Powder Technology 345 (2019) 501–508

Contents lists available at ScienceDirect

Powder Technology

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

Synthesis of Ce-doped magnetic biochar for effective Sb(V) removal:

Performance and mechanism
Li Wang a,⁎, Jingyi Wang a, Zixuan Wang a,b, Jiangtao Feng a, Shanshan Li a, Wei Yan a,⁎
a
Department of Environmental Science and Engineering, State Key Laboratory of Multiphase Flow in Power Engineering, School of Energy and Power Engineering, Xi'an Jiaotong University, Xi'an
710049, China
b
Department of Civil and Environmental Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA

a r t i c l e i n f o a b s t r a c t

Article history: A novel Ce-doped magnetic biochar was successfully synthesized by chemical co-precipitation (Ce/FeCP-BC) and
Received 29 August 2018 solvothermal methods (Ce/FeST-BC) for efficient Sb(V) adsorption. The Ce/FeCP-BC exhibited a high adsorption
Received in revised form 6 January 2019 capacity of 25.0 mg/g, which was two times and one order of magnitude higher than that of Ce/FeST-BC and
Accepted 7 January 2019
un-doped BC, respectively. The excellent adsorption performance of Ce/FeCP-BC maintained over a wide pH
Available online 8 January 2019
range or in the presence of coexisting anions. The adsorption results indicated that co-precipitation method
Keywords:
was superior to solvothermal method and Ce oxide was the main contribution to Sb(V) adsorption enhancement.
Ce-doped The combined results of TEM, XRD, FTIR and XPS revealed that Ce atom was successfully doped into the Fe3O4
Biochar structure and abundant hydroxyl groups were formed on the surface of Ce/FeCP-BC. The magnetic performance
Sb(V) adsorption decreased after Ce doping, but Ce/FeCP-BC still showed good separation potential. Based on the results of batch
Mechanism experiments and physiochemical analyses, the underlying mechanisms controlling Sb(V) adsorption on Ce/
FeCP-BC involved the inner-sphere surface complexation, hydrogen bonding, electrostatic attraction and ligand
exchange. Among them, the ligand exchange and the formation of Ce–O–Sb complex were the main contribu-
tions to the significant Sb(V) adsorption increment on Ce/FeCP-BC. All the results implied that Ce/FeCP-BC could
be used as a promising adsorbent for antimonate decontamination.
© 2019 Elsevier B.V. All rights reserved.

1. Introduction of Sb(V) are greater than Sb(III) in Sb polluted wastewater that mostly
stays in an oxidizing environment [5].
Antimony (Sb) contamination in aqueous systems has been causing An extensive literature survey reveals that numerous techniques
growing concern due to environmental threats posed from its bio- have been applied to antimony removal from water environment
accumulation, poor degradability and environmental persistence [1]. [5,6], among which adsorption is normally deemed to be a superior
The elevated concentrations of antimony in water environment (ppm technique because of its high efficiency, cost effectivity and operational
level) are normally derived from the production of various industrial simplicity [6–8]. Recently, biochar as a novel adsorbent has received
products such as flame retardants, semiconductors, batteries and plas- considerable interest for the removal of heavy metals from wastewater
tics as well as the process of mining, smelting, and metallurgy [2]. due to its low cost and eco-friendly features as well as unique properties
Long-term exposure to antimony could result in various illnesses such such as large surface area, high porosity, and various surface functional
as skin rashes, myalgia, diarrhea, abdominal colic and cardiotoxicity groups [9–11]. Various types of biomass (e.g., agricultural waste, wood
[3]. To minimize its potential threats, the United States Environmental chips, algae, manure, sewage sludge, etc.) have been reported to be
Protection Agency (USEPA, 1979) regulated the stringent limit of 6 used as precursors for BC preparation under oxygen-limited pyrolysis
μg/L, with 5 μg/L set in EU and China, as the maximum permissible anti- conditions [12–14]. Despite that, the surface of the pristine biochar
mony level in drinking water [4]. In addition, Sb(III) and Sb(V) are the (BC) is generally negative charged, which is not favorable for the ad-
oxidation states of Sb in the natural environment, with the usual sorption of anionic contaminants [15]. Moreover, the large-scale use of
forms of Sb(OH)3 and [Sb(OH)6]− in relatively reducing and oxidizing BC is quite limited because of the difficulties entailed in the separation
water environment respectively [2]. Although the toxicity of Sb(III) is of powdered adsorbents [16]. Therefore, suitable modification of BC is
10 times greater than that of Sb(V), the mobility, stability and solubility required.
Magnetic separation is an excellent option for easy separation of ad-
sorbents from aqueous solutions. Fe3O4 is one of the most commonly
⁎ Corresponding authors. used magnetic materials to endow the BC with magnetic property
E-mail addresses: wangli-2015@xjtu.edu.cn (L. Wang), yanwei@xjtu.edu.cn (W. Yan). [17–19]. However, its specific adsorption affinity towards Sb(V) still

https://doi.org/10.1016/j.powtec.2019.01.022
0032-5910/© 2019 Elsevier B.V. All rights reserved.
502 L. Wang et al. / Powder Technology 345 (2019) 501–508

remains to be improved. Recently, various transition metal elements until the pH reached 7. Then the products were separated and dried in
have gained increasing attention for the adsorption of Sb(V) from aque- an oven at 60 °C for 12 h. All the products were ground to pass through
ous solutions [20–22]. For instance, a Fe\\Zr bimetal oxide fabricated a 200 mesh sieve and stored in sealed bottles respectively for further
via co-precipitation method exhibited significant enhancement in Sb use.
(V) removal compared to the referenced ferric oxide [23]. In contrast The supernatant was filtrated and the residual concentrations of Ce
to the pure Fe3O4, Cu-Fe3O4 nanoparticle prepared by solvothermal and Fe were detected by inductively coupled plasma emission spec-
method increased Sb(V) adsorption capacity from 7.09 mg/g to trometer (ICPE-9000, Shimadzu Company, Japan). Results showed that
30.92 mg/g [24]. all the Ce and Fe were loaded on the BC’ s surface as no Ce or Fe ions
Cerium (Ce), one of the most abundant and the least expensive rare were detected. For comparison, un-doped Fe-BC was also synthesized
earth metals, has been proved that the presence of Ce ions makes adsor- using the same method as Ce/FeCP-BC without adding CeCl3·7H2O. All
bent surface more positively charged, which would contribute to attract the chemicals used in this work are of analytical grade and solutions
the negative adsorbates [25] such as arsenate, chromate, phosphate and were prepared with deionized water.
fluoride [26,27]. Besides, numerous hydroxyl groups could be normally
formed on the surface of Ce-doped adsorbents and displaced by certain 2.2. Batch experiments
anions, which would lead to the excellent specific adsorption perfor-
mance [28]. It has been demonstrated that the Ce-doped Fe3O4 showed The adsorption performances of the modified BCs were investigated
a dramatically higher As(V) adsorption capacity than pure Fe3O4 [29]. through a series of batch experiments. The effects of initial Sb
As Sb and As both locate in VA group with similar structure, Ce could (V) concentration (10.0–100.0 mg/L), contact time, initial solution pH
be used as a good doping element to improve Sb removal efficiency. (2.0–12.5) and coexisting anions (Cl−, SO42−, NO3−, HCO3−, H2PO4−) on
However, pure metal oxide nanoparticles are inclined to agglomerate the removal of Sb(V) were studied successively. The stock solution of
during the synthesis and adsorption process, which may decrease 500 mg/L Sb(V) ions was prepared by dissolving KSbO6H6 (1.079 g) in
their removal efficiency and stability [30]. Given these issues, combining 1.0 L of deionized water. The solutions of required concentrations for
Ce-doped Fe3O4 and BC matrix is an excellent option for facilitating the subsequent tests were freshly prepared before use by diluting the
dispersion degree of metal oxides, Sb(V) adsorption performance as stock solution with deionized water and adjusting pH using HCl and
well as easy separation. Until now, no systematic study concerning the NaOH solutions (0.001 to 2.0 mol/L).
adsorption performance of Ce-doped magnetic BC in Sb(V) removal The batch adsorption equilibrium experiments were performed in
from aquatic environment has been published yet. 50 mL polypropylene centrifuge tubes by shaking 0.02 g of each adsor-
Chemical co-precipitation and solvothermal methods are two com- bent in 20 mL Sb(V) solution with a certain pH and concentration. All
monly used methods for the development of metal oxide nanoparticles the adsorption experiments were carried out in a thermostatic shaker
or metal modified BCs as extensively reported in literature [23,24], lead- (150 rpm) at 25 °C for 15 h to reach equilibrium. For the adsorption ki-
ing to different morphologies of the materials. In view of the above con- netic experiment, 0.2 g each adsorbent was added into 200 mL Sb
siderations, the main aims of the research are to 1) synthesize Ce-doped (V) solution with an initial concentration of 50 mg/L. About 2.0 mL of
magnetic BCs via two different methods (co-precipitation and the solution was taken out at predetermined time intervals (1, 3, 7,
solvothermal method); 2) compare and evaluate their aqueous Sb 10, 15, 21, 28, 36, 45, 55, 70, 90, 120, 150, 180, 240, 300, 360, 450,
(V) adsorption performances as well as mechanisms involved through 600 min). Afterwards, each sample was separated by a 0.45 μm mem-
a series of batch experiments and BET, zeta potential, XRD, FTIR and brane filter. The residual Sb(V) concentration was also measured by
XPS analyses. ICPE, in which the detection limit of Sb was 0.03 mg/L. The adsorption
amount of Sb(V) at equilibrium was calculated as: Qe = (C0\\Ce)V/W,
2. Materials and methods where Qe represents the adsorption amount (mg Sb(V)/g adsorbent)
at equilibrium, C0 and Ct are Sb(V) concentrations at the initial moment
2.1. Adsorbent preparation and equilibrium (mg/L), respectively; V is the solution volume (L); and
W is the adsorbent mass (g). The adsorption amount at time t, that is, Qt
Phragmites australis (PA), a common hydrophyte, was used as the (mg/g), was calculated as: Qt = (C0-Ct)V/W, where C0 and Ct (mg/L) are
biomass precursor for BC preparation. PA was collected from Chan Ba Sb(V) concentrations at the initial moment and time t, respectively; V is
wetland, Shaanxi province, China. The biomass was first cleaned with the solution volume (L); and W is the adsorbent mass (g). The adsorp-
deionized water and dried at 80 °C for one day. Then the sample was tion isotherms were analyzed by Langmuir, Freundlich and Dubinin–
pulverized into small pieces and heated in a microwave oven (MKX- Radushkevich isotherm (D-R) models, their equations were shown in
M1B, Qingdao, China) at 700 W (600 °C) for 20 min under N2 atmo- Eqs. (S1)–(S3). The batch kinetic data were calculated using pseudo-
sphere. After that, the sample was cooled down to room temperature first order, pseudo-second order, elovich and intra-particle diffusion
and ground to sieve through 200 meshes [31,32], then the pristine BC models by Eqs. (S4)–(S7).
was acquired.
The Ce-doped magnetic BCs were synthesized by two methods: 2.3. Adsorbent characterization
chemical co-precipitation (Ce/FeCP-BC) and solvothermal method (Ce/
FeST-BC). Specifically, CeCl3·7H2O (0.46 g), FeCl3·6H2O (0.76 g) and The morphologies and element compositions of the adsorbents were
FeCl2·4H2O (0.27 g) in a molar ratio of 0.3 (Ce:Fe) were dissolved in characterized on a scanning electron microscope (SEM, Gemini SEM
50 mL deionized water under vigorous stirring to form a homogeneous 500, Zeiss, Germany) and a transmission electron microscopy (TEM,
solution. Then 2.0 g of BC with 10 mL ethyl alcohol as the dispersing JEM-2100Plus, Japan) with an energy dispersive X-ray spectrometer
agent was added to the solution and stirred at room temperature for (EDX). The X-ray diffraction (XRD) patterns of the adsorbents were col-
2 h. The pH of the suspension was adjusted to 10 by adding about lected on a diffractometer (XRD, PANalytical, the Netherlands) with Cu
2.5 mL ammonia solution (28% in H2O) dropwise. In the synthesis of Ka radiation of 1.5418 Å in the 2θ range of 10°-90°. The average size of
Ce/FeCP-BC sample, the mixture was shaken in a water bath shaker the crystallite was calculated according to the Scherrer's equation: Ds
(THZ-82) at 80 °C (180 rpm) for another 2 h. In the synthesis of Ce/ = Kλ/βcosθ, where Ds is the mean size of the ordered (crystalline) do-
FeST-BC sample, the mixture was transferred into a stainless Teflon mains; K is a constant that depends on the morphology of crystallites,
lined autoclave. The autoclave was sealed and maintained at 180 °C usually assumed as unity; λ is the X-ray wavelength; β is the half
for 12 h [25]. After being naturally cooled down to room temperature, width of diffraction peak; θ is the Bragg's angle. The magnetic properties
the two products were washed several times with deionized water of the adsorbents were measured by using a vibrating sample
 L. Wang et al. / Powder Technology 345 (2019) 501–508 503

magnetometer (VSM, MPMS-SQUID VSM-094 Quantum, USA). The spe- FeST-BC could be better fitted by Freundlich model, while that of Fe-BC
cific surface area, pore volume and pore size of the adsorbents were cal- was better matched by Langmuir model. This indicated that the adsorp-
culated using the Brunauer-Emmett-Teller (BET) and Barrett-Joyner- tion of Sb(V) on Ce/FeCP-BC and Ce/FeST-BC occurred via a chemical and
Halenda (BJH) method on a Builder SSA-4200 (Beijing, China) at 77 K. multiple layer process, while on Fe-BC was a physical and monolayer
The zeta potentials of the adsorbents were determined by a zeta poten- process. Compared to Fe-BC and BC, the value of 1/n (1/n b 0.5) indi-
tial analyzer (Zetasizer Nano ZS90, Malvern, UK). Fourier transform in- cated that Sb(V) ions were favorably adsorbed onto the surface of Ce/
frared spectroscopy (FTIR, Bruker, Germany) was used to analyze the FeCP-BC and Ce/FeST-BC [37]. The D-R model could identify whether
functional groups of the adsorbents in a KBr pellet in the range of 400 the adsorption is rate-limited by physical or chemical absorption
and 4000 cm−1. The surface elemental compositions and functional based on the value of mean adsorption energy E (kJ/mol) [10]. The E
groups were further quantified by X-ray photoelectron spectroscopy values of Ce/FeCP-BC and Ce/FeST-BC were 11.54 and 8.49 kJ/mol, respec-
(XPS, ESCALAB Xi+, Thermo Fisher Scientific, USA) employing Al Kα ra- tively, while that of Fe-BC was lower than 8.0 kJ/mol. This further con-
diation. The C 1 s signal at 284.8 eV was used as the reference for firmed that the adsorption of Sb(V) on Ce/FeCP-BC and Ce/FeST-BC was
analysis. chemical adsorption in nature while on Fe-BC was physical
adsorption [10].
3. Results and discussion
3.2. Adsorption kinetics
3.1. Adsorption isotherms
The study of adsorption kinetics is essential to evaluate the adsorp-
The equilibrium adsorption isotherm is crucial to assess the adsorp- tion efficiency as well as to expound the mechanism of the adsorption
tion performance and affinity towards Sb(V). Fig. 1 illustrates Sb reaction. As can be seen in Fig. 2A, the adsorption of Sb(V) on Ce/FeCP-
(V) adsorption isotherms on Ce/FeCP-BC, Ce/FeST-BC, Fe-BC and BC. BC and Ce/FeST-BC was rapid in the first 160 min, with about 80% of
The adsorption capacity (Qe) improved with the increase of Sb equilibrium adsorption capacity achieved because of the substantial
(V) concentrations for the larger driving force arisen from higher con- binding sites and high concentration gradient. Then the adsorption
centration gradient [33]. It was clear that Ce/FeCP-BC exhibited the rate gradually slowed down and equilibrium was reached at 600 min.
highest Qe in conditions of all the Sb(V) concentrations, followed by Therefore, 900 min was applicable for other equilibrium experiments.
Ce/FeST-BC, Fe-BC and BC. The Qe gradually reached a plateau and The Sb(V) adsorption capacity of Ce/FeCP-BC displayed approximately
achieved a maximum capacity of 25.0 mg/g for Ce/FeCP-BC, which was twice as much as that of Ce/FeST-BC.
almost two times and one order of magnitude higher than that of Ce/ Four classic adsorption kinetic models, the pseudo-first order,
FeST-BC and Fe-BC, respectively. This indicated that chemical co- pseudo-second order, Elovich and intraparticle diffusion models were
precipitation method was more effective than solvothermal method applied to fit the experimental data, with their parameters recorded in
and the doped Ce played a crucial role during the Sb(V) adsorption in Table S2. For both Ce/FeCP-BC and Ce/FeST-BC, the adsorption kinetic
contrast to Fe. The Sb(V) adsorption capacity on Ce/FeCP-BC was higher data can be best fitted by the pseudo-second order model due to the
compared to other adsorbents as reported in literature, such as highest correlation coefficients (R2) and the closest values between
10.0 mg/g for MnFe2O4 [16], 7.5 mg/g for α-FeOOH [34], 11.0 mg/g for the calculated adsorption capacities (Qe,cal) and the experimental values
α-Fe2O3 [35], and 4.26–24.48 mg/g for biochar derived from Soybean (Qe,exp). This indicated that chemisorption was the rate-limiting step of
stover [36], which indicated that Ce/FeCP-BC could be selected as a Sb(V) adsorption on Ce/FeCP-BC and Ce/FeST-BC. The intraparticle diffu-
promising candidate for Sb(V) removal from aqueous solution. sion model (Fig. 2B) revealed that the adsorption process occurred in
Three equilibrium adsorption isotherm equations, Langmuir, two steps: boundary layer diffusion on the surface of the adsorbents
Freundlich and Dubinin–Radushkevich isotherm (D-R), were applied and intra-particle diffusion when the binding sites on the outer surface
to evaluate the nature of the adsorption process and the relevant pa- reached saturation [28]. As linear portion did not pass through the ori-
rameters were provided in Table S1. The Langmuir model supposes gin, the adsorption of Sb(V) on Ce/FeCP-BC and Ce/FeST-BC encompassed
that the adsorption process occurs at homogeneous surfaces with ion other steps.
exchange as the dominating mechanism, while the Freundlich model
suggests multilayer heterogeneous adsorptions and adsorption- 3.3. Effect of pH and coexisting anions on Sb(V) adsorption
complexation interactions dominate the sorption process. According
to the correlation coefficient, the isotherm data of Ce/FeCP-BC and Ce/ The effects of pH on the adsorption of Sb(V) on Ce/FeCP-BC and Ce/
FeST-BC were investigated and the results were depicted in Fig. 3A. It
was evident that the adsorption capacity decreased with the increase
of solution pH, with Ce/FeCP-BC maintaining higher adsorption capacity
than Ce/FeST-BC over a wide range of pH (2−12). The results regarding
the optimum pH for Sb(V) removal were in agreement with literature
reports [16,25,38]. The solution pH has a profound effect on the adsorp-
tion performance for it would influence the surface charge of adsorbents
as well as metallic species [9]. Ce/FeCP-BC and Ce/FeST-BC possessed
pHPZC values of 5.8 and 4.5, respectively (Table 1), and the [Sb(OH)6]−
is the only dominating species at pH N 2.0. At low pH b pHPZC, the hy-
droxyl groups on the adsorbent surface was protonated and formed
an electron positive potential, which facilitated the adsorption of an-
ionic [Sb(OH)6]− via electrostatic attraction. When the solution pH is
higher than pHPZC, the adsorption of Sb(V) was suppressed due to the
electrostatic repulsion between the negative charged surface sites and
[Sb(OH)6]− as well as the competition from hydroxyls (OH−).
Several anions (Cl−, SO42−, NO3−, HCO3−, H2PO4−) often coexist in nat-
Fig. 1. Adsorption isotherms for Sb(V) adsorption on the adsorbents. The dots and lines
represent the experimental data and fitting curves of calculated models, respectively.
ural water and can compete with Sb(V) for the surface active sites of the
Experimental conditions: initial C0 10–100 mg/L, contact time 15 h, dosage 1.0 g/L, adsorbents. Therefore, it is necessary to investigate the effect of
initial pH 7.5, temperature 25 ± 1 °C. coexisting anions on Sb(V) adsorption of Ce/FeCP-BC and Ce/FeST-BC
504 L. Wang et al. / Powder Technology 345 (2019) 501–508

Fig. 2. Effect of contact time (A) and intraparticle diffusion kinetic model (B) for Sb(V) adsorption on Ce/FeCP-BC and Ce/FeST-BC. Experimental conditions: initial C0 50 mg/L, dosage 1.0 g/L,
initial pH 7.5, temperature 25 ± 1 °C.

(Fig. 3B). For both Ce/FeCP-BC and Ce/FeST-BC, the presence of Cl−, SO42− The porous structure analysis of all the adsorbents was determined
or NO3− had negligible influence on the adsorption capacity due to the and the parameters were presented in Table 1. The SBET and Vtot of the
less binding energy than Sb(V) [28]. However, HCO3– and H2PO4− exhib- Fe-BC were higher compared to those of BC, owing to the introduction
ited inhibitory effects on Sb(V) removal, for which a significant reduc- of Fe ions. The Ce/FeCP-BC and Ce/FeST-BC had similar BET surface area
tion of 50% and 74% were found with the presence of phosphate for to Fe-BC, but lower average pore diameters. The introduction of Ce
Ce/FeCP-BC and Ce/FeST-BC, respectively. The decrease can be attributed ions could render the surfaces to be rough and folded, thus decreasing
to the competition for adsorption active sites between HCO3−/H2PO4− the average pore diameter consequentially. The zeta potentials of the
and Sb(V). The influence of H2PO4− on the decline of Sb(V) removal adsorbent suspensions were measured (Fig. S2) and their point of
was more obvious than HCO3−, which was likely due to the higher affin- zero charges (pHPZC) were provided in Table 1. The BC had a low
ity of H2PO4− towards Ce active sites [39,40]. pHPZC of 2.3, which was not favorable for Sb(V) adsorption, hence the
adsorption of Sb(V) on BC in neutral solution was almost negligible.
The pHPZC of Fe-BC was found to be 4.3 and the value increased to
3.4. Characterization of the modified Biochars 4.5–5.8 after the introduction of Ce ions. The Ce/FeCP-BC had higher
pHPZC than that of Ce/FeST-BC, thereby resulting in higher adsorption ca-
The morphologies and structures of Fe-BC, Ce/FeCP-BC and Ce/FeST- pacity due to the stronger electrostatic attraction for negative adsorbate
BC were characterized by SEM and TEM. The surface of Fe-BC was rela- Sb(V) in acid and neutral solutions.
tively smooth with distinct macroporous structure (Fig. 4). After modi- The crystal structures of Fe-BC, Ce/FeCP-BC and Ce/FeST-BC before
fication of different methods, small differences were observed in the and after Sb(V) adsorption were characterized and their XRD patterns
surface morphology of Ce/FeCP-BC and Ce/FeST-BC. A group of large par- were illustrated in Fig. 5A. The Fe-BC had five major peaks at 30.10o,
ticles were deposited on the surface of Ce/FeCP-BC, which made the sur- 35.45o, 43.08 o, 56.98o and 62.57o, which attributed to (220), (311),
face rough and irregular. In comparison, some small crystals, which (400), (511), and (440) crystal planes of magnetite, respectively [41].
were clearly visible at high magnification, were formed on the surface The strong and sharp reflections of XRD peaks in Fe-BC suggested the
of Ce/FeST-BC. The TEM-EDX mapping images showed that Fe and Ce el- formation of magnetite crystallites in amorphous biochar matrix. How-
ements were homogeneously dispersed in the whole Ce/FeCP-BC and ever, after modified with Ce ions, the magnetite displayed weaker dif-
Ce/FeST-BC, with the weight percentage of 12.13%, 5.68% and 9.22%, fraction intensity peaks in Ce/FeCP-BC and Ce/FeST-BC, with their
4.55%, respectively (Table 1). All these confirmed that Fe and Ce nano- positions changed, indicating that the Ce ion addition significantly influ-
particles were successfully loaded on BC's surface. enced the crystallization of magnetite. No obvious peaks of cerium

Fig. 3. Effect of solution pH (A) and coexisting anions (B) for Sb(V) adsorption on Ce/FeCP-BC and Ce/FeST-BC. Experimental conditions in A: dosage 1.0 g/L, temperature 25 ± 1 °C, initial Sb
(V) C0 50 mg/L, and contact time 15 h; Experimental conditions in B: dosage 1.0 g/L, temperature 25 ± 1 °C, initial Sb(V) and anions C0 50 mg/L, initial pH 7.5 and contact time 15 h.
 L. Wang et al. / Powder Technology 345 (2019) 501–508 505

Table 1 ions was contributed to the disrupted crystallinity of magnetite, which

General characteristic of the adsorbents. was consistent with XRD analysis. The solvothermal method could pro-
BC Fe-BC Ce/FeCP- BC Ce/FeST-BC vide a more reductive condition, which was more suitable for the for-
C (wt%) 68.00 77.90 53.92 61.90
mation of crystallinity than co-precipitation method. Therefore, Ce/
Fe (wt%) – 7.02 12.13 9.22 FeST-BC had higher Ms. than Ce/FeCP-BC. Although the Ms. decreased
Ce (wt%) – – 5.68 4.55 after doping with ions, it was in the range as reported in literature
a
SBET (m2/g) 111.2 232.69 230.65 269.9 (0.05–68.2 emu/g).
b
Vtot (cm3/g) 0.19 0.28 0.19 0.22
c The FTIR spectra of the samples before and after Sb(V) adsorption
Dp (nm) 6.98 4.78 3.33 3.28
d
pHPZC 2.3 4.3 5.8 4.5 were obtained to determine the presence of various functional groups
(311) Peak – 35.448 35.581 35.645 on the adsorbents. The Fe-BC, Ce/FeCP-BC and Ce/FeST-BC had similar in-
Position (°) frared absorption bands as displayed in Fig. 5B. All the adsorbents
(311) peak shift (°) – 0.006 −0.127 −0.191 contained large amounts of –OH as located between 3500 and
Crystalline size (nm) – 13.0 17.3 23.3
Saturated magnetization (emu/g) 0.00 5.00 0.93 2.11
3304 cm−1 [43,44]. In the spectrum of Fe-BC, a strong characteristic
a
peak at 580 cm−1 for the Fe\\O vibration of magnetite was observed,
SBET: BET surface area was determined by using the Brunauer–Emmett–Teller (BET)
further confirming the formation of magnetite on BC's surface [25]. In
theory;
b
Vtot: total pore volume was determined using density functional theory (DFT) the spectrum of Ce/FeCP-BC and Ce/FeST-BC, no obvious Ce band was
method; identified probably due to low Ce loading content. Besides, the Fe\\O
c
Dp: average pore diameter was calculated as Dp = 4Vtot/SBET; band almost disappeared after doping Ce ions via either co-
d
pHPZC: Point of zero charge of adsorbent. precipitation or solvothermal method, indicating that the crystal struc-
ture of magnetite was destroyed [45]. After Sb(V) adsorption, the spec-
oxide were observed in the samples of Ce/FeCP-BC and Ce/FeST-BC, indic- trum of Fe-BC did not change much except that the intensity of Fe\\O
ative of the probable amorphous forms of cerium oxide. The crystalline band markedly weakened, revealing formation of Fe–O–Sb complexes
sizes of Ce/FeCP-BC and Ce/FeST-BC were higher than that of Fe-BC on Fe-BC surfaces as has been previously observed in literature [46].
(Table 1), suggesting that Ce ions were doped into the lattice of magne- For Ce/FeCP-BC and Ce/FeST-BC, the intensity of –OH band at
tite as Ce3+ ions (115.0 pm) had larger ionic radius than Fe3+ ions 3048 cm−1 significantly increased and the peak position showed an ob-
(78.5 pm). The combined form of the spinel structure in a lattice cell vious red-shift after Sb(V) adsorption, which could be caused by the –
was assumed based on relevant researches (Fig. 6.) [25,42]. After the ad- OH in the adsorbed Sb(V) species.
sorption of Sb(V) ions, the peak intensity of Fe-BC, Ce/FeCP-BC and Ce/ The surface functional groups on the adsorbents as well as the mech-
FeST-BC further weakened slightly, implying chemical interactions be- anisms were further elucidated by XPS analysis. The full survey XPS
tween Ce/Fe and Sb(V) to realize enhanced adsorption. spectrum illustrated that the main compositions of Fe-BC, Ce/FeCP-BC
The magnetic performance of modified BCs was characterized by and Ce/FeST-BC were C and O (Fig. 5C), which were consistent with
magnetic hysteresis loop (Fig. 1S) and their saturated magnetizations EDS analysis (Table 1). Fe element was detected in all three adsorbents
were listed in Table 1. The magnetic saturations (Ms) of Ce/FeCP-BC and a new peak corresponding to Ce element was observed on Ce/FeCP-
and Ce/FeST-BC were 0.93 and 2.11 emu/g, respectively, which were BC and Ce/FeST-BC, which further validated that Ce ions were success-
lower than that of Fe-BC. The reduction in Ms. after doping with Ce fully loaded on BC’ s surface. The newly emerged Sb 3d spectra at

Fig. 4. The first line is the SEM images of Fe-BC, Ce/FeCP-BC and Ce/FeST-BC, the second and third lines are TEM of Ce/FeCP-BC and Ce/FeST-BC and their C, Fe, Ce elemental mapping images.
506 L. Wang et al. / Powder Technology 345 (2019) 501–508

Fig. 5. XRD patterns (A), FTIR spectra (B) and XPS spectra (C) of the adsorbents before and after Sb(V) adsorption for Fe-BC, Ce/FeCP-BC and Ce/FeST-BC. The Fe 2p3/2 and Fe 2p1/2 spectra
(D) and Ce 3d5/2, Ce 3d3/2 spectra of Ce/FeCP-BC (E) and Ce/FeST-BC (F) before and after Sb(V) adsorption.

540.5 eV after Sb(V) adsorption implied the successful Sb loading on the After the Sb(V) adsorption, the percentage of M–OH in Ce/FeCP-BC dra-
adsorbents. matically decreased to 60.56%, while the ratio of M–O almost increased
The high resolution scan of O 1 s spectra of Fe-BC, Ce/FeCP-BC and Ce/ in the same proportion as M–OH. This was likely caused by the replace-
FeST-BC could be deconvoluted into three peaks of lattice oxygen, M–O ment of hydroxyl groups on Sb(V) species via ligand exchange, with the
(528.9–530.6 eV), hydroxyl groups, M–OH (531.6–533.0 eV) and formation of M–O–Sb after adsorption [47]. Although an obvious reduc-
adsorbed water (533.0–535.6 eV) [28,47] as shown in Fig. S3. The rela- tion was also observed for M–OH in Fe-BC, the ratio of M-O did not
tive content of each component was calculated based on the peak area change after adsorption. This indicated that no Fe–O–Sb complex was
and was presented in Table 2. The Fe-BC and Ce/FeCP-BC had similar ox- formed and M–OH decrease was caused by the formation of hydrogen
ygen distributions and possessed considerable amounts of M–OH with bonding with Sb(V) species. For Ce/FeST-BC, the oxygen species showed
high ratios of 72.32% and 72.33%, respectively. The Ce/FeST-BC had no obvious changes after adsorption, indicating that other mechanisms
lower content of M–OH but higher content of M-O compared to the were involved.
other two adsorbents, indicating more cerium oxides were formed in The high resolution XPS spectra of Fe 2p before and after ad-
the course of solvothermal synthesis than co-precipitation process. sorption were displayed in Fig. 5D. The Fe-BC contained two
 L. Wang et al. / Powder Technology 345 (2019) 501–508 507

Fig. 6. Schematic diagram of the formation of Ce/FeCP-BC and its mechanisms for Sb(V) adsorption.

major peaks at approximately 711.1 and 724.8 eV [48,49], which generated on the surface of Ce/FeCP-BC after the co-precipitation
were assigned to Fe 2p3/2 and Fe 2p1/2 of standard magnetite, re- method. In the enlarged view of Ce/FeCP-BC, one Ce3+ ion replaced
spectively. The Ce/FeST-BC exhibited similar Fe 2p peaks but with Fe3+ at octahedral sites in a lattice structure, which was speculated ac-
lower intensity compared to Fe-BC, suggesting lower magnetite cording to relevant researches in CeIII-doped Fe3O4 magnetic particles
content after doped with Ce ions via solvothermal method. For [25] and cobalt ferrites [51]. After the Sb(V) adsorption, the abundant
Ce/Fe CP -BC, the intensity of Fe 2p 3/2 decreased and its position presence of –OH on Ce/FeCP-BC was replaced by the Sb(V) species via li-
shifted slightly to higher binding energy sides, indicating that Ce gand exchange [47] in accordance with the decreased content of M–OH
ions were doped into magnetite structure via co-precipitation species, and the subsequently formed inner-sphere surface complex can
method [50]. After Sb(V) adsorption, the Fe 2p spectra of all the be demonstrated by the increased content of M–O species as well as the
adsorbents did not change much, which further confirmed that positive shifts of Ce 3d XPS spectra. Meanwhile, the –OH on the surface
Fe–O–Sb complex was not formed during the adsorption process. of Ce/FeCP-BC could act as H-donor and formed hydrogen bonding with
The XPS spectrum of Ce 3d in both Ce/FeCP-BC and Ce/FeST-BC exhib- the oxygen atom of the Sb(V) complex (H-acceptor) [52]. At low solu-
ited two major peaks of Ce 3d5/2and Ce 3d3/2 that located at 885.8 and tion pH, the deprotonation of hydroxyl groups could adsorb the nega-
904.1 eV, respectively (Fig. 5 E and F) [25], indicating that the dominat- tive adsorbate [Sb(OH)6]− via electrostatic attraction. Among these
ing chemical state of Ce was Ce3+. A weak peak located at 916.8 eV mechanisms, the ligand exchange and the formation of Ce–O–Sb com-
(Ce5+) also existed, possibly due to the oxidization by the oxygen con- plex were the main contributions to the significant Sb(V) adsorption in-
taining groups on BC's surface during the synthesis process. After Sb crement on Ce/FeCP-BC.
(V) adsorption, the peaks of Ce 3d in Ce/FeCP-BC shifted to higher bind-
ing energy, implying the formation of new Ce–O–Sb inner-sphere com- 4. Conclusions
plex. However, no apparent changes were found in Ce/FeST-BC. This was
probably because the Ce element was crystallized during the In this work, two methods (chemical co-precipitation and
solvothermal process, which was not beneficial for the formation of solvothermal method) were used to synthesize Ce-doped magnetic bio-
Ce–O–Sb complex and resulted in a low Sb(V) adsorption capacity. chars. The chemical co-precipitation method was more effective than
Therefore, the formation of Ce–O–Sb complex was the main contribu- solvothermal method in this system for Sb(V) removal, with Ce/FeCP-
tion to the Sb(V) adsorption enhancement in Ce/FeCP-BC. BC exhibiting a higher adsorption capacity of 25.0 mg/g than Ce/FeST-
BC and Fe-BC. The Sb(V) adsorption capacity on Ce/FeCP-BC was high
3.5. Adsorption mechanisms of Ce/FeCP-BC at acid conditions due to the electrostatic attraction. The presence of
coexisting anions had negligible effect on Sb(V) adsorption except
Based on the synthesis procedure of Ce/FeCP-BC and the results from HCO3– and H2PO4−. The combined results of physiochemical character-
the adsorption performance and physiochemical analyses, the mecha- izations demonstrated that Ce atoms were successfully doped into mag-
nisms involved in the Sb(V) adsorption on Ce/FeCP-BC were illustrated netite structure and abundant hydroxyl groups were formed on the
in Fig. 6. The adsorption capacity in Ce/FeCP-BC was one order of magni- surface of Ce/FeCP-BC, which played a dominant role in the adsorption
tude higher than that of Fe-BC, indicating the crucial role of Ce element enhancement. Based on the analyses, the adsorption mechanisms in-
in Sb(V) adsorption enhancement. The combined results of XRD, FTIR cluded the formation of Ce–O–Sb inner-sphere surface complex, hydro-
and XPS analyses revealed that Ce ions were far more likely to be gen bonding, electrostatic attraction and ligand exchange, and the
doped into the structure of magnetite and large amounts of –OH were inner-sphere complexation was the dominant contribution. The high
Sb(V) adsorption performance indicated that Ce/FeCP-BC could be
used as a promising adsorbent candidate for Sb(V) removal in aqueous
Table 2 solution.
Distribution (%) of oxygen forms of the adsorbents before and after Sb(V) adsorption.

Samples M-O M-OH H2O (ads) Acknowledgements

528.9–530.6 eV 531.6–533.0 eV 533.0–535.6 eV
This work was supported by the Natural Science Basic Research Plan
Fe-BC 11.66 72.32 16.02
Ce/FeCP-BC 11.65 72.33 16.02
in Shaanxi Province of China [grant number 2018JQ4014], the Funda-
Ce/FeST-BC 21.22 49.58 29.20 mental Research Funds for the Central Universities [grant number
Fe-BC + Sb 11.70 64.77 23.52 xjj2017172], Shaanxi Postdoctoral Science Foundation funded project
Ce/FeCP-BC + Sb 25.06 60.56 14.38 [grant number 2017BSHEDZZ65] and China Postdoctoral Science Foun-
Ce/FeST-BC + Sb 21.54 50.90 27.56
dation funded project [grant number 2016M592798].
508 L. Wang et al. / Powder Technology 345 (2019) 501–508

Appendix A. Supplementary data [26] H. Qiu, C. Liang, J. Yu, Q. Zhang, M. Song, F. Chen, Preferable phosphate sequestration
by nano-La(III) (hydr)oxides modified wheat straw with excellent properties in re-
generation, Chem. Eng. J. 315 (2017) 345–354.
Supplementary data to this article can be found online at https://doi. [27] P.K. Mishra, R. Kumar, P.K. Rai, Surfactant-freeone-pot synthesis of CeO2, TiO2 and
org/10.1016/j.powtec.2019.01.022. Ti@Ce oxide nanoparticles for the ultrafast removal of Cr(VI) from aqueous media,
Nanoscale 10 (2018) 7257–7269.
[28] Z.P. Wen, J. Ke, J.L. Xu, S. Guo, Y.L. Zhang, R. Chen, One-step facile hydrothermal syn-
References thesis of flowerlike Ce/Fe bimetallic oxides for efficient as(V) and Cr(VI) remedia-
tion: performance and mechanism, Chem. Eng. J. 343 (2018) 416–426.
[1] L. Han, H. Sun, K.S. Ro, K. Sun, J.A. Libra, B. Xing, Removal of antimony (III) and cad- [29] Y. Zhang, M. Yang, X.M. Dou, H. He, D.S. Wang, Arsenate adsorption on an Fe-Ce bi-
mium (II) from aqueous solution using animal manure-derived hydrochars and metal oxide adsorbent: role of surface properties, Environ. Sci. Technol. 39 (2005)
pyrochars, Bioresour. Technol. 234 (2017) 77–85. 7246–7253.
[2] M. Filella, N. Belzile, Y.W. Chen, Antimony in the environment: a review focused on [30] Q. Zhou, B. Liao, L. Lin, W. Qiu, Z. Song, Adsorption of Cu(II) and Cd(II) from aqueous
natural waters I. Occurrence, Earth-Sci. Rev. 57 (2002) 125–176. solutions by ferromanganese binary oxide–biochar composites, Sci. Total Environ.
[3] J. Luo, X. Luo, J. Crittenden, J. Qu, Y. Bai, Y. Peng, J. Li, Removal of antimonite (Sb(III)) 615 (2018) 115–122.
and antimonate (Sb(V)) from aqueous solution using carbon nanofibers that are [31] L. Wang, W. Yan, C. He, H. Wen, Z. Cai, Z.X. Wanga, Z.Z. Chen, W.F. Liu, Microwave-
decorated with zirconium oxide (ZrO2), Environ. Sci. Technol. 49 (2015) assisted preparation of nitrogen-doped biochars by ammonium acetate activation
11115–11124. for adsorption of acid red 18, Appl. Surf. Sci. 433 (2018) 222–231.
[4] Y.-Y. Wang, H.-Y. Ji, H.-H. Lu, Y.-X. Liu, R.-Q. Yang, L.-L. He, S.-M. Yang, Simultaneous [32] L. Wang, Z. Chen, H. Wen, Z. Cai, C. He, Z. Wang, W. Yan, Microwave assisted mod-
removal of Sb(III) and Cd(II) in water by adsorption onto a MnFe2O4 biochar nano- ification of activated carbons by organic acid ammoniums activation for enhanced
composite, RSC Adv. 8 (2018) 3264–3273. adsorption of acid red 18, Powder Technol. 323 (2018) 230–237.
[5] J. Li, B. Zheng, Y. He, Y. Zhou, X. Chen, S. Ruan, Y. Yang, C. Dai, L. Tang, Antimony con- [33] J.J. Zhao, Y.Z. Niu, B. Ren, H. Chen, S.X. Zhang, J. Jin, Y. Zhang, Synthesis of
tamination, consequences and removal techniques: a review, Ecotoxicol. Environ. Schiff base functionalized superparamagnetic Fe3O4 composites for effective
Saf. 156 (2018) 125–134. removal of Pb(II) and Cd(II) from aqueous solution, Chem. Eng. J. 347
[6] S.Y. Pu, H. Ma, A. Zinchenko, W. Chu, Novel highly porous magnetic hydrogel beads (2018) 574–584.
composed of chitosan and sodium citrate: an effective adsorbent for the removal of [34] D. Li, X. Hu, Y. Sun, S. Su, A. Xia, H. Ge, Goethite (α-FeOOH) nanopowders synthe-
heavy metals from aqueous solutions, Environ. Sci. Pollut. Res. 24 (2017) sized via a surfactant-assisted hydrothermal method: Morphology, magnetic prop-
16520–16530. erties and conversion to rice-like α-Fe2O3 after annealing, RSC Adv. 5 (2015).
[7] L. Wang, H. Wen, L. Li, Scale dependence of surface complexation capacity and rates [35] X. Guo, Z. Wu, M. He, X. Meng, X. Jin, N. Qiu, J. Zhang, Adsorption of antimony onto
in heterogeneous media, Sci. Total Environ. 635 (2018) 1547–1555. iron oxyhydroxides: adsorption behavior and surface structure, J. Hazard. Mater.
[8] Q. Shi, A. Terracciano, Y. Zhao, C. Wei, C. Christodoulatos, X. Meng, Evaluation of 276 (2014) 339–345.
metal oxides and activated carbon for lead removal: Kinetics, isotherms, column [36] M. Vithanage, A.U. Rajapaksha, M. Ahmad, M. Uchimiya, X. Dou, D.S. Alessi, S.O.
tests, and the role of co-existing ions, Sci. Total Environ. 648 (2019) 176–183. Yong, Mechanisms of antimony adsorption onto soybean stover-derived biochar
[9] J. Ifthikar, X. Jiao, A. Ngambia, T. Wang, A. Khan, A. Jawad, Q. Xue, L. Liu, Z.Q. Chen, in aqueous solutions, J. Environ. Manag. 151 (2015) 443–449.
Facile One-pot synthesis of sustainable carboxymethyl chitosan - sewage sludge [37] L. Wang, D. Hu, X.K. Kong, J.G. Liu, X.H. Li, K. Zhou, H.G. Zhao, C.H. Zhou, Anionic
biochar for effective heavy metal chelation and regeneration, Bioresour. Technol. polypeptide poly(gamma-glutamic acid)-functionalized magnetic Fe3O4-GO-(o-
262 (2018) 22–31. MWCNTs) hybrid nanocomposite for high-efficiency removal of Cd (II), Cu(II) and
[10] M.K. Luo, H. Lin, B. Li, Y.B. Dong, Y.H. He, L. Wang, A novel modification of lignin on Ni(II) heavy metal ions, Chem. Eng. J. 346 (2018) 38–49.
corncob-based biochar to enhance removal of cadmium from water, Bioresour. [38] L. Wang, J.-m. Wang, R. Zhang, X.-g. Liu, G.-x. Song, X.-f. Chen, Y. Wang, J.-l. Kong,
Technol. 259 (2018) 312–318. Highly efficient as(V)/Sb(V) removal by magnetic sludge composite: synthesis,
[11] Q. Yin, H. Ren, R. Wang, Z. Zhao, Evaluation of nitrate and phosphate adsorption on characterization, equilibrium, and mechanism studies, RSC Adv. 6 (2016)
Al-modified biochar: Influence of Al content, Sci. Total Environ. 631-632 (2018) 42876–42884.
895–903. [39] J. Liu, J. Cao, Y. Hu, Y. Han, J. Zhou, Adsorption of phosphate ions from aqueous solu-
[12] M.B. Ahmed, J.L. Zhou, H.H. Ngo, W. Guo, M. Chen, Progress in the preparation and tions by a CeO2 functionalized Fe3O4@SiO2 core-shell magnetic nanomaterial,
application of modified biochar for improved contaminant removal from water Water Sci. Technol. 76 (2017) 2867–2875.
and wastewater, Bioresour. Technol. 214 (2016) 836–851. [40] Y. Feng, H. Lu, Y. Liu, L. Xue, D.D. Dionysiou, L. Yang, B. Xing, Nano-cerium oxide
[13] Z. Ren, X. Xu, X. Wang, B. Gao, Q. Yue, W. Song, L. Zhang, H. Wang, FTIR, Raman, and functionalized biochar for phosphate retention: preparation, optimization and rice
XPS analysis during phosphate, nitrate and Cr(VI) removal by amine cross-linking paddy application, Chemosphere 185 (2017) 816–825.
biosorbent, J. Colloid Interface Sci. 468 (2016) 313–323. [41] J. Ifthikar, J. Wang, Q. Wang, T. Wang, H. Wang, A. Khan, A. Jawad, T. Sun, X. Jiao, Z.
[14] M. Vithanage, A.U. Rajapaksha, M. Ahmad, M. Uchimiya, X. Dou, D.S. Alessi, Y.S. Ok, Chen, Highly efficient lead distribution by magnetic sewage sludge biochar:
Mechanisms of antimony adsorption onto soybean stover-derived biochar in aque- sorption mechanisms and bench applications, Bioresour. Technol. 238 (2017)
ous solutions, J. Environ. Manag. 151 (2015) 443–449. 399–406.
[15] X. Hu, Z. Ding, A.R. Zimmerman, S. Wang, B. Gao, Batch and column sorption of ar- [42] D.H.K. Reddy, Y.-S. Yun, Spinel ferrite magnetic adsorbents: alternative future mate-
senic onto iron-impregnated biochar synthesized through hydrolysis, Water Res. rials for water purification? Coord. Chem. Rev. 315 (2016) 90–111.
68 (2015) 206–216. [43] H. Liu, P. Dai, J. Zhang, C.L. Zhang, N. Bao, C. Cheng, L. Ren, Preparation and evalua-
[16] Z. Qi, T.P. Joshi, R. Liu, Y. Li, H. Liu, J. Qu, Adsorption combined with superconducting tion of activated carbons from lotus stalk with trimethyl phosphate and tributyl
high gradient magnetic separation technique used for removal of arsenic and anti- phosphate activation for lead removal, Chem. Eng. J. 228 (2013) 425–434.
mony, J. Hazard. Mater. 343 (2018) 36–48. [44] A.M. Ebrahim, E. Rodriguez-Castellon, J. Maria Montenegro, T.J. Bandosz, Effect of
[17] S.Q. Shi, J.K. Yang, S. Liang, M.Y. Li, Q. Gan, K.K. Xiao, J.P. Hu, Enhanced Cr(VI) re- chemical heterogeneity on photoluminescence of graphite oxide treated with S-/
moval from acidic solutions using biochar modified by Fe3O4@SiO2-NH2 particles, N-containing modifiers, Appl. Surf. Sci. 332 (2015) 272–280.
Sci. Total Environ. 628-629 (2018) 499–508. [45] A.S. Kumar, C. Sharma, Synthesis, characterization and catalytic wet air oxidation
[18] E.B. Son, K.M. Poo, J.S. Chang, K.J. Chae, Heavy metal removal from aqueous solutions property of mesoporous Ce1-xFexO2 mixed oxides, Mater. Chem. Phys. 155
using engineered magnetic biochars derived from waste marine macro-algal bio- (2015) 223–231.
mass, Sci. Total Environ. 615 (2018) 161–168. [46] K.A. McComb, D. Craw, A.J. McQuillan, ATR-IR spectroscopic study of antimonate ad-
[19] Q. Yang, X. Wang, W. Luo, J. Sun, Q. Xu, F. Chen, J. Zhao, S. Wang, F. Yao, D. Wang, X. sorption to iron oxide, Langmuir 23 (2007) 12125–12130.
Li, G. Zeng, Effectiveness and mechanisms of phosphate adsorption on iron- [47] Y. Yu, C. Zhang, L. Yang, J.P. Chen, Cerium oxide modified activated carbon as an ef-
modified biochars derived from waste activated sludge, Bioresour. Technol. 247 ficient and effective adsorbent for rapid uptake of arsenate and arsenite: material
(2018) 537–544. development and study of performance and mechanisms, Chem. Eng. J. 315
[20] L.-K. Wu, Y.-Y. Li, H.-Z. Cao, G.-Q. Zheng, Copper-promoted cementation of antimony (2017) 630–638.
in hydrochloric acid system: a green protocol, J. Hazard. Mater. 299 (2015) 520–528. [48] J. Lu, X. Jiao, D. Chen, W. Li, Solvothermal synthesis and characterization of Fe3O4
[21] R. Liu, W. Xu, Z. He, H. Lan, H. Liu, J. Qu, T. Prasai, Adsorption of antimony(V) onto and γ-Fe2O3 nanoplates, J. Phys. Chem. C 113 (2009) 4012–4017.
Mn(II)-enriched surfaces of manganese-oxide and Fe-Mn binary oxide, [49] X. Zhang, Y. Niu, X. Meng, Y. Li, J. Zhao, Structural evolution and characteristics of the
Chemosphere 138 (2015) 616–624. phase transformations between α-Fe2O3, Fe3O4 and γ-Fe2O3 nanoparticles under
[22] L. Wang, J. Wang, Z. Wang, C. He, W. Lyu, W. Yan, L. Yang, Enhanced antimonate (Sb reducing and oxidizing atmospheres, Cryst. Eng. Comm. 15 (2013) 8166–8172.
(V)) removal from aqueous solution by La-doped magnetic biochars, Chem. Eng. J. [50] L. Fang, R. Liu, J. Li, C. Xu, L.-Z. Huang, D. Wang, Magnetite/Lanthanum hydroxide for
354 (2018) 623–632. phosphate sequestration and recovery from lake and the attenuation effects of sed-
[23] X. Li, X. Dou, J. Li, Antimony(V) removal from water by iron-zirconium bimetal iment particles, Water Res. 130 (2018) 243–254.
oxide: performance and mechanism, J. Environ. Sci. 24 (2012) 1197–1203. [51] S. Zhang, H. Niu, Y. Cai, X. Zhao, Y. Shi, Arsenite and arsenate adsorption on
[24] Z. Qi, H. Lan, T.P. Joshi, R. Liu, H. Liu, J. Qu, Enhanced oxidative and adsorptive capa- coprecipitated bimetal oxide magnetic nanomaterials: MnFe2O4 and CoFe2O4,
bility towards antimony by copper-doping into magnetite magnetic particles, RSC Chem. Eng. J. 158 (2010) 599–607.
Adv. 6 (2016) 66990–67001. [52] F. Wang, H. Sun, X. Ren, K. Zhang, Sorption of naphthalene and its hydroxyl substi-
[25] Z. Qi, T.P. Joshi, R. Liu, H. Liu, J. Qu, Synthesis of Ce(III)-doped Fe3O4 magnetic parti- tutes onto biochars in single-solute and bi-solute systems with propranolol as the
cles for efficient removal of antimony from aqueous solution, J. Hazard. Mater. 329 co-solute, Chem. Eng. J. 326 (2017) 281–291.
(2017) 193–204.