid in water (A)
and methanol (B) was eluted at 0.2 mL min-1.
6
 Electron paramagnetic resonance (EPR) spectra were recorded by a Bruker
EMX-10/12 spectrometer (German), and 5,5-dimethyl-1-pyrroline N-oxide (DMPO)
was used as a spin-trapping reagent for radical detection.
Total organic carbon (TOC) contents in the reaction samples were determined
using a TOC analyzer (OI-1030D, USA).
2.5 Toxicity evaluation
The acute and chronic aquatic toxicity of BP-3 and its degradation products by
PS/CoFe2O4 was predicted by ECOSAR program (version 1.11) (US EPA, 2012). The
estimation of the hazard indices was performed on three organisms, including the 96 h
half lethal concentration (LC50) for fish, the 48 h LC50 for Daphnid and the 96 h half
effective concentration (EC50) for green algae. For the compound with multiple
classifications, the most conservative toxicity level was used for the selection of
ECOSAR class [32].
3. Results and Discussion
3.1 Characterization of CoFe2O4
Since the CoFe2O4 catalyst calcined at 200 oC showed the best catalytic
performance (discussed below), this material was chosen for detailed characterization.
As shown in Fig. 1, the catalyst particles were in the nanometer level. According to
EDS analysis, the catalyst was composed of Co, Fe and O, and their molar ratio was
nearly 1 : 2 : 4, suggesting that the chemical formula was CoFe2O4. Furthermore,
element maps indicated that the distribution of these elements was even. The results
from TEM (Fig. 2) showed the catalyst particles have nearly spherical morphology
with a diameter of 10 to 20 nm.
The structural evolution of CoFe2O4 samples calcined at different temperatures
was studied by the XRD spectrum (Fig. 3). The sample treated at 100 oC was almost
amorphous, as there was no obvious diffraction peaks in its diffractogram. For the
sample at 200 oC and 300 oC, a few diffraction peaks in the X-ray diffraction pattern
were indexed into the cubic CoFe2O4 (JCPDS Card No. 22-1086, Fd-3m (227)),
suggesting they have a semi-crystalline structure. To be specific, the diffraction peaks
at 2θ = 32.4°, 35.4°, 57.4° and 62.5° were individually assigned to crystal plane (220),
7
(311), (711) and (440). When the calcining temperature was 400 oC, the material
crystallized and all reflections observed correspond well to a pure cubic phase of
CoFe2O4 with a type spinel structure [33]. As confirmed by the XRD spectra of
samples at 500 and 600 oC, the increase in the calcination temperature further
improved the crystallinity of the desired phase. Varma et al. [34] found that the
formation of catalyst was mainly dependent on calcination temperature, and there was
no obvious peak in the XRD spectrum of CoFe2O4 when the calcination temperature
was below 300 oC. By contrast, a spinel structure of CoFe2O4 can be formed above
400 oC [35]. At calcination temperature > 600 oC, the sample exhibited a single phase
of polycrystalline spinel structure with no preferred crystallite orientation [36], and
metal oxides like Fe2O3 and Co 3O4 were introduced into CoFe2O4 particles [35].
Fig. 4 and Fig. 5 show the XPS spectra of the CoFe2O4 catalysts calcined at
different temperatures, and all photoelectron lines recorded were calibrated to the C 1s
line at 284 eV. The binding energy of O 1s, Fe 2p and Co 2p in the sample prepared at
200 oC were located at 529.850 eV, 712.700 eV and 779.870 eV, respectively (Fig S7
Survey), consistent with the results of Xu and Meng [37,38]. All XPS of Co 2p
regions can be fitted into four contributions: Co 2p3/2, Co 2p1/2 and their shakeup
satellites (Fig. 4A) [39,40]. In addition, the intensity ratios of Co 2p3/2 and Co 2p1/2
were very similar in the six samples. The peaks at 779.905 eV and 795.005 eV in the
samples calcined at 100, 500 and 600 oC were ascribed to the binding energies of Co
2p3/2 and Co 2p1/2, respectively. Since the low-spin Co3+ cation gives rise to much
weaker satellite features than does high-spin Co2+ with unpaired valence 3d electron
orbitals, the presence of a large amount of Co 2+ species in the sample can be
confirmed by the intense Co 2p 3/2 shakeup satellite and the symmetry of Co 2p1/2
peaks [39,41]. The peaks at 779.605 eV, 781.705 eV and 795.005 eV in the samples
calcined at 200, 300 and 400 oC were ascribed to the binding energies of Co 2p 3/2, Co
2p3/2 and Co 2p1/2, respectively. The two Co 2p3/2 peaks were assigned to Co (II) in
octahedral sites and Co (III) in octahedral sites, respectively [42].
Likewise, the peaks at 710.005 eV, 712.205 eV and 723.505 eV were assigned to
Fe 2p 3/2, Fe 2p3/2 and Fe 2p1/2, respectively, demonstrating Fe3+ species existed in
8
more than one coordination environment. The tetrahedral site (Fe3+A) was observed at
higher binding energy, while the octahedral site (Fe3+B) at lower binding energy. As
shown in Fig. 4B, the ratio of Fe3+A and Fe3+B was nearly 1:1 at 100 oC, suggesting
total Fe3+ cations lie in the two sites of CoFe2O4 nanoparticles [43]. Moreover, we can
see that the content of Fe3+A decreased with increasing temperature. The shoulder on
the lower binding energy side of the main peak indicated the occurrence of a small
amount of Fe (II) [44].
The O 1s XPS spectra of the six CoFe2O4 nanoparticle catalysts were shown in
Fig. 5. The peak at 529.095 eV is typical of the metal-oxygen bond (Olatt), while the
peak at 531.065 eV is commonly associated with OH- belonging to the defect-oxide or
hydroxyl-like group (Oads) [45-48]. As the calcination temperature elevated, the
content of Oads decreased but the Olatt content increased. The temperature of 100 oC
can only remove the physically adsorbed water [49]. The high calcination temperature
decreased the catalyst activity, which was also reported by Wang et al. [50].
According to the nitrogen adsorption/desorption isotherm (Fig S1), the BET
specific surface area and pore size of CoFe2O4 caclined at 200 oC were calculated as
207.3167 m2 g-1 and 7.47448 nm, respectively. The calcination temperature has a
significant influence on the porous structure of the catalyst. With increasing
calcination temperature, the particle size becomes larger, and correspondingly the
surface area decreased (Fig. 6).
The FT-IR spectrum of the catalysts prepared at different temperatures was
shown in Fig. 7. The peaks at 580 cm-1 in the six samples were assigned to the Fe-O
group [51]. Because of the relatively low absorption wavenumbers, the Co-O
vibration was not observed. Although the peaks were observed at 1616 and 3397 cm-1,
especially the catalyst at 100 oC, the stretching vibration became weaker and weaker
with the increasing calcining temperatures, suggesting there was rare adsorbed water
in the catalysts prepared at relatively higher temperatures, in agreement with the O 1s
results.
<Fig. 1.>
<Fig. 2.>
9
<Fig. 3.>
<Fig. 4.>
<Fig. 5.>
<Fig. 6.>
<Fig. 7.>
3.2 Catalytic degradation of BP-3 by CoFe2O4 activated PS oxidation
The performance of the CoFe2O4 catalyst was evaluated under different reaction
conditions (Fig. 8). As we all know, calcining temperature is crucial for catalysts and
can partly decide their properties. In this study, the catalyst calcined at different
temperatures (100 - 600 ºC) was compared for their removal efficiency of BP-3 (Fig.
8A). When the concentration of catalyst is 250 mg L-1, the removal efficiency of BP-3
after 6 h was 59% for 200 ºC CoFe2O4. By contrast, the other catalysts presented little
effect on BP-3 degradation as compared with PS alone, and the removal rate after 6 h
were 35%, 31%, 29%, 24% and 21% for CoFe2O4 calcined at 100, 300, 400, 500 and
600 ºC, respectively. The reason was that different calcination temperatures resulted
in different structures of catalysts. As suggested by Varma et al. [34], the formation of
catalyst was mainly dependent on calcination temperature. At a lower cacination
temperature, the larger surface area of the smaller-size particles could provide more
active sites to produce more radicals, thus accelerating BP-3 degradation. For 100 ºC
CoFe2O4, the peak at 3397 cm-1 in the FT-IR spectrum suggests that the catalyst was
not completely dehydrated. The metal hydroxides have a weak ability to activate PS,
thus leading to a lower degradation efficiency of BP-3. Since CoFe2O4 calcined at 200
ºC showed the best catalytic performance, kinetic tests were performed using this
material in the following study.
As shown in Fig. 8B, the degradation efficiency of BP-3 was improved
dramatically with the increasing catalyst loading. To be specific, the removal rates
after 6 h were 49%, 59% and 75% at the catalyst concentration of 100, 250 and 500
mg L-1, respectively, while only 18% of BP-3 was degraded by single PS oxidation.
The research results of Sathishkumar et al. [27] also showed that the reaction rate
constant of reactive red 120 in the PMS/CoFe2O4/TiO2 system was nearly three times
10
than that without CoFe2O4/TiO2. It was likely that the increased number of active sites
at a higher catalyst loading could catalyze PS to generate more SO4-• and •OH to
degrade BP-3.
Temperature is a vital parameter in PS system, which will determine the degree
of PS activation. Persulfate will generate two sulfate radicals through scission of the
peroxide bond after absorbing heat energy [52]. The effect of reaction temperature on
BP-3 decomposition in the PS/CoFe2O4 system was investigated between 25 oC and
40 oC (Fig. 8C). As seen, the degradation of BP-3 via PS/CoFe2O4 was temperature
dependent, and the degradation rate was significantly enhanced with increasing
temperature. For example, only 44% of BP-3 was removed at 25 oC after 3 h, while
complete degradation of BP-3 was achieved at 40 oC. Except for T = 40 oC, the BP-3
removal followed a pseudo-first-order degradation pattern (R2 > 0.91), and the
reaction rate constant (Kobs) increased from 0.0036 min-1 at 25 oC to 0.01 min-1 at 35
o
C (Fig S2). According to Arrhenius equation (Eq. 1), the calculated Ea was 77.98 kJ
mol-1. More reactive species (sulfate or hydroxyl radicals) may be produced at higher
temperatures, resulting in faster degradation of target pollutants [53-55].
ln Kobs = ln A – Ea / RT (1)
where A is the Arrhenius constant, Ea is the activation energy (kJ mol-1), R is the
universal gas constant (8.314×10 -3 kJ mol-1 K-1), and T is the absolute temperature
(K).
The influence of pH on the degradation of BP-3 was examined over a broad pH
range of 3.0 - 12.0 (Fig. 8D). Generally, the degradation of BP-3 increased with
increasing pH, in agreement with the research of Gago-Ferrero et al. [6]. In the whole
reaction time, the pseudo-first-order reaction constant increased from 0.0006 min-1 at
pH 3.0 to 0.0028 min-1 at pH 9.0 (Fig S3). The pH-dependent stability of BP-3
molecule can contribute to its enhanced removal at high pH. Blüthgen et al. [10]
reported that benzophenone-derivatives were more stable under acidic conditions.
Moreover, BP-3 mainly existed in the anion form at pH > pKa (8.06) [6], leading to an
accelerated degradation. Besides, Sun et al. [56] reported that at low pH (pH < 5), H+
scavenged SO4-• and •OH according to Eq. 2 and 3, respectively. We also noticed that
11
at pH = 12.0, the remaining BP-3 concentration quickly dropped to < 40% in the first
0.5 h. This may be attributed to the predominant •OH, which exhibits a slightly higher
redox potential than SO4-• in the alkaline medium [57]. Kolthoff and Miller [51]
suggested that under activation, PS would decompose into two SO4-•, which were
then transformed into •OH through subsequent reactions with OH-. Additionally, base
activation, which is the most common activator of PS for in situ chemical oxidation
(ISCO) applications [58], was also involved in the reaction process. When base was
used in excess, highly chlorinated methanes and ethanes could be successfully
destroyed by base-activated persulfate technology [59].
H+ + SO4-• + e- HSO4-• (2)
H+ + •OH + e- H2O (3)
The degradation of BP-3 at different PS doses (the molar ratios of BP-3 : PS are
1:100, 1:250, 1:500 and 1:1000) were performed in this work. From Fig. 8E, we can
see an increasing removal efficiency with the increasing PS dosages, especially at the
molar ratio of 1:500. To be specific, the removal efficiency after 6 h was 37, 47%,
75% and 91% at the BP-3: PS molar ratios of 1:100, 1:250, 1:500 and 1:1000,
respectively. However, a proportional relationship was not found between the removal
rate and PS concentration, possibly due to the SO4-• could be scavenged by PS [60].
Since TOC removal reflects the mineralization of organic compounds, the TOC
content was quantified during BP-3 degradation (Fig. 8F). The mineralization degree
of the reaction solutions all increased slowly with the reaction time. It is obvious that
the addition of catalyst leads to a higher TOC reduction than single PS oxidation, and
lower TOC content was observed with increasing catalyst load. To be specific, the
TOC removal after 6 h was 29%, 41% and 52% for control, 100 mg L-1 and 500 mg
L-1 catalyst, respectively. The improved mineralization degree further illustrated that
CoFe2O4 has a good catalytic performance.
<Fig. 8.>
3.3 Stability and reusability of CoFe2O4
In order to evaluate the stability and reusability of CoFe2O4, ion leakage and
recycle tests were conducted and the results were shown in Fig. 9. After 6h, Co 2+
12
leakage was 0.301 mg L-1, more than Fe3+ (0.220 mg L-1). This may be because the
solution pH at this time was about 6.50, Fe3+ starts precipitating at pH=3, while
precipitation of Co2+ occurs at pH ≥ 7. Hence, the concentrations of Co2+ and Fe3+ in
the reaction solution were far lower than the value (2.0 mg L-1) limited by the
European Union, suggesting the use of the CoFe2O4 catalyst would not cause
secondary pollution.
As illustrated in Fig. 9B, the 6 h removal rate of BP-3 in five cycles was 75%,
74%, 73%, 70% and 66%, respectively. The slightly decreased degradation efficiency
was likely due to catalyst poisoning that the adsorption of substances like byproducts
on catalyst surfaces was able to induce changes in surface morphology and block
reactions [61], as confirmed by the SEM, TEM and BET analyses of the recycled
catalyst (Text S1). Similar result was reported by Su et al. [62]. In addition, catalyst
loss in sampling process may also contribute to the decreased removal rate. Overall,
no obvious changes were observed in the recycled sample (Text S1), suggesting the
CoFe2O4 catalyst possessed high stability during the reaction process.
<Fig. 9.>

3.4 Catalytic mechanism of CoFe2O4

According to previous studies, both SO4-• and •OH existed in the PS activation
system [54]. To further verify this result, EPR test was performed. As can be seen
from Fig. 10A, SO4-• and •OH were detected in the system of PS/CoFe2O4, suggesting
the simultaneous presence of the two radicals.
To determine the dominant radical responsible for BP-3 degradation, tert-butyl
alcohol (TBA) was used as scavengers to quench •OH, and ethanol (EtOH) was
selected to quench both radicals [22,54]. As shown in Fig. 10B, the degradation
efficiency of BP-3 by PS/CoFe2O4 decreased significantly from 75% to 17% in the
presence of EtOH, and the removal was 63% when TBA existed in the system. These
data showed that these two radicals could contribute to the degradation of BP-3 by
CoFe2O4 activated PS, but SO4-• was the major reactive radical species involved in
the reaction process. This was similar to the results of Ding et al. [15] who used

13
nanoscaled magnetic CuFe2O4 as a heterogeneous catalyst of peroxymonosulfate to
degrade tetrabromobisphenol A, and Zhong et al. [63] who degraded 1,4-dioxane by
PS/Fe. Feng et al. [14] also reported that SO4-• dominated in the degradation of
flumequine during PS activation by Fe3O4/MWCNTs/PHQ catalyst.
In view of dominant SO4-• in the system, the catalytic mechanism of PS
activation by CoFe2O4 was summarized in Fig. 11. Electron transfer on the surface of
CoFe2O4 was an important mechanism for SO4-• generation. Francesco et al. [64]
suggested that SO4-• preferred to attack organic matters through electron transfer
reaction, while •OH was more likely to participate in hydrogen abstraction or addition
reactions. Basically, Co2+ on the surface of CoFe2O4 can activate PS to generate SO4-•
(Eq. 4). The resulting Co 3+ can be reduced to Co2+ by HSO5-, which was produced
from hydrolysis of S2O82- under acidic condition (Eq. 5-6). The change of Co2+ into
Co3+ was suggested by the Co 2p XPS spectra of the CoFe2O4 catalyst before and
after reaction (Text S2). Moreover, Li et al. [65] and Ren et al. [66], have also
demonstrated the involvement of Co 2+-Co3+-Co2+ redox processes in the catalytic
oxidation reaction. On the other hand, BP-3 was deprotonated to the corresponding
organic matter radical by Fe3+ on the surface of the catalyst, and Fe3+ was reduced to
Fe2+ (Eq. 7, RH represents BP-3) [67,68]. The R• can be directly oxidized into the
byproducts by Fe3+ (Eq. 8), while the Fe2+ can react with PS to form SO4-• and Fe3+
(Eq. 9). In this sense, the synergistic catalytic cycle of Co2+-Co3+-Co2+ and
Fe3+-Fe2+-Fe3+ was built, which could activate PS to generate SO4-• continuously. In
addition, since •OH partially contributed to the degradation of BP-3, it should be
noted that SO4-• can react with H2O to generate •OH, even at a low rate constant (k =
2.0×10-3 s−1, Eq. 10) [69].
≡Co2+ + S2O82- SO4-• + Co3+ + SO42- (4)
S2O82- + H2O HSO5- + HSO4- (5)
Co3+ + HSO5- Co2+ + SO5•- + H+ (6)
≡Fe3+ + RH ≡Fe2+ +R• + H+ (7)
≡Fe3+ + R• Intermediates + Fe2+ (8)
≡Fe2+ + S2O82- SO4-• + Fe3+ + SO42- (9)
14
SO4-• + H2O •OH + SO42- + H+ (10)
<Fig. 10.>
<Fig. 11.>
3.5 Identification of intermediates and reaction pathways
The degradation intermediates of BP-3 via thermally catalytic PS oxidation were
tentatively identified by LC-TOF-MS. The new peaks in the total ion current (TIC)
chromatogram (Fig. 12) suggest the generation of BP-3 oxidation products, and we
can see that the peaks of most byproducts appeared from 11 to 19 min. A total of 15
products belonging to thirteen nominal masses were detected, and their structural
assignments were performed according to the TIC chromatogram (Fig. 12) and their
respective MS/MS spectra (Fig S8). The fragmentation ions in the MS/MS spectra
provide direct information for structural identifications. For example, four fragments
ions 212.04, 211.03, 184.05 and 121.02 were observed in BP-3 (227.07),
corresponding to the loss of CH3, O, CH3+CO and CH3+C6H3O group from the parent
ion, respectively. In P2, sequential loss of OH and C6H7O (or C7H5O) produced
fragment ions at m/z 198.04, 103.93 and 92.97, and this compound was thus identified
as 2,4-dihydroxybenzophenone (BP-1). The deprotonated ion at m/z 233.04 (P7) is
proposed as the benzene ring opening product, which was supported by five daughter
ions at m/z 205.05, 177.05, 161.02, 146.03 and 129.03, corresponding to the
successive losses of CO, CO, O, CH3 and OH fragments. Especially, the structure of
P1 was further verified by the standard solution of 2-methoxy-xanthone (Fig S9),
which was synthesized by Qu et al. [70]. Two pairs of isomers, namely, P5A/P5B and
P12A/P12B, were observed since they have the same parent ions but different
fragmentation patterns.
<Fig. 12.>
Based on the identified transformation intermediates, the reaction pathways of
BP-3 in the PS/CoFe2O4 system were tentatively proposed in Fig. 13. Pathway I is
initiated by HO• radicals attack on the benzene ring of BP-3, leading to the formation
of C-C bond cleavage product P8 and hydroxylation products P5A and P5B. Then,
P5A was transformed into P1 via hydroxyl condensation reaction, and P5B could
15
undergo consecutive hydroxylation reaction to generate P6 and P10. In this process,
demethylation of P5B produced P3, which was further converted into P9 through
direct oxidation. Moreover, reactive radicals (HO• and/or SO4-•) attack induced
benzene ring opening of P6 to yield P7, which was then transformed into P12B
through direct oxidation and into P13 through hydroxylation and subsequent benzene
ring opening reaction (P7→P12A→P13). In pathway II, the methoxy group of BP-3
was attacked by the reactive radicals, and the demethylation reaction occurred to give
2,4-dihydroxybenzophenone (BP-1, P2). We should note that BP-1 has been detected
as the intermediate of BP-3 in various treatments, such as ozonation and peroxone
oxidation [71], heterogeneous photocatalytic degradation [72], UV/H2O2 [73] and
ozonation [74]. Hydroxylation of P2 yielded P4, which can also be generated from
P5B as result of demethylation. The •OH radicals can attack the neighboring carbon
atom to the carbonyl group (—C(O)-C) in the more substituted benzene ring of P4,
causing the bond cleavage to form P11. In short, the degradation of BP-3 mainly
involves hydroxylation, demethylation, direct oxidation and benzene ring opening
reaction.
<Fig. 13.>
3.6 Toxicity evaluation
The ECOSAR program, which was developed by the US Environmental
Protection Agency, can be used to predict the toxicity of organic compounds. The
operating principle of toxicity prediction is based on some available data of organic
chemicals by using structure-activity relationships (SARs) (US EPA, 2012). In this
paper, the toxicity of BP-3 and its degradation intermediates to three typical aquatic
species were evaluated by ECOSAR and classified according to the Globally
Harmonized System of Classification and Labelling of Chemicals (United Nations,
2011). As seen in Table S2, the acute and chronic toxicity of all reaction products
(except for P1 and P5A) is much less than that of BP-3. Despite the slightly high
toxicity of P1 and P5A, their concentrations are limited, and a large amount of BP-3
was transformed into compounds with low or even no toxicity. This suggested that
BP-3 degradation by PS/CoFe2O4 was very significant for the safety of aquatic
16
environment.
4. Conclusions
Results of this work showed that CoFe2O4 could be used as a heterogeneous
catalyst to effectively activate PS for the degradation of BP-3.
Under the conditions of [BP-3]0 : [PS]0 = 1 : 1000, initial pH = 7.0 ± 0.1, T = 25
o
C and catalyst load = 500 mg L-1, 91% of BP-3 (1.31 µM) was removed in 6 h by PS/
CoFe2O4 treatment.
Increasing the catalyst loading, reaction temperature and solution pH would
accelerate the reaction.
According to radical scavenging tests and EPR analysis, both sulfate and
hydroxyl radicals contributed to the decomposition of BP-3 in the CoFe2O4 catalyzed
PS oxidation system, while SO4-• was the dominant radical species for the
degradation.
15 degradation products were detected by mass spectrometry analysis, and
hydroxylation, demethylation, direct oxidation and benzene ring opening are mainly
involved in the reaction process.
Toxicity evaluation by the ECOSAR program showed that the acute and chronic
aquatic toxicity of all oxidation intermediates (except for P1 and P5A) was much
lower than that of BP-3.
Acknowledgements
This research was financially supported by the National Natural Science
Foundation of China (No. 21607073; 21577063; 21377051;), the Natural Science
Foundation of Jiangsu Province (No. BK20160651) and the China Postdoctoral
Science Foundation (No. 2016M590445).
References

[1] M.S. Díaz-Cruz, D. Barceló, Chemical analysis and ecotoxicological effects of

organic UV-absorbing compounds in aquatic ecosystems, TrAC-Trend Anal.
Chem. 28 (2009) 708-717.
[2] R. Rodil, J.B. Quintana, E. Concha-Graña, P. López-Mahía, S.
Muniategui-Lorenzo, D. Prada-Rodriguez, Emerging pollutants in sewage, surface
and drinking water in Galicia (NW Spain), Chem. 86 (2012) 1040-1049.

17
[3] M.E. Balmer, H.R. Buser, M.D. Müller, T. Poiger, Occurrence of some organic UV
filters in wastewater, in surface waters, and in fish from Swiss lakes, Environ. Sci.
Technol. 39 (2005) 953-962.
[4] K. Fent, A. Zenker, M. Rapp, Widespread occurrence of estrogenic UV-filters in
aquatic ecosystems in Switzerland, Environ. Pollut. 158 (2010) 1817-1824.
[5] M.S. Díaz-Cruz, P. Gago-Ferrero, M. Llorca, D. Barceló, Analysis of UV filters in
tap water and other clean waters in Spain, Anal. Bioanal. Chem. 402 (2012)
2325-2333.
[6] P. Gago-Ferrero, M.S. Díaz-Cruz, D. Barceló, Fast pressurized liquid extraction
with in-cell purification and analysis by liquid chromatography tandem mass
spectrometry for the determination of UV filters and their degradation products in
sediments, Anal. Bioanal. Chem. 400 (2011) 2195-2204.
[7] M. Schlumpf, B. Cotton, M. Conscience, V. Haller, B. Steinmann, W.
Lichtensteiger, In vitro and in vivo estrogenicity of UV screens, Environ. Health
Persp. 109 (2001) 239-244.
[8] R.S. Ma, B. Cotton, W. Lichtensteiger, M. Schlumpf, UV filters with antagonistic
action at androgen receptors in the MDA-kb 2 cell transcriptional-activation assay,
Toxicol. Sci. 74 (2003) 43-50.
[9] T. Suzuki, S. Kitamura, R. Khota, K. Sugihara, N. Fujimoto, S. Ohta, Estrogenic
and antiandrogenic activities of 17 benzophenone derivatives used as UV
stabilizers and sunscreens, Toxicol. Appl. Pharm. 203 (2005) 9-17.
[10] N. Blüthgen, S. Zucchi, K. Fent, Effects of the UV filter benzophenone-3
(oxybenzone) at low concentrations in zebrafish (Danio rerio), Toxicol. Appl.
Pharm. 263 (2012) 184-194.
[11] S.H. Yuan, P. Liao, A.N. Alshawabkeh, Electrolytic manipulation of persulfate
reactivity by iron electrodes for trichloroethylene degradation in ground water,
Environ. Sci. Technol. 48 (2014) 656-663.
[12] T. Zhang, Y. Chen, Y.R. Wang, J.L. Roux, Y. Yang, J.P. Croué, Efficient
peroxydisulfate activation process not relying on sulfate radical generation for
water pollutant degradation, Environ. Sci. Technol. 48 (2014) 5868-5875.
[13] X.X. He, A.A. de la Cruz, K.E. O'Shea, D.D. Dionysiou, Kinetics and
mechanisms of cylindrospermopsin destruction by sulfate radical-based advanced
oxidation processes, Water Res. 63 (2014) 168-178.
[14] M.B. Feng, R.J. Qu, X.L. Zhang, P. Sun, Y.X. Sui, L.S. Wang, Z.Y. Wang,
Degradation of flumequine in aqueous solution by persulfate activated with
common methods and polyhydroquinone-coated magnetite/multi-walled carbon
nanotubes catalysts, Water Res. 85 (2015) 1-10.
[15] Y.B. Ding, L.H. Zhu, N. Wang, H.Q. Tang, Sulfate radicals induced degradation
of tetrabromobisphenol A with nanoscaled magnetic CuFe2O4 as a heterogeneous
catalyst of peroxymonosulfate, Appl. Catal. B: Environ. 129 (2013) 153-162.
[16] G.D. Fang, J. Gao, D.D. Dionysiou, C. Liu, D.M. Zhou, Activation of persulfate
by quinones: Free radical reactions and implication for the degradation of PCBs,
Environ. Sci. Technol. 47 (2013) 4605-4611.
[17] S.G. Huling, B.E. Pivetz, In-situ Chemical Oxidation (EPA/600/R-06/072). US

18
 EPA. 2006.
[18] A. Tsitonaki, B. Petri, M. Crimi, H. MosbÆk, R.L. Siegrist, P.L. Bjerg, In situ
chemical oxidation of contaminated soil and groundwater using persulfate: A
review, Crit. Rev. Env. Sci. Tec. 40 (2010) 55-91.
[19] A. Ghauch, A.M. Tuqan, N. Kibbi, Ibuprofen removal by heated persulfate in
aqueous solution: A kinetics study, Chem. Eng. J. 197 (2012) 483-492.
[20] M.M. Ahmed, S. Chiron, Solar photo-Fenton like using persulphate for
carbamazepine removal from domestic wastewater, Water Res. 48 (2014)
229-236.
[21] I. Michael-Kordatou, M. Iacovou, Z. Frontistis, E. Hapeshi, D.D. Dionysiou, D.
Fatta-Kassinos, Erythromycin oxidation and ERY-resistant Escherichia coli
inactivation in urban wastewater by sulfate radical-based oxidation process under
UV-C irradiation, Water Res. 85 (2015) 346-358.
[22] Y.F. Ji, C. Ferronato, A. Salvador, X. Yang, J.M. Chovelon, Degradation of
ciprofloxacin and sulfamethoxazole by ferrous-activated persulfate: Implications
for remediation of groundwater contaminated by antibiotics, Sci. Total Environ.
472 (2014) 800-808.
[23] G.D. Fang, D.D. Dionysiou, S.R. Al-Abed, D.M. Zhou, Superoxide radical
driving the activation of persulfate by magnetite nanoparticles: Implications for
the degradation of PCBs, Appl. Catal. B: Environ. 129 (2013) 325-332.
[24] H. Nguyen, S.A. El-Safty, Meso- and macroporous Co3O4 nanorods for effective
VOC gas sensors, J. Phys. Chem. C. 115 (2011) 8466-8474.
[25] H. Wang, J. Huang, C. Wang, D. Li, L. Ding, Y. Han, Immobilization of glucose
oxidase using CoFe2O4/SiO2 nanoparticles as carrier, Appl. Surf. Sci. 257 (2011)
5739-5745.
[26] J.H. Li, I. Levin, J. Slutsker, V. Provenzano, P.K. Schenck, Self-assembled
multiferroic nanostructures in the CoFe2O4-PbTiO3 system, Appl. Phys. Lett. 87
(2005) 072909.
[27] P. Sathishkumar, R.V. Mangalaraja, S. Anandan, M. Ashokkumar, CoFe2O4/TiO2
nanocatalysts for the photocatalytic degradation of Reactive Red 120 in aqueous
solutions in the presence and absence of electron acceptors, Chem. Eng. J. 220
(2013) 302-310.
[28] S. Gao, Z.Y. Li, X.L. Zhang, Preparation and characterization of a novel
core-shell structure CoFe2O4/TiO2 magnetic nano-photocatalyst, Chinese J. Funct.
Mater. Dev. 15 (2009) 201-205.
[29] R. Loos, G. Locoro, S. Contini, Occurrence of polar organic contaminants in the
dissolved water phase of the Danube River and its major tributaries using
SPE-LC-MS2 analysis, Water Res. 44 (2010) 2325-2335.
[30] Y. Ji, L. Zhou, Y. Zhang, C. Ferronato, M. Brigante, G. Mailhot, J.M. Chovelon,
Photochemical degradation of sunscreen agent 2-phenylbenzimidazole-5-sulfonic
acid in different water matrices, Water Res. 47 (2013) 5865-5875.
[31] J. Chen, R.J. Qu, X.X. Pan, Z.Y. Wang, Oxidative degradation of triclosan by
potassium permanganate: Kinetics, degradation products, reaction mechanism,
and toxicity evaluation, Water Res. 103 (2016) 215-223.

19
[32] J.M. Kuang, J. Huang, B. Wang, Q.M. Cao, S.B. Deng, G. Yu, Ozonation of
trimethoprim in aqueous solution: Identification of reaction products and their
toxicity, Water Res. 47 (2013) 2863-2872.
[33] International Center for Diffraction Data (ICDD), Powder Diffraction Database,
Pattern 22-1086, 1994.
[34] P.C.R. Varma, R.S. Mann, D. Banerjee, M.R. Varma, K.G. Suresh, A.K. Nigam,
Magnetic properties of CoFe2O4 synthesized by solid state, citrate precursor and
polymerized complex methods: A comparative study, J. Alloy. Compd. 453 (2008)
298-303.
[35] S.M. Montemayor, L.A. García-Cerda, J.R. Torres-Lubián, Preparation and
characterization of cobalt ferrite by the polymerized complex method, Mater. Lett.
59 (2005) 1056-1060.
[36] J.G. Lee, J.Y. Park, Y.J. Oh, C.S. Kim, Magnetic properties of CoFe2O4 thin films
prepared by a sol-gel method, J. Appl. Phys. 84 (1998) 2801-2804.
[37] L.J. Xu, J.L. Wang, Fenton-like degradation of 2,4-dichlorophenol using Fe3O4
magnetic nanoparticles, Appl. Catal. B: Environ. 123-124 (2012) 117-126.
[38] Y.D. Meng, D.R. Chen, X.L. Jiao, Synthesis and characterization of CoFe2O4
hollow spheres, Eur. J. Inorg. Chem. 25 (2008) 4019-4023.
[39] C. Altavilla, E. Ciliberto, Decay characterization of glassy pigments: An XPS
investigation of smalt paint layers, Appl. Phys. A. 79 (2004) 309-314.
[40] A. Amri, Z.T. Jiang, C.Y. Yin, A. Fadli, M.M. Rahman, S. Bahri, H. Widjaja, N.
Mondinos, T. Herawan, M.M. Munir, G. Priyotomo, Structural, optical, and
mechanical properties of cobalt copper oxide coatings synthesized from low
concentrations of sol-gel process, Phys. Status Solidi A 213 (2016) 3205-3213.
[41] A. Amri, Z.T. Jiang, X.L. Zhao, Z.H. Xie, C.Y. Yin, N. Ali, N. Mondinos, M.M.
Rahman, D. Habibi, Tailoring the physicochemical and mechanical properties of
optical copper-cobalt oxide thin films through annealing treatment, Surf. Coat.
Tech. 239 (2014) 212-221.
[42] Z.Z. Liu, S.J. Yang, Y.N. Yuan, J. Xu, Y.F. Zhu, J.J. Li, F. Wu, A novel
heterogeneous system for sulfate radical generated through sulfite activation on a
CoFe2O4 nanocatalyst surface, J. Harzard Mater. 324 (2017) 583-592.
[43] Z.J. Gu, X. Xiang, G.L. Fan, F. Li, Facile synthesis and characterization of cobalt
ferrite nanocrystals via a simple reduction-oxidation route, J. Phys. Chem. C. 112
(2008) 18459-18466.
[44] I. Nedkov, R.E. Vandenberghe, T. Marinova, P. Thailhades, T. Merodiiska, I.
Avramova, Magnetic structure and collective Jahn-Teller distortions in
nanostructured particles of CuFe2O4, Appl. Surf. Sci. 253 (2006) 2589-2596.
[45] F. Jiao, H. Frei, Nanostructured cobalt oxide clusters in mesoporous silica as
efficient oxygen-evolving catalysts, Angew. Chem. Int. Ed. 48 (2009) 1873-1876.
[46] J. Rosen, G.S. Hutchings, F. Jiao, Ordered mesoporous cobalt oxide as highly
efficient oxygen evolution catalyst, J. Am. Chem. Soc. 135(2013) 4516-4521.
[47] Y. Zhang, J.W. Huang, Y. Ding, Porous Co3O4/CuO hollow polyhedral nanocages
derived from metal-organic frameworks with heterojunctions as efficient
photocatalytic water oxidation catalysts, Appl. Catal. B: Environ. 198 (2016)

20
 447-456.
[48] J. Deng, Y.J. Chen, Y.A. Lu, X.Y. Ma, S.F. Feng, N.Y. Gao, J. Li, Synthesis of
magnetic CoFe2O4/ordered mesoporous carbon nanocomposites and application in
Fenton-like oxidation of rhodamine B, Environ. Sci. Pollut. Res. 2017, DOI
10.1007/s11356-017-8941-5.
[49] Y.D. Meng, D.R. Chen, X.L. Jiao, Synthesis and characterization of CoFe2O4
hollow spheres, Eur. J. Inorg. Chem. 25 (2008) 4019-4023.
[50] Y. Wang, L. Zhou, X. Duan, H. Sun, E.L. Tin, W. Jin, S. Wang, Photochemical
degradation of phenol solutions on Co3O4 nanorods with sulfate radicals, Catal.
Today 258 (Part 2) (2015) 576-584.
[51] E. Manova, B. Kunev, D. Paneva, I. Mitov, L. Petrov, Mechano-synthesis,
characterization, and magnetic properties of nanoparticles of cobalt ferrite,
CoFe2O4, Chem. Mater. 16 (2004) 5689-5696.
[52] I.M. Kolthoff, I.K. Miller, The chemistry of persulfate. I. The kinetics and
mechanism of the decomposition of the persulfate ion in aqueous medium, J. Am.
Chem. Soc. 73 (1951) 3055-3059.
[53] J. Deng, Y.S. Shao, N.Y. Gao, Y. Deng, S.Q. Zhou, X.H. Hu, Thermally activated
persulfate (TAP) oxidation of antiepileptic drug carbamazepine in water, Chem.
Eng. J. 228 (2013) 765-771.
[54] M.H. Nie, Y. Yang, Z.J. Zhang, C.X. Yan, X.N. Wang, H.J. Li, W.B. Dong,
Degradation of chloramphenicol by thermally activated persulfate in aqueous
solution, Chem. Eng. J. 246 (2014) 373-382.
[55] C.Q. Tan, N.Y. Gao, Y. Deng, N. An, J. Deng, Heat-activated persulfate oxidation
of diuron in water, Chem. Eng. J. 203 (2012) 294-300.
[56] J. Sun, M. Song, J. Feng, Y. Pi, Highly efficient degradation of ofloxacin by UV/
Oxone/Co 2+ oxidation process, Environ. Sci. Pollut. Res. 19 (2011) 1536-1543.
[57] C.J. Liang, Z.S. Wang, C.J. Bruell, Influence of pH on persulfate oxidation of
TCE at ambient temperatures, Chem. 66 (2007) 106-113.
[58] O.S. Furman, A.L. Teel, R.J. Watts, Mechanism of base activation of persulfate,
Environ. Sci. Technol. 44 (2010) 6423-6428.
[59] P.A. Block, R.A. Brown, D. Robinson, Novel activation technologies for sodium
persulfate in situ chemical oxidation, the Fourth International Conference on the
Remediation of Chlorinated and Recalcitrant Compounds, Columbus, OH, USA.
2004.
[60] R.E. Huie, C.L. Clifton, Rate constants for hydrogen abstraction reactions of the
sulfate radical, SO4-• alkanes and ethers, Int. J. Chem. Kinet. 21 (1989) 611-619.
[61] J.L. Oudar, D. Hulin, A. Migus, A. Antonetti, F. Alexandre, Subpicosecond
spectral hole burning due to nonthermalized photoexcited carriers in GaAs, Phys.
Rev. Lett. 55 (1985) 2074.
[62] S.N. Su, W.L. Guo, Y.Q. Leng, C.L. Yi, Z.M. Ma, Heterogeneous activation of
Oxone by CoxFe3-xO4 nanocatalysts for degradation of rhodamine B, J. Hazard.
Mater. 244-245 (2013) 736-742.
[63] H. Zhong, M.L. Brusseau, Y.K. Wang, N. Yan, L. Quig, G.R. Johnson, In-situ
activation of persulfate by iron filings and degradation of 1,4-dioxane, Water Res.

21
 83 (2015) 104-111.
[64] M. Francesco, C. Attilio, Electron-transfer processes: Peroxydisulfate, a useful
and versatile reagent in organic chemistry, Acc. Chem. Res. 16 (1983) 27-32.
[65] X. Li, Z. Wang, B. Zhang, A.I. Rykov, M.A. Ahmed, J. Wang, FexCo3-xO4
nanocages derived from nanoscale metal-organic frameworks for removal of
bisphenol A by activation of peroxymonosulfate, Appl. Catal. B: Environ. 181
(2016) 788-799.
[66] Y. Ren, L. Lin, J. Ma, J. Yang, J. Feng, Z. Fan, Sulfate radicals induced from
peroxymonosulfate by magnetic ferrospinel MFe2O4 (M = Co, Cu, Mn, and Zn) as
heterogeneous catalysts in the water, Appl. Catal. B: Environ. 165 (2015)
572-578.
[67] A.J. Jafari, B. Kakavandi, N. Jaafarzadeh, R.R. Kalantary, M. Ahmadi, A.A.
Babaei, Fenton-like catalytic oxidation of tetracycline by AC@Fe3O4 as a
heterogeneous persulfate activator: Adsorption and degradation studies, J. Ind.
Eng. 45 (2017) 323-333.
[68] C.J. Liang, C.P. Liang, C.C. Chen, pH dependence of persulfate activation by
EDTA/Fe (III) for degradation of trichloroethylence, J. Contam. Hydrol. 106
(2009) 173-182.
[69] O.P. Chawla, R.W. Fessenden, Electron spin resonance and pulse radiolysis
studies of some reactions of SO4•-. J. Phys. Chem. 79 (1975) 2693-2700.
[70] R.J. Qu, Q. Zhang, X.S. Zhang, Z.Y. Wang, A comprehensive study on infrared
spectra of 2-hydroxyxanthone, Spectrosc. Lett. 45 (2012) 240-245.
[71] P. Gago-Ferrero, K. Demeestere, M.S. Díaz-Cruz, D. Barceló, Ozonation and
peroxone oxidation of benzophenone-3 in water: Effect of operational parameters
and identification of intermediate products, Sci. Tot. Environ. 443 (2013) 209-217.
[72] H. Zúñiga-Benítez, C. Aristizábal-Ciro, G.A. Peñuela, Heterogeneous
photocatalytic degradation of the endocrine-disrupting chemical Benzophenone-3:
Parameters optimization and by-products identification, J. Environ. Manage. 167
(2016) 246-258.
[73] P. Gong, H.X. Yuan, P.P. Zhai, Y.C. Xue, H.J. Li, W.B. Dong, Investigation on the
degradation of benzophenone-3 by UV/H2O2 in aqueous solution, Chem. Eng. J.
277 (2015) 97-103.
[74] P. Neta, V. Madhavan, H. Zemel, R.W. Fessenden, Rate constants and mechanism
of reaction of SO4•- with aromatic compounds, J. Am. Chem. Soc. 99 (1977)
163-164.

22
Fig .1. SEM image (A) and EDS spectrum (B) of the pristine CoFe2O4 and element
maps of Co, Fe and O (C).

23
Fig. 2. TEM image of the pristine CoFe2O4 with different scale bar.

24
Fig. 3. XRD patterns of CoFe2O4 caclined at different temperatures and the standard
spectra in JCPDS card.

25
Fig. 4. XPS spectra of Co 2p (A) and Fe 2p (B) of CoFe2O4 calcined at different
temperatures.

26
Fig. 5. XPS spectra of O 1s of CoFe2O4 calcined at different temperatures.

27
A 250 B
20

BET surface area (m g )

-1
2 200

Pore size (nm)

16
150

12
100

50 8

0 4
100 200 300 400 500 600 100 200 300 400 500 600
o o
Temperature ( C) Temperature ( C)

Fig. 6. BET surface area (A) and pore size (B) of CoFe2O4 catalyst calcined at
different temperatures.

28
Fig. 7. FT-IR spectra of CoFe2O4 prepared at different temperatures.

29
Fig. 8. Influence of catalyst calcining temperatures (A), catalyst loadings (B), reaction
temperature (C), solution pH (D), BP-3 : PS molar ratios (E) and evolution of TOC (F)
on the degradation of BP-3 in the PS/CoFe2O4 system. Typical reaction conditions:
[BP-3]0 = 1.31 µM, [BP-3]0 : [PS]0 = 1 : 500, initial pH = 7.0 ± 0.1, T = 25 oC,
catalyst load = 500 mg L-1. Note that catalyst load is 250 mg L-1 for panel A.

30
Fig. 9. Ion leaching (A) and recycle test (B) of CoFe2O4. Conditions: [BP-3]0 = 1.31
µM, [BP-3]0 : [PS]0 = 1 : 500, initial pH = 7.0 ± 0.1, T = 25 ºC, catalyst load = 500
mg L-1.

31
Fig. 10. (A) EPR spectrum of DMPO-OH and DMPO-SO4 adducts in BP-3 solution
after 10 min of reaction. (B) Kinetic tests of BP-3 in the presence of radical quenchers.
● means •OH adduct, ○ means SO4-• adduct. Conditions: [BP-3]0 = 1.31 µM, [BP-3]0 :
[PS]0 = 1 : 500, initial pH = 7.0 ± 0.1, T=25 ºC, catalyst load = 500 mg L-1.

32
Fig. 11. Schematic diagram of the possible oxidation mechanism of BP-3 by
PS/CoFe2O4.

33
Fig.12. Total ion current (TIC) chromatogram of BP-3 reaction solutions at different
times in negative mode (A) and positive mode (B).

34
Fig. 13. Transformation pathways of BP-3 in the PS/CoFe2O4 system.

35
Graphical Abstract

36
Highlights

CoFe2O4 is an effective catalyst to activate PS for the degradation of BP-3.

Kinetics and mechanisms of BP-3 degradation by PS/CoFe2O4 were studied.

SO4-• was the dominant radical for BP-3 decomposition.

15 transformation products were identified by LC-TOF-MS.

Hydroxylation, demethylation, direct oxidation, benzene ring opening were

observed.

37