Chemical Engineering Journal 379 (2020) 122324

Contents lists available at ScienceDirect

Chemical Engineering Journal

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

Heterogeneous Fenton-like degradation of tetracyclines using porous T

magnetic chitosan microspheres as an efficient catalyst compared with two
preparation methods
Xiaoyang Lia, Kangping Cuia, , Zhi Guoa, , Tingting Yanga, Yong Caoa, Yinping Xiangb,
⁎ ⁎

Huanhuan Chenc, Mufan Xia

a
School of Resources and Environmental Engineering, Hefei University of Technology, Hefei 230009, PR China
b
College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China
c
College of Oceanography, Hohai University, Nanjing 210098, China

HIGHLIGHTS GRAPHICAL ABSTRACT

• Magnetic chitosan beads with two

preparation methods were synthesized
and compared.
• Magnetic chitosan beads was for the
first time proposed as a Fenton-like
catalyst.
• High degradation activity and reusa-
bility to tetracyclines were in-
vestigated.
• Detailed reaction pathways and me-
chanisms in the TC degradation pro-
cess were proposed.

ARTICLE INFO ABSTRACT

Keywords: The construction of heterogeneous Fenton-like catalyst with high efficiency is crucial for wastewater treatment.
Heterogeneous Fenton-like catalyst In this study, magnetite nanoparticles were successfully embedded into chitosan beads to form a porous structure
Magnetite nanoparticles (Fe3O4-Cs) by in-situ or two-step method, and it was put forward as a kind of Fenton-like catalyst to degrade
Chitosan tetracyclines for the first time. SEM, TEM, XRD, FT-IR, XPS, TG and BET technologies were used to characterize
In-situ method
its properties. After comparing the tetracycline hydrochloride (TC) degradation efficiencies of two prepared
Tetracyclines
beads, the in-situ method was selected for subsequent studies. The influential factors of TC degradation effi-
ciency were investigated, including iron content, pH value, the amount of H2O2, reaction temperature, catalyst
dosage and initial TC concentration. The results showed that under the optimal conditions, about 96.0% of TC
(48.09 mg/L) and 68.3% of total organic carbon (TOC) were degraded within 20 min and 120 min, respectively.
The degradation efficiencies for the other two tetracyclines: oxytetracycline (OTC) and doxycycline hyclate
(DOTC), have been explored as well. And the degradation process followed pseudo first order kinetic equation.
In addition, these Fe3O4-Cs beads with saturated magnetization could be easily reused for 6 cycles without
significantly reducing catalytic activity. High degradation performance of beads was attributed to the stable
porous structure, abundant active sites and possible synergistic effects between two components, which pro-

⁎
Corresponding authors.
E-mail addresses: [email protected] (K. Cui), [email protected] (Z. Guo).

https://doi.org/10.1016/j.cej.2019.122324
Received 12 May 2019; Received in revised form 8 July 2019; Accepted 23 July 2019
Available online 24 July 2019
1385-8947/ © 2019 Elsevier B.V. All rights reserved.
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

moted the formation of hydroxyl radicals (HO%) for TC degradation. The possible reaction pathways and me-
chanism on TC degradation were discussed technically. In short, these results indicated that macroscopic Fe3O4-
Cs beads (3–4 mm) with high efficiency, excellent stability and reusability exhibited a superior Fenton-like
process on the removal of tetracyclines for future application.

1. Introduction Biopolymers, such as chitosan (Cs), bio-char and alginate, are re-
cognized as efficient supporters for catalytic applications. Compared
Antibiotics are increasingly used in the medical industry because of with traditional supports, they have the advantages of low toxicity, low
their broad-spectrum antimicrobial activity and low cost [1,2]. In cost, good biocompatibility, high availability and abundance [32].
particular, the wide applications of tetracycline antibiotics, such as Among them, chitosan is regarded as one of the most effective biopo-
tetracycline (TC), oxytetracycline (OTC) and doxycycline hyclate lymers due to its superior adsorption capacity, well stability of metal
(DOTC), make them the second most used antibiotics worldwide [3–5]. ions and strong affinity to transition metals [33]. Besides, there are
However, these tetracyclines are difficult to biodegrade, and only a abundant active amino and hydroxyl functional groups on the polymer
small fraction of them can be actually metabolized/absorbed by hu- chain of chitosan, which makes chitosan highly chemical active [34].
mans or animals [6]. Therefore, a large amount of tetracyclines were All these advantages promote chitosan to become an excellent support
excreted directly into the environment, which will lead to a water carrier.
ecological risk and the spread of resistance genes in water environment In this study, magnetite nanoparticles (Fe3O4 NPs) were successfully
[7,8]. Thus, it is highly urgent to explore a new technology to degrade embedded into chitosan beads through two methods to form a porous
tetracycline effectively. structure (Fe3O4-Cs). The in-situ co-precipitation method was able to
Over the past two decades, a lot of studies have been carried out on synthesize nanoscale Fe3O4, which could distribute evenly on the sur-
the removal of tetracycline, including physical adsorption [9], biolo- face of beads. For comparison, the two-step method was also applied to
gical treatment [10], and chemical degradation [11,12]. Among them, compare the degradation activity of microspheres. It is the first time
advanced oxidation processes (AOPs) are treated to be an available that Fe3O4-Cs has been used in the Fenton-like field for the treatment of
technology to eliminate and mineralize tetracyclines in wastewater tetracyclines wastewater. This study is aim to determine the optimal
[13]. AOPs technologies, such as O3/H2O2, UV/O3, UV/H2O2, Fe2+/ degradation conditions of tetracyclines wastewater (including the
H2O2, and UV/TiO2, can degrade tetracycline by continuously produ- amount of TC, H2O2 content, pH value, iron content, reaction tem-
cing hydroxyl radicals (HO%) [14–16]. These technologies can degrade perature and catalyst dosage), explore the degradation mechanism of
most organic pollutants in groundwater and wastewater, and the redox TC, and verify the reusability of Fe3O4-Cs beads. Our study provides a
potential is 1.9–2.7 V vs. NHE (normal hydrogen electrode) [17]. As potential way for the removal of tetracyclines in wastewater.
one of the AOPs technologies, Fenton reaction has attracted more and
more attention. The traditional Fenton reaction is a mixture of Fe2+ and 2. Experiments
H2O2 in acidic solution, which produces hydroxyl radicals (HO%) as the
main active groups to attack organic pollutants (Eqs. (1)) [18]. Al- 2.1. Materials and chemicals
though it has simple operation, high catalytic activity and no selectivity
with organic and inorganic compounds [19], it also has some short- Iron chloride hexahydrate (FeCl3·6H2O, ≥98%), Chitosan flakes
comings: (i) non-recyclable, (ii) acidic pH (pH < 3), (iii) the produc- (80.0–95.0% deacetylated), hydrochloric acid (HCl, 36.0–38.0%),
tion of solid sludge [20], which seriously affect the practical application ethylene glycol (HOCH2CH2OH, ≥99%), sodium acetate (CH3COONa,
of homogeneous Fenton. ≥99%) and polyethylene glycol (HO(CH2CH2O)nH) were acquired
To solve these difficulties, the application of heterogeneous cata- from Sinopharm Chemical Reagent Co., Ltd, China. Oxytetracycline
lysts has become a research hotspot in recent years due to its low cost, (OTC, ≥98%), tetracycline hydrochloride (TC·HCl, ≥98%), doxycy-
mild operating conditions, easy recovery and no sludge formation [21]. cline hyclate (DOTC, ≥98%), tert-Butanol (C4H10O, ≥99%) and p-
These catalysts are mainly divided into two types [22]: one is the Fe, Benzoquinone (BQ, ≥99%) were purchased from Shanghai Maclin
Mn, Ce and other metal oxides, and the other is a compound of carrier Chemistry Co., Ltd, China. Other chemicals, such as sodium hydroxide
materials loaded with iron compounds [23–25]. To date, different (NaOH, ≥98%), sodium citrate (C6H5Na3O7, 98%) and hydrogen per-
metal oxides and iron-immobilized materials, such as MnFe2O4/bio- oxide (H2O2, 30 wt%), were supplied from Shanghai Aladdin Chemistry
char [26], Fe3O4@C [27], Fe@Bacillus subtilis [28], MIL-100(Fe)@ Co., Ltd, China. And all reagents are analytical grade.
Fe3O4/CA [29] and so on, have shown a certain removal efficiency of
tetracyclines. However, most of these catalysts are still facing the dif- 2.2. Synthesis of porous magnetic chitosan microspheres (Fe3O4-Cs)
ficulty of relatively poor activity, and require light, electricity, or ul-
trasound to help enhance their activity. Besides, the inevitable metal 2.2.1. Method 1 (The in-situ method)
leaching and agglomeration of nanoparticles are also important issues The in-situ method means that the Fe3O4 NPs in Fe3O4-Cs micro-
to reduce the activity and stability of catalysts [21]. Magnetite nano- spheres was fabricated by in-situ co-precipitation of ferrous and ferric
particles (Fe3O4 NPs), as a common iron-based catalyst, has attracted ions (Eqs. (2)) [31,35]. In short, 1.0 g chitosan was weighed into 40 mL
numerous attentions because of its simple preparation, superior cata- 1% acetic acid solution. And then the chitosan was completely dis-
lytic efficiency, strong superparamagnetism, easy recovery in external solved by mechanical stirring for about 30 min, following by 5 mL
magnetic field and stable oxidation [30,31]. However, due to the un- Fe3+/Fe2+ solution (the concentration ratio of Fe3+: Fe2+, e.g. 0.2:0.1,
saturation and instability of the nanoparticles surface, the size of Fe3O4 0.8:0.4 and 1.4:0.7 M). Then, the mixed solution was dropped into a
NPs significantly increases during the synthesis process, which makes soaking solution with sodium hydroxide/sodium citrate (NaOH 1.25 M,
Fe3O4 NPs easy to aggregate, and is not conducive to the formation of CH3COONa 0.1 M) by a peristaltic pump at a rate of 50 drops per
HO% by activating H2O2 [18]. Fortunately, above problems can be ef- minute. At the moment when the mixed solution of chitosan and iron is
fectively solved by using biopolymers as supporter to prepare hybrid in contact with the soaking solution, Fe3O4 NPs will form on the surface
biocomposite. of microspheres through in situ co-precipitation reaction of Fe2+ and
Fe3+ (Eqs. (2)). After magnetic separation, the macroscopic beads
Fe2 + + H2 O2 Fe3 + + OH + HO· (1)
whose size range from 3 to 4 mm, were cleaned with pure water and

2
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

ethanol, and then freeze-dried for 24 h to form a porous structure [31]. 2.5. Analysis methods
For comparison, chitosan microspheres without the addition of Fe3+/
Fe2+ solution were also prepared under the same experimental condi- The concentrations of TC, OTC and DOTC were detected by high-per-
tions. For convenience, these prepared samples were named as 2Fe3O4- formance liquid chromatography (HPLC; Agilent 1200 Series). The op-
Cs, 8Fe3O4-Cs and 14Fe3O4-Cs and Cs, respectively. timum mobile phase for TC test was a mixed solution of 0.1 M oxalic acid,
methanol and acetonitrile (Voxalic acid:Vmethanol:Vacetonitrile = 67:11:22), and
Cs - NH+3 + Fe 2+ + 2Fe3 + + 8OH - Cs - NH2 + Fe3 O4 + 4H2 O (2) the analytical wavelength was 360 nm. A different mobile phase including
acetonitrile and ultrapure water (Vacetonitrile:Vultrapure = 20:80) was used for
OTC test with a wavelength of 355 nm. Besides, the mobile phase for DOTC
test consisted of methanol and 0.02 M phosphoric acid (Vmethanol:Vphosphoric
2.2.2. Method 2 (The two-step method)
acid = 20:80) with a analytical wavelength of 346 nm. The degradation ef-
The two-step method includes the preparation of Fe3O4 and the
ficiency of total organic carbon (TOC) in solution was used to evaluate the
crosslinking of Fe3O4 and chitosan. Fe3O4 NPs were prepared by
degree of mineralization of TC by TOC/TN analysis (Multi N/C 3100). The
synthesis or purchase. For synthesized Fe3O4, the Fe3O4 was produced
amount of iron was measured using inductively coupled plasma mass
via a common and simple solvothermal method. FeCl3·6H2O (1.35 g),
spectrometry (ICP-MS; Agilent 7500) whose detection limit is estimated to
sodium acetate (3.6 g) and polyethylene glycol (1.0 g) were added into
1PPB. The total leached iron in reaction solution was measured by atomic
ethylene glycol (40 mL) to form a dark brown solution. The solution
absorption spectrophotometry (AAS; AA140, VARIAN). H2O2 was measured
was vigorously stirred using for about 30 min by a magnetic stirrer, and
by the potassium titanium (IV) oxalate spectrophotometric method at
next sealed in a stainless steel autoclave lined with polytetra-
400 nm [37]. Dynamic changes in absorbance at a wavelength range of
fluoroethylene. After heating in an autoclave and holding at 200 °C for
200–700 nm during the reaction were detected by spectrophotometer
24 h, the autoclave was cooled to room temperature. This product was
(UV–Vis; Hitachi, Japan). HO% radicals were detected by electron para-
rinsed with pure water and ethanol, and then dried at 60 °C under va-
magnetic resonance spectra (ESR; Bruker, Germany), and 50 mM of DMPO
cuum for about 8 h [36]. Secondly, the newly synthesized Fe3O4 was
was used as the scavenger. The degradation intermediates were analyzed
cross-linked with Cs. 1 g of Cs and an appropriate amount of Fe3O4 were
using high-performance liquid chromatography-mass spectrometry (HPLC-
dissolved into 40 mL of 2% acetic acid liquor, and mechanically stirred
MS; Agilent 1290/6460, USA) equipped with a linear ion trap interface with
for about 30 min to mix the solution thoroughly. The following specific
an HPLC and quaternary pump, more details about the analysis of HPLC-MS
steps were consistent with the in-situ method described previously.
are available in the Supporting Information.

2.3. Characterization of Fe3O4-Cs synthesized by two methods

3. Results and discussion
The internal porous network and surface morphologies of Fe3O4-Cs
were analyzed by SU-8020 scanning electron microscopy (SEM; 3.1. Characterization of the beads
Hitachi, Tokyo) and JEM-2100F transmission electron microscopy
(TEM; JEOL, Tokyo). The compositions phases of the synthesized ma- Fe3O4-Cs microspheres were synthesized by two traditional
terials were identified by X'Pert PRO MPD X-ray diffraction (XRD) methods. For in-situ method (Fig. 1A), the mixed solution of chitosan
analysis (PANalytical, Netherlands) working at 40 kV and 40 mA in the and Fe3+/Fe2+ ions were dropwise added to an alkaline solution, and
2θ range of 10–70° with 6° min−1 scan rate. The structures of as-syn- then the in-situ co-precipitation of Fe2+ and Fe3+ ions occurred (Eqs.
thesized catalysts were studied by fourier transform infrared spectro- (2)), which led to the formation of the Fe3O4 NPs on the chitosan layers.
scopy (FTIR; Bruker, German) using potassium bromide tablet method. For two-step method (Fig. 1B), the Fe3O4-Cs beads were synthesized by
The X-ray photoelectron spectroscopy (XPS; Shimadzu, Japan) was used crosslinking Fe3O4 NPs with chitosan in the presence of sodium citrate.
to detect the surface chemical states, and the C1s, O1s and Fe2p spectra What’s more, for comparison, the Fe3O4 NPs in two-step beads were
were studied by Casaxps 2.3.19 software. Thermodynamic effects of obtained by hydrothermal method (Fig. 1C) or purchased (Fig. 1D). In
Fe3O4 NPs loaded in Chitosan microspheres were measured using addition, As shown in Fig. 1E and F, Fe3O4-Cs microspheres prepared by
thermogravimetric analyzer (TGA; STA449F3, Netzsch, German) from
25 to 800 °C. The nitrogen adsorption-desorption isotherm at 77 K was
used to analyze the specific surface area of Fe3O4-Cs beads by the
Brunauer-Emmett-Teller (BET) analysis (Autosorb-IQ3, Quantachrome,
USA), and the pore structures were evaluated.

2.4. Reaction procedures

For the degradation studies of TC, the Fenton-like process was

conducted in a sealed conical flask. First, 0.1 g Fe3O4-Cs were weighed
into 200 mL tetracycline hydrochloride (TC) solution. Then, the de-
gradation reaction started once the H2O2 was added to the mixed so-
lution. The suspension reacted for 120 min in a water bath shaker at
200 rpm and preset temperature. During the process, 3 mL aliquots
were collected at given time intervals and filtered to remove suspended
sludge. The concentration of TC was measured at 360 nm by high-
performance liquid chromatography (HPLC). The TC solution had an
initial concentration of 0.1 mM (48.09 mg/L), and the range of H2O2
concentration was 0–40 mM (0, 5, 10, 20, 30, and 40 mM). The initial Fig. 1. Dried Fe3O4-Cs beads prepared by in-situ method (A), and two-step
pH value was regulated using 0.5 M NaOH or H2SO4 solution. During method (B), synthesized Fe3O4 (C) and purchased Fe3O4 (D) after separation in
the reaction period, 1 mL of samples, which was quenched using excess a magnetic field; Fe3O4-Cs beads with external magnetic field by in-situ method
tert-Butanol (20 μL), was taken from solution for further studies. (E), and two-step method (F).

3
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

(A) (B)

(D) Spectrum 1
(C)

Before

Spectrum 2

After

(E) (F)

(G) (H)

500nm
d=0.297nm(220)
C O Fe

Fig. 2. SEM images of Fe3O4-Cs beads synthesized by two-step methods (A), or in-situ method before (B) and after (C) reaction; Energy Dispersive Spectrum (EDS)
(D), TEM images (inset: diameter distribution of Fe3O4 in in-situ beads) (E) (F), HRTEM image of (G), and elemental mapping (H) of Fe3O4-Cs beads.

two preparation methods can be attracted by the external magnetic beads. In addition, the energy dispersive spectroscopy (EDS) in Fig. 2D
field, indicating that they have good magnetism. demonstrates the existence of C, O, N, and Fe elements in Fe3O4-Cs
The morphologies and microstructures of as-synthesized Fe3O4-Cs beads before and after reaction, and the percentage of Fe decreased
beads were analyzed by SEM and TEM technologies. Compared to the slightly from 20.5% to 19.5%. More detailed atomic percentages of
SEM image (Fig. 2A) of Fe3O4-Cs beads prepared by two-step method, Fe3O4-Cs beads are listed in Table S1. The TEM images of in-situ beads
the Fe3O4-Cs synthesized by in-situ method in Fig. 2B possesses more in Fig. 2(E, F) and Fig. S1(A, B) show that many tiny Fe3O4 NPs with a
numerous, more uniform and smaller pore networks, which can provide diameter of approximately 5.97 nm, were evenly dispersed on the
extensive active sites with H2O2 and abundant electronic transport beads’ surface. The result indicates the agglomeration of Fe3O4 NPs was
pathways. In Fig. 2C, some cracks and blocked pores are found on the effectively inhibited under the support action of chitosan. Notably, the
surface of used beads, which affect the reusability and stability of average diameter of Fe3O4 in in-situ beads was much smaller than that

4
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

of two-step beads, whose average size was approximately 180.37 nm and C]O, respectively. The O1s pattern of XPS is shown in Fig. 4C. The
(Fig. S1(C, D) in Supporting information). By calculation, the distance formation of Fe3O4 leads to the appearance of characteristic peak at
between two adjacent lattice fringes is 0.297 nm (Fig. 2G, S2), which is 529.6 eV, which corresponds to lattice oxygen (O2−) [42,43]. The
consistent with the (2 2 0) lattice of Fe3O4 crystal [38]. Simultaneously, binding energy at 531.2 eV was approach to that at the surface ad-
the TEM elemental mapping of Fe3O4-Cs reveals that these components sorption oxygen type (OH−). Besides, the peak of 532.8 eV may relate
of C, Fe and O are evenly formed on the surface (Fig. 2H). to the oxygen bound species of C-O. What’s more, eight peaks are ob-
The phase structures of naked and chitosan-bound Fe3O4 beads in served for 14Fe3O4-Cs in the Fe2p spectrum (Fig. 4D). These four
two methods were evaluated by XRD patterns, and the results were binding energies at 709.2, 711.2, 722.8 and 724.8 eV correspond to
depicted in Fig. 3A. Six characteristic peaks for Fe3O4 (2θ = 30.1, 35.5, Fe2+2p3/2, Fe3+2p3/2, Fe2+2p1/2 and Fe3+2p1/2, respectively,
43.1, 53.4, 57.1 and 62.6°), which can be found in all samples, are manifesting that the catalyst consists of Fe2+ and Fe3+ species. And the
attributed to their lattices (2 2 0), (3 3 1), (4 0 0), (4 2 2), (5 1 1) and other four peaks belong to the satellite peaks of Fe2+ or Fe3+ [44]. The
(4 4 0) [38]. These peaks can be confirmed by the standard card (JCPDS above results confirm the fact that Fe3O4-Cs composites were success-
75-0033), which indicates that the synthesized Fe3O4 is pure with a fully synthesized.
spinel structure. Furthermore, the diffraction peaks in all samples are The thermal stability of Cs and Fe3O4-Cs beads synthesized via two
sharp and wide, indicating that Fe3O4 has a small size and good crys- methods were evaluated by TGA analysis. As shown in Fig. 4E, three
tallinity. From Fig. 3A (c–g), a strong amorphous phase peak similar weight losses are found on the thermo-gravimetric curves of Cs
(2θ = 15–25°) demonstrates that Fe3O4-Cs beads were coated with an and Fe3O4-Cs beads, respectively. The first stage in the range of
amorphous phase of chitosan on the surface. Compared with the spec- 25–200 °C is owed to the evaporation of free water and bound water in
trum of pristine 14Fe3O4-Cs beads, the used 14Fe3O4-Cs beads show six the microspheres. The weight reduction of chitosan beads, which ap-
similar characteristic peaks, which preliminarily proves the reusability peared in the second stage at 200–300 °C, is attributed to depolymer-
of 14Fe3O4-Cs beads. In short, the crystal structure of Fe3O4 was not ization and degradation of the acetylation and deacetylation units of
destroyed by the chelation reaction. the polymers, dehydration of the sugar rings and imidization of the
FT-IR was used to further research the characteristic bands of these amide groups, and the reason for the weight reduction of Fe3O4-Cs in
synthesized catalysts, as displayed in Fig. 3B. The characteristic peak at the range of 200–320 °C is the same with it. In the third stage of tem-
3444 cm−1 is considered as the stretching vibration of OeH and NeH perature above 320 °C, chitosan will be carbonized completely, and the
groups in chitosan. More specifically, these stretching and bending weight corresponding to 800 °C is the residual carbon of the final pro-
bands of NeH (at 1648 and 1595 cm−1) and CeO (at 1421 and duct of pyrolysis. Moreover, two additional weight losses of Fe3O4-Cs at
1379 cm−1, from residual acetylated moieties of chitin) can further 570–640 °C and 660–710 °C are also observed, respectively, which is
demonstrate the presence of chitosan. The peaks observed at 2920 and attributed to the reaction between Fe3O4 with carbon to form elemental
2877 cm−1 are ascribed to CeH stretching vibrations. Other featured iron at high temperature [45]. The initial decomposition temperature of
absorption peaks are formed at 1031 cm−1 (OeH bending), 1074 cm−1 Fe3O4-Cs microspheres is higher than that of chitosan microspheres,
(CeOH stretching vibrations) and 1151 cm−1 (OeH bending) because manifesting that the presence of Fe3O4 can markedly improve the
of the unique saccharide structure [39]. In this spectrum of Fe3O4-Cs thermal stability of catalyst.
beads, the Fe-O stretching vibration of Fe3O4 can be identified by the The specific surface area and porosity of these as-prepared beads
characteristic peak at 568 cm−1, manifesting that Fe3O4 was embedded were characterized via nitrogen adsorption/desorption isotherms. As
successfully into the microspheres, and it is consistent with the above displayed in Fig. 4F, based on the classification of the International
XRD results. Besides, this adsorption peak of Fe-O at 568 cm−1 gradu- Union of Pure and Applied Chemistry (IUPAC), the N2 adsorption-des-
ally broadens with the increase of iron content in the beads, and the orption isotherm was determined to be type IV of the H3 hysteresis
NeH absorption peak at 3444 cm−1 of chitosan occurs a slight shift. cycle [46]. The type IV isotherm is related to the capillary condensation
This indicates the occurrence of metal-NH2 bonds [40,41]. occurring in the mesopores, and the H3 hysteresis line represents the
To further identify the chemical characteristics of Fe3O4-Cs beads, fractured pores. The surface areas of 2Fe3O4-Cs, 8Fe3O4-Cs and
the XPS analysis was conducted. Fig. 4A showed the wide-scan XPS 14Fe3O4-Cs are 31.928, 23.397 and 20.569 m2/g, respectively, which
spectrum of Fe3O4-Cs. It is easy to find that the Fe3O4-Cs beads mainly are less than chitosan beads (50.044 m2/g) and Fe3O4-Cs beads via two-
consist of C, O, N, and Fe, and their atomic content is 40.80%, 29.70%, step method (57.984 m2/g). This may be attributed to that the forma-
5.82% and 21.80% respectively, which is consistent with the results of tion of Fe3O4 NPs blocked part of the mesoporous space in the synthesis
SEM. For C1s in Fig. 4B, this peak can be divided into three parts of process of microspheres. In comparison to the raw 14Fe3O4-Cs beads,
284.8, 286.4 and 288.0 eV, attributing to the vibration of CeC, CeOeC the surface area of used 14Fe3O4-Cs (19.355 m2/g) only exhibited a

(A) (B)

Fig. 3. The XRD patterns (A) and FT-IR spectra (B) of synthesized or purchased Fe3O4, Cs and Fe3O4-Cs microspheres.

5
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

(A) (B)

(C) (D)

(E) (F)

Fig. 4. XPS spectra of wide-scan (A), C1s (B), O1s (C) and Fe2p (D) for 14Fe3O4-Cs; TG analysis of Cs and Fe3O4-Cs beads (E); Nitrogen adsorption–desorption
isotherms and the corresponding pore size distributions curve (inset) of Cs and Fe3O4-Cs beads (F).

(A) (B)

Fig. 5. The comparison of Fe3O4-Cs beads prepared by two methods (A); Time evolution of TC removal under different conditions (B). (Reaction conditions: pH, 3.0;
initial TC, 48.09 mg/L; catalyst, 500 mg/L; H2O2, 10 mM; temperature, 40 °C).

6
 X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

negligible decrease, indicating the porous structure of the beads was abundant cavities and open channels for the diffusion of mediums, and
not damaged after reaction. Table 2 showed the detailed properties of is conductive to the solid-liquid interaction within the carbon matrix
Cs and Fe3O4-Cs beads based on N2 adsorption/desorption isotherms. [21]. In this Fe3O4-Cs/H2O2 system, the removal rate of TC was re-
The pore size distribution of prepared beads (inset of Fig. 4F) shows markably accelerated than that of Fe3O4/H2O2 system, and almost
that it contains numerous mesopores with a diameter of about 4 nm and 96.0% of TC was removed within 20 min. On the one hand, the support
a wide distribution by Barrett-Joyner-Halenda (BJH) method. Further- of chitosan biopolymer could prevent the agglomeration of Fe3O4 NPs
more, the existence of interior nanoporous space makes it have rela- effectively; On the other hand, the decomposition of H2O2 can be ac-
tively large and wide pore distribution [47]. tivated by the strong synergy between Fe-based groups and carbon
matrix, thereby increasing the degradation efficiency. Then sufficient
3.2. Comparison of catalytic activities of two prepared beads hydroxyl radicals (HO%) will be released [26,48], which demonstrates
the catalyst’ primary reactive groups of heterocatalytic effect in the
The degradation efficiencies of two prepared microspheres at initial removal of TC in the latter mechanism analysis. Moreover, some re-
TC concentration of 0.1 mM (48.09 mg/L) were compared in the Fe3O4- searchers have explored the oxidation degradation of TC by various
Cs/H2O2 system. As displayed in Fig. 5A, more than 96.0% of TC could Fenton-like catalysts in the past several years. As seen in Table 1, al-
be rapidly removed by in-situ 14Fe3O4-Cs beads within 20 min, whereas though the initial TC concentration is maximum compared with the
the two-step beads with the same iron content only degraded 44.3% of other catalysts, such as MnFe2O4/bio-char [26], Fe3O4@C [27], Fe@
TC within the same reaction time (synthesized Fe3O4). For comparison, Bacillus subtilis [28] and MIL-100(Fe)@Fe3O4/CA [29], the Fe3O4-Cs
the influences of synthesized and purchased Fe3O4 in two-step method beads can still remove a higher percentage of TC in the shortest time. In
01 were also explored, and less catalytic performance of purchased Fe3O4 addition, the H2O2 utilization efficiency of 14Fe3O4-Cs beads was the
can be easily obtained in Fig. 5A. The maximum removal efficiency of best among these catalysts, which only 0.217 mol H2O2 was required
Prerna Sainiin-situ method stabilized around 96.0% after 20 min, whereas the re- for 1.0 g TC removal. The excellent performance indicates that Fe3O4-Cs
moval efficiency of two-step method began to stabilize after 8 h (Table beads are better Fenton-like catalyst for TC degradation compared with
S2). others.
02 As can be seen in Fig. S3A, H2O2 was decomposed rapidly and
21/03/2024
completely within 120 min in the in-situ beads/H2O2 system, whereas 3.4. Influences of experimental conditions on the degradation of
Prerna Saini
only 24.6% H2O2 could be decomposed within the same time in the tetracyclines
two-step beads/H2O2 system. This phenomenon further confirms the
stronger catalytic activity of the in-situ beads. Besides, the removal rate 3.4.1. Effects of iron content in catalyst
of TOC is an important index to evaluate the performance of catalyst. As The amount of surface active sites (iron) of beads will affect the
shown in Fig. S3B, the 68.3% TOC removal rate caused by in-situ beads catalytic degradation efficiency of TC [49]. In order to better under-
within 120 min is clearly higher than that of two-step beads, indicating stand the effects of iron content, the microspheres with different iron
the in-situ beads have stronger mineralization ability for TC. In addi- contents (Fig. S4) were dissolved using a combination of concentrated
tion, the leached iron is considered to be an important factor in iden- hydrochloric acid and hydrogen peroxide at first. Then the content of
tifying the stability of Fe-based catalysts [18]. As shown in Fig. S3C, the iron in the microspheres was measured using ICP-MS with an accuracy
leached iron of two-step beads (2.684 ± 0.153 mg/L) is about four of 1 PPB (Table 2). Moreover, XRD, FT-IR and BET analytical methods
times as much as that of in-situ beads (0.753 ± 0.050 mg/L) after were also used to evaluate the effects. As shown in Fig. 6A, when the
120 min, demonstrating the better stability of in-situ beads. iron content in per 10 mg mass catalyst increased from 0.0078 to
The more excellent catalytic activity of in-situ method for TC de- 0.0417 mmol, the TC removal efficiency measured within 20 min in-
gradation is attributed to the following synergistic effects: (i) Fe3O4 in creased from 30.2% to 70.7%. These results reveal that the iron content
two-step method may be encapsulated by chitosan resulting in fewer of the catalyst has the vital function in the control of reaction rate. On
active sites, However, in in-situ method, Fe3O4 can be precipitated on the one hand, when the initial iron content increases, the Fe3O4-Cs
the surface of porous microspheres by in situ precipitation. (ii) The size beads possesses higher crystalline structure observed from the XRD
of Fe3O4 in in-situ beads (about 5.97 nm) is much smaller than that in patterns (Fig. 3A). On the other hand, the higher the iron content is, the
two-step beads (about 180.37 nm). Therefore, the Fe3O4-Cs beads syn- more active sites and the larger specific surface area the reaction has
thesized by in-situ method were selected for further studies [36]. (Table 2), which can obviously affect the activity of catalyst. Based on
above analysis, the samples with the highest iron content (14Fe3O4-Cs)
3.3. Fenton-like catalytic performance of Fe3O4-Cs beads were selected for further studies.

To assess the catalytic activity of Fe3O4-Cs beads synthesized via in- 3.4.2. Effects of pH
situ method in the presence of H2O2, batch experiments were carried The pH of the reaction solution has an influence on the surface
out in different systems (Fig. 5B). As observed in Fig. 5B, the removal charge characteristics, adsorption capacity and electron transfer cap-
rate of TC is only 18.5% in the alone H2O2 system, indicating that the ability of the fabricated Fe3O4-Cs beads, thereby affecting their catalytic
catalytic effect of H2O2 is quite weak without catalyst. For alone Fe3O4- degradation rate [50]. Considering the significant influence of pH on
Cs beads, about 9.5% of TC was removed because of the adsorption of Fenton-like process, batch TC degradation experiments were conducted
Fe3O4-Cs microspheres. This adsorption capacity of Fe3O4-Cs beads at different initial pH (Fig. 6B). As the pH value varied from 3 to 7, the
originates from its unique mesoporous structure, which provides TC removal efficiency within 120 min dropped sharply from 96.2% to

Table 1
Comparison of the performance of the 14Fe3O4-Cs beads with other Fenton-like catalysts.
Fenton-like Catalyst [TC]0a (ppm) TC Conversion (%) H2O2 concentration (mM) Reaction Time (min) H2O2 Consumption (mol)/1.0 g TC removal Ref.

MnFe2O4/bio-char 40 95.0 100 120 2.632 [26]

Fe3O4@C 20 79.3 5 120 0.315 [27]
Fe@Bacillus subtilis 25 93.5 400 100 17.112 [28]
MIL-100(Fe)@Fe3O4/CA 10 85.0 10 180 1.176 [29]
14Fe3O4-Cs beads 48.09 96.0 10 20 0.217 This Work

7
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

Table 2
The amount of iron in 10 mg mass of catalyst, and the BET results of beads.
Samples [Fe] (mmol/10 mg mass of catalyst) SBET (m2g−1) Total Pore Volume (cm3g−1) Average Pore Diameter (nm)

The in-situ method Cs beads 0 50.044 0.077 3.454

2Fe3O4-Cs 0.0078 ± 0.0006 31.928 0.059 3.825
8Fe3O4-Cs 0.0252 ± 0.001 23.397 0.063 3.643
14Fe3O4-Cs 0.0417 ± 0.0008 20.569 0.054 3.817
used 14Fe3O4-Cs 0.0408 ± 0.0006 19.355 0.050 3.648
The two-step method 0.0420 ± 0.0004 57.984 0.128 3.296

25.5%, indicating that the pH of the solution has a crucial effect on the hydrated ferrous complexes [51]. What’s more, when the pH is too
degradation reaction. When the pH is higher than 4, the catalytic ability high, the H2O2 tends to decompose into O2 instead of forming HO%
of Fe3O4 NPs significantly decreases by forming the hydrated iron and radicals, which is critical for TC degradation. In general, the optimum

(A) (B)

(C) (D)

(E) (F)

Fig. 6. The effect of various parameters on the TC degradation in Fe3O4-Cs catalyzed Fenton-like system: (A) iron content, (B) initial pH value, (C) H2O2 con-
centration, (D) temperature, (E) catalyst dosage, (F) initial TC concentration. (Reaction conditions: pH 3.0; initial TC, 48.09 mg/L; catalyst, 500 mg/L; H2O2, 10 mM;
temperature, 25 °C).

8
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

pH of most iron-based catalysts is about 3 [52,53], while the Fe3O4-Cs dosage was 700 mg/L, the degradation percentage of TC within 20 min
beads can still degrade TC efficiently at the pH of 4.0, illustrating that was 88.5%, which was less than that of 500 mg/L. The following rea-
the catalyst is capable of operating over a wider pH range. sons can explain this phenomenon: on the one hand, when the dosage of
Fe3O4-Cs beads rises from 300 mg/L to 500 mg/L, the increasing iron
3.4.3. Effects of H2O2 concentration content leads to the increase of active sites, which is consistent with the
The effects of H2O2 concentration were also investigated (Fig. 6C). variation of the TC degradation efficiency. On the other hand, a further
In the absence of H2O2, only 10% of TC was removed within 120 min. rise of the catalyst dosage results in a large amount of HO% radicals’
When the H2O2 concentration increased from 5 to 10 mM, the TC de- formation. Since the highly active HO% radicals have a very short life-
gradation efficiency improved from 44.6% to 70.7% within 20 min, time (about 1 ns), a large number of HO% radicals begin to quench ra-
which indicates that the degradation rate of TC raised with the increase pidly because of the rapid reactions between HO% radicals [55]. As a
of the H2O2 concentration. But when the concentration of H2O2 ex- result, the degradation efficiency reduced. The main reactions are as
ceeded 10 mM, the degradation performance of TC decreased slightly. It follows (Eqs. (6)–(8)):
can be explained that when the concentration is below the critical
Fe3+ + H2 O2 Fe2 + + HO2 ·+ H+ (6)
value, the amount of HO% radicals produced by the catalytic reaction
increases with the increase of H2O2 concentration. On the contrary, Fe 2 + + H2 O2 Fe3 + + HO· +OH (7)
when the H2O2 concentration is above the critical value, the generated
HO% can be trapped by excess H2O2 to form less active HO2% radical HO· + HO· H2 O2 (8)
[54] (Eqs. (3)–(5)), resulting in loss of the activity in TC removal. To further realize the reaction kinetics of TC removal, the pseudo-
H2 O2 + HO· HO2 · +H2 O (3) first-order model was conducted, as expressed by the following formula
(Eqs. (9)):
HO2 ·+ HO· H2 O+ O2 (4)
ln (C0 /C) = k t (9)
HO· + HO· H2 O2 (5) −1
where k indicates the apparent first-order rate constant (min ), t is the
reaction time (min), C0 is the initial concentration and C is the con-
3.4.4. Effects of reaction temperature centration at reaction time t. The removal of TC followed a pseudo-first-
To investigate the effects of temperature on TC degradation in order kinetic model with k = 0.053, 0.077, 0.127, 0.144 and
Fenton-like reaction, the temperature was set at 10, 25, and 40 °C, re- 0.124 min−1 for 300–700 mg/L, respectively (Fig. S5A). When the
spectively. As shown in Fig. 6D, the TC degradation rate showed an amount of catalyst is more than 500 mg/L, the rate constant didn't
obvious positive feedback with the rise of temperature. For instance, change a lot. Based on the above discussion, the optimal dosage of
the removal percentage of TC at 10, 25 and 40 °C within 120 min was Fe3O4-Cs catalyst was determined to be 500 mg/L.
84.2%, 96.2% and 96.6%, respectively. The catalytic reaction at the
temperature of 40 °C could reach the end point within only 20 min, 3.4.6. Effects of TC concentration
quicker than 25 °C (at least 60 min). The reason is probably that higher Further studies on the effects of TC concentration over Fe3O4-Cs
temperature may increase the generation rate of HO% radicals, thereby beads were also conducted. According to previous literature [56], the
accelerating the degradation of antibiotics. Thus, the optimal tem- concentration of antibiotics in actual pharmaceutical wastewater is
perature for this experiment is 40 °C, but considering the temperature of usually hundreds mg/L. So the concentration of TC in the range of
actual wastewater and there is little difference in TC removal efficiency 0.10–0.25 mM (48.09–120.23 mg/L) was selected in the experiment.
at 25 °C and 40 °C, the subsequent experiments were carried out at Fig. 6F compared the degradation abilities at various initial con-
25 °C. centrations and indicated that more than 90% of TC (0.10–0.25 mM)
could be degraded within 120 min, demonstrating the extremely high
3.4.5. Effects of catalyst dosage catalytic activity of Fe3O4-Cs beads. As displayed in Fig. S5B, the de-
Batch experiments were carried out to study the effects of catalyst gradation rate constant (k) decreases from 0.127 min−1 at 0.10 mM TC
dosage (300–700 mg/L) on TC removal. The percentage of TC de- to 0.031 min−1 at 0.15 mM TC, 0.026 min−1 at 0.20 mM TC, and
gradation rate increased from 61.7% to 96.0% within 20 min as the 0.017 min−1 at 0.25 mM TC. This can be explained by the fact that this
catalyst dosage rose from 300 to 500 mg/L (Fig. 6E). However, the TC heterogeneous process takes place in the boundary region between solid
degradation efficiency didn’t change a lot and even decreased when this and liquid phase, and this process includes the adsorption of organic
catalyst dosage rose further from 500 to 700 mg/L. When the catalyst pollutants and H2O2 on the catalyst surface. The rise of initial TC

(A) (B)

Fig. 7. Effectiveness for the degradation of various tetracyclines (A), and pseudo-first-order model based on the various tetracyclines (B). (Reaction conditions: pH
3.0; initial tetracyclines, 50 mg/L; catalyst, 500 mg/L; H2O2, 10 mM; temperature, 25 °C).

9
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

concentration indicates competitive adsorption of TC and H2O2, thus

suppressing the adsorption of H2O2. In addition, the aggregation of
intermediate products on the catalyst surface also affects the adsorption
of H2O2 so that the degradation rate dropped as the TC concentration
increased. While 500 mg/L of Fe3O4-Cs beads can still degrade up to
120.23 mg/L TC within 120 min, which is significantly stronger than
previous studies [26–29].

3.5. Degradation activities of 14Fe3O4-Cs to three kinds of tetracyclines

The general application of the 14Fe3O4-Cs catalyst to different kinds

of tetracyclines was further examined. Based on the concentration of
tetracyclines in actual wastewater [56], the initial concentration of
tetracyclines was set at 50 mg/L in this section. As shown in Fig. 7A, it
was a remarkable fact that all three tetracyclines, including Tetra-
cycline hydrochloride (TC), Oxytetracycline (OTC), and doxycycline
hyclate (DOTC) could be degraded over 90% within 120 min by the
catalytic reaction. The Fenton-like degradation kinetics of TC, OTC and Fig. 9. UV–vis absorption spectra over 14Fe3O4-Cs during the reaction time.
DOTC are displayed in Fig. 7B. This rate (k) of TC and DOTC is similar, (Reaction conditions: pH 3.0; initial TC, 48.09 mg/L; catalyst, 500 mg/L; H2O2,
both around 0.035 min−1, which is faster than that of OTC 10 mM; temperature, 25 °C).
(0.026 min−1). The different molecular structures can account for the
difference of degradation kinetics, and the structural formulas of TC, components. The total iron concentration was 116.7 mg/L in 14Fe3O4-
OTC and DOTC are shown in Fig. S6. The OTC has one more hydroxyl Cs/H2O2 system. As shown in Fig. 8B, the amount of leached iron in this
group than TC and DOTC, making it easier to form an association. bulk solution after each cycle gradually decreased from 0.75 to
Therefore, it is difficult to be adsorbed on the surface of Fe3O4-Cs beads, 0.25 mg/L, and the leached iron after the first cycle (inset of Fig. 8B) is
thereby reducing its degradation rate [57]. approximately 0.75 mg/L, accounting for only 0.64% of 116.7 mg/L. As
reported in previous literature [58], a strong chelating reaction be-
3.6. Reusability and stability of the 14Fe3O4-Cs beads tween the amino groups of chitosan and iron in the structure of Fe3O4-
Cs occurred during the synthesis process, which can obviously reduce
The reusability and stability of catalyst are significant indicators in the leached iron and the agglomeration of Fe3O4 NPs. Due to the ex-
practical applications, which are also the important significance of cellent reusability and stability, the Fe3O4-Cs beads are promising for
preparing large particle size spheres of Fe3O4-Cs. To evaluate the reu- water treatment application.
sability of Fe3O4-Cs beads, six cycles were conducted in Fenton-like
process. After each reaction, the beads were separated from the system 3.7. Reaction pathways of TC degradation
via imposing an external magnetic field, and further freeze-dried for
24 h. During each cycle, H2O2 and TC with the same concentration were 3.7.1. Dynamic degradation process of TC
added repeatedly, and the experimental results are presented in Fig. 8A. UV–vis spectra from 200 to 500 nm during the TC degradation
More than 85% of TC could be further degraded within 120 min after process within 120 min are illustrated in Fig. 9. The result shows that
six times of catalyst usage, indicating that it is an excellent catalyst for the characteristic peak of TC at 360 nm gradually decreased to ~ 0
heterogeneous Fenton-like reaction, especially for tetracyclines de- within 80 min, which indicates that most of TC was degraded during
gradation. After six cycles, the TOC degradation efficiency within the reaction process. Meanwhile, the absorption peak at 270 nm de-
120 min reduced from 68.3% to 52.4%, only decreased by 15.9%. It was creased gradually with a tiny shift from 270 to 261 nm, while the
further proved that Fe3O4-Cs beads had good catalytic performance. baseline in the UV region increased. These results suggest that the
The stability of 14Fe3O4-Cs was also determined by the amount of conjugated structure of tetracycline hydrochloride is rapidly destroyed
leached iron, which directly reflected the stability of the catalytic and may be decomposed into smaller molecular fragments. In addition,

(A) (B)

Fig. 8. Removal efficiency of TC and TOC (A), and the leached iron (inset of the leached iron variation in the first cycle) (B) during the TC degradation in six
consecutive runs. (Reaction conditions: pH 3.0; initial TC, 48.09 mg/L; catalyst, 500 mg/L; H2O2, 10 mM; temperature, 25 °C; reaction time per cycle, 120 min).

10
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

the absorbance in the 230–240 nm region rose in the first 60 min and After that, TC 2 was decomposed into TC 3 (m/z = 410) by condensa-
then reduced rapidly from 60 to 120 min, which reveals the generation tion reaction between amino groups and hydroxyl radicals to form the
and disappearance of some intermediates. fifth ring. For pathway III, TC 4 (m/z = 417) was formed by the de-
methylation of the dimethylamino group by low N-C bond energy at C1.
Then, HO% radicals continued to attack TC 4, leading to two decom-
3.7.2. Analysis of TC degradation intermediates position pathways of TC 4. On the one hand, TC 5 (m/z = 374) was
As shown in Fig. 8A, about 68.3% TOC of TC solution was degraded formed by the deamidation of TC 4. On the other hand, TC 6 (m/
after 120-min reaction, which means that TC can be mineralized into z = 362) was formed by dehydration of TC 4 in C6 and loss of acyla-
harmless final products in the Fe3O4-Cs/H2O2 system. Then, these in- mino groups in C2. Cleavage of carboatomic ring A formed TC 7 (m/
termediates of TC degradation were measured and analyzed by HPLC- z = 318), and further loss of TC 7 carboxyl groups formed TC 10 (m/
MS analysis. Eleven compounds including m/z = 427, 410, 417, 374, z = 274). In pathway Ⅳ, TC degraded to TC 8 (m/z = 405), due to the
405, 306, 362, 318, 274 and 114 (Fig. S7) were found and listed in the loss of acylamino group of TC. Then the fracture of carboatomic ring A
Table S3. Based on the molecular structure of products and previous of TC 8 lead to the formation of TC 9 (m/z = 306). Moreover, TC 10 (m/
research on TC degradation, four possible evolution pathways were z = 274) was resulted from combined actions, including dehydration at
proposed in Fig. 10 [59,60]. C11, deamination and loss of hydroxyl groups [62]. These results show
During the whole Fenton-like degradation process, there are two that TC can be degraded by ring-opening, and then oxidized to small
main reasons for the production of these intermediates: loss of func- molecule acids in Fe3O4-Cs/H2O2 Fenton-like system.
tional groups and ring opening reactions. For pathway Ⅰ, the production
of TC 1 (m/z = 475) was attributed to the fracture of CeC double-bond,
the rearrangement of HO% radicals at C5 and the oxidation of N-methyl 3.8. Fenton-like degradation mechanism
to N-aldehyde group [61]. For pathway Ⅱ, the elimination reaction of
C11a-C12 single-bond in TC 1 resulted in the formation of C11a-C12 It is well known that HO% radicals are the main active groups and
double-bond, thereby resulting in the formation of TC 2 (m/z = 427). play a crucial role in the heterogeneous Fenton-like reaction [63]. To

10 11 12 1
9 11a 2
8 3
7 6 5 4

Fig. 10. Proposed degradation pathways of TC in the Fe3O4-Cs/H2O2 system.

11
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

(A) (B)

Fig. 11. Effect of radical scavengers on the degradation of TC (A). ESR spectra of DMPO–HO% adducts that formed over time with Fenton-like degradation of a
suspension of 14Fe3O4-Cs/H2O2 (B). (Reaction conditions: pH 3.0; catalyst, 500 mg/L; H2O2, 10 mM; temperature, 25 °C).

further identify this dominant reactive species in catalytic reaction of (Fig. 11B). Obviously, four adsorption peaks of HO% radicals with an
Fe3O4-Cs, free radical capture experiments were carried out by the intensity value of 1:2:2:1 are observed in this ESR spectrum of 14Fe3O4-
addition of enough tert-Butanol (100 mM), KI (10 mM), and p-Benzo- Cs/H2O2 system, while no significant characteristic peaks are observed
quinone (BQ, 1 mM). All HO% radicals, including the HO% radicals in DMPO system. This result indicates that HO% radicals were success-
generated by ^Fe (HO%surf) and the HO% radicals generated by iron ions fully formed on the surface of Fe3O4-Cs beads.
in solution (HO%free), can be effectively captured [64,65], while the KI On the basis of the previous experimental results and discussion, a
only scavenge the HO%surf radicals. p-Benzoquinone can be used as a possible mechanism for H2O2 decomposition catalyzed by Fe3O4-Cs
capture for the formation of HO2% in the reaction process [66]. As beads is proposed and shown in Scheme 1. As mentioned above, Fe3O4-
displayed in Fig. 11A, the removal percentage of TC reduced from Cs microspheres with high activity and porous network structure
96.8% to 14.5% and 29.9% within 120 min after the addition of ex- (average pore diameter was 3.817 nm) can rapidly enrich TC and H2O2
cessive tert-Butanol and KI, respectively. This demonstrated that the into the interior materials. Firstly, TC are adsorbed onto the beads’
HO% radicals produced by heterogeneous Fenton-like process played a surface from the bulk solution through π-π action between the benzene
main role in the TC degradation process. Also, when BQ (HO2% radicals rings of TC and mesoporous Fe3O4-Cs beads. Then, once H2O2 is added,
inhibitor) was mixed, the removal percentage of TC reduced slightly the Fe3O4-Cs catalyst, containing a large number of Fe3O4 NPs, can
from 96.8% to 87.8%, indicating that HO2% might be involved in TC catalyze H2O2 to form HO%surf radicals to attack TC through in-
degradation. tramolecular electron transfer process. The powerful chelating reaction
After 20-min catalytic degradation, DMPO was selected as a sca- between the amino groups of chitosan and iron in the structure of
venger to further identify the species of HO% by ESR in this reaction Fe3O4-Cs can effectively prevent the precipitation of large amount of

Scheme 1. Schematic illustration showing the main steps of H2O2 activation and TC degradation in the Fe3O4-Cs/H2O2 Fenton-like system.

12
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

iron ions, thereby enhancing the stability of microspheres. In addition, References

the strong conductivity of mesoporous Fe3O4-Cs beads not only pro-
motes the electron transfer between ^Fe2+ and ^Fe3+ species, but [1] F. Deng, L.N. Zhao, X.B. Luo, S.L. Luo, D.D. Dionysiou, Highly efficient visible-light
also provides more active sites for Fenton-like process [67]. Meanwhile, photocatalytic performance of Ag/AgIn5S8 for degradation of tetracycline hydro-
chloride and treatment of real pharmaceutical industry wastewater, Chem. Eng. J.
the few leached iron can also initiate the composition of H2O2 through 333 (2018) 423–433.
chain reaction to produce HO%free. Finally, TC is rapidly mineralized into [2] S.L. Wang, H. Wang, Adsorption behavior of antibiotic in soil environment: a cri-
inorganic compounds attacked via HO%surf radicals on Fe3O4-Cs beads’ tical review, Front. Environ. Eng. Sci. 9 (2015) 565–574.
[3] J. Sun, W.J. Xu, P. Yang, N. Li, Y. Yuan, H.G. Zhang, Y.J. Wang, X.N. Ning,
border and HO%free radicals in solution. Y.P. Zhang, K.L. Chang, Y.P. Peng, K.F. Chen, Enhanced oxytetracycline removal
On the basis of previous discussion, the process for the enhanced coupling with increased power generation using a self-sustained photo-bioelec-
Fenton-like activity of Fe3O4-Cs microspheres can be described as fol- trochemical fuel cell, Chemosphere 221 (2019) 21–29.
[4] E. Bagda, E. Yabas, N. Karakus, Feasibility of tetracycline, a common antibiotic, as
lows (Eqs. (10)–(17)): chelating agent for spectrophotometric determination of UO22+ after cloud point
extraction, J. Radioanal. Nucl. Chem. 299 (2014) 1813–1820.
Fe 2 + + H2 O2 Fe3 + + HO·surf + OH (10) [5] G. Cheng, K.G. Karthikeyan, Interaction of tetracycline with aluminum and iron
hydrous oxides, Environ. Sci. Technol. 2005 (39) (2005) 2660–2667.
[6] J.Y. Cao, L.D. Lai, B. Lai, G. Yao, X. Chen, L.P. Song, Degradation of tetracycline by
Fe3+ + H2 O2 Fe2 + + HO2 ·surf + H+ (11)
peroxymonosulfate activated with zero-valent iron: performance, intermediates,
toxicity and mechanism, Chem. Eng. J. 364 (2019) 45–56.
Fe3+ + HO2 ·surf Fe2 + + O2 + H+ (12) [7] Z. Guo, K.P. Cui, G.M. Zeng, J.J. Wang, X.P. Guo, Silver nanomaterials in the nat-
ural environment: an overview of their biosynthesis and kinetic behavior, Sci. Total
Environ. 643 (2018) 1325–1336.
HO·surf + TC·HCl CO2 + H2 O (13) [8] Y.H. Chen, J.Q. Su, J.Y. Zhang, P. Li, H.J. Chen, B. Zhang, K.Y. Gin, Y.L. He, High-
throughput profiling of antibiotic resistance gene dynamic in adrinking water river-
reservoir system, Water Res. 149 (2019) 179–189.
Fe2 + + H2 O2 Fe3 + + HO· free + OH (14)
[9] J. Wan, G.M. Zeng, D.L. Huang, L. Hu, P. Xu, C. Huang, R. Deng, W.J. Xue, C. Lai,
C.Y. Zhou, K.X. Zheng, X.Y. Ren, X.M. Gong, Rhamnolipid stabilized nano-chlor-
Fe3 + + H2 O2 Fe2 + + HO2 · free + H+ (15) apatite: synthesis and enhancement effect on Pb-and Cd-immobilization in polluted
sediment, J. Hazard. Mater. 343 (2018) 332–339.
[10] H.F. Xiong, D.L. Zou, D.D. Zhou, S.S. Dong, J.W. Wang, B.E. Rittmann, Enhancing
Fe3 + + HO2 · free Fe2 + + O2 + H+ (16) degradation and mineralization of tetracycline using intimately coupled photo-
catalysis and biodegradation (ICPB), Chem. Eng. J. 316 (2017) 7–14.
[11] J.J. Xue, S.S. Ma, Y.M. Zhou, Z.W. Zhang, M. He, Facile photochemical synthesis of
HO· free + TC·HCl CO2 + H2 O (17)
Au/Pt/g-C3N4 with plasmon-enhanced photocatalytic activity for antibiotic de-
gradation, ACS Appl. Mater. Inter. 7 (2015) 9630–9637.
[12] M.H. Khan, H. Bae, J.Y. Jung, Tetracycline degradation by ozonation in the aqueous
4. Conclusion phase: proposed degradation intermediates and pathway, J. Hazard. Mater. 181
(2010) 659–665.
[13] X. Jiang, Y.H. Guo, L.B. Zhang, W.J. Jiang, R.Z. Xie, Catalytic degradation of tet-
A novel Fenton-like catalyst Fe3O4-Cs beads with high dispersion of racycline hydrochloride by persulfate activated with nano Fe0 immobilized meso-
ultrafine Fe3O4 nanoparticles was synthesized by in-situ method. Due to porous carbon, Chem. Eng. J. 341 (2018) 392–401.
[14] V. Homem, L. Santos, Degradation and removal methods of antibiotics from aqu-
the porous network structure, the larger surface area and the less lea- eous matricese-a review, J. Environ. Manage. 92 (2011) 2304–2347.
ched iron of Fe3O4-Cs beads, the as-synthesized samples not only had [15] Z.M. Liu, M.F. Zhu, Z. Wang, H. Wang, C. Deng, K. Li, Effective degradation of
enhanced Fenton-like degradation efficiency and mineralization of TC, aqueous tetracycline using a nano-TiO2/carbon electrocatalytic membrane,
Materials 9 (2016) 364.
but also exhibited superior catalytic activity of OTC and DOTC in the [16] H.T. Luu, K. Lee, Degradation and changes in toxicity and biodegradability of tet-
presence of H2O2. The Fe3O4-Cs beads showed better degradation effi- racycline during ozone/ultraviolet-based advanced oxidation, Water Sci. Technol.
ciency in a wide range of conditions than previous literatures. 70 (2014) 1229–1235.
[17] S.Y. Yang, Y.Y. Chen, H.Z. Xu, P. Wang, Y.H. Liu, M.D. Wang, A novel advanced
Moreover, the effects of six crucial parameters on the TC degradation
oxidation technology based on activated persulfate, Prog. Chem. 20 (2008)
were discussed. The HO% radicals generated on Fe3O4-Cs beads' surface 1433–1438.
played a major role in TC removal process. In addition, the macroscopic [18] R.L. Liu, Y.M. Xu, B.L. Chen, Self-assembled nano-FeO(OH)/reduced graphene
Fe3O4-Cs beads (3–4 mm) with excellent reusability and stability can be oxide aerogel as a reusable catalyst for photo-Fenton degradation of phenolic or-
ganics, Environ. Sci. Technol. 52 (2018) 7043–7053.
recycled for 6 times with a little leached iron, indicating that Fe3O4-Cs [19] H.J.H. Fenton, Oxidation of tartaric acid in presence of iron, J. chem. Soc. Trans. 65
beads are a kind of superior Fenton-like catalyst for tetracyclines re- (1894) 899–910.
moval. [20] B. Iurascu, I. Siminiceanu, D. Vione, M.A. Vicente, A. Gild, Phenol degradation in
water through a heterogeneous photo–Fenton process catalyzed by Fe-treated la-
ponite, Water Res. 43 (2009) 1313–1322.
[21] J.T. Tang, J.L. Wang, Fenton-like degradation of sulfamethoxazole using Fe-based
Declaration of Competing Interest
magnetic nanoparticles embedded into mesoporous carbon hybrid as an efficient
catalyst, Chem. Eng. J. 351 (2018) 1085–1094.
The authors declare that they have no known competing financial [22] D.L. Jiang, T.Y. Wang, Q. Xu, D. Li, S.C. Meng, M. Chen, Perovskite oxide ultrathin
interests or personal relationships that could have appeared to influ- nanosheets/g-C3N4 2D–2D heterojunction photocatalysts with significantly en-
hanced photocatalytic activity towards the photodegradation of tetracycline, Appl.
ence the work reported in this paper. Catal. B: Environ. 201 (2017) 617–628.
[23] R.S. Ribeiro, A.M.T. Silva, J.L. Figueiredo, J.L. Faria, H.T. Gomes, Catalytic wet
peroxide oxidation: a route towards the application of hybrid magnetic carbon
Acknowledgements nanocomposites for the degradation of organic pollutants. A review, Appl. Catal. B:
Environ. 187 (2016) 428–460.
This study was financially supported by the Changfeng County- [24] M. Cheng, G.M. Zeng, D. Huang, C. Lai, P. Xu, C. Zhang, Y. Liu, Hydroxyl radicals
based advanced oxidation processes (AOPs) for remediation of soils contaminated
Hefei University of Technology Industrial Innovation Guidance Fund with organic compounds: a review, Chem. Eng. J. 284 (2016) 582–598.
Key Project, the National Natural Science Foundation of China [25] Y. Sun, Z.X. Yang, P.F. Tian, Y.Y. Sheng, J. Xu, Y.F. Han, Oxidative degradation of
(51809068), and the Fundamental Research Funds for the Central nitrobenzene by a Fenton-like reaction with Fe-Cu bimetallic catalysts, Appl. Catal.
B: Environ. 244 (2019) 1–10.
Universities (JZ2018HGBZ0311). [26] C. Lai, F.L. Huang, G.M. Zeng, D.L. Huang, L. Qin, M. Cheng, C. Zhang, B.S. Li, H. Yi,
S.Y. Liu, L. Li, L. Chen, Fabrication of novel magnetic MnFe2O4/bio-char composite
and heterogeneous photo-Fenton degradation of tetracycline in near neutral pH,
Appendix A. Supplementary data Chemosphere 224 (2019) 910–921.
[27] B. Kakavandi, A. Takdastan, N. Jaafarzadeh, M. Azizi, A. Mirzaei, A. Azari,
Supplementary data to this article can be found online at https:// Application of Fe3O4@C catalyzing heterogeneous UV-Fenton system for tetra-
cycline removal with a focus on optimization by a response surface method, J.
doi.org/10.1016/j.cej.2019.122324.

13
X. Li, et al. Chemical Engineering Journal 379 (2020) 122324

Photoch. Photobiol. A 314 (2016) 178–188. encapsulated in boron and nitrogen codoped carbon nanotubes as synergistic cat-
[28] P. Zheng, B. Bai, W.S. Guan, H.L. Wang, Y.R. Suo, Degradation of tetracycline hy- alysts for Fenton-like reaction, Water Res. 101 (2016) 281–291.
drochloride by heterogeneous fenton-like reaction using Fe@Bacillus subtilis, RSC [49] E.G. Garrido-Ramírez, J.F. Marco, N. Escalona, M.S. Ureta-Zanartu, Preparation and
Adv. 00 (2013) 1–3. characterization of bimetallic Fe–Cu allophane nanoclays and their activity in the
[29] H.U. Rasheed, X.M. Lv, S.Y. Zhang, W. Wei, N. Ullah, J.M. Xie, Ternary MIL-100(Fe) phenol oxidation by heterogeneous electro-Fenton reaction, Micropor. Mesopor.
@Fe3O4/CA magnetic nanophotocatalysts (MNPCs): magnetically separable and Matter. 225 (2016) 303–311.
Fenton-like degradation of tetracycline hydrochloride, Adv. Powder Technol. 29 [50] H.J. Kim, T. Phenrat, R.D. Tilton, G.V. Lowry, Effect of kaolinite, silica fines and pH
(2018) 3305–3314. on transport of polymer-modified zero valent iron nanoparticles in heterogeneous
[30] S.T. Xing, Z.C. Zhou, Z.C. Ma, Y.S. Wu, Characterization and reactivity of Fe3O4/ porous media, J. Colloid. Interf. Sci. 370 (2012) 1–10.
FeMnOx core/shell nanoparticles for methylene blue discoloration with H2O2, Appl. [51] Y.B. Si, G.D. Fang, J. Zhou, D.M. Zhou, Reductive transformation of 2,4-di-
Catal. B: Environ. 107 (2011) 386–392. chlorophenoxyacetic acid by nanoscale and microscale Fe3O4 particles, J. Environ.
[31] R. Soumaya, A. Amrane, F. Fourcade, A. Assadi, L. Monser, N. Adhoum, Electro- Health. Sci. 45 (2010) 233–241.
Fenton catalyzed with magnetic chitosan beads for the removal of Chlordimeform [52] P. Bautista, A.F. Mohedano, J.A. Casas, J.A. Zazo, J.J. Rodriguez, Highly stable Fe/
insecticide, Appl. Catal. B: Environ. 226 (2018) 346–359. γ-Al2O3 catalyst for catalytic wet peroxide oxidation, J. Chem. Technol. Biotechnol.
[32] R.J. White, R. Luque, V.L. Budarin, J.H. Clark, D.J. Macquarrie, Supported metal 86 (2010) 497–504.
nanoparticles on porous materials, Methods and applications, Chem. Soc. Rev. 38 [53] W. Zhong, J.L. Wang, Degradation of sulfamethazine using Fe3O4-Mn3O4/reduced
(2009) 481–494. graphene oxide hybrid as Fenton–like catalyst, J. Hazard. Mater. 324 (2017)
[33] Y.M. Lee, W.J. Lee, Degradation of trichloroethylene by Fe(II) chelated with cross- 653–664.
linked chitosan in a modified Fenton reaction, J. Hazard. Mater. 178 (2010) [54] G.V. Buxton, C.L. Greenstock, W.P. Helman, A.B. Ross, Critical Review of rate
187–193. constants for reactions of hydrated electrons, hydrogenatoms and hydroxyl radicals
[34] K.X. Wang, H. Ma, S.Y. Peng, C. Yan, M.T. Wang, J. Yu, X.K. Wang, W. Chu, (·OH/O–) in aqueous solution, J. Phys. Chem. Ref. Data 17 (1988) 513–886.
A. Zinchenko, Hybrid porous magnetic bentonite-chitosan beads for selective re- [55] J. Wang, C. Liu, J. Li, R. Luo, X.R. Hu, X.Y. Sun, J.Y. Shen, W.Q. Han, L.J. Wang, In-
moval of radioactive cesium in water, J. Hazard. Mater. 362 (2019) 160–169. situ incorporation of iron-copper bimetallic particles in electrospun carbon na-
[35] Z. Guo, G.M. Zeng, K.P. Cui, A.W. Chen, Toxicity of environmental nanosilver: nofbers as an efficient Fenton catalyst, Appl. Catal. B-Environ. 207 (2017) 316–325.
mechanism and assessment, Environ. Chem. Lett. 17 (2019) (2018) 319–333, [56] Y.H. Chen, P. Li, Y.S. Huang, K.F. Yu, H.J. Chen, K.P. Cui, Q.L. Huang, J.Y. Zhang,
https://doi.org/10.1007/s10311-018-0800-1. K.Y.H. Gin, Y.L. He, Environmental media exert a bottleneck in driving the dy-
[36] H. Deng, X.L. Li, Q. Peng, X. Wang, J.P. Chen, Y.D. Li, Monodisperse magnetic namics of antibiotic resistance genes in modern aquatic environment, Water Res.
single-crystal ferrite microspheres, Angew. Chem. Int. Ed. 44 (2010) 2782–2785. 162 (2019) 127–138.
[37] R.M. Sellers, Spectrophotometric determination of hydrogen peroxide using po- [57] J.S. Jeong, W.H. Song, W.J. Cooper, J. Jung, J. Greaves, Degradation of tetracycline
tassium titanium(IV) oxalate, Analyst 105 (1980) 950. antibiotics: mechanisms and kinetic studies for advanced oxidation/reduction
[38] Z.J. Yan, X.B. Jiang, Y. Dai, W. Xiao, X.C. Li, N.X. Du, G.H. He, Pulverization control processes, Chemosphere 78 (2010) 533–540.
by confining Fe3O4 nanoparticles individually into macropores of hollow carbon [58] Z. Guo, W.D. Ding, X.W. Liu, Z. Sun, L. Wei, Two-dimensional black phosphorus: a
spheres for high-performance Li-Ion batteries, ACS Appl. Mater. Interfaces 10 new star in energy applications and the barrier to stability, Appl. Mater. Today 14
(2018) 2581–2590. (2019) 51–58.
[39] Stoilova, H. Penchev, T. Ruskov, I. Spirov, N. Manolova, H. Rashkov, One–pot [59] H.R. Dong, Z. Jiang, C. Zhang, J.M. Deng, K.J. Hou, Y.J. Cheng, L.H. Zhang,
preparation of magnetic chitosan beads, Bulgarian Chem. Commun. 40 (2008) G.M. Zeng, Removal of tetracycline by Fe/Ni bimetallic nanoparticles in aqueous
491–497. solution, J. Colloid. Interf. Sci. 513 (2018) 117–125.
[40] Z. Abdeen, S.G. Mohammad, M.S. Mahmoud, Adsorption of Mn (II) ion on polyvinyl [60] W.X. Wang, K.J. Xiao, L. Zhu, Y.R. Yin, Z.M. Wang, Graphene oxide supported ti-
alcohol/chitosan dry blending from aqueous solution, Environ. Nanotechnol. tanium dioxide & ferroferric oxide hybrid, a magnetically separable photocatalyst
Monit. Manage. 3 (2015) 1–9. with enhanced photocatalytic activity for tetracycline hydrochloride degradation,
[41] L. Jin, R.B. Bai, Mechanisms of lead adsorption on chitosan/PVA hydrogel beads, RSC Adv. 7 (2017) 21287–21297.
Langmuir 18 (2002) 9765–9770. [61] X.D. Zhu, Y.J. Wang, R.J. Sun, D.M. Zhou, Photocatalytic degradation of tetra-
[42] M. Piumetti, D. Fino, N. Russo, Mesoporous manganese oxides prepared by solution cycline in aqueous solution by nanosized TiO2, Chemosphere 92 (2013) 925–932.
combustion synthesis as catalysts for the total oxidation of VOCs, Appl. Catal. B: [62] W.R. Chen, C.H. Huang, Transformation kinetics and pathways of tetracycline an-
Environ. 163 (2015) 277–287. tibiotics with manganese oxide, Environ. Pollut. 159 (2011) 1092–1100.
[43] V.I. Nefedov, Y.V. Salyn, G. Leonhardt, R. Scheibe, A comparison of different [63] L. Zhou, W. Song, Z.Q. Chen, G.C. Yin, Degradation of organic pollutants in was-
spectrometers and charge corrections used in X-ray photoelectron spectroscopy, J. tewater by bicarbonate activated hydrogen peroxide with a supported cobalt cat-
Electron Spectrosc. Relat. Phenom. 10 (1977) 121–124. alyst, Environ. Sci. Technol. 47 (2013) 3833–3839.
[44] G. Bhargava, I. Gouzman, C.M. Chun, T.A. Ramanarayanan, S.L. Bernasek, [64] L.J. Xu, J.L. Wang, Magnetic nanoscaled Fe3O4/CeO2 composite as an efficient
Characterization of the “native” surface thin film on pure polycrystalline iron: a Fenton-like heterogeneous catalyst for degradation of 4-chlorophenol, Environ. Sci.
high resolution XPS and TEM study, Appl. Surf. Sci. 253 (2007) 4322–4329. Technol. 46 (2012) 10145–10153.
[45] C.F. Zheng, H.L. Zheng, Y.J. Wang, Y.J. Song, Y.Y. An, H.X. Liu, S. Liu, Modified [65] S.T. Martin, A.T. Lee, M.R. Hoffmann, Chemical mechanism of inorganic oxidants
magnetic chitosan microparticles as novel superior adsorbents with huge “force inthe TiO2/UV process: increased rates of degradation of chlorinated hydrocarbons,
field” for capturing food dyes, J. Hazard. Mater. 367 (2019) 492–503. Environ. Sci. Technol. 29 (1995) 2567–2573.
[46] A.R. Paniego, Reporting physisorption data for gas-solid systems with special re- [66] Y.M. Lin, D.Z. Li, J.H. Hu, G.C. Xiao, J.X. Wang, W.J. Li, X.Z. Fu, Highly efficient
ference to the determination of surface-area and porosity, Pure Appl. Chem. 57 photocatalytic degradation of organic pollutants by PANI-modified TiO2 composite,
(1985) 603–619. J. Phys. Chem. C 116 (2012) 5764–5772.
[47] J. Zhang, F. Shi, J. Lin, D. Chen, J. Gao, Z. Huang, X. Ding, C. Tang, Self-assembled [67] J.C. Espinosa, S. Navalon, A. Primo, M. Moral, J.F. Sanz, M. Alvaro, H. Garcia,
3-D architectures of BiOBr as a visible light-driven photocatalyst, Chem. Mater. 20 Graphenes as efficient metal-free Fenton catalysts, Chem. Eur. J. 21 (2015)
(2008) 2937–2941. 11966–11971.
[48] Y.J. Yao, H. Chen, J.C. Qin, G.D. Wu, C. Lian, J. Zhang, S.B. Wang, Iron

14
Annotations

Heterogeneous Fenton-like degradation of tetracyclines using

porous magnetic chitosan microspheres as an efficient
catalyst compared with two preparation methods
Li, Xiaoyang; Cui, Kangping; Guo, Zhi; Yang, Tingting; Cao, Yong; Xiang,
Yinping; Chen, Huanhuan; Xi, M.

01 Prerna Saini Page 7

20/3/2024 9:45

02 Prerna Saini Page 7

20/3/2024 9:45