High Catalytic Activity of Fe_3−xCu_xO₄/Graphene Oxide (0 ≤ x ≤ 0.1) Nanocomposites as Heterogeneous Fenton Catalysts for p-Nitrophenol Degradation

Abstract

In order to improve the catalytic properties of Fe₃O₄ nanoparticles in wastewater treatment, the Cu-doped Fe₃O₄/graphene oxide (Fe_3−xCu_xO₄/GO) nanocomposites were prepared by a modified co-precipitation method and used as heterogeneous catalyst for p-Nitrophenol (p-NP) degradation. The effect of the GO and Cu contents in the nanocomposites was investigated. Compared with the unsupported Fe₃O₄ nanoparticles, the Fe₃O₄/GO nanocomposites have obviously improved catalytic performance, especially for the nanocomposite with 6.25 wt.% of the GO content. Furthermore, the catalytic efficiency is greatly improved by doping Cu in the nanocomposite. The Fe_3−xCu_xO₄/GO nanocomposite achieves the best catalytic property in our catalyst system when the x value is about 0.075. Under the optimal reaction condition (0.8 g L⁻¹ of catalyst dosage, 15 mmol L⁻¹ of initial H₂O₂ concentration, 3.0 of pH value, and 30 °C of temperature), the p-NP conversion and chemical oxygen demand removal efficiencies in 120 min for the Fe_2.925Cu_0.075O₄/GO nanocomposite are about 98.4% and 74.7%, respectively. And the p-NP conversion efficiency is still as high as 96.2% after four recycles under the optimum condition. The results clearly show that the Fe_2.925Cu_0.075O₄/GO nanocomposite has outstanding catalytic properties for the p-NP degradation.

1 Introduction

With the rapid development of economy, the environmental pollution has become more and more serious, especially for the organic pollution. Phenolic compounds have been widely used in dyes, explosives, preservatives, pesticides, plasticizers, pharmaceuticals, and leather preservatives as raw materials or intermediates, such as p-Nitrophenol (p-NP) (Wan et al. 2016). But the p-NP is a very high-stable, water-soluble, strongly toxic, and poorly biodegradable molecular (Wang et al. 2017a, b; Zhao et al. 2010). According to the index of the United States Environmental Protection Agency (USEPA), the p-NP concentration in water environment must be lower than 10 ng L⁻¹ in order to ensure our safety (Chen et al. 2017). Therefore, it is of great significance to find some efficient methods to degrade p-NP. So far, there are many technologies to treat the p-NP containing wastewater, such as Fenton degradation (Wan et al. 2016), photocatalytic degradation, adsorption, membrane filtration, and biological treatment (Yu et al. 2016).

In recent years, Fenton process has been widely used to treat the phenolic wastewater because of highly active hydroxyl radical (·OH) (Tian et al. 2017). Due to the strong oxidation ability of hydroxyl radical (E₀ = 2.8 V), most organic moleculars could be mineralized into inorganic substances (CO₂, H₂O, etc.) or transformed into biodegradable organic molecules (Wang et al. 2017a, b; Gonzalez-Olmos et al. 2012). Fenton process can be divided into homogeneous and heterogeneous Fenton according to catalyst state. The heterogeneous process uses the solid catalyst and has many merits, such as relatively smaller amount of iron sludge (Pugazhenthiran et al. 2011) and relatively higher H₂O₂ utilization efficiency. Natural minerals have been widely applied as heterogeneous catalysts due to their large reserves and low price, such as goethite (α-FeOOH) (Kwan and Voelker 2003), hematite (α-Fe₂O₃) (Huang et al. 2001), and magnetite (Fe₃O₄) (Matta et al. 2007) etc. But the natural magnetite generally has low catalytic ability. So, a great deal of interest has been attracted to the artificial Fe₃O₄-based catalysts due to their high catalytic activity, low cost, easily separation, and recovery (Pereira et al. 2012; Zhu et al. 2017; Zhang et al. 2009). However, Fe₃O₄ nanoparticles (NPs) easily aggregate into large particles that will inevitably damage the catalytic activity. So, immobilizing Fe₃O₄ on support is an effective way to solve this problem (Zubir et al. 2014).

Carbon materials (carbon nanotube (Xu et al. 2016), graphene (Zhang et al. 2009; Guo et al. 2013), and activated carbon (Kakavandi et al. 2016)) have been used as Fe₃O₄ supports. Graphene oxide (GO) is particularly considered as one of the most promising materials for nanoparticle support due to its outstanding properties: (1) GO has a special two-dimensional lamellar structure with high surface area and catalytic activity (Yu et al. 2016); (2) GO has many oxygen-containing functional groups (epoxides, ketones, and carboxyl radicals (Wu et al. 2014)) that could provide the nucleation sites for metal ions (Zubir et al. 2014); (3) furthermore, GO has a higher electron transport capacity than the metal oxides and other materials (Ullah et al. 2014). As results, the aggregation phenomenon of the nanoparticles decreases and the catalytic activity of the composites will become high due to the synergistic effect between Fe₃O₄ NPs and GO (Zubir et al. 2014). Up to now, some Fe₃O₄/GO nanocomposites have been gotten as catalyst for phenol oxidation (Yu et al. 2016) and nitroarene degradation (He et al. 2013), which show good catalytic abilities.

Many studies have shown that the co-existence of M^{n +}/M⁽ⁿ⁺¹⁾⁺ redox pairs in the octahedral sites of the spinel-structure metal oxides play a very important role in the formation of OH (Barros et al. 2016). Magnetite has a typical spinel structure (Menini et al. 2008) and could have more actively catalytic sites after isostructural substitution by other transition metals (Moura et al. 2005). As results, many Fe_3−xM_xO₄ materials have been prepared by introducing different metal elements (Co, Cr, Ti, Mn, and so on) and used for the decolorization of methylene blue (Nejad and Jonsson 2004; Magalhães et al. 2007; Yang et al. 2009; Costa et al. 2006). Although having the “couple” (Cu²⁺/Cu⁺), copper used alone has a very low catalytic activity as Fenton catalyst. But the Cu²⁺ ions could obviously enhance the catalytic activity of Fe²⁺ or Fe³⁺ after doping into the catalyst system (Barros et al. 2016). Therefore, it is possible to effectively improve the activity of the Fe₃O₄-based catalyst by doping Cu with an appropriate content. Up to now, there was no literature about the preparation of Fe_3−xCu_xO₄/GO nanocomposite as heterogeneous Fenton catalyst for organic pollutant degradation. In this work, we prepared the Fe_3−xCu_xO₄/GO nanocomposites for p-NP degradation and investigated the effect of the GO and Cu contents in the nanocomposites on the catalytic performances. And the possible catalytic mechanism was also discussed.

2 Experimental Procedure

2.1 Reagents and Materials

Sulfuric acid (H₂SO₄, 98%), potassium permanganate (KMnO₄), ferric chloride, ferrous chloride, copper chloride, p-NP, and hydrogen peroxide (H₂O₂, 30%, w/w) were all purchased from Tianjin Fuchen Chemical Reagents Factory (China). All of the above chemicals and reagents were of analytical grade and used without further purification. Expanded graphite (EG) was bought from Qingdao Yanhai Carbon Materials Co., Ltd.

2.2 Preparation of GO

The GO was prepared by a modified Hummer’s method (Sun and Fugetsu 2013). First, 15 g of KMnO₄ and 5 g of EG were mixed in a 500-mL round-bottom flask and kept in an ice-water bath. Then, 100 mL of concentrated sulfuric acid (98%) was dropwise added into the mixture and stirred for 0.5 h. After that, the mixture was agitated at room temperature for 1 h to form a foam-like intermediate spontaneously. Then, deionized water (400 mL) was slowly added and stirred vigorously for 1 h. Following that, the flask was stirred at 90 °C for 1 h. After reaction, the mixture was cooled to room temperature, and an appropriate amount of H₂O₂ was dropped with agitation. And a dark yellow suspension was formed. The resultant was centrifuged and washed several times with 5% HCl aqueous solution and further with deionized water to remove impurities. Finally, the obtained slurry was kept in low temperature for further use.

2.3 Synthesis of Fe₃O_4, Fe₃O₄/GO, and Fe_3−xCu_xO₄/GO Nanocomposites

The Fe₃O₄ NPs were prepared by a modified co-precipitation method. The detail steps were as follows: Firstly, 350 mL of deionized water was added into a 500-mL round-bottom flask that was kept in a pre-heated water bath (80 °C); secondly, 0.664 g of FeCl₂·4H₂O and 1.751 g of FeCl₃·6H₂O were added into the deionized water with stirring under N₂ atmosphere; thirdly, 3 M NaOH solution was dropwise added into the mixture until the pH value reached to 9~10, and black precipitate was produced instantly; fourthly, the reactant was further stirred at 80 °C for 4 h; fifthly, the Fe₃O₄ sample was collected with a magnet and washed several times with deionized water and anhydrous ethanol; finally, the sample was dried in a vacuum oven at 60 °C for 12 h.

The Fe₃O₄/GO nanocomposites were prepared according to the above processes except that the obtained GO was added into deionized water with agitation in the first step and further sonicated for 30 min. A series of the Fe₃O₄/GO nanocomposites were prepared according to the GO dosages. The preparation of the Fe_3−xCu_xO₄/GO nanocomposites was based on that of the Fe₃O₄/GO. The only difference was that a certain amount of CuCl₂⋅4H₂O was added to the previous system. And a series of the Fe_3−xCu_xO₄/GO (x = 0, 0.025, 0.05, 0.075, and 0.1) nanocomposites were prepared.

2.4 Characterization

The crystal structures of the samples were characterized by X-ray powder diffractometer (XRD, Rigaku D/max-2400, Japan) with a Cu Kα radiation source. The morphologies of the samples were analyzed by high-resolution transmission electron microscope (HRTEM, JEOL JEM2100, Japan). The N₂ adsorption–desorption isotherms of the samples were conducted at − 196 °C (JW-BK100, China). The specific surface area (SSA) and pore size distribution were calculated by Brunauer–Emmett–Teller method and Barreet–Juyner–Halenda (BJH) method, respectively. The surface states of the samples were identified by X-ray photoelectron spectroscopy (XPS, Thermo Scientific ESCALAB 250 XI, America). The surface functional groups of the samples were detected by Fourier transform-infrared spectroscopy (FT-IR, Thermo Scientific NICOLET 6700, America).

The p-NP concentration was measured by UV–Vis spectroscopy (YU-1810, Beijing Purikinje General Co., China) at wavelength of 400 nm. The chemical oxygen demand (COD) was analyzed by dichromate method (Tian et al. 2017). The concentrations of the total dissolved Fe and Fe²⁺ were tested by 1,10-phenatroline spectrophotometric method (Wan et al. 2016). And the H₂O₂ concentration in aqueous solution was quantified by UV–Vis spectroscopy with titanium oxalate at wavelength of 395 nm (Tian et al. 2017).

2.5 Catalytic Activity Measurement

The catalytic performance of these samples was evaluated using p-NP as a model pollutant in aqueous solution. The experiments were conducted in a 150-mL conical flask with agitation. The detail experimental process was as follows: first, 0.1 g of the catalyst sample (Fe₃O₄, Fe₃O₄/GO, or Fe_3−xCu_xO₄/GO catalyst) was added into 100 mL of the simulated p-NP wastewater (100 mg L⁻¹), and the initial pH value was adjusted to 3.0 by H₂SO₄ or NaOH solution (1 M); then, 10 mmol L⁻¹ of H₂O₂ was added with agitation for p-NP degradation. After some time, about 5 mL of the solution was taken out for analysis. Furthermore, the catalyst’s reusability was also evaluated. The used catalyst was collected with an external magnet, and repeatedly washed with deionized water and ethanol for three times. After dried in vacuum oven, the catalyst was used for the next cycle under the same condition. All the experiments were repeated at least two times.

3 Results and Discussions

3.1 Characterization of the Samples

Figure 1 shows the XRD patterns of the samples. The characteristic peak of the GOs appears at about 12.5° of 2θ (Guo et al. 2013). The diffraction peaks of the pure Fe₃O₄ locate at 2θ of about 18.1°, 30.2°, 35.5°, 43.2°, 53.8°, 57.2°, and 62.8° that are indexed to (111), (220), (311), (400), (422), (511), and (440) lattice planes of Fe₃O₄ with a cubic spinel structure according to the standard JCPDS data (Card No. 19-0629). For the as-prepared Fe₃O₄/GO, no obvious diffraction peaks of the GOs were observed. The reasons may be the following: (1) small amounts of GOs in the nanocomposites (3.22 wt.%, 6.25 wt.%, and 9.09 wt.%), (2) much weaker diffraction intensity of GOs than that of Fe₃O₄ NPs, and (3) well dispersion of Fe₃O₄ NPs on GO sheets (Guo et al. 2013). Compared with Fe₃O₄/GO, the NPs in the Fe_3−xCu_xO₄/GO have relatively wider diffraction peaks, which may be originating from their smaller sizes (He et al. 2013; Bahadur et al. 2017).

Fig. 1

XRD patterns of a GO, Fe₃O₄ NPs, and Fe₃O₄/GO nanocomposites and b Fe_3-xCu_xO₄/GO nanocomposites

The TEM images of these nanocomposites and corresponding NP size-distribution histograms are shown in Fig. 2. More than 200 of the nanoparticles for each sample were measured for the statistical analysis. The Fe₃O₄/GO (6.25 wt.%) sample has a large number of the Fe₃O₄ aggregates. For the Cu-doped nanocomposites, the nanoparticle aggregations obviously decrease. The mean nanoparticle sizes of the nanocomposites are about 9.84 nm (Fe₃O₄/GO), 7.39 nm (Fe_2.975Cu_0.025O₄/GO), 4.46 nm (Fe_2.95Cu_0.05O₄/GO), 7.75 nm (Fe_2.925Cu_0.075O₄/GO), and 6.68 nm (Fe_2.9Cu_0.1O₄/GO), respectively. Compared with Fe²⁺ ions, Cu²⁺ ions have better inclination for the octahedral positions and smaller ionic radius (0.73 Å for Cu²⁺ and 0.78 Å for Fe²⁺ in octahedral coordination) (Barros et al. 2016). It may lead to the difference in average particle size.

Fig. 2

TEM images of nanocomposites and their corresponding particle size distribution histograms. a, b Fe₃O₄/GO (6.25 wt.%). c, d Fe_3−xCu_xO₄/GO (x = 0.025). e, f Fe_3−xCu_xO₄/GO (x = 0.05). g, h) Fe_3−xCu_xO₄/GO (x = 0.075). i, j Fe_3−xCu_xO₄/GO (x = 0.1)

The N₂ adsorption–desorption isotherms of the Fe_3−xCu_xO₄/GO nanocomposites are shown in Fig. S1. And the corresponding SSAs and pore structure parameters are listed in Table S1. It can be seen that the SSAs of the composites become higher with increasing of the GO content that may be the large SSA of the GOs. The SSAs of the nanocomposites become high after Cu doping. The corresponding SSAs increase from 137.2 m² g⁻¹ (Fe₃O₄/GO) to 142.0 m² g⁻¹ (Fe_3−xCu_xO₄/GO, x = 0.025), 220.5 m² g⁻¹ (x = 0.05), 223.9 m² g⁻¹ (x = 0.075), and 224.0 m² g⁻¹ (x = 0.1), respectively. Those data may reflect the states of the catalysts to some extent that the Cu-doped nanocomposites have less aggregates comparing with Fe₃O₄/GO.

The XPS analysis was used to study the surface states of the nanocomposites before and after use. As shown in Fig. S2 (a)(d), the XPS full survey spectra reveal that the Fe_2.925Cu_0.075O₄/GO contain C, O, Fe, and Cu elements (binding energy, 283.5 (C1s); 529.1 (O1s); 709.8 (Fe2p); and 932.7 eV (Cu2p), respectively). For the original Fe_2.925Cu_0.075O₄/GO, the binding energy values at 710.1 and 712.0 eV or 717.5 and 724.3 eV indicate the presence of Fe²⁺ and Fe³⁺, respectively. And the binding energy of Cu2p3/2 and Cu2p1/2 are located at 932.3 and 933.6 eV, respectively (Barros et al. 2016). For the used Fe_2.925Cu_0.075O₄/GO, the Fe2p peaks appear at 710.0 and 711.7 eV or 717.4 and 724.2 eV, respectively. However, there is no obvious peak of Cu2p as shown in Fig. S2 (f). This result indicates the significant wastage of the Cu element in Fenton reaction.

The FT-IR spectra of p-NP and Fe_2.925Cu_0.075O₄/GO before and after four reuses are shown in Fig. S3 (a) and (b), respectively. Compared with the raw Fe_2.925Cu_0.075O₄/GO, there are some changes for the used sample. The peaks belonging to O–H, C–H, C=C, C–O, and C=O functional groups have some slight differences. The peaks of these functional groups are red shifted and sharped, especially for of the aromatic C=C bonding. And two new peaks at 1101 and 1404 cm⁻¹ appear in the FT-IR spectrum of the used catalyst, which may be contributed to the intermediate organic molecules originating from the p-NP degradation. Furthermore, the peaks at 584 and 578 cm⁻¹ can be attributed to Fe–O in Fe₃O₄, suggesting the connection of Fe₃O₄ with the –COO– functional group on the edge of the GO sheets (Guo et al. 2013), which is consistent with the results of TEM analysis.

3.2 Effects of Parameters on the Degradation of p-NP

3.2.1 Effect of Catalysts on Degradation of p-NP

The catalytic activity of as-prepared catalysts is evaluated using p-NP as the stimulated pollutant in aqueous solution (30 °C of temperature, 3.0 of pH value, 100 mg L⁻¹ of p-NP concentration, 1.0 g L⁻¹ of catalyst dosage, and 10 mmol L⁻¹ of H₂O₂ concentration). When only H₂O₂ is added without catalyst in the reaction system, the p-NP removal efficiency is just only 1.12% as shown in Fig. 3a. It indicates a weak oxidizing ability of H₂O₂ to p-NP molecular. The p-NP removal efficiency is also as low as 16.3% for Fe₃O₄/GO without H₂O₂, which may be the composite’s absorption. When Fe₃O₄-H₂O₂ or Fe₃O₄/GO-H₂O₂ system is used, the p-NP removal efficiency reaches to 80% and 96.9% in 120 min, respectively. It is the generation of reactive species (⋅OH) (E₀ = 2.8 V). Figure 3b shows the H₂O₂ consumption with the treatment time. H₂O₂ is almost never consumed without catalyst. After adding Fe₃O₄ or Fe₃O₄/GO, the H₂O₂ consumption rates are about 44.3% and 54.1% in 120 min, respectively.

Fig. 3

Effect of catalysts on a p-NP removal efficiency and b H₂O₂ decomposition

3.2.2 Effect of GO Contents on Degradation of p-NP

Figure 4a shows the effect of the GO contents in Fe₃O₄/GO on the p-NP degradation. With increasing of the GO content from 0 to 6.25 wt.%, the p-NP removal rate increases from about 80 to 96.9%. The reasons may be the following: (1) agglomeration reduction of Fe₃O₄ NPs due to GO support and (2) synergistic effect between GO and Fe₃O₄ NPs (Zubir et al. 2014; Yu et al. 2016). But when the GO content increases to about 9.09 wt.%, the p-NP removal efficiency reduces to 88.3%. The following are possible: 1) re-stacking of some GO sheets through the p–p interactions and (2) covering the catalytic sites of Fe₃O₄ NPs by superfluous GOs. Usually, in a certain range, the more Fe₃O₄ exists, the more H₂O₂ will be consumed and the higher the p-NP degradation efficiency will be achieved. However, if the GO content is too low or too high, the synergistic effect between the two components will become weak (He et al. 2013). Figure. 4b indicates the effect of the GO content on the H₂O₂ consumption rate. It is clear that the H₂O₂ consumption rate is relatively higher for 6.25 wt.% of the GO content in our research system. The reaction could be considered as a pseudo-first-order. And the pseudo-first-order rate constant (k) is calculated according to Eq. (1) (Wang 2008; Wang et al. 2017a, b):

$$ \mathrm{l}n\left({C}_t/{C}_0\right)=- kt $$

(1)

where, C_t is the p-NP concentration at reaction time t (mg/L); C₀ is the initial p-NP concentration (mg/L), and C is a constant. The obtained data are listed in Table 1. The rate constants of the catalysts with different GO contents (0, 3.22, 6.25, and 9.09 wt.%) are 0.0186, 0.0250, 0.0427, and 0.0249 min⁻¹, respectively. The data demonstrate that the catalyst with 6.25 wt.% of GO content has highest catalystic ability among the above said samples.

Fig. 4

Effect of GO content on a p-NP removal efficiency and b H₂O₂ decomposition

Table 1 Calculated pseudo-first-order rate constant of samples

3.2.3 Effect of Cu Doping on Degradation of p-NP

Further, the effects of Cu doping in the catalyst on the p-NP degradation were studied. As shown in Fig. 5a, the p-NP degradation rates obviously become fast after the Cu doping. The p-NP removal efficiencies of all the Cu doped catalysts exceed 95.0% in 20 min, while that of the Fe₃O₄/GO catalyst is only 57.9%. With increasing the Cu content, the p-NP removal efficiency in 20 min increases and reaches a maximal value at 0.075 of x value. It may be that more Cu²⁺ ions replace Fe²⁺ ions in the octahedral sites and decrease the Fe-active sites. Compared with the Fe₃O₄/GO catalyst, the H₂O₂ consumptions of the Fe_3−xCu_xO₄/GO catalysts increase significantly as shown in Fig. 5b. But the H₂O₂ utilization efficiency will decrease for the catalyst with excessive Cu doping content. The results show that appropriate Cu doping in the catalyst is conductive to the improvement of the catalytic properties. In the following studies, the Fe_2.925Cu_0.075O₄/GO catalyst was used.

Fig. 5

Effect of Cu doping on a p-NP degradation and b H₂O₂ decomposition

3.2.4 Effect of the Catalyst Dosage on Degradation of p-NP

The effect of catalyst dosage on the p-NP degradation was further studied (100 mg L⁻¹ of initial p-NP concentration, 10 mmol L⁻¹ of H₂O₂ concentration, 3.0 of pH value, and 30 °C of temperature). Figure 6 displays the effect of the catalyst dosage on the p-NP degradation efficiency. As the catalyst dosage increases from 0.1 to 1.5 g L⁻¹, the p-NP removal rate increases and reaches to a maximal value (97.8%) at 0.8 g L⁻¹ of the catalyst dosage, and then slightly decreases. The catalyst agglomeration is possible. Despite this, the p-NP degradation efficiencies still exceed 96.0%. Therefore, the following experiments are carried out at 0.8 g L⁻¹ of the catalyst dose.

Fig. 6

Effect of catalyst dosage on p-NP degradation

3.2.5 Effect of Initial pH on Degradation of p-NP

The pH value is an important parameter in wastewater treatment. Although Fe²⁺ remains in dissolved form even at neutral condition, Fe³⁺ will forms ferric hydroxide sludge if the pH value is larger than 4. Thus, a strict acidic condition (pH < 4) is usually required for practical applications (Bokare and Choi 2014). Fe(OH)⁺ will form at low pH value (2–4) and may have higher active catalytic activity than Fe²⁺ in Fenton reaction (Malik and Saha 2003; Wang 2008). So, the experimental conditions are as follows: 100 mg L⁻¹ of initial p-NP concentration, 0.8 g L⁻¹ of catalyst dosage, 10 mmol L⁻¹ of H₂O₂ concentration, and 30 °C of temperature and 2–3.5 of pH value. It can be seen from Fig. 7 that the p-NP removal efficiencies all exceed 97% in 10 min in 2.0 to 3.0 of pH value. When the pH value is 3.5, the removal efficiency decreases to about 90.0% in 120 min. Therefore, the relatively optimal pH value is about 3.0 in our researched scope.

Fig. 7

Effect of initial pH on p-NP degradation

3.2.6 Effect of H₂O₂ Concentration on Degradation of p-NP

Following, the effect of H₂O₂ concentration on p-NP degradation was also investigated (an initial p-NP concentration of 100 mg L⁻¹, a catalyst dosage of 0.8 g L⁻¹, pH value of 3.0, and temperature of 30 °C). As shown in Fig. 8, the p-NP conversion and COD removal efficiencies increase with increasing of the H₂O₂ concentration and reach to the maximal values (98.4% and 74.7%) at 15 mmol L⁻¹. When the concentration is as high as 20 mmol L⁻¹, there is no obvious enhancement of the p-NP degradation and COD removal efficiency. Excessive H₂O₂ can lead an unprofitable H₂O₂ consumption in the Fenton process. Therefore, subsequent experiments are conducted at 15 mmol L⁻¹ of the H₂O₂ concentration.

Fig. 8

Effect of initial H₂O₂ concentration on p-NP removal efficiency

3.3 Iron Leaching and H₂O₂ Decomposition

The iron leaching and H₂O₂ consumption were also studied under the optimal reaction conditions (100 mg L⁻¹ of initial p-NP concentration, 15 mmol L⁻¹ of H₂O₂ concentration, 3.0 of pH, and 30 °C of temperature). As shown in Fig. 9a, the concentration of total dissolved iron increases with prolonging the reaction time, and about 11.0 mg L⁻¹ of dissolved iron is released into solution within 120 min. In the first 20 min, the concentrations of ferrous ion and total dissolved iron have a similar change trend. And ferrous ions account for the majority of total dissolved iron. It may be the transformation of Fe³⁺ into Fe²⁺ by Cu⁺, which is benefit for reversible conversion between Fe²⁺ and Fe³⁺ in octahedron structure (Barros et al. 2016). The ferrous ion concentration increases rapidly in the first 20 min, which corresponds to the rapid p-NP degradation as shown in Fig. 9b. After 40 min, the H₂O₂ consumption rate obviously decreases with the reduction of Fe²⁺ ion. The H₂O₂ utilization efficiency (E) is a very important index to evaluate the nanocomposite’s catalytic ability (Gao et al. 2015). It is defined as the amount of removed COD corresponding to the consumption of the unit mass H₂O₂ and calculated according to Eq. (2) (Wan et al. 2016):

$$ E\left(\%\right)=\frac{\varDelta \left[\mathrm{COD}\right]}{0.47\times \varDelta \left[\mathrm{H}2\mathrm{O}2\right]}\times 100 $$

(2)

Where, Δ[COD] is the removed COD value (mg L⁻¹); Δ[H₂O₂] is the decomposed H₂O₂ amount in this process (mg L⁻¹), and 0.47 is the conversion factor (Bokare and Choi 2014). Here, Δ[COD] and Δ[H₂O₂] were about 120 mg L⁻¹ and 484.5 mg L⁻¹, respectively. Therefore, the H₂O₂ utilization efficiency is about 52.7%.

Fig. 9

a H₂O₂ decomposition, iron dissolution, and b p-NP conversion and COD removal during p-NP degradation under the optimum reaction conditions

3.4 Stability and Reusability of the Catalyst

The stability and reusability of the Fe_2.925Cu_0.075O₄/GO nanocomposite were also investigated. The experiments were carried out under the optimal condition in our research range (100 mg L⁻¹ of initial p-NP concentration, 0.8 g L⁻¹ of catalyst dosage, 15 mmol L⁻¹ of initial H₂O₂ concentration, 3.0 of pH, and 30 °C of temperature). Figure 10 shows that the p-NP degradation efficiency is still more than 96.0% after four recycles. However, the corresponding COD removal efficiency drops from 74.7 to only 38.7%. It is possible the Cu wastage in nanocomposite. Figure S4 displays that the Fe_2.925Cu_0.075O₄/GO nanocomposite could be easily separated with an external magnet from water. It is very important for practical application.

Fig. 10

Reusability of Fe_2.925Cu_0.075O₄/GO nanocomposite

3.5 Catalytic Mechanism of Fe_2.925Cu_0.075O₄/GO Nanocomposite on p-NP Degradation

The possible mechanism of the p-NP degradation in Fe_2.925Cu_0.075O₄/GO-H₂O₂ system is shown in Fig. 11. Firstly, H₂O₂ adsorbed on the surface of nanocomposite due to synergistic adsorption of Fe_2.925Cu_0.075O₄ NPs and GOs. Then, H₂O₂ was transformed into ⋅OH and OH⁻ on the Fe-active and Cu-active sites by Eqs. (3) and (4). And Fe³⁺ and Cu²⁺ reacted with H₂O₂ to form Fe²⁺, Cu⁺, and ⋅OOH (Eqs. (5) and (6)). And the synergistic effect between Fe and Cu ions in the octahedral sites of the spinel structure promoted the production of ⋅OH. Cu⁺ could react with Fe³⁺ to regenerate Fe²⁺ (Eq. (7)), which could accelerate the conversion of Fe³⁺ into Fe²⁺. The Cu doping plays a positive role in the enhancement of the p-NP degradation efficiency. Furthermore, GO may accelerate electron transmission and H₂O₂ absorption during the reactions. The formed ⋅OH reacted with p-NP molecular and its intermediate products (Eqs. (8) and (9)).

$$ {\mathrm{Fe}}^{2+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Fe}}^{3+}+\cdot \mathrm{OH}+{\mathrm{O}\mathrm{H}}^{-} $$

(3)

$$ {\mathrm{Cu}}^{+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Cu}}^{2+}+\cdot \mathrm{OH}+{\mathrm{O}\mathrm{H}}^{-} $$

(4)

$$ {\mathrm{Fe}}^{3+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Fe}}^{2+}+\cdot \mathrm{OOH}+{\mathrm{H}}^{+} $$

(5)

$$ {\mathrm{Cu}}^{2+}+{\mathrm{H}}_2{\mathrm{O}}_2\to {\mathrm{Cu}}^{+}+\cdot \mathrm{OOH}+{\mathrm{H}}^{+} $$

(6)

$$ {\mathrm{Cu}}^{+}+{\mathrm{Fe}}^{3+}\to {\mathrm{Cu}}^{2+}+{\mathrm{Fe}}^{2+} $$

(7)

$$ \mathrm{p}-\mathrm{NP}+\cdot \mathrm{OH}\to \mathrm{intermediate}\ \mathrm{p}\mathrm{roducts}+{\mathrm{H}}_2\mathrm{O} $$

(8)

$$ \mathrm{intermediate}\ \mathrm{products}+\cdot \mathrm{OH}\to {\mathrm{CO}}_2+{\mathrm{H}}_2\mathrm{O} $$

(9)

Fig. 11

Possible mechanism of p-NP degradation in heterogeneous Fe_2.925Cu_0.075O₄/GO-H₂O₂ system

4 Conclusion

The Fe₃O₄/GO and Fe_3−xCu_xO₄/GO (0 ≤ x ≤ 0.1) nanocomposites were prepared by a co-precipitation method and used as heterogeneous Fenton catalyst for the p-NP degradation. The Fe₃O₄/GO containing 6.25 wt.% of GO has the best catalytic properties among the samples with different GO contents. The catalytic performances of the nanocomposites were remarkably improved after doping Cu, especially for Fe_2.925Cu_0.075O₄/GO. The p-NP degradation rate in 20 min for Fe_2.925Cu_0.075O₄/GO is about 97.8%, while that for Fe₃O₄/GO is only 57.9% at the same reaction condition. Under the optimal reaction condition, the p-NP conversion and COD removal efficiencies for Fe_2.925Cu_0.075O₄/GO are 98.4% and 74.7%, respectively. The corresponding H₂O₂ decomposition reaches to 96.7%. The p-NP removal rate for Fe_2.925Cu_0.075O₄/GO is still as high as 96.0% after four reuse, while the COD removal rate gradually decreases. We believe that the Fe_2.925Cu_0.075O₄/GO catalyst is a promising heterogeneous Fenton catalyst in phenolic wastewater treatment. The synergetic catalytic effect is the key for the high COD removal rate. But how to immobilize the Cu ingredient in the nanocomposite is a very important work in the next work.