Decolorization of Acid Orange II dye by peroxymonosulfate activated with magnetic Fe₃O₄@C/Co nanocomposites

Abstract

New metallic cobalt loaded magnetic nanocomposites (Fe₃O₄@C/Co) were prepared by calcination and in situ reduction of the Co²⁺-impregnated magnetic carbon nanoparticles of Fe₃O₄@C at 900 °C under a nitrogen atmosphere. Fourier transform infrared spectroscopy (FT-IR), powder X-ray diffraction (XRD), vibrating sample magnetometry, transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy (XPS) were employed to characterize the properties of the composites and the loading of metallic Co⁰ in the material was clarified. The catalytic properties of the nanocomposites Fe₃O₄@C/Co in activating peroxymonosulfate (PMS) for decolorization of Acid Orange II (AO II) dye in aqueous solution were investigated. Fe₃O₄@C/Co nanocomposites exhibited high activity in PMS activation for AO II decolorization. Complete decolorization of AO II solution could be achieved in 20 min within near neutral pH range (pH 6.4–8.5), while a complete decolorization occurred in 40 min for the as-prepared solution without pH adjustment (pH 4.05). The effects of several parameters including pH, catalyst load, PMS concentration and reaction temperature on the catalytic activity were also investigated. Sulfate free radicals activated from PMS were proposed to be the dominant active species in the “Fe₃O₄@C/Co + PMS” system for AO II decolorization. The catalytic and decolorization mechanism was suggested. The catalyst Fe₃O₄@C/Co could be recycled easily by a magnet with good reusability. This study provides a promising method for the activation of “green” oxidant, PMS, by the new magnetic nanocomposites for environmental remediation and oxidation catalysis.

---

1. Introduction

Synthetic dyes are among the main organic pollutants and they are present in wastewater streams from many industrial sectors including textile, leather, cosmetics, paper, electronics, plastic, pharmaceutical, food and other chemical usages.^1,2 Among all the synthetic dyes, azo dyes occupy the largest and most important position of dyes for industrial applications, while an estimated 10–15% of the total world production of dyes is lost during the dyeing process and is released in the textile effluents.^3–5 The release of azo dyes into the environment would be harmful to aquatic life and human health because of their toxic, mutagenic and carcinogenic characteristics.^1,5 Therefore, the degradation and decolorization of wastewater contaminated with azo dyes have aroused worldwide interest.^5,6 Unfortunately, due to their biological resistance and chemical stability, azo dyes pollutants can hardly be treated using classical wastewater treatment technologies.⁷ Therefore, more powerful wastewater treatment methods are highly demanded. Advanced oxidation processes (AOPs) have been used in recent decades as an effective treatment option for the removal of emerging recalcitrant organic contaminants.^8,9 The AOPs are based on highly reactive species such as hydroxyl (HO˙) and/or sulfate (SO₄⁻˙) radicals. These radicals are capable of degrading the organic molecules and yielding intermediates and CO₂. Radical species can be produced mainly from the thermal, UV and transition metal activation of oxidants such as hydrogen peroxide, persulfate (PS) and peroxymonosulfate (PMS).^10–12 As one of the popular AOPs, Fenton reaction using HO˙ radicals activated from H₂O₂ is able to efficiently decompose various organic contaminants, but it suffers from some demerits such as low pH range (2–4) and large amount of sludge produced.¹³ Compared with HO˙ radicals (1.8–2.7 V), SO₄⁻˙ possesses a higher redox potential (2.5–3.1 V) with a longer life time and exhibits a higher degradation efficiency for most organic pollutants over a wide pH range (4 < pH < 9). PMS also has a higher oxidizing potential (1.82 V) than H₂O₂ (1.76 V), and it intervenes in degradation processes in a more efficient way than PS does.⁴ PMS has been widely used in oxidative degradation of organic pollutants due to its higher stability and ease of storage, transport and handling compared to H₂O₂.^14,15 PMS is a triple salt of potassium monopersulfate (2KHSO₅·KHSO₄·K₂SO₄), a product manufactured by DuPont, and commercially available as OXONE®. PMS is considered to be a green oxidant because it is benign, as are most of the byproducts of its reactions. The decomposition rate of organic compounds with PMS at room temperature is low, but it can be enhanced through activation by photolysis or thermolysis (eqn (1)) or by using transition metals such as manganese, iron and cobalt (eqn (2)), thereby producing highly reactive SO₄⁻˙ and HO˙ radicals.^4,10

HSO₅⁻ + metal ion (Mⁿ⁺) → Mⁿ⁺¹ + SO₄⁻˙ + OH⁻ (2)

Homogeneous reactions using cobalt ions to activate PMS were proven of high efficiency for oxidation of organic compounds.¹⁶ However, the loss of cobalt from the homogeneous processes leads to metal pollution, which may cause several health problems.^10,17 Heterogeneous activation of peroxymonosulfate would provide a good solution and many efforts have been carried out for developing heterogeneous oxidation systems with high activity. For example, several investigations have been reported in using cobalt oxides,^18,19 supported cobalt compounds,^8,20 and Co-exchanged zeolites,²¹ etc. as heterogeneous catalysts for activation of PMS. Many materials, such as metal oxide, silica, carbon, zeolite, polymer resins, and nanocomposites, have been developed as catalyst support for preparing heterogeneous catalyst. Compared to polymer and silica shells, carbon shells exhibit much higher stability in rigorously environments including acid or base media, as well as at high temperatures and pressures.²² Moreover, hydrophilic groups can be generated on the surface of carbon shell, making hybrid nanoparticles being negatively charged to further enhance their colloidal stability and biocompatibility and load other metal nanoparticles on their surfaces.²³ Therefore, carbon materials have been regarded as promising candidates of supporter for heterogeneous catalyst.^20,24,25 On the other hand, magnetic Fe₃O₄ loaded materials have numerous advantages such as high separation efficiency, simple manipulation process, and easy operation conditions. Magnetic separation can remove nanoparticles in water simply by using an external magnetic field, indicating great potentials for practical applications in wastewater treatment technology.^13,26 Core/shell magnetic carbon nanospheres have recently attracted intense attention owing to their unique magnetic response, low cytotoxicity and highly functionable surface.^13,22 For example, Wang et al. prepared magnetic Mn catalysts using Fe₃O₄ as the magnetic core, carbon spheres as barrier and manganese oxide nanoparticles as functional component. These magnetically separable catalysts demonstrated to be effective for activation of PMS in producing oxidative radicals for phenol degradation.¹³ Magnetic carbon core/shell nanospheres were also employed by Wang et al. to load Co₃O₄ by an impregnation method, and the prepared magnetic carbon sphere supported Co₃O₄ composites exhibited high activity in PSM activation for phenol degradation with 100% conversion within 30 min.⁸ Metal nanoparticles are highly attractive in catalytic reactions. Li et al. prepared multifunctional Ag-loaded Fe₃O₄@C by loading Ag nanocrystals onto the surface carbon shell of Fe₃O₄@C nanospheres. The synthesized Ag–Fe₃O₄@C nanocomposite exhibited excellent catalytic activity toward organic dye with 98% of RhB conversion within 20 min in the presence of NaBH₄.²² Gong et al. reported the synthesis of magnetic core/shell Co@C spheres via a one-pot approach through catalytic carbonization of mixed PP/PE/PS by Co₃O₄. The core/shell Co@C spheres showed high performance in the photo-degradation of Congo Red (CR). It was demonstrated that the outer carbon shell promoted the degradation of CR and served as a protective layer for cobalt core to improve acid resistance, while the inner cobalt core accelerated the decomposition of H₂O₂ into radicals, which catalyzed the degradation of CR.²⁵ It is of note that the copper nanoparticles loaded Fe₃O₄@C/Cu magnetic core/shell microspheres exhibit much higher photocatalytic activity for degradation of methylene blue (MB) in comparison with Fe₃O₄@CuO under visible light irradiation and in the presence of H₂O₂. It was proposed that the Cu nanocrystals can exhibit the collective oscillation of the conduction electrons upon interaction with electromagnetic radiation, which can effectively promote visible light absorption.²⁷ Yao et al. recently reported the preparation of Co–graphene hybrids with metal cobalt nanoparticles anchored on graphene nanosheets and the hybrids exhibit high catalytic activity in the degradation of Orange II.²⁸ However, the reports of Fe₃O₄@C/metal magnetic nanocomposites for catalytic removal of organic pollutants using PMS are still very limited. To the best of our knowledge, no investigation on metallic cobalt loaded Fe₃O₄@C magnetic nanocomposites catalyst, namely Fe₃O₄@C/Co, for PMS activation has been yet reported for the decolorization of organic dyes solution.

We report herein the preparation of a new nanocomposite of magnetic carbon nanospheres supported cobalt, Fe₃O₄@C/Co, via the approach as described in Scheme 1. The magnetic nanocomposite, Fe₃O₄/C, was first prepared by an ultrasonic irradiation-hydrothermal method. And the cobalt is supported onto Fe₃O₄/C by impregnation method followed by in situ reduction with heat treatment under nitrogen atmosphere. The Fe₃O₄@C/Co catalyst was characterized and its catalytic performances and magnetic separation/reusability were investigated for the decolorization of Acid Orange II solution with peroxymonosulfate.

Scheme 1 Preparation procedure of Fe₃O₄@C/Co composites.

2. Experimental

2.1 Materials

Acid Orange II (AO II) and PMS (Oxone®, DuPont's triple salt: 2KHSO₅·KHSO₄·K₂SO₄) were purchased from Sigma-Aldrich Chemicals. D-Glucose (C₆H₁₂O₆·H₂O), Co(OAc)₂·4H₂O, FeCl₃·6H₂O, FeSO₄·7H₂O, Co₃O₄ (30 nm), ammonium hydroxide (25%), sulfuric acid, sodium hydroxide, methanol, ethanol and tert-butyl alcohol (TBA) were purchased from Sinopharm Chemical Reagents Co. Ltd, China. All chemicals and reagents were of analytical grade and were used as received without further purification. All aqueous samples were prepared by using doubly distilled water (DDW).

2.2 Preparation of magnetic nanoparticles (MNPs) Fe₃O₄@C

Magnetic carbon shell nanoparticles MNPs Fe₃O₄@C were synthesized via modification of literature methods.^8,29 FeCl₃·6H₂O (5.40 g, 0.02 mol), FeSO₄·7H₂O (2.78 g, 0.01 mol) and D-glucose (7.92 g, 0.04 mol) were separately added into 50 mL of an aqueous HCl solution (pH = 1.5). The mixed solution was then added dropwise into 40 mL of an ammonia solution (3.5 mol L⁻¹) at 60 °C under ultrasound irradiation. After reaction for 40 min, the mixture was transferred into a Teflon-lined autoclave (120 mL) and treated in an oven at 180 °C for 18 h. After cooling down to room temperature, the black suspension was collected by magnetic separation, washed with DDW to neutral pH, followed by washing twice with methanol. The obtained black powder was dried at 60 °C, and it was labelled as Fe₃O₄@C.

MNPs Fe₃O₄ was also prepared by ultrasonic irradiation reaction according to above procedure without addition of D-glucose and a hydrothermal treatment. The black precipitate separated after ultrasound reaction was named as MNPs Fe₃O₄, and it was used for comparison in materials characterization and the control experiments.

2.3 Preparation of Fe₃O₄@C/Co

Magnetic carbon nanospheres supported cobalt (Fe₃O₄@C/Co) was synthesized by an impregnation method followed by heat treatment.⁸ A mixture containing 1.0 g of Fe₃O₄@C and 0.34 g of Co(OAc)₂·4H₂O in 50 mL of ethanol was stirred for 30 min at room temperature. The temperature was increased to 50 °C to evaporate ethanol whilst stirring. The dried mixture was then calcinated in a muffle furnace at 900 °C for 1 h in nitrogen atmosphere. The sample was then washed with DDW and ethanol, respectively, followed by drying in an oven at 80 °C. The samples were denoted as Fe₃O₄@C/Co. The Co content in the Fe₃O₄@C/Co sample was determined to be 14%.

When the amount of Co(OAc)₂·4H₂O was changed to 0.17 g and 0.68 g, respectively, while the amount of Fe₃O₄@C and the preparation procedures remained unchanged, Fe₃O₄@C/Co with Co-content of 7.6% and 30%, respectively, were obtained. The samples were designated as Fe₃O₄@C/Co-7.6% and Fe₃O₄@C/Co-30%, respectively.

2.4 Characterization of materials

FT-IR spectra were recorded using an AVATAR 360 spectrometer (Nicolet, USA), and KBr pellets were used for solid samples. X-ray powder diffraction (XRD) patterns were recorded on an Empyrean diffractometer (PANalytical, B.V Company) with a Cu Kα radiation generated at 40 kV and 40 mA. A JEM-2010 transmission electron microscope (TEM) (Japanese Electronic Optical Co. Ltd, Japan) was used to obtain the information about particle size and morphology of the samples. Surface area and porous structure parameters were characterized by Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods through N₂ adsorption and desorption at 77 K using Quantachrome Autosorb-1-C-TCD Automated Gas Sorption System (Quantachrome Instruments, USA). X-ray photoelectron spectroscopy (XPS) was recorded on a VG Multilab 2000 spectrometer (Thermo Electron Corporation) with Al Kα radiation as the exciting source (300 W). Charging effects were corrected by adjusting the binding energy of C 1s to 284.6 eV. The static magnetic properties were examined at room temperature by means of a vibrating sample magnetometer (XPMS XL-7, Quantum Design, USA). Concentration of the dye in aqueous solution was determined using a Varian Cary 50 Scan UV-Vis spectrophotometer (Varian, USA). pH measurements were made by using a Starter-3C digital pH meter (OHAUA Instrument (Shanghai) Co. Ltd, China). Cobalt content in solution was analyzed by atomic absorption spectrometer (TAS-990, Beijing Purkinje General Instrument Co. Ltd, China). Ultrasound irradiation was carried out in an ultrasound clean bath (SB-5200D, Ningbo Scientz Biotechnology Co. Ltd, China) operating at a power of 200 W.

2.5 Catalytic decolorization experiments in “Fe₃O₄@C/Co + PMS” system

Decolorization experiments were performed in a 100 mL flask at ambient temperature and under normal laboratory light. After the desired amount of organic dye in 50 mL of the aqueous solution was added into the flask, the reaction was initiated by adding Fe₃O₄@C/Co and PMS (Fe₃O₄@C/Co samples with Co-content of 14% were used unless otherwise stated). The flask was shaken on a shaker at a speed of 300 rpm. Since PMS is an acidic oxidant, the addition of PMS led to a significant decrease of pH, and the experiment was conducted at acidic medium (pH 4.05, no adjustment).³⁰ For studying the effect of initial solution pH on the decolorization of AO II, H₂SO₄ (20 mM) and NaOH (20 mM) were used to adjust the solution pH after acidic salt PMS was added into the solution. For the catalyst recycling experiment, the catalyst was separated by external magnetic field and washed twice with DDW after each recycle, followed by adding a fresh solution of AO II and PMS to start the next reaction. To monitor the decolorization process of AO II, solution samples were taken out at given time intervals and measured immediately on a Varian Cary 50 Scan UV-Vis spectrophotometer under the maximum absorption wavelength (484 nm). All the measurements were conducted in triplicate with a mean deviation lower than 0.01 to ensure the reproducibility of the experimental results.

3. Results and discussion

3.1 Preparation of the composites

Magnetic nanocomposites Fe₃O₄@C and Fe₃O₄@C/Co were prepared by the approach as illustrated in Scheme 1 via modification of literature methods.^8,29 Fe₃O₄ nanoparticles were prepared by ultrasonic reaction in ammonium solution using Fe²⁺ and Fe³⁺ as iron source. A thin layer of carbon was coated on the surface of the magnetite particles through the feasible carbonization of glucose under hydrothermal condition.^8,13,31 The functional groups on the surface of carbon shell, such as –COOH, –OH and –C=O, were a benefit to immobilize the catalytic active species.^22,31 The metallic ions from Co(OAc)₂ can be adsorbed by hydrophilic groups on the surface to Fe₃O₄@C in ethanol, followed by reduction by carbon or other reducing agents formed in situ during the calcination process at 900 °C under nitrogen, generating Co⁰ nanoparticles loaded Fe₃O₄@C composites, Fe₃O₄@C/Co.

The characterization of XRD and magnetic property in later part confirmed that zero valence cobalt (Co⁰) was the main species deposited on the carbon shell upon the calcination. The proposed formation of mainly Co⁰ rather than cobalt oxides (Co_xO_y) during calcination at 900 °C under nitrogen may be attributed to the fact that the organic ligands in the carbon shell were decomposed at high temperature into carbon and different kinds of gases including hydrogen, carbon monoxide, together with other fragments, which have strong reducibility and thus the cobalt ions or cobalt oxides were reduced to Co⁰ metal particles.^27,31 However, as reported by Wang et al., Co₃O₄ loaded Fe₃O₄@C/Co₃O₄ nanoparticles were obtained by calcining Fe₃O₄@C and Co²⁺ in air atmosphere.⁸ Similarly, Zhang et al. found that core–shell structured magnetic Fe₃O₄@C/Cu or Fe₃O₄@CuO nanocomposites could be synthesized via direct calcination of MOF based materials of magnetic Fe₃O₄@HKUST-1 in N₂ or air atmosphere, respectively.²⁷ Chu et al. entrapped Co²⁺ (Fe³⁺) into polyacrylonitrile (PAN) template microspheres, and pyrolysis of the template microspheres containing these metal complexes at 800 °C for 1 h under a nitrogen atmosphere produced metal nanocrystals (Co⁰ or Fe⁰) embedded in the carbon microsphere.³² Gong et al. also prepared magnetic core/shell Co@C spheres using mixed plastics (consisting of polypropylene, polyethylene and polystyrene) as carbon source and it reduced Co₃O₄ to Co⁰ during the carbonization at 700 °C under nitrogen atmosphere.²⁵ Shi et al. found that the valence state of cobalt oxide and the reduction degree of graphene oxide (GO) of Co₃O₄/GO composites varied according to different calcination conditions. When the temperature was increased to greater than 200 °C, Co₃O₄ was reduced to CoO by the C or CO from the carbon sources of graphene or graphite under N₂ atmosphere. When the temperature rose to about 700 °C, the reduction reaction occurred in-depth, and Co₃O₄ and CoO would be reduced to metallic Co. It is of note that the Co₃O₄/GO composites catalysts calcined at 900 °C exhibited high activity for the decolorization of Orange II with PMS.³³

3.2 Characterization of the composites

The FT-IR spectra for Fe₃O₄@C/Co and its precursors are shown in Fig. 1. The strong band at 580 cm⁻¹ is a characteristic signal of the Fe–O vibrations, the peaks at 1450 cm⁻¹, 1640 and 1745 cm⁻¹ are attributed to C=C, O–H, and C=O vibrations, reflecting the aromatization and carbonization of the glucose during hydrothermal treatment.^22,34 The absorption at 3440 cm⁻¹ implies the existence of residual hydroxyl groups. The peaks at 1000–1400 cm⁻¹ correspond to the C–OH stretching, O–H bending and C–O–C vibrations.^8,22 The absorptions at 2845 cm⁻¹ and 2925 cm⁻¹ are characteristics of aliphatic C–H stretching vibrations.³⁵ Fig. 1 shows that the spectrum curve of Fe₃O₄@C/Co is smoother than that of Fe₃O₄@C. As shown in Fig. 1, the peaks in the regions of 1000–1400 cm⁻¹ and 2845–2950 cm⁻¹ disappeared or weakened after calcination, indicating that the surface feature of Fe₃O₄@C was reconstructed after cobalt loading. Comparison of the FT-IR spectra indicated that some of the C=O and –OH functional groups in Fe₃O₄@C were also removed during the heat treatment.⁸

Fig. 1 FT-IR spectra of Fe₃O₄@C/Co and its precursors.

X-ray powder diffraction patterns for Fe₃O₄, Fe₃O₄@C and Fe₃O₄@C/Co are shown in Fig. 2. Six characteristic peaks of Fe₃O₄ were observed at 2θ = 30.2°, 35.5°, 43.3°, 53.6°, 57.2° and 62.7°.^8,29 The characteristic peaks for Fe₃O₄ were also observed in Fe₃O₄@C and Fe₃O₄@C/Co, suggesting that the magnetic Fe₃O₄ nanospheres were successfully encapsulated and further hydrothermal and calcination treatment did not affect the original crystallinity of Fe₃O₄ structure.^8,13 No diffraction peaks corresponding to the graphite (2θ = 26°) are observed, indicating that most of the carbon prepared with this method was amorphous.^8,13 XRD pattern for Fe₃O₄@C/Co shows a sharp diffraction peak of metallic cobalt (2θ = 44.5°), confirming that crystalline Co⁰ nanoparticles are embedded in the carbon matrix.^35,36 The XRD pattern for Co⁰ in Fe₃O₄@C/Co corresponds to (111) diffraction of the fcc (face-centered cubic) cobalt phase.^37–39 The presence of ferromagnetic Co metal particles is also confirmed by magnetization measurements as described in later part. It should be pointed out that, although XPS measurement found Co_xO_y, no peaks associated with reflections of Co_xO_y (CoO, Co₂O₃ or Co₃O₄) were observed in the XRD diffraction.^8,17,36 This suggests that only a small amount of cobalt oxides are present on the surfaces of the Fe₃O₄@C/Co composite, which are formed by air passivation.³² Nevertheless, XRD analysis confirmed that cobalt in the Fe₃O₄@C/Co composites mainly presented as zero valence Co nanoparticles with a crystalline structure, and this is in accordance with the analysis in Section 3.1.

Fig. 2 XRD patterns of Fe₃O₄@C/Co and its precursors.

Fig. 3 shows the room temperature magnetic hysteresis loops of the samples. All the samples exhibited typical ferromagnetic behavior. The saturation magnetization (M_s) values for the Fe₃O₄, Fe₃O₄@C and Fe₃O₄@C/Co nanocomposites are 62.2, 37.0, and 71.7 emu g⁻¹, respectively. The M_s value of Fe₃O₄@C is much lower than that of Fe₃O₄ nanoparticles, which is ascribed to the existence of nonmagnetic carbon in the total mass.²² The coercivity of Fe₃O₄@C/Co is much higher than that of Fe₃O₄ and Fe₃O₄@C, and this may be caused by the enhanced shape and magnetic anisotropy in the sample of Fe₃O₄@C/Co.^40–42 It is noteworthy that the open hysteresis loops characteristics was also observed for Co⁰/C core/shell sphere prepared from carbonization of mixed plastics and Co₃O₄ at 700 °C in N₂.²⁵ The M_s of sample Fe₃O₄@C/Co is impressively high, it is higher than that of Fe₃O₄ and almost twice that of Fe₃O₄@C. It is known that cobalt itself is ferromagnetic, but its oxides (Co₃O₄ and CoO) are antiferromagnetic at room temperature.^32,35,41 Therefore, the high M_s of Fe₃O₄@C/Co strongly supports that the loaded cobalt was mainly metallic Co⁰ as also proved by the XRD characterization. Only a small amount of antiferromagnetic cobalt oxide may present on the surface of the nanocomposites (from XPS result in later part), and the bulk carbon matrix prevents ferromagnetic cobalt from oxidation, which makes the final products air-stable.³²

Fig. 3 Magnetization curves measured at room temperature for Fe₃O₄@C/Co and its precursors (inset showing magnetic separation of Fe₃O₄@C/Co).

The higher M_s ensured better magnetic response of the magnetic nanocomposites towards an external magnetic field. The magnetic separability of Fe₃O₄@C/Co nanocomposites was tested in water by placing a magnet near the glass bottle. It was found that the samples were completely attracted to a magnet within a very short time and the solution became clear and transparent (inset in Fig. 3), confirming convenient separation of the Fe₃O₄@C/Co nanocomposites from liquids by using an external magnetic field.

TEM was applied to reveal the dimension and surface morphology of the as-prepared samples. Fig. 4(a) shows that the nanoparticles of Fe₃O₄ are generally homogeneous and spherical with the diameters mainly ranging from 10 to 15 nm. After carbonization, the TEM image of the obtained products Fe₃O₄@C (Fig. 4(b)) shows core–shell structures with an amorphous grey carbon coating layer on the surface. TEM of Fe₃O₄@C particles showed high aggregation which could be assigned to the inter-molecular magnetic dipolar interaction⁸ and the self-interaction of the surrounding layer of hydrophilic polar groups.⁴³ Compared with Fe₃O₄, the size of many Fe₃O₄@C particles was vividly increased due to the agglomeration of Fe₃O₄ inside the nanospheres and surface growth of carbon on the shell. Fig. 4(c) shows that Fe₃O₄@C can maintain their roughly spherical shapes after calcination of the Co²⁺-impregnated composites at 900 °C under N₂ forming Fe₃O₄@C/Co, but the Fe₃O₄@C/Co particles were also agglomerated and irregular carbon surfaces were observed. The cobalt nanoparticles cannot be observed directly because of the low contrast between the Co nanoparticles and the carbon matrix.³² The surface of Fe₃O₄@C/Co is not smooth and it may be due to the aggregation of Co crystals.³⁵

Fig. 4 TEM images of nanoparticles (a) Fe₃O₄, (b) Fe₃O₄@C, and (c) Fe₃O₄@C/Co.

The specific surface area and pore volume of the samples were determined using the nitrogen sorption technique, with typical isotherms shown in Fig. 5 and the corresponding parameters of porous structure listed in Table 1. The surface area of Fe₃O₄@C/Co is much higher than that of Fe₃O₄@C while the pore volume and pore size of Fe₃O₄@C/Co are smaller than those of Fe₃O₄@C. The isotherms can be ascribed to type IV with a distinct hysteresis loop, indicating the mesoporous structures of Fe₃O₄@C and Fe₃O₄@C/Co samples.^8,42 Fig. 5 shows that Fe₃O₄@C/Co has a narrower pore size distribution than Fe₃O₄@C and this may be ascribed to that calcination can regulate the pore size by desorption of organic substances from the carbonaceous matter and partial oxidation of carbon.⁸

Fig. 5 Nitrogen adsorption–desorption isotherm and the corresponding pore size distribution (inset) of (a) Fe₃O₄@C, and (b) Fe₃O₄@C/Co.

Table 1 Parameters of porous structure of the samples

Sample	BET surface area (m^2 g^−1)	Pores volume (cm^3 g^−1)	Average pore diameter (nm)
Fe3O4@C	35.6	0.34	38
Fe3O4@C/Co	96.9	0.22	9

---

XPS measurement was applied to investigate the oxidation states and surface chemical compositions of Fe₃O₄@C/Co composites and the results are shown in Fig. 6. The wide survey spectrum shows that the sample consists of carbon, oxygen, iron and cobalt.⁴⁴ The spectrum of Co 2p is shown in Fig. 6(b). Gaussian–Lorentzian curve fitting of Co 2p3/2 and Co 2p1/2 was performed to describe the oxidation states. The typical Co 2p3/2 and Co 2p1/2 at 780.7 eV and 796.4 eV of cobalt oxides (Co_xO_y) are found. The satellite peaks at 786.0 eV and 803.0 eV are characteristic of the Co²⁺ and Co³⁺ phase.^32,33,36 The appearance of cobalt oxides in the surface layer can further be verified by inspecting the O 1s spectrum (Fig. 5(c)). A peak located at 529.8 eV is found, which is related to the binding energy of the cobalt oxides.³² The absence of any peak at 778 eV suggested the absence of metallic Co 2p3/2.⁴⁵ XPS spectra of the metallic cobalt anchored graphene composites, Co–graphene, prepared separately by Yao et al.²⁸ and Chen et al.⁴⁶ via different approaches also indicated that the surface cobalt was present as cobalt oxide since the cobalt on the surface of graphene sheet was easily oxidized due to the small particle size and high surface area.^28,46 It is worthy of noting that due to the limited penetration depth of XPS (<about 10 nm), XPS data reflect only the surface information of the hybrid materials, which inflated the contribution of Co_xO_y on the surface of Fe₃O₄@C/Co products. However, according to XRD measurements, there are no peaks associated with the reflections of Co_xO_y. Based on the results of XRD and magnetic measurement, it is rational to assume that only a very small amount of cobalt oxides is present on the surfaces of the Fe₃O₄@C/Co composite, which are formed by air passivation (XPS results). The carbon matrix prevents metallic cobalt nanoparticles inside the bulk Fe₃O₄@C/Co samples from oxidation, making the final products air-stable.³²

Fig. 6 X-ray photoelectron spectroscopy of Fe₃O₄@C/Co nanocomposites: (a) and (d) survey spectrum; (b) and (e) Co 2p peak; (c) and (f) O 1s peak. (a)–(c) for pristine samples, (d)–(f) for recycled samples.

3.3 Catalytic activity of “Fe₃O₄@C/Co + PMS” system

The catalytic performances of different samples and composites are shown in Fig. 7. It can be seen that Fe₃O₄@C/Co displayed high efficiency for AO II decolorization with a removal of 99% after 40 min in the presence of PMS, while the removal rate was about 2% using only PMS, indicating that PMS could not degrade the azo dye directly and not be activated by daylight irradiation. For the experiment with sole Fe₃O₄@C, only 25% of AO II was removed in 40 min while about 50% removal of AO II was achieved in 2 hours, probably due to the adsorption ability of Fe₃O₄@C.^11,47 For the experiment with only Fe₃O₄@C/Co, 16% of AO II was removed in 2 hours. Table 1 shows that Fe₃O₄@C/Co sample possesses higher surface area and lower pore volume than Fe₃O₄@C, therefore, we may assume that the lower removal ability of Fe₃O₄@C/Co compared with Fe₃O₄@C may due to the supported cobalt which occupies the specific surface and pore volume on the nanocomposites. On the other hand, compared with Fe₃O₄@C/Co, Fe₃O₄@C has more groups such as –OH and C=O on the surface of the carbon shell, and they might also contribute to the increased adsorption via hydrogen bond interaction with AO II. The results in Fig. 7 show that the adsorption of Fe₃O₄ and Co₃O₄ for AO II was very small while both Fe₃O₄ and Co₃O₄ can activate PMS for the decolorization of AO II. About 80% and 30% of AO II was removed in 2 h, respectively, when Co₃O₄ and Fe₃O₄, respectively, were used as catalyst for PMS activation. The catalytic efficiency of Co₃O₄ was much higher than that of Fe₃O₄.⁴⁸ It was also reported that Fe₃O₄ exhibited minor activity for PMS activation in degradation of phenol.¹³ It has been reported that activated carbon could activate PMS to degrade azo dye Acid Orange 7 in aqueous solution.^49,50 In our work, it was found that 55% of AO II was removed in 40 min and about 85% of AO II could be degraded in 2 hour by using Fe₃O₄@C to activate PMS. This suggests that carbon on the Fe₃O₄@C plays more role than Fe₃O₄ in activating PMS, and this is due to that carbon is an electron catalyst and contains functional groups that can activate PMS producing reactive SO₄⁻˙ radicals.^49,51 It was seen that the AO II decolorization rate using different catalysts under the same operation decreased in the order: Fe₃O₄@C/Co > Fe₃O₄@C > Co₃O₄ > Fe₃O₄. Therefore, the results here strongly suggest that the loaded Co species in Fe₃O₄@C/Co has a specific function in making Fe₃O₄@C/Co as a highly efficient catalyst for activating PMS to produce more reactive SO₄⁻˙ and/or HO˙ radicals, resulting in high removal rate of AO II in aqueous solution. These results clearly indicated that Fe₃O₄@C/Co + PSM system is an efficient catalytic system for remediation of dyes wastewater.

Fig. 7 Decolorization of AO II with different systems: (1) PMS; (2) Fe₃O₄; (3) Co₃O₄; (4) Fe₃O₄@C/Co; (5) Fe₃O₄ + PMS; (6) 0.08 mg L⁻¹ Co(II) + PMS; (7) Fe₃O₄@C; (8) Co₃O₄ + PMS; (9) Fe₃O₄@C + PMS; (10) Fe₃O₄@C/Co + PMS. Conditions: Fe₃O₄ or Co₃O₄ or Fe₃O₄@C or Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C.

For the “Fe₃O₄@C/Co + PSM” system, the concentration of Co in the reaction solution was detected to be 0.08 mg L⁻¹, which was contributed to the leached Co from Fe₃O₄@C/Co. As a comparison, homogeneous catalysis by Co(II) ion (with a concentration of 0.08 mg L⁻¹) was further investigated in the same reaction system, as shown in Fig. 7. Compared to the activity of Fe₃O₄@C/Co catalysts, Co leaching has a minor contribution to the activity in PMS activation, indicating that the decolorization process is heterogeneously conducted on Fe₃O₄@C/Co.¹¹

In order to further ascertain the efficiency of Fe₃O₄@C/Co + PMS system for AO II decolorization, UV-Vis spectroscopy was used to identify the products during the decolorization. Representative UV-Vis spectra changes observed during the reaction are depicted in Fig. 8. For the AO II solution before reaction, it shows a main absorption bands at 484 nm, corresponding to the n–π* transition of the azo form, and another two bands at 230 and 310 nm in the ultraviolet region, which is attributed to the π–π* transition of the benzoic and naphthalene ring, respectively.^11,30 Addition of Fe₃O₄@C/Co and PMS into the aqueous solution caused the absorption bands of the dye in the visible region to decrease with time and finally to disappear, indicating the destruction of its chromophoric structure in the vicinity of the azo-linkage. At the same time, the decrease of the two bands at 230 and 310 nm was observed, due to the opening of the benzene and naphthalene ring.^11,30 The absorption band around 250 nm increased with the reaction time, and this may be attributed to the formation of some aromatic intermediates such as naphthalene ring (e.g. 1-amino-2-naphthol) and benzene ring (e.g. sulfanilamide).^5,52

Fig. 8 Representative UV-Vis spectra changes during AO II decolorization by Fe₃O₄@C/Co + PMS system. Conditions: Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C.

3.4 Effect of reaction parameters on AO II decolorization with Fe₃O₄@C/Co + PMS system

To further elucidate the efficiency of Fe₃O₄@C/Co + PSM system for AO II decolorization, the effect of different operating parameters, including the concentrations of Fe₃O₄@C/Co and PMS, the temperature and initial solution pH was investigated.

The influence of initial solution pH on AO II decolorization is shown in Fig. 9(a). The dye solution with addition of PMS had an acidic pH value of 4.05 without adjustment. For the test of pH effect, the pH value of the dye solution was adjusted with diluted H₂SO₄ and NaOH solutions. The results showed that 99% of AO II degraded in 40 min when the solution pH was not adjusted (pH = 4.05), while almost complete decolorization of AO II was achieved in 20 min when the initial solution pH was in the near neutral region (pH 6.4–8.5). When the initial solution pH value was decreased to 2.1 or increased to 10.9, the AO II decolorization rate became much slower. However, a removal of ca. 90% could still be achieved within 2 h, implying that the catalyst still exhibited a good catalytic activity.¹¹ When pH < 3, the metallic cobalt could be dissolved and the activation would behave the same as that of homogenous Co²⁺ ions. Excess Co²⁺ ions will react with SO₄⁻˙ radicals, resulting in poor utilization of PMS. In addition, the activation at acid condition may also produce too much SO₄⁻˙ radicals, which enhance radical–radical scavenger reactions.¹⁵ Under basic conditions, SO₄⁻˙ radicals can convert into HO˙ radicals through eqn (3), and SO₄²⁻ is the major abundant product in the PMS oxidation system, which may play a role in scavenging HO˙ radicals and lead to low AO II removal efficiency.^15,53 The results found here are consistent with the pH range (4 < pH < 9) reported for higher organic degradation efficiency by SO₄⁻˙ free radicals.⁴ These phenomena are also in accordance with the results reported by Liang et al. that a higher activity under neutral conditions was obtained when they investigated the influence of pH on persulfate oxidation of trichloroethylene at ambient temperature.⁵³ Thus, the high catalytic efficiency of Fe₃O₄@C/Co for activation of PMS in neutral pH range can find promising applications in the practical wastewater treatment.

SO₄⁻˙ + OH⁻ → SO₄²⁻ + HO˙ (3)

Fig. 9 Influence of initial solution pH (a), Fe₃O₄@C/Co dosages (b), PMS concentration (c), reaction temperature (d), cobalt content (%) in Fe₃O₄@C/Co (e), and pH variation during reaction process (f) on AO II decolorization with Fe₃O₄@C/Co + PMS system. Conditions: (a) Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C; (b) PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C; (c) Fe₃O₄@C/Co 0.2 g L⁻¹, AO II 20 mg L⁻¹, 25 °C; (d), (e) and (f) Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹.

The influence of Fe₃O₄@C/Co catalyst dosage on the decolorization of AO II by Fe₃O₄@C/Co + PMS system is illustrated in Fig. 9(b). The results indicate that the decolorization of AO II was significantly influenced by the dosage of catalyst. The decolorization efficiency at 30 min was enhanced from 69% to 99% when the catalyst dosage was increased from 0.1 g L⁻¹ to 0.3 g L⁻¹. This fact is evidently attributed to the increased active sites on the surface of Fe₃O₄@C/Co catalyst for reaction with PMS, which will generate more reactive radicals. In addition, the dye adsorbed on the catalyst increased with increase of Fe₃O₄@C/Co catalyst dosage, which could also enhance the decolorization rate.^8,30,54 The decolorization efficiency increased relatively slower when the catalyst dosage increased from 0.2 g L⁻¹ to 0.3 g L⁻¹, which may be mainly attributed to possible scavenging of radicals by the reaction between excess active sites and the sulfate and hydroxyl radicals.^54,55

The decolorization of AO II over Fe₃O₄@C/Co was further studied at varying PMS concentrations. As shown in Fig. 9(c), AO II was slowly degraded in the presence of 0.05 g L⁻¹ of PMS, the slow removal of AO II was probably due to lack of sufficient oxidant amount.^30,54 When PMS concentration was increased to 0.1 g L⁻¹, the decolorization efficiency increased significantly, as more active radicals could be produced under a high concentration of PMS.^30,54 However, further increase in PMS concentration to 0.2 g L⁻¹ did not result in the same increasing trend of the decolorization efficiency. The fact may be due to that the reaction proceeds via radical mechanism, and this was possibly due to the self-quenching of sulfate and hydroxyl radicals by PMS.^8,30

The effect of reaction temperature on AO II decolorization is shown in Fig. 9(d). It is seen that reaction temperature dramatically affected the decolorization efficiency of AO II. A general trend can be observed that higher temperature will result in an increase in the decolorization rate. When reaction took place at 20 °C, almost a complete AO II decolorization was achieved in ca. 80 min while the time would be reduced to ca. 40 min at 25 °C. The duration would be further reduced to 25 min when the reaction temperature was increased to 35 °C. The results revealed that the PMS activation process was an endothermic reaction, and higher temperature would drive the reaction to produce more active sulfate and hydroxyl radicals and thus enhance the AO II decolorization efficiency.^8,54

The effect of cobalt content in the catalyst Fe₃O₄@C/Co on AO II decolorization was also evaluated and the result is shown in Fig. 9(e). Increase of Co-content in the catalyst would enhance the decolorization efficiency of AO II and it can be attributed to the increased active sites on the catalyst.

Variation of pH during the reaction processes was also monitored by adjusting reaction solution pH after addition of PMS, and the results are shown in Fig. 9(f). It was found that the pH remained essentially unchanged with only a slight variation for the reaction with initial pH of 2.10 and 4.05. However, for the reaction with initial pH of 8.50, the pH decreased with the increase of reaction time. The pH decreased markedly to about 4.3 at 30 min (reaction finished according to Fig. 9(a)) and remained unchanged for the followed reaction process. Similar phenomenon was also observed in the degradation of AO II by “Fe_3−xM_xO₄ (Cr, Mn, Co, Ni) + PMS” system⁵⁶ and “Fe–Co/SBA-15 + PMS” system.⁴ When the pH was raised, OH⁻/H₂O would be oxidized by SO₄⁻˙ and converted into HO˙, generating H⁺ and consuming OH⁻ (eqn (3) and (4)). On the other hand, the dissociation of HSO₄⁻ (eqn (5)), which was formed as a result of the reaction between hydroxyl radical and sulfate radical (eqn (6)), would further release H⁺ and decrease the solution pH.^4,56

SO₄⁻˙ + H₂O → SO₄²⁻ + HO˙ + H⁺ (4)

HSO₄⁻ → SO₄²⁻ + H⁺ (5)

SO₄⁻˙ + HO˙ → HSO₄⁻ + 1/2O₂ (6)

3.5 Activation mechanism of PMS on Fe₃O₄@C/Co

It has been proven that several types of reactive radicals, such as sulfate, peroxy-sulfate and hydroxyl radicals, could be generated by PMS under activation of transition metals. SO₄⁻˙ and HO˙ could attack the organic molecules, but SO₅⁻˙ cannot due to its low redox potential (1.1 V). To identify the dominating oxidizing species responsible for AO II decolorization in the “Fe₃O₄@C/Co + PMS” system, ethanol and tert-butyl alcohol (TBA) were used to perform quenching studies. Ethanol is widely used as the scavenger of SO₄⁻˙ and HO˙, while TBA is an effective quenching agent for HO˙.^50,54 It can been seen from Fig. 10 that the addition of 1.0 M ethanol significantly inhibited AO II decolorization, leading to a decrease in the AO II removal efficiency to 14% compared to the 98% AO II removal efficiency that occurred in 40 min without addition of ethanol. When 1.0 M TBA was added to the solution, the decolorization of AO II declined to 75% in 40 min. Thus, the inhibition of AO II decolorization caused by EtOH is much stronger than that caused by TBA, indicating that sulfate radicals, SO₄⁻˙, were the dominant active species controlling the oxidation reaction in the “Fe₃O₄@C/Co + PMS” system for decolorization of AO II.^30,54 Results in Fig. 10 show that HO˙ radicals might also present, but its contribution was minor compared with SO₄⁻˙ radicals.³⁰ Fig. 10 implies that the decolorization efficiency decreased when the AO II decolorization reaction was conducted under N₂ atmosphere, and this result suggests that oxygen (from air) is involved in the decolorization process activated by Fe₃O₄@C/Co. According to the mechanisms proposed by Shi et al.³³ and Yao et al.²⁸ for the PSM activation by cobalt-loaded graphene hybrids, the mechanism of PMS activation with Fe₃O₄@C/Co catalyst is hereby proposed. At first, metallic cobalt (Co⁰) nanoparticles in Fe₃O₄@C/Co composites can release Co²⁺ at certain conditions (eqn (7) and (8)), which was employed to activate PMS to generate SO₄⁻˙ for the decolorization of AO II in water (eqn (9)). Meanwhile, Co⁰ can also activate PMS to generate SO₄⁻˙ (eqn (10)). PMS can reduce Co³⁺ to Co²⁺ and generate SO₅⁻˙ (eqn (11)). Thus, Co²⁺ exhibits redox cycling in the presence of PMS and produces SO₄⁻˙, similar to iron ions in a Fenton-like reaction.²⁸ The highly reactive SO₄⁻˙ radicals consequently react with AO II and its intermediate products to accomplish the full degradation of organic components (eqn (12)). The reactions are listed as follows:^5,8,33

Fe₃O₄@C/Co⁰ → Fe₃O₄@C/Co²⁺ + 2e⁻ (7)

Fe₃O₄@C/Co⁰ + 0.5O₂ + H₂O → Fe₃O₄@C/Co²⁺ + 2OH⁻ (8)

Fe₃O₄@C/Co²⁺ + HSO₅⁻ → Fe₃O₄@C/Co³⁺ + SO₄⁻˙ + OH⁻ (9)

Fe₃O₄@C/Co⁰ + 2HSO₅⁻ → Fe₃O₄@C/Co²⁺ + 2SO₄⁻˙ + 2OH⁻ (10)

Fe₃O₄@C/Co³⁺ + HSO₅⁻ → Fe₃O₄@C/Co²⁺ + SO₅⁻˙ + H⁺ (11)

SO₄⁻˙ + AO II → intermediates → CO₂ + H₂O (12)

Fig. 10 Effect of EtOH, TBA and N₂ atmosphere on AO II decolorization with the Fe₃O₄@C/Co + PMS system. Conditions: Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C.

3.6 Recycling and reusability of Fe₃O₄@C/Co catalyst

The recyclability of catalysts is an important factor for more economical processes.^22,27 Therefore, the recyclability of catalyst Fe₃O₄@C/Co was investigated in repeated cycles of reaction. At the end of the reaction, the catalyst was separated from the reaction medium using a magnet. The catalyst was washed several times with DDW and reused for next cycle of AO II decolorization. After repeating for six times, the Fe₃O₄@C/Co composite was still stable and exhibited a high catalytic activity with the AO II decolorization efficiency of 93% in the sixth run (Fig. 11).

Fig. 11 The decolorization efficiency of AO II during different cycle use of Fe₃O₄@C/Co. Conditions: Fe₃O₄@C/Co 0.2 g L⁻¹, PMS 0.1 g L⁻¹, AO II 20 mg L⁻¹, 25 °C, reaction time 40 min.

The separated catalyst Fe₃O₄@C/Co after the sixth reaction run was further examined by using FT-IR, XRD, vibrating sample magnetometer and XPS. Compared with the images of the pristine catalyst, FT-IR (Fig. 1), XRD (Fig. 2), Magnetization curves (Fig. 3) and XPS (Fig. 6) of the recycled catalyst did not show any obvious changes. All these analyses suggest the excellent stability and reusability of Fe₃O₄@C/Co composite. Comparison of XPS, XRD and magnetization between the pristine and recycled catalyst indicate that metallic cobalt presents in the bulk Fe₃O₄@C/Co composite and the relative content of Co⁰ remains unchanged, while the surface composition of Fe₃O₄@C/Co composite was mainly cobalt oxides as observed from the XPS for the pristine Fe₃O₄@C/Co composite. Therefore, the oxidation states of the cobalt post catalysis were essentially the same as in the pristine catalyst. We may assume that the catalytic activation of PMS was performed mainly by the metallic cobalt nanoparticles inside the porous Fe₃O₄@C/Co composite.

In recent years, dye degradation has been studied using metal or metal oxide nanoparticles supported on carbon and/or graphene. A preliminary and brief comparison of oxidation removal of AO II by PMS activated with different heterogeneous cobalt catalysts is given in Table 2. A few examples of degradation of methylene blue (MB) dye and H₂O₂ or peroxydisulfate (PS) as oxidant are also included in Table 2 for comparison. It can be seen from Table 2 that the performance of Fe₃O₄@C/Co is comparable to the other materials and it also exhibits some other advantages. Therefore, the nanocomposite catalyst Fe₃O₄@C/Co might be one of the promising and feasible materials for oxidation removal of organic pollutants.

Table 2 Comparison of some heterogeneous cobalt catalysts for oxidation removal of dyes with PMS

Catalyst	Dye	Reaction conditions	Time used for >99% removal	Advantages (A)/disadvantages (D)	Ref.
Fe3O4@C/Co^0	AO II	AO II 20 mg L^−1, catalyst loading 0.2 g L^−1, PMS/AO II = 5	40 min (pH0 = 4), 20 min (pH0 6.4)	(A): feasible preparation, high efficiency, magnetic separation, slight Co leaching, (D): relatively higher catalyst loading	This work
Co^0–graphene	AO II	AO II 60 mg L^−1, catalyst loading 0.01 g L^−1, PMS/AO II = 3.3, pH0 = 7	30 min	(A): small catalyst loading, high efficiency, magnetic separation, slight Co leaching, (D): relatively expensive	28
Co3O4/graphene oxide	AO II	AO II 70 mg L^−1, catalyst loading 0.1 g L^−1, PMS/AO II = 17.4, pH0 = 7	6 min	(A): high efficiency, slight Co leaching, (D): relatively expensive, separation relatively difficult, higher PMS amount	33 and 57
Mn3O4–rGO	AO II	AO II 30 mg L^−1, catalyst loading 0.05 g L^−1, PMS/AO II = 50, pH0 = 7	120 min (64% at pH 4)	(A): small catalyst loading, (D): higher PMS amount, longer reaction time, relatively expensive	58
ZnFe2O4–rGO	AO II	AO II 20 mg L^−1, catalyst loading 0.1 g L^−1, PMS/AO II = 25, pH0 = 7	150 min	(A): magnetic separation, effective under various salt, (D): lower efficiency, higher PMS amount, light irradiation required	59
MnFe2O4–rGO hybrid	AO II	AO II 20 mg L^−1, catalyst loading 0.05 g L^−1, PMS/AO II = 25	120 min	(A): magnetic separation, small catalyst loading, small leaching of metal, (D): longer reaction time, high PMS amount	60
ZnFe2O4–C3N4 hybrid	AO II	AO II 10 mg L^−1, catalyst loading 0.5 g L^−1, H2O2 0.1 M, pH0 = 7	240 min	(A): magnetic separation, feasible preparation, (D): lower efficiency, high catalyst loading, light irradiation required	61
Fe–Co/GAC (granular activated carbon)	AO II	AO II 50 mg L^−1, catalyst loading 0.8 g L^−1, PS/AO II = 10, pH0 = 5.8	60 min	(A): feasible preparation, (D): high catalyst loading, leaching for Co 6.28% (while Fe 0.76%), ultrasound assistance	62
CuFe2O4/activated carbon	MB	MB 20 mg L^−1, catalyst loading 0.2 g L^−1, PMS/MB = 100, pH0 = 5	60 min	(A): magnetic separation, feasible preparation, (D): higher PMS amount, relatively high catalyst loading	51
Fe3O4/Mn3O4/rGO	MB	MB 50 mg L^−1, catalyst loading 0.1 g L^−1, PMS/MB = 6, pH0 = 7	30 min	(A): magnetic separation, high efficiency, (D): complicated preparation, relatively expensive	54

---

4. Conclusions

Metallic Co loaded magnetic nanocomposite Fe₃O₄@C/Co were effectively synthesized from the precursor of Fe₃O₄@C impregnated with Co²⁺ through in situ reduction of Co²⁺ to Co⁰ during the high temperature calcination process under nitrogen atmosphere. The nanocomposites were characterized and Fe₃O₄@C/Co exhibited high performance in activating PMS for decolorization of AO II dye in aqueous solution. The dye solution could be completely decolorized in 40 min within a wide pH range (pH 4.1–8.5). Increase of catalyst load, PMS concentration and reaction temperature promoted the dye decolorization efficiency. The quenching studies confirmed that sulfate radicals were the primary species produced during the catalytic decomposition of PMS. Fe₃O₄@C/Co nanocomposites also exhibited good stability and reusability for 6 cycles use without deterioration in its performance. This study provides a feasible approach for removal of organic pollutants by magnetically separable catalyst. Current work is continuing in our laboratory to investigate the applications of the catalyst for the treatment of complex wastewater containing organic contaminants.

Acknowledgements

We are very grateful to the reviewers for their comments and suggestions. This work was financially supported by the National Natural Science Foundation of China (Grant No. 21277106 and 21101122); Collaborative Innovation Plan of Hubei Province for Key Technology of Eco-Ramie Industry (2014-8); Scientific and Technological Innovation Plan of Wuhan Textile University (Grant No. 153031).

References

1. M. T. Yagub, T. K. Sen, S. Afroze and H. M. Ang, Adv. Colloid Interface Sci., 2014, 209, 172.
2. H. Lin, H. Zhang, X. Wang, L. Wang and J. Wu, Sep. Purif. Technol., 2014, 122, 533.
3. X. Chen, W. Wang, H. Xiao, C. Hong, F. Zhu, Y. Yao and Z. Xue, Chem. Eng. J., 2012, 193–194, 290.
4. C. Cai, H. Zhang, X. Zhong and L. Hou, J. Hazard. Mater., 2015, 283, 70.
5. X. Wang, L. Wang, J. Li, J. Qiu, C. Cai and H. Zhang, Sep. Purif. Technol., 2014, 122, 41.
6. K. Singh and S. Arora, Crit. Rev. Environ. Sci. Technol., 2011, 41, 807.
7. R. Ganesh, G. D. Boardman and D. Michelsen, Water Res., 1994, 28, 1367.
8. Y. Wang, H. Sun, H. M. Ang, M. O. Tadé and S. Wang, Chem. Eng. J., 2014, 245, 1.
9. A. Rastogi, S. R. Al-Abed and D. D. Dionysiou, Appl. Catal., B, 2009, 85, 171.
10. J. Sharma, I. M. Mishra, D. D. Dionysiou and V. Kumar, Chem. Eng. J., 2015, 276, 193.
11. L. Duan, B. Sun, M. Wei, S. Luo, F. Pan, A. Xu and X. Li, J. Hazard. Mater., 2015, 285, 356.
12. Y. Tao, Q. Ni, M. Wei, D. Xia, X. Li and A. Xu, RSC Adv., 2015, 5, 44128.
13. Y. Wang, H. Sun, H. M. Ang, M. O. Tadé and S. Wang, J. Colloid Interface Sci., 2014, 433, 68.
14. G. Chen, X. Si, J. Yu, H. Bai and X. Zhang, Appl. Surf. Sci., 2015, 330, 191.
15. J. Yan, L. Zhu, Z. Luo, Y. Huang, H. Tang and M. Chen, Sep. Purif. Technol., 2013, 106, 8.
16. Q. Yang, H. Choi, Y. Chen and D. D. Dionysiou, Appl. Catal., B, 2008, 77, 300.
17. G. Zhou, H. Sun, S. Wang, H. M. Ang and M. O. Tadé, Sep. Purif. Technol., 2011, 80, 626.
18. Q. Yang, H. Choi, S. R. Al-Abed and D. D. Dionysiou, Appl. Catal., B, 2009, 88, 462.
19. M. Stoyanova, I. Slavova, S. Christoskova and V. Ivanova, Appl. Catal., A, 2014, 476, 121.
20. P. R. Shukla, S. Wang, H. Sun, H. M. Ang and M. Tadé, Appl. Catal., B, 2010, 100, 529.
21. P. Shukla, S. Wang, K. Singh, H. M. Ang and M. O. Tadé, Appl. Catal., B, 2010, 99, 163.
22. W. H. Li, X. P. Yue, C. S. Guo, J. P. Lv, S. S. Liu, Y. Zhang and J. Xu, Appl. Surf. Sci., 2015, 335, 23.
23. S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki and R. Ryoo, Nature, 2001, 412, 169.
24. G. Liu, F. He, J. Zhang, L. Li, F. Li, L. Chen and Y. Huang, Appl. Catal., B, 2014, 150–151, 515.
25. J. Gong, J. Liu, X. C. Chen, Z. Jiang, X. Wen, E. Mijowska and T. Tang, J. Mater. Chem. A, 2014, 2, 7461.
26. S. Linley, T. Leshuk and F. X. Gu, Clean: Soil, Air, Water, 2013, 41, 1152.
27. Y. F. Zhang, L. G. Qiu, Y. P. Yuan, Y. J. Zhu, X. Jiang and J. D. Xiao, Appl. Catal., B, 2014, 144, 863.
28. Y. Yao, C. Xu, J. Qin, F. Wei, M. Rao and S. Wang, Ind. Eng. Chem. Res., 2013, 52, 17341.
29. N. Wang, L. Zhu, D. Wang, M. Wang, Z. Lin and H. Tang, Ultrason. Sonochem., 2010, 17, 526.
30. S. Luo, L. Duan, B. Sun, M. Wei, X. Li and A. Xu, Appl. Catal., B, 2015, 164, 92.
31. G. Yu, B. Sun, Y. Pei, S. Xie, S. Yan, M. Qiao, K. Fan, X. Zhang and B. Zong, J. Am. Chem. Soc., 2010, 132, 935.
32. Y. Chu, P. Zhang, J. Hu, W. Yang and C. Wang, J. Phys. Chem. C, 2009, 113, 4047.
33. P. Shi, X. Wang, X. Zhou, Y. Min, J. Fan and W. Yao, RSC Adv., 2015, 5, 34125.
34. X. Sun and Y. Li, Angew. Chem., Int. Ed., 2004, 43, 597.
35. R. Qiao, X. L. Zhang, R. Qiu, Y. Li and Y. S. Kang, J. Phys. Chem. C, 2007, 111, 2426.
36. D. Banerjee, R. V. Jagadeesh, K. Junge, M.-M. Pohl, J. Radnik, A. Brückner and M. Beller, Angew. Chem., Int. Ed., 2014, 53, 4359.
37. Z. J. Liu, Z. Y. Yuan, W. Zhou, L. M. Peng and Z. Xu, Phys. Chem. Chem. Phys., 2001, 3, 2518.
38. J. A. Díaz, M. Martínez-Fernández, A. Romero and J. L. Valverde, Fuel, 2013, 111, 422.
39. K. Maaz, S. Karim, M. Usman, A. Mumtaz, J. Liu, J. L. Duan and M. Maqbool, Nanoscale Res. Lett., 2010, 5, 1111.
40. T. N. Narayanan, M. M. Shaijumon, P. M. Ajayan and M. R. Anantharaman, Nanoscale Res. Lett., 2010, 5, 164.
41. L. Duan, S. Jia, R. Cheng and L. Zhao, Chem. Eng. J., 2011, 173, 233.
42. C. He, S. Qiu, X. Wang, J. Liu, L. Luan, W. Liu, M. Itoh and K. I. Machida, J. Mater. Chem., 2012, 22, 22160.
43. T. Poursaberi and M. Hassanisadi, Clean: Soil, Air, Water, 2013, 41, 1208.
44. L. Fu, Z. Liu, Y. Liu, B. Han, P. Hu, L. Cao and D. Zhu, Adv. Mater., 2005, 17, 217.
45. T. Preethi, B. Abarna, K. N. Vidhya and G. R. Rajarajeswari, Ceram. Int., 2014, 40, 13159.
46. Y. Chen, Q. Wang, C. Zhu, P. Gao, Q. Ouyang, T. Wang, Y. Ma and C. Sun, J. Mater. Chem., 2012, 22, 5924.
47. Z. Zhang and J. Kong, J. Hazard. Mater., 2011, 193, 325.
48. H. Sun, G. Zhou, S. Liu, H. M. Ang, M. O. Tadé and S. Wang, ACS Appl. Mater. Interfaces, 2012, 4, 6235.
49. J. Zhang, X. Shao, C. Shi and S. Yang, Chem. Eng. J., 2013, 232, 259.
50. S. Yang, T. Xiao, J. Zhang, Y. Chen and l. Li, Sep. Purif. Technol., 2015, 143, 19.
51. W. D. Oh, S. K. Lua, Z. Dong and T. T. Lim, J. Hazard. Mater., 2015, 284, 1.
52. G. Quan, W. Sun, J. Yan and Y. Lan, Water, Air, Soil Pollut., 2014, 225, 2195.
53. C. J. Liang, Z. S. Wang and C. J. Bruell, Chemosphere, 2007, 66, 106.
54. B. Yang, Z. Tian, B. Wang, Z. Sun, L. Zhang, Y. Guo, H. Li and S. Yan, RSC Adv., 2015, 5, 20674.
55. T. D. Nguyen, N. H. Phan, M. H. Do and K. T. Ngo, J. Hazard. Mater., 2011, 185, 653.
56. G. Wei, X. Liang, Z. He, Y. Liao, Z. Xie, P. Liu, S. Ji, H. He, D. Li and J. Zhang, J. Mol. Catal. A: Chem., 2015, 398, 86.
57. P. Shi, X. Dai, H. Zheng, D. Li, W. Yao and C. Hu, Chem. Eng. J., 2014, 240, 264.
58. Y. Yao, C. Xu, S. Yu, D. Zhang and S. Wang, Ind. Eng. Chem. Res., 2013, 52, 3637.
59. Y. Yao, J. Qin, Y. Cai, F. Wei, F. Lu and S. Wang, Environ. Sci. Pollut. Res., 2014, 21, 7296.
60. Y. Yao, Y. Cai, F. Lu, F. Wei, X. Wang and S. Wang, J. Hazard. Mater., 2014, 270, 61.
61. Y. Yao, Y. Cai, F. Lu, J. Qin, F. Wei, C. Xu and S. Wang, Ind. Eng. Chem. Res., 2014, 53, 17294.
62. C. Cai, L. Wang, H. Gao, L. Hou and H. Zhang, J. Environ. Sci., 2014, 26, 1267.

---