Construction of ZnFe₂O₄/rGO composites as selective magnetically recyclable photocatalysts under visible light irradiation

Graphene oxide (GO) was prepared from flake graphite by a modified Hummers method [39]. Figure 1 illustrates the scheme of the preparation of ZnFe₂O₄/rGO composites. Briefly, 60 mg of GO was dispersed into 40 ml of H₂O with sonication for 60 min. Meanwhile, 0.297 g Zn(NO₃)₂ · 6H₂O and 0.808 g Fe(NO₃)₃ · 9H₂O (molar ratio of Fe³⁺ versus Zn²⁺ was 2:1) were dissolved in 20 ml of H₂O at room temperature. The above two solutions were mixed together and stirred for 1 h. After that, 10 ml aqueous solution containing 120 mg of ascorbic acid was added dropwise to the above mixture for the reduction of GO. After raising the water bath temperature to 80 °C the mixture solution was stirred for another 4 h, resulting in a black rGO solution. Then, 0.5 g urea was slowly added to the obtained solution followed by stirring at 90 °C until a black gel was obtained. The gel was dried in a vacuum at 60 °C overnight and then calcined at 300 °C, 400 °C, 500 °C and 600 °C for 3 h in N₂ to obtain the ZnFe₂O₄/rGO (ZFG) composites. The samples calcined at 300 °C, 400 °C, 500 °C and 600 °C are denoted as ZFG300, ZFG400, ZFG500 and ZFG600, respectively. For comparison, pure ZnFe₂O₄ was also prepared using the same procedures without GO.

Zoom In Zoom Out Reset image size

The phase structure was analysed by an x-ray diffractometer (XRD; Rigaku, Ultima IV). The morphology of products was determined by transmission electron microscopy (TEM, JEM-2010F). The optical properties of the samples were investigated by UV–vis diffuse reflectance spectra (DRS) using a spectrophotometer (PE lammda750S). The elemental composition was investigated by x-ray photoelectron spectroscopy (XPS; Thermo Escalab 250XI). The chemical components of the samples were determined by Fourier transform infrared (FT-IR, Nicolet-6700) spectra and Laser Raman spectroscopy (Horiba Evolution). The magnetic properties was measured in a vibrating sample magnetometer (VSM; Quantum design, PPMS-9). BET surface area and pore size distribution were investigated by nitrogen adsorption-desorption studies at 77 K using a surface area and porosimetry system (Micromeritics, ASAP 2460).

The photocatalytic performance of the as-prepared photocatalysts was evaluated by degrading organic dyes (MB, MO and RhB) at a neutral pH. A 300 W halogen tungsten lamp with cut off filter (λ > 420 nm) was employed as the visible light source and a 300 W xenon lamp as the simulated solar light source. For each photocatalytic test, 25 mg of catalyst was added into 50 ml of aqueous dye solution with an initial concentration of 20 mg  L⁻¹. Before starting the illumination, the mixture was stirred for 60 min in the dark to reach the adsorption-desorption equilibrium between the dye and catalyst. The lamp was turned on after adding 0.1 ml, 0.3 ml, 0.5 ml, or 0.7 ml of H₂O₂ to the above reaction mixture, and no method was further employed to remove H₂O₂. At given intervals, 3 ml of liquid was withdrawn and magnetically separated to remove all catalysts. The residual concentration was monitored using the spectrophotometer (PE lammda750S). The photocatalytic activity was calculated by applying the following equation:

$$\begin{eqnarray}&&{\rm{Degradation}}\,{\rm{efficiency}}=\displaystyle \frac{{C}_{0}-C}{{C}_{0}}\times 100 \% \end{eqnarray} \tag{ 1 }$$

where C₀ is the concentration at the adsorption equilibrium and C is the residual concentration at different illumination intervals.

To investigate the stability of catalyst, the catalyst was collected, washed with water and absolute ethanol, and dried at 70 °C. The reuse study of the catalyst was carried out under identical conditions to the above.

The phase structure was analyzed by XRD as shown in figure 2(a). The strong peak at 2θ = 10.91° (GO) is attributed to the (001) plane of GO. The rGO shows a broad peak at around 25°, which suggests that rGO was prepared successfully [40]. All the peaks for ZnFe₂O₄ index well to crystalline ZnFe₂O₄ with a cubic spinel structure. As shown in ZFG500, the peaks at 2θ = 18.03°, 29.76°, 35.08°, 36.80°, 42.78°, 53.00°, 56.52°, 62.10°, 70.48° and 73.51° correspond to the (111), (220), (311), (222), (400), (422), (511), (440), (620) and (533) crystal planes of ZnFe₂O₄ (JCPDS NO. 22-1012), respectively. Meanwhile, the peaks observed at 2θ = 31.54°, 34.22°, 36.00°, 47.37° and 67.71° can be attributed to crystalline ZnO (JCPDS NO. 36-1451). The diffraction peak of 41.88° in ZFG500 is associated with the (200) crystal plane of FeO (JCPDS NO. 06-0615). Of note, no typical peaks of rGO (002) or GO (001) were observed in ZFG500, due to the fact that GO was reduced to rGO and the rGO sheets were exfoliated by ZnFe₂O₄ NPs [41, 42]. Figure 2(b) displays the patterns of ZFG300, ZFG400, and ZFG600. It can be clearly seen that the diffraction peak intensity of ZnFe₂O₄ enhances gradually with the increasing calcination temperature.

Zoom In Zoom Out Reset image size

Based on the XRD results, the crystal sizes of ZnFe₂O₄ were estimated from the FWHM (full width at half maximum) of the diffraction peaks of (311) by the well known Scherrer equation and summarized in table 1. The grain sizes of ZnFe₂O₄ in composites were increased from 27.2 nm to 37.4 nm with the increasing calcination temperature. Because the negative charge environment of the GO surface serves as anchoring and nucleation sites, positively charged metal ions are required to participate in the nucleation and growth of ZnFe₂O₄, which ultimately leads to a bigger crystal size of ZnFe₂O₄ in ZnFe₂O₄/rGO composites than pure ZnFe₂O₄ [7].

The chemical components of samples were confirmed by FT-IR and Raman spectroscopy. Figure 3(a) shows the FT-IR spectra of the materials. The absorption bands of GO at 3369 cm⁻¹, 1734 cm⁻¹, 1224 cm⁻¹ and 1055 cm⁻¹ are indexed to the stretching vibrations of O–H of water, C=O of carboxyl and carbonyl, C–O of carboxyl and C–OH, respectively [40, 43, 44]. Usually, the absorption band of GO at 1620 cm⁻¹ is attributed to the bending vibration of O–H of adsorbed water [45]. For pure ZnFe₂O₄, the absorption bands observed at 549 cm⁻¹ and 423 cm⁻¹ can be assigned to the tetrahedral Zn²⁺ (Zn–O) stretching vibration and octahedral Fe³⁺ (Fe–O) stretching vibration, respectively. The absorption band at 1627 cm⁻¹ is related to the O–H bending vibration of water. In the case of ZFG500, the absorption bands for oxygen functional groups of GO are markedly weakened or disappear, implying the substantial reduction of GO after the reduction reaction. The weak absorption bands located at 1382 cm⁻¹ and 1124 cm⁻¹ are attributed to the bending vibration of O–H and stretching vibration of C–O–C, respectively. Meanwhile, new bands of Zn–O and Fe–O stretching vibrations are observed at 562 cm⁻¹ and 436 cm⁻¹ respectively, which suggests the existence of ZnFe₂O₄. Furthermore, a weak band appearing at 1582 cm⁻¹ is due to the conjugate C=C skeletal stretching vibration of rGO in the ZFG500 [8]. Combined with the results of XRD, it further proves that the ZnFe₂O₄/rGO composites were successfully prepared.

Zoom In Zoom Out Reset image size

The Raman spectrum of ZFG500 is depicted in figure 3(b). The two prominent peaks at 1322 cm⁻¹ (D peak) and 1589 cm⁻¹ (G peak) are attributed to rGO. The I_D/I_G intensity ratio could be a measure of disorder degree and the average size of the sp² domains in graphite materials [46, 47]. The I_D/I_G ratio for ZFG500 is about 1.5, which indicates the presence of localized sp³ defects in the rGO network. The three active Raman modes (A_1g + E_g + F_2g) in the 100 ∼ 1000 cm⁻¹ region can be ascribed to the spinel ZnFe₂O₄ phase, indicating the presence of ZnFe₂O₄ in the ZFG500 composite.

The morphology of ZFG500 and pure ZnFe₂O₄ were investigated by TEM. The image of pure ZnFe₂O₄ is shown in figure 4(a). Due to the aggregation of bare ZnFe₂O₄ NPs, the observed particle size is larger than the grain size of the XRD calculation. As can be seen from figure 4(b) for the TEM graph of the representative ZFG500, ZnFe₂O₄ NPs were densely grown on the crumpled rGO substrates without aggregation. Figure 4(c) clearly demonstrated that the rGO sheets were decorated homogeneously by ZnFe₂O₄ NPs with a diameter of 20 ∼ 40 nm, which is consistent with the results obtained by the Dubye-Scherrer formula. The high resolution TEM image of ZFG500 is presented in figure 4(d). It shows that the crystal lattice has distance of 0.252 nm and 0.296 nm, which correspond to (311) and (220) planes of ZnFe₂O₄ crystals, respectively.

Zoom In Zoom Out Reset image size

Figure 5 shows the high magnification TEM images of ZFG300 (a), ZFG400 (b), ZFG500 (c), and ZFG600 (d). The ZnFe₂O₄ NPs and layered rGO were observed in samples calcined at 300 °C, 400 °C, 500 °C and 600 °C. It could be presumed from the darker colour portions of images that there are different degrees of stacking for the layered rGO in the composites. The ZFG300, ZFG400, ZFG500 and ZFG600 samples have ZnFe₂O₄ particle sizes of 15 ∼ 35 nm, 20 ∼ 40 nm, 20 ∼ 40 nm, and 25 ∼ 50 nm, respectively, which are consistent with the average grain sizes calculated by XRD.

Zoom In Zoom Out Reset image size

The chemical state of the elements in ZFG500 before and after photocatalytic reactions were measured by XPS. It is obvious that the elements of Zn, Fe, C and O coexist in ZFG500, figure 6(a). The peaks of Zn 2p_1/2 and Zn 2p_3/2 are located at 1044.0 eV and 1020.9 eV (figure 6(b)), indicating that the Zn (II) oxidation state exists in the composite [48]. Simultaneously there is no significant change in the binding energy of Zn 2p before and after reactions, which shows that zinc ions have good stability on the catalyst surface. Figure 6(c) shows Fe 2p_1/2 peak located at 724.4 eV, and four peaks at 709.7 eV, 710.9 eV, 712.0 eV and 713.7 eV were fitted for the Fe 2p_3/2 [49, 50]. The peak value of 709.7 eV was ascribed to the existence of Fe (II) oxidation, and the other three peaks were assigned to the Fe (III) oxidation [51]. In addition, the peaks at 732.7 eV and 718.7 eV were attributed to shake-up satellite structures [52]. Based on the corresponding peak area, the proportion of Fe (II) area decreased from 23% to 19.3% (figure 6(d)) after five repeated reaction runs, possibly due to the reaction of Fe (II) with ·OH to form Fe (III) [53]. Figure 6(e) displayed three main peaks of O 1s with binding energy of 531.3 eV, 530.2 eV and 529.3 eV, which could be assigned to the residual oxygen-containing groups (epoxy and hydroxyl), adsorbed water, and lattice oxygen bonding with Fe and Zn (denoted as Fe–O and Zn–O), respectively [40, 47]. The peaks of C 1s were found at 288.2 eV, 285.3 eV, 284.0 eV and 283.5 eV in figure 6(f), which could be attributed to C=O bonds, C–O bonds (epoxy and hydroxyl), C–C bonds, and C=C bonds, respectively [54, 55].

Zoom In Zoom Out Reset image size

The optical absorption properties of catalysts is shown in figure 7. As shown in figure 7(a), ZnFe₂O₄ has remarkable absorbance in the 200 ∼ 700 nm range. In comparison with ZnFe₂O₄, ZFG500 exhibited significantly improved visible absorption capability after the introduction of rGO. The band gap energy (E_g) of the photocatalyst can be estimated according to the Kubelka-Munk equation [32]:

$$\begin{eqnarray}&&{(\alpha h\nu )}^{n}={\rm{k}}(h\nu -{E}_{{\rm{g}}})\end{eqnarray} \tag{ 2 }$$

where α is the absorption coefficient, hv is the energy of the incident photon, E_g is the band gap energy (eV), and n is a constant, respectively. From figure 7(b), the E_g of bare ZnFe₂O₄ was estimated to be 1.9 eV. Because of the strong absorption of ZFG500 in the visible region, the E_g of ZFG500 reduces to 1.6 eV. Literature reports that the red shift of the absorption band in the visible light region was beneficial to enhance the photocatalytic activity of the catalysts [56].

Zoom In Zoom Out Reset image size

The magnetic property is vital for a recyclable photocatalyst. The hysteresis loops of ZnFe₂O₄ and ZFG500 were measured by VSM at room temperature, as shown in figure 8. The saturation magnetization (Ms) of ZFG500 is 47.7 emu  g⁻¹ while the coercivity is 107.6 Oe, revealing the soft magnetic properties of the ZFG500 composite. However, due to the small grain size, pure ZnFe₂O₄ did not show the hysteresis loop, indicating that pure ZnFe₂O₄ has paramagnetic properties [45]. As seen in the inset picture, the ZFG500 catalyst was quickly separated by an external magnet.

Zoom In Zoom Out Reset image size

The N₂ adsorption-desorption isotherms and the BJH pore size distribution of ZnFe₂O₄/rGO composites are shown in figure 9. The N₂ adsorption-desorption isotherms of ZnFe₂O₄/rGO composites can be assigned to type IV with H3 hysteresis loops observed at relative pressure of 0.43 ∼ 1.0, which reveal the presence of mesoporous and slit-like pores in the composites [57]. The BET surface areas were calculated to be 36.5 m²  g⁻¹ for ZFG300, 32.6 m²  g⁻¹ for ZFG400, 31.1 m²  g⁻¹ for ZFG500 and 21.0 m²  g⁻¹ for ZFG600, respectively. Noticeably, the higher calcination temperature, the smaller specific surface area of the samples obtained. The phenomenon can be interpreted by the fact that the particle size of ZnFe₂O₄ increases with the increasing of calcination temperature, resulting in a decrease in the specific surface area of the materials. Furthermore, the pore size distribution of all ZnFe₂O₄/rGO products was estimated to be about 3 ∼ 4 nm from the desorption curves of the isotherms.

Zoom In Zoom Out Reset image size

The photocatalytic removal of MB under H₂O₂/visible light systems using ZnFe₂O₄/rGO composites and bare ZnFe₂O₄ as catalysts was investigated. The adsorption properties of different catalysts for MB were first evaluated. After dark adsorption for 1 h, the adsorption efficiency of MB by ZFG300, ZFG400, ZFG500 and ZFG600 are 49%, 43%, 38% and 34%, respectively (figure 10(a)), which was supported by the experimentally observed results of BET surface area measurement of the ZnFe₂O₄/rGO composites. On the other hand, bare ZnFe₂O₄ showed only 23% adsorption capacity after 1 h. It can be explained that during the adsorption process, the dye molecules were transferred from the bulk solution to the catalyst surface and the sp² hybridized structure of rGO allows preferential absorption of aromatic components contained in MB molecules through non-covalent π-π stacking interactions [58, 59].

Zoom In Zoom Out Reset image size

After the dark adsorption, the adsorption-desorption equilibrium solutions were used as the starting solution, and the concentration changes of MB against the irradiation time was evaluated as shown in figure 10(b). MB showed approximately 15% degradation after 60 min under visible light irradiation using bare ZnFe₂O₄, while MB degradation rate was 22% with the help of H₂O₂. This is because the dissociation of H₂O₂ under visible light (420 < λ < 550 nm) produces free hydroxyl radicals (·OH), which attack the MB framework and initiate MB degradation [60, 61]. However, the photocatalytic activity of ZnFe₂O₄ was improved to 32% after 60 min by the introduction of H₂O₂ under visible light, due to the formation of more ·OH radicals by the reaction of ZnFe₂O₄ with H₂O₂ or the direct oxidation of MB by the photo-induced holes (h⁺). From figure 10(b), a very small amount of MB was degraded by ZFG500-H₂O₂ in dark. As is well known, the Fenton reaction could promote the formation of ·OH radicals to some extent [53]. However, the recovery of Fe²⁺ through reaction (10) is extremely slow [62], and there is no UV light in the reaction system to be employed for the generation of Fe²⁺ through reaction (3).

$$\begin{eqnarray}&&{{\rm{Fe}}}^{3+}+{\rm{UV}}+{{\rm{OH}}}^{-}\to {{\rm{Fe}}}^{2+}+\cdot {\rm{OH}}\end{eqnarray} \tag{ 3 }$$

The MB degradation rate by the ZFG500 catalyst is 26% after 180 min with visible light, and was significantly enhanced when adding H₂O₂ to the reaction system. The degradation rate of MB was 65% after irradiation for 15 min and reached 98% after irradiation for 180 min. During irradiation, the photo-generated electrons (e⁻) can immediately be transferred from the conduction band of ZnFe₂O₄ to rGO and then react with H₂O₂ to generate massive ·OH radicals, which serve as a strong oxidant for the degradation of MB dye. Meanwhile, the h⁺-e⁻ pairs recombination was dramatically suppressed in the catalysts due to the faster electron-migration process, and then the photocatalytic efficiency was enhanced. Of note, the H₂O₂-assisted ZnFe₂O₄/rGO catalysts exhibited similar photocatalytic activity, indicating that H₂O₂ and rGO play a key role in the photocatalytic reactions. The results suggests that rGO not only prevent the agglomeration of ZnFe₂O₄ NPs as a support but also promote the separation of photo-generated carriers.

To further investigate the photocatalytic processes, the kinetics for the degradation of MB by ZnFe₂O₄/rGO catalysts were carried out. A second-order kinetics model was used to describe the photocatalytic processes as following [62]:

$$\begin{eqnarray}&&\left(\displaystyle \frac{1}{C}\right)-\left(\displaystyle \frac{1}{{C}_{0}}\right)={\rm{kt}}\end{eqnarray} \tag{ 4 }$$

where t is the visible light exposure time, k is the second-order apparent rate constant, C₀ is the concentration at adsorption equilibrium, and C is the residual concentration of MB. The apparent rate constant k is the basic kinetics parameter which can reflect the photocatalytic activity of different catalysts [63], and the fitting results of second-order kinetics are shown in figure 11(a). The k were calculated to be 0.0106 L mg⁻¹  min⁻¹ for ZFG300, 0.0054 L  mg⁻¹  min⁻¹ for ZFG400, 0.0209 L  mg⁻¹  min⁻¹ for ZFG500 and 0.0133 L  mg⁻¹  min⁻¹ for ZFG600. The ZFG500 has the fastest rate of apparent rate constant, indicating that ZFG500 has the relatively best photocatalytic performance. From the kinetics analysis, the photocatalytic activity of the catalyst decreases first and then increases gradually with the increase of calcinations temperature, and the best catalytic effect is obtained at calcination temperature of 500 °C. It can be assumed that the photocatalytic activity of the catalyst is not only related to the specific surface area of the catalyst, but also to the crystallinity of ZnFe₂O₄ NPs.

Zoom In Zoom Out Reset image size

The effect of the amount of ZFG500 catalyst and the initial MB concentration on the degradation kinetics of MB were firstly investigated. As shown in figures 11(b) and (c), the degradation reactions followed second-order kinetics. The maximum apparent rate constant was obtained when 0.5 g  L⁻¹ of ZFG500 catalyst was used. The light scattering factor with an increase in the density of the suspension may be the reason for the observed decrease in the photocatalytic activity when increasing the catalyst concentration of above 0.5 g  L⁻¹ [64]. In addition, as the initial concentration of MB increases, the apparent rate constant decreases.

Next, the degradation of MB by catalysts calcined at different temperatures was also studied under simulated solar light. Figure 11(d) shows the fitting results of second-order kinetics for different catalysts. The photocatalytic reaction apparent rate constants (k) on ZFG300, ZFG400, ZFG500, and ZFG600 are in turn 0.0520 L  mg⁻¹  min⁻¹, 0.0544 L  mg⁻¹  min⁻¹, 0.0828 L  mg⁻¹  min⁻¹, and 0.0613 L  mg⁻¹  min⁻¹. The results show that the k values of different catalysts are on the same order of magnitude. ZFG500 still shows the best photocatalytic activity in simulated solar light, which demonstrates that the optimum calcination temperature is 500 °C. Compared with visible light catalysis, the photocatalytic efficiency of the catalysts is significantly enhanced in simulated solar light because the extra ultraviolet light promotes the progress of photocatalysis.

Finally, the effect of H₂O₂ dosage and the universality of the obtained catalysts for the removal of different types of dye were determined (figures 12(a)–(c)). It can be seen that the degradation efficiency of the photocatalyst on dyes first increases and then decreases with the increasing H₂O₂ dosage. The decrease of degradation efficiency may be ascribed to the unfavourable consumption of ·OH radicals by the excess H₂O₂, a reaction that scavenges ·OH radicals and generates ·HO₂, the latter of which would also react with ·OH to form H₂O and O₂ [60, 65]. When the dosage of H₂O₂ in the reaction system was 0.3 ml, maximum degradation efficiencies were obtained after 180 min under visible light. Figure 12(d) shows the digital images of colour changes (MB, MO and RhB) during the visible light irradiation. It can be seen that as the illumination time increases, the dyes solutions are obviously discoloured. At the same time, the experimental results revealed that the ZFG500 photocatalyst has better efficiency on cationic dyes (MB and RhB) which are much easier to be adsorbed on the anionic catalysts than anion dyes (MO).

Zoom In Zoom Out Reset image size

To identify the main reactive oxidants responsible for the degradation of MB, trapping experiments were performed. Silver nitrate (AgNO₃), isopropyl alcohol (IPA) and sodium oxalate (Na₂C₂O₄) were used to capture the photo-generated e⁻, ·OH radicals and photo-generated h⁺, respectively [53, 65]. As depicted in figure 13, in the absence of scavengers, MB was almost completely degraded by ZFG500. The degradation efficiency of MB decreased significantly with the addition of AgNO₃ because of the capture of photo-generated e⁻ by ${\rm{N}}{{\rm{O}}}_{3}^{-},$ which leads to a decrease in the generation of ·OH radicals. The degradation of MB also decreased considerably with the addition of IPA while Na₂C₂O₄ slightly decelerated the photocatalytic reaction. It should be pointed out that the addition of IPA has a greater negative impact on photocatalytic degradation than AgNO₃, indicating that ·OH radicals are the major species responsible for the oxidation of MB. Moreover, the ·OH radicals are mainly derived from the reaction of photo-generated e⁻ with H₂O₂, and the Fenton reaction also contributes.

Zoom In Zoom Out Reset image size

The possible expected routes to achieve an enhanced photocatalytic degradation of MB using ZnFe₂O₄/rGO photocatalysts under visible light exposure are discussed as follows (see also figure 14).

Zoom In Zoom Out Reset image size

3.8.1. Semiconductor mediated photo-degradation process

When ZnFe₂O₄ was radiated by visible light with the photon energy higher or equal to the band gaps of ZnFe₂O₄, the electrons in the valence band could be excited to the conduction band with simultaneous generation of the same amount of holes (h⁺) in the valence band (reaction (5)). The conduction band electrons (e⁻) immediately transferred from ZnFe₂O₄ to the surface of rGO sheets (reaction (6)) through internal diffusion and surface migration because the conduction band of graphene (−0.75 V versus normal hydrogen electrode (NHE)) [66] is more positive than that of ZnFe₂O₄ (−1.54 V versus NHE) [67]. It can be seen from figures 10(b) and 12 that the presence of H₂O₂ had a significant influence on the catalytic efficiency and the trapping experiments (figure 13) proved that the hydroxyl radicals (·OH) are the main active species catalysed by visible light. H₂O₂ in the solutions were continuously adsorbed to the catalyst surface and participated in the photocatalytic reaction. The e⁻ transferred from the conduction band of ZnFe₂O₄ to rGO could be captured by H₂O₂ (E⁰ (H₂O₂/·OH) = +0.38 V versus NHE) [60] to generate ·OH radicals (reaction (7)), rather than reacting with dissolved O₂ to form $\cdot {{\rm{O}}}_{2}^{-}$ (E⁰ (O₂/$\cdot {{\rm{O}}}_{2}^{-}$) = −0.33 V versus NHE) [52]. Finally, the generated ·OH radicals and the h⁺ left in the valence band of ZnFe₂O₄ acted as strong oxidants to initiate the degradation of organic dyes in aqueous solution (reaction (8)). By this method, the recombination of h⁺-e⁻ pairs was effectively suppressed, and the dye was effectively degraded.

$$\begin{eqnarray}&&{{\rm{ZnFe}}}_{2}{{\rm{O}}}_{4}+h\nu \to {{\rm{ZnFe}}}_{2}{{\rm{O}}}_{4}\left({h}^{+}+{e}^{-}\right)\end{eqnarray} \tag{ 5 }$$

$$\begin{eqnarray}&&\,{{\rm{ZnFe}}}_{2}{{\rm{O}}}_{4}\left({h}^{+}+{e}^{-}\right)+{\rm{rGO}}\to {{\rm{ZnFe}}}_{2}{{\rm{O}}}_{4}\left({h}^{+}\right)+{\rm{rGO}}\left({e}^{-}\right)\end{eqnarray} \tag{ 6 }$$

$$\begin{eqnarray}&&{\rm{rGO}}\left({e}^{-}\right)+{{\rm{H}}}_{2}{{\rm{O}}}_{2}\to \cdot {\rm{OH}}+{{\rm{OH}}}^{-}+{\rm{rGO}}\end{eqnarray} \tag{ 7 }$$

$$\begin{eqnarray}&&{{\rm{ZnFe}}}_{2}{{\rm{O}}}_{4}\left({h}^{+}\right)+\cdot {\rm{OH}}+\text{dye}\,\to {\rm{degradation}}\,{\rm{products}}\end{eqnarray} \tag{ 8 }$$

3.8.2. Fenton process [62, 68]

From figure 10(b), the degradation of MB was observed in the dark with the catalysis of ZFG500 and H₂O₂, indicating that the active species independent of the visible light were produced. The H₂O₂ adsorbed on the catalyst surface were directly trapped by the Fe³⁺ to form Fe²⁺ (reaction (9)), which could react with H₂O₂ to produce ·OH radicals and Fe³⁺ (reaction (10)). However, the recovery of Fe²⁺ through reaction (9) is extremely slow. Upon illumination of visible light, the reaction (7) consumed most of the H₂O₂ to generate ·OH radicals, so the production of ·OH radicals through the Fenton process was limited by the competitive reaction on the surface of catalyst. Lastly, the dye was slightly degraded by the generated ·OH radicals (reaction (11)).

$$\begin{eqnarray}&&{{\rm{H}}}_{2}{{\rm{O}}}_{2}+{{\rm{Fe}}}^{3+}\to {{\rm{Fe}}}^{2+}+\cdot {{\rm{HO}}}_{2}+{{\rm{H}}}^{+}\end{eqnarray} \tag{ 9 }$$

$$\begin{eqnarray}&&{{\rm{Fe}}}^{2+}+{{\rm{H}}}_{2}{{\rm{O}}}_{2}\to {{\rm{Fe}}}^{3+}+\cdot {\rm{HO}}+{{\rm{OH}}}^{-}\end{eqnarray} \tag{ 10 }$$

$$\begin{eqnarray}&&\cdot {\rm{OH}}+\text{dye}\,\to {\rm{degradation}}\,{\rm{products}}\end{eqnarray} \tag{ 11 }$$

In order to examine the stability and reusability of the as-prepared photocatalysts, ZFG500 was selected for a cyclic decomposition experiment. As shown in figure 15(a), the XRD analysis of the ZFG500 catalyst before and after five repetitive reactions was carried out. It was found that the main diffraction peaks of the recycled sample have no obvious change, which indicated that the ZnFe₂O₄/rGO composites have stable crystal structures. The repeated photocatalytic results are given in figure 15(b). After 180 min of light irradiation, the degradation rate of MB is still over 90% after five cycles, indicating a stable photocatalytic ability of the ZnFe₂O₄/rGO catalysts.

Zoom In Zoom Out Reset image size