Novel and Green Synthesis of Nitrogen-Doped Carbon Cohered Fe₃O₄ Nanoparticles with Rich Oxygen Vacancies and Its Application

Abstract

A one-pot and green synthesis methodology was successfully designed to prepare nitrogen-doped carbon (NC) cohered Fe₃O₄ nanoparticles with rich oxygen vacancies (Fe₃O₄-OVs/NC). The preparation was achieved via cold-atmospheric-pressure air plasma using Fe₂O₃ nanoparticles as the only precursor, and pyridine as the carbon and nitrogen source. Systematic characterization results of the as-prepared Fe₃O₄-OVs/NC confirmed the transition from Fe₂O₃ to Fe₃O₄, along with the generation of oxygen vacancies, while preserving the original needle-like morphology of Fe₂O₃. Moreover, the results indicated the formation of the NC attaching to the surface of the formed Fe₃O₄ nanoparticles with a weight percent of ~13.6%. The synthesized nanocomposite was further employed as a heterogeneous Fenton catalyst to remove phenol from an aqueous solution. The material has shown excellent catalytic activity and stability, demonstrating a promising application for wastewater treatment.

1. Introduction

Ferro ferric oxide nanoparticles (referred to as Fe₃O₄ NPs) have attracted much attention recently due to their low cost, unique magnetism and other excellent physicochemical properties [1], such as biocompatibility [2], environmental friendliness [3], high theoretical discharge specific capacity [4], corrosion resistance [5] and peroxidase-like activity [6]. Concomitantly, it has been explored that Fe₃O₄ NPs had a great potential in heterogeneous catalysis [7], medicines [8], batteries and supercapacitors [1,9,10], microwave absorbers [11], wastewater treatments [12], and so on. With the development of mechanistic research, it has been found that surface oxygen vacancies (OVs) dramatically influenced the physical and chemical properties of metal oxides, and thus played a significant role in the energetics, kinetics and catalytic activity [13,14,15]. Therefore, some investigation has been pursued to increase oxygen vacancies of Fe₃O₄ NPs to improve the photocatalytic and electrocatalytic properties [16,17], Fenton activities [18], microwave absorption [19], etc. However, the oxygen vacancy of Fe₃O₄ NPs is usually generated under conditions at high temperature and high pressure or via complicated procedures [16,17,18,19], which are energy-intensive or have safety concerns and, thus, are not environmentally friendly.

On the other hand, Fe₃O₄ NPs tend to aggregate into large particles due to the nano-size effects. Aggregation can result in the reduction of specific surface areas, and then inevitably cause a decrease in dispersion, therefore, a decrease in catalytic activity and stability of Fe₃O₄ NPs. This is a great challenge for the application of Fe₃O₄ NPs [20,21,22]. Loading Fe₃O₄ NPs onto suitable substrates to form nanocomposites is a viable and effective method for inhibiting aggregation. Therefore, it has attracted considerable research attention. Various carbon materials [4,5,9,17,19,23,24], organic polymers [25,26,27], and other materials, such as ZnO, CeO₂, TiO₂, and metal-organic frameworks [12,18,22,28] have been widely studied as supports for Fe₃O₄ NPs. For the carbon material supports, nitrogen doping has been proven to be an efficient means to greatly enhance their physio-chemical properties. Compared with pure carbon materials, the electron transport characteristics, chemical reactivity and stability of nitrogen-doped carbon (NC) can be significantly improved due to N doping [29,30,31,32,33]. However, as far as we know, previous preparation conditions of the NC materials used are either harsh, complex, or both, such as high temperatures (typically 500–1000 °C), high pressure, inert gas conditions, or using complex repolymerization precursors [30,33].

In the current study, a novel and green methodology for the preparation of nitrogen-doped carbon (NC) cohered Fe₃O₄ NPs with surface oxygen vacancies (OVs), by a one-pot process via the assistance of cold plasma at normal pressure in air atmosphere, was developed. The resulted catalyst was used for the degradation of phenol in a Fenton-like reaction system. The effects of the initial solution pH, initial H₂O₂ concentration and catalyst dosage on phenol degradation were investigated. The relationship between the morphologic structure and the catalytic performance for the degradation of phenol was discussed.

2. Results and Discussion

2.1. Microstructural, Morphological and Physio-Chemical Characterization

The XRD patterns of Fe₃O₄-OVs/NC, Fe₃O₄-TR and the precursor α-Fe₂O₃ NPs are illustrated in Figure 1a. The diffraction peaks of Fe₂O₃ NPs are in line with pure α-hematite Fe₂O₃ (JCPDS card no. 89–0597). The diffraction peaks, located at 2θ = 24.13°, 33.12°, 35.60°, 40.82°, 49.41°, 54.00°, 57.51°, 62.38° and 63.95°, are corresponding to the lattice planes of (012), (104), (110), (113), (024), (116), (018), (214) and (300) of pure α-Fe₂O₃, respectively [34,35]. In more detail, however, for the α-Fe₂O₃ NPs, the intensity ratios of the (104) to (104) and (110) to (104) peaks are both bigger than those in previous reports [34]. This should be attributed to the oriented growth of the needle-shaped α-Fe₂O₃ NPs used in this study. Jian and co-workers [35] have also observed similar phenomena for their α-Fe₂O₃ samples with different shapes. In the cases of Fe₃O₄-OVs/NC and Fe₃O₄-TR, five distinctly characteristic peaks belonging to the face-centered-cubic (fcc) Fe₃O₄ (JCPDS card no. 19-0629) are successfully observed. They are located at 2θ = 30.10°, 35.42°, 43.05°, 56.94°, and 62.52°, and could be indexed to the crystal faces of (220), (311), (400), (511) and (440), respectively [5,26]. Notably, no peaks at 2θ =33.12° and 40.82° are detected for Fe₃O₄-OVs/NC and Fe₃O₄-TR. This means that the two peaks which are attributed to the (104) and (113) crystal faces of α-Fe₂O₃ disappeared completely after the DBD plasma or thermal reduction treatment. That is, the Fe₃O₄ nanoparticles in Fe₃O₄-OVs/NC and Fe₃O₄-TR should be pure Fe₃O₄. Additionally, there is a broad diffraction peak centered at around 2θ = 22° for Fe₃O₄-OVs/NC, indicating the presence of amorphous carbon [28]. Moreover, compared with Fe₃O₄-TR, the intensity of all diffraction peaks assigned to Fe₃O₄ are much weaker for Fe₃O₄-OVs/NC. This could be due to the formation of the carbon cladding on the surface of the formed Fe₃O₄ NPs. In addition, from the full width at half maximum of the (110) diffraction peak of α-Fe₂O₃, and the (311) diffraction peaks of Fe₃O₄-TR and Fe₃O₄-OVs/NC, the average sizes of the crystallites were estimated to be 12 nm (α-Fe₂O₃), 18 nm (Fe₃O₄-TR) and 14 nm (Fe₃O₄-OVs/NC) using Scherrer’s equation, respectively. The crystal size of Fe₃O₄ obtained by conventional thermal reduction became obviously larger than its precursor, α-Fe₂O₃.

To further confirm the formation of carbon by the method that we proposed in this study, the Raman spectra were collected and presented in Figure 1b. Two distinct characteristic D (about 1320 cm⁻¹) and G (about 1570 cm⁻¹) bands of carbon can be observed. The D band is ascribed to the edges, defects and structurally disordered carbon, while the G band originates from in-plane vibrations of sp2-hybridized carbon atoms. It also can be seen that the two bands are both broad, indicating that the formed carbon exists in amorphous forms [5,30]. The TG curves (Figure 1c) display the thermal decomposition of Fe₃O₄-TR and Fe₃O₄-OVs/NC in an air atmosphere. The weight loss before 120 °C is attributed to the release of physically adsorbed water [36]. However, for Fe₃O₄-TR, in the range between 120 °C and 650 °C, with the temperature increase, its weight increases obviously until the temperature reaches 300 °C, and then remains constant. The final weight gain is 3.6%. This is in line well with the theoretical weight gain (3.5%) of Fe₂O₃ during the transformation from Fe₃O₄. On the contrary, the temperature-programmed treatment for Fe₃O₄-OVs/NC, in a range from 120 °C to 650 °C, is a weight loss process, and the final weight loss reached 11.8%. This results from the synergistic effect of the transformation from Fe₃O₄ to Fe₂O₃ and the combustion of carbon. Based on the above data, the formed nitrogen-doped carbon in Fe₃O₄-OVs/NC is calculated to be ~14.8 wt.%. Moreover, the analysis of the C and N elements for Fe₃O₄-OVs/NC was also performed in this study. The results show that the contents of C and N are 12.2 wt.% and 1.4 wt.%, which agree fairly well with that derived from the TG analysis. Note, however, that the carbon combustion mainly occurs between 250 °C and 450 °C in this case, which is well consistent with the previous report by Liu, et al. [37]; however, the temperature range is much lower than that of the combustion of pure carbon [31]. This results from the iron oxides acting as catalysts, playing an important role in promoting carbon combustion. The magnetic properties of Fe₃O₄-TR and Fe₃O₄-OVs/NC at room temperature were characterized by VSM. As presented in Figure 1d, the saturation magnetization (Ms) values of Fe₃O₄-TR and Fe₃O₄-OVs/NC were 52 emu·g⁻¹ and 45 emu·g⁻¹, respectively. The smaller Ms value of Fe₃O₄-OVs/NC compared with Fe₃O₄-TR should be attributed to the existence of nitrogen-doped carbon (NC) reducing the content of Fe₃O₄ in Fe₃O₄-OVs/NC. Nonetheless, the magnetic response is still enough for Fe₃O₄-OVs/NC to meet the need for magnetic separation.

To evaluate the morphology and microstructure of the samples, TEM images were collected and shown in Figure 2. For the precursor α-Fe₂O₃ NPs, the TEM image clearly exhibits that the morphology is a needle-shaped nanostructure, and the average length and diameter are ~117 nm and ~14 nm, respectively (Figure 2a). After thermal reduction under H₂ atmosphere, the formed Fe₃O₄ nanoparticles present an irregularly short worm-like shape with severe aggregation and adhesion (Figure 2b). This indicates that the needle-shaped structure of α-Fe₂O₃ NPs was totally damaged during the transformation process from Fe₂O₃ to Fe₃O₄. However, as shown in Figure 2c, the morphology of Fe₃O₄, which was obtained from the α-Fe₂O₃ NPs via the treatment of DBD plasma with pyridine, is well retained without obvious structural damages. In addition, Figure 2d clearly shows that the Fe₃O₄ NPs are adhered by membranous substances, which are identified to be NC according to Figure 1a,b and Figure 3a below. From both the high-resolution TEM images of Fe₃O₄-TR (Figure 2d) and Fe₃O₄-OVs/NC (Figure 2e), a lattice fringe spacing of 0.243 nm is observed, which is assigned to the (222) plane of fcc Fe₃O₄ [38]. Moreover, lattice fringes with d = 0.193 nm and 0.253 nm are also detected in Figure 2d, which match well with the (331) and (311) lattice planes of cubic Fe₃O₄ [39]. Noteworthily, in the case of Fe₃O₄-OVs/NC, all the lattice spacings are 0.243 nm in Figure 3e, and given this, it is speculated that the main exposed lattice planes in Fe₃O₄-OVs/NC should be (222) faces.

XPS spectra for Fe₃O₄-OVs/NC were collected to investigate the surface chemical composition. Obvious peaks at about 285, 399, 530 and 711 eV can be observed in the survey spectrum of Fe₃O₄-OVs/NC (Figure 3a), which confirms the existence of C, N, O, and Fe. Furthermore, the survey spectrum also demonstrates that the nitrogen content is up to 4.0 wt.%. The high-resolution N1s spectrum of Fe₃O₄-OVs/NC (Figure 3b) contained three deconvolution peaks whose binding energies are 398.5, 400.2, and 402.4 eV, corresponding to pyridinic N, pyrrolic N and N, which coordinates to Fe, respectively [40]. The Fe2p spectrum in Figure 3c shows, as expected, peaks of Fe 2p_1/2 (~724.1 eV), Fe 2p _3/2 (~710.3 eV), and a satellite peak (~718.5 eV). The Fe 2p_1/2 and Fe 2p_3/2 peaks can further be deconvoluted into four contributory peaks at 710.7, 713.6, 722.9 and 725.2 eV. The peaks at 725.2 eV and 722.9 eV correspond to the binding energies of 2p_1/2 of Fe(III) and Fe(II), respectively, meanwhile, the peaks at 713.6 eV and 710.7 eV can be assigned to the 2p _3/2 of Fe(II) ion and Fe(III) ion, respectively. The satellite peak also indicates the coexistence of the Fe(II )and Fe(III) ions [41]. These conclusions are consistent with the previous report and further verify the existence of Fe₃O₄ [40,41], which is agreeable with the XRD results. Furthermore, it is known that the EPR measurement can provide strong evidence to investigate the presence of oxygen vacancies. As shown in Figure 3d, the EPR signals at a g-value of 2.003 are clearly observed for both samples, Fe₃O₄-TR and Fe₃O₄-OVs/NC, which demonstrates the surface electron trapping sites caused by the O vacancy [16,17]. However, compared with Fe₃O₄-TR, Fe₃O₄-OVs/NC shows a much stronger EPR response, indicating the existence of much more O vacancies.

2.2. Catalytic Performance of Fe₃O₄-TR and Fe₃O₄-OVs/NC

In order to evaluate the performance of Fe₃O₄-TR and Fe₃O₄-OVs/NC for catalysis in a Fenton-like process, both catalysts were tested under different conditions. It is shown in Figure 4a,b, that it is obvious the removal of phenol is highly pH-dependent for both samples. With the increase in pH, the efficiency of phenol degradation decreases correspondingly, and at a pH of around 3.0, the two samples show optimal performance for the catalytic phenol degradation. These are in agreement with previous study results for the degradation of organic contaminants with Fe₃O₄-based catalysts [24,27]. This is because, at high pH (>3.0), inert oxide film might form on the surface of the Fe₃O₄ nanoparticles, which would prevent the active Fe sites from catalytically decomposing H₂O₂, generating sufficient reactive radicals [27]. Further, from Figure 4a, it can be seen that 0.2g L⁻¹ phenol could be completely removed only in 60 min under the initial H₂O₂ concentration of 20 mmol L⁻¹ at pH 3.0 using a 0.4 g L⁻¹ dosage of Fe₃O₄-OVs/NC. In contrast, it needs as long as 120 min to reach complete phenol removal under the same reaction condition. Moreover, leaving the other parameters unchanged, while increasing the pH value to 5.0, the maximum removal of phenol is only 47% with Fe₃O₄-TR. On the other hand, however, the removal efficiency is still up to 98% for Fe₃O₄-OVs/NC. This indicates that the catalytic performance of the as-prepared nanocomposite, Fe₃O₄-OVs/NC is remarkable in a wide range of pH values. The great improvement in the catalytic performance of Fe₃O₄-OVs/NC compared with Fe₃O₄-TR should be attributed to its small Fe₃O₄ nanoparticles and abundant oxygen vacancies [42]. Moreover, the NC formed during the cold plasma treatment should also play a significant role in the catalytic performance improvement of Fe₃O₄-OVs/NC [43,44]. In addition, the optimum reaction conditions for the catalytic removal of phenol with Fe₃O₄-OVs/NC have also been discussed, by changing one parameter and leaving the others unchanged, and the results are shown in Figure 4c,d. The H₂O₂ concentration dramatically affects the removal efficiency of phenol (Figure 4c). The phenol degradation efficiency is rapidly increased from 40% to 100% as the concentration of H₂O₂ increases from 5 to 10 mmol L⁻¹, and the time to the equilibrium state is also significantly reduced from about 250 min to 100 min. However, with the further increase in the H₂O₂ concentration, the degradation efficiency no longer changes notably. Therefore, the optimal concentration of H₂O₂ is confirmed to be 20 mmol L⁻¹. The effect of the Fe₃O₄-OVs/NC dosage on the degradation of phenol, with the 20 mmol L⁻¹ initial H₂O₂ at pH 3.0, is shown in Figure 4d. The degradation of phenol was significantly enhanced by the addition of Fe₃O₄-OVs/NC, even when the amount of catalyst is as low as 0.05 g L⁻¹, which demonstrates that Fe₃O₄-OVs/NC has an outstanding catalytic ability to activate the decomposition of H₂O₂. However, when the dosage of Fe₃O₄-OVs/NC is further increased from 0.4 to 0.5 g L⁻¹, the increase in phenol degradation efficiency diminishes. Given the above, the optimal dosage of Fe₃O₄-OVs/NC in the selected reaction condition is 0.4 g L⁻¹.

The reusability of Fe₃O₄-OVs/NC (Figure 4e) was examined under the optimum conditions of a Fe₃O₄-OVs/NC dosage of 0.4 g L⁻¹, initial H₂O₂ dosage of 20 mmol L⁻¹, and pH of 3.0. After four Fenton reaction recycles, the catalytic degradation efficiency of phenol for Fe₃O₄-TR sharply decreases from 100% to 33%. For Fe₃O₄-OVs/NC, however, as the number of cycles increases, phenol degradation efficiency decreases much slower in comparison with Fe₃O₄-TR. After four recycles, phenol degradation efficiency still reaches nearly 87.5%, suggesting that the reused Fe₃O₄-OVs/NC still remains highly active and Fe₃O₄-OVs/NC possesses much better stability than Fe₃O₄-TR.

3. Materials and Methods

3.1. Material Preparation

For the preparation of Fe₃O₄-OVs/NC nanocomposite, the reagents applied, including needle-shaped α-Fe₂O₃ NPs (≥99 wt.%) and pyridine (99.5 mol.%) were both supplied by Tianjin Zhiyuan Chemical Reagent Co. (Tianjin, China), and used without further purification. Distilled water was used in all the experiments. The cold plasma used in this study was generated by a dielectric barrier discharge (DBD) plasma generator (CTP-2000 K), which was purchased from Corona Laboratory, Nanjing, China. The structure of the generator and the operation of the system have been described in detail elsewhere [45]. Briefly, a chamber was constructed by two pieces of quartz plates with a diameter and thickness of 90 mm and 2 mm, respectively, and a quartz ring with a height, internal diameter and external diameter of 8 mm, 50 mm and 54 mm, respectively. The two quartz plates served as the upper and lower surfaces of the chamber, which sandwiched the quartz ring as the chamber’s wall. The chamber was fixed by two columnar steel electrodes with a height and diameter of 50 mm and 54 mm, respectively, by squeezing the two quartz plates during the plasma treatment. The preparation procedures of the Fe₃O₄-VOs/NC nanocomposites are as follows. Firstly, the mixture of α-Fe₂O₃ NPs and pyridine was obtained via the incipient wet impregnation method. Further, 0.3 g of α-Fe₂O₃ NPs were introduced into the quartz chamber of the DBD plasma generator and then mixed with 200 μL of pyridine. Secondly, the chamber was fixed with two steel plate electrodes, and then the mixture in the chamber was treated with DBD plasma for 3 min. The average voltage applied was 100 V and the average power was 250 W. Static air was directly used as the medium gas for the DBD plasma generation. Thirdly, the processed mixture was grounded after the plasma treatment, then mixed with another 200μL of pyridine, followed by another treatment of DBD plasma for 3 min again. This process was repeated 5 times. Finally, the resulted sample was washed with an ultrasonic washer, and further dried in a vacuum oven at 50 °C. The obtained black nanocomposite was designated as Fe₃O₄-OVs/NC. For purposes of comparison, the α-Fe₂O₃ NPs were conventionally thermally reduced at 598 K under a hydrogen atmosphere for 1 h by following the reported methods with slight changes in the holding time [46]. The as-prepared black sample was denoted as Fe₃O₄-TR.

3.2. Material Characterization

The compositions’ phases of the as-prepared materials were identified by a Bruker D8 (Bruker Co., Billerica, MA, USA) advance X-ray diffractometer (XRD) equipped with Cu Kα radiation (λ = 1.5418 Å) and operated at 30 kV and 40 mA with a 4°∙min⁻¹ scan rate. To demonstrate the presence of the carbon in Fe₃O₄-OVs/NC, Raman spectra were collected by a JY-HR800 laser confocal micro-Raman spectrometer (Horiba Jobin Yvon, Paris, France) with a 532 nm wavelength laser. Thermal-gravimetric analyses (TGA) for the two samples were carried out on a Shimadzu TGA-50 (Shimadzu Co., Kyoto, Japan) at a heating rate of 10 °C·min⁻¹ from an ambient temperature to 700 °C in air with a flow rate of 25 cm³∙min⁻¹. Elemental analysis was carried out by using a Vario MACRO cube CHNS analyzer (Elementar Analysensysteme GmbH, Hanau, Germany) to analyze C and N in Fe₃O₄-OVs/NC. The magnetic properties of the two samples were measured by a 7400-S vibrating sample magnetometer (Lakeshore Inc., Columbus, OH, USA) at room temperature. The morphological information of the two samples was surveyed using a JEM- 2100F (JEOL Ltd., Tokyo, Japan)high-resolution transmission electron microscopy (HTEM) working at 200 kV. The surface chemical compositions and chemical states were examined by X-ray photoelectron spectroscopy (XPS), performed on a Thermo ESCALAB250Xi X-ray Photoelectron spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a monochromatic Al Kα (hv = 1486.6 eV) radiation source. The carbonaceous C1s line at 284.6 eV was taken as the reference for the calibration of binding energies (BEs). Electron paramagnetic resonance (EPR) spectra were conducted on a Bruker EMX EPR spectrometer (Bruker Co., Billerica, MA, USA) to monitor the oxygen vacancies of samples.

3.3. Catalytic Degradation Experiments

The potential of Fe₃O₄-OVs/NC used as a catalyst for the removal of phenol in a Fenton-like process was examined. For comparison, the corresponding performance tests of Fe₃O₄-TR, prepared by conventional thermal reduction, were also carried out. The entire experiments in this study were performed in the dark to eliminate the effect of photo-decomposition of H₂O₂ and in a water-bath thermostatic oscillator with continuous shaking at 40 ± 2 °C. The initial concentration of phenol and the volume of the reaction solution were 200 mg L⁻¹ and 100 mL, respectively. The solution pH was initially adjusted by using 0.1 M HCl or 0.1 M NaOH. The reactions were initiated by adding the desired dosage of H₂O₂ to a pH-adjusted reaction solution containing the target substance and a specific amount of catalyst. Samples were withdrawn from the reaction mixture at given time intervals and filtered immediately through a filter membrane (0.22 μm) for subsequent analysis.

The concentration of phenol was measured by a high-performance liquid chromatography (HPLC) method using the Agilent 1200 Series equipment (Agilent Technologies Inc., Santa Clara, CA, USA) coupled with a Zorbax Eclipse XDB-C18 column (3.5 µm, 2.1 mm × 100 mm) and a UV diode array detector. The column temperature was maintained at 30 °C, and the UV detector wavelength was set at 270 nm. A mixture of acetonitrile and water with 70:30 v/v was used as the mobile phase at a flow rate of 1.0 mL∙min⁻¹.

4. Conclusions

Cold-atmospheric-pressure air plasma was successfully introduced in this study for the preparation of nitrogen-doped carbon (NC) cohered Fe₃O₄ nanoparticles with rich oxygen vacancies. The plasma-treatment time was only 15 min, the process temperature was less than 150 °C [34], and the Fe₂O₃ nanoparticles and pyridine were the only two precursors used. Overall, the synthesis methodology has shown its simple, green, fast and ‘one-pot’ properties. XRD, Raman and XPS characterizations confirmed the formation of Fe₃O₄ and NC. The EPR result indicated that the as-prepared sample possessed much more O vacancies than the thermally reduced Fe₃O₄. Moreover, the TEM images showed that the generated Fe₃O₄ nanoparticles were cohered by NC and basically keep the original needle-like morphology of the precursor, Fe₂O₃. This provides the possibility of achieving the flexible shape control of the Fe₃O₄ NPs just by selecting the suitable morphologic structure of the precursor, Fe₂O₃ NPs. The synthesized nanocomposite was further used as a heterogeneous catalyst for the degradation of phenol in a Fenton-like reaction system. The experimental results demonstrated that 0.2 g L⁻¹ of phenol could be completely removed only in 60 min with Fe₃O₄-OVs/NC under optimal conditions, such as a catalyst dosage of 0.4 g L⁻¹, initial H₂O₂ concentration of 20 mmol L⁻¹, and pH of 3.0. Even when pH was up to 5.0, about 98% of the phenol can still be eliminated by 0.4 g L⁻¹ of the as-prepared nanocomposite within 240 min. In addition, under optimal conditions, the nanocomposite showed favorable reusability with 87.5% of phenol removal efficiency after four cycles. Compared with the Fe₃O₄ that was prepared by traditional thermal reduction, the synergistic effect of smaller particle size, more regular structure, much more O vacancies, and NC agglutination should be the reasons for the as-prepared nanocomposite possessing outstanding catalytic performance for phenol removal.