3.1. XRD and morphology analysis
Through X-ray diffraction (XRD) analysis, the pure ZFO sample showed 5 obvious characteristic diffraction peaks at 2θ = 29.9, 35.2, 42.8, 56.6 and 62.1 °, corresponding to ZFO (2 2 0), (3 1 1), (1 0 0), (5 1 1) and (4 4 0) crystal planes, which coincide with the diffraction peak positions of the ZFO JCPDS standard card (79-1150). The pure BOI sample has 5 distinct characteristic diffraction peaks at 2θ = 29.7, 31.7, 45.5, 51.5 and 55.3 °, corresponding to BOI (012), (110), (020), (114) and (122) crystal planes, which coincide with the diffraction peak positions of the BOI JCPDS standard card (PDF#73-2062). The X-ray diffraction pattern of ZFO/BOI composite material shows that in addition to the significant diffraction peaks of BOI, there are also ZFO diffraction peaks, and no impurity phase is observed, which indicates that ZFO and BOI have been successfully recombined.
Figure 2a shows the SEM image of pure ZFO. The hydrothermally synthesized ZFO has a uniform spherical structure. Fig. 2d is the SEM image of the composite ZFO/BOI-2. It can be seen that the composite material is a flake BOI with uniformly attached ZFO pellets. Fig. 2b shows the TEM image of the ZFO/BOI-2 sample. The structure composed of flakes and small balls can be observed, consistent with the SEM image of ZFO/BOI. It can be seen from the mark in Fig. 2c that the enlarged view of the circular area is Fig. 2f, and its lattice spacing is 0.25 nm, which corresponds to the (311) crystal plane of ZFO. The enlarged view of the square area in Fig. 2c is Fig. 2g, and its lattice spacing is 0.28 nm, which corresponds to the (110) crystal plane of BOI. Fig. 2e is a selected area of electron diffraction (SAED), showing a diffraction ring composed of many homogeneous and tiny crystals. The cyan diffraction rings represent the (2 2 0), (3 1 1), (5 1 1) and (4 4 0) crystal planes of ZFO crystals, and the yellow diffraction rings represent the (012), (1 1 0), (0 2 0) and (1 2 2) crystal planes, consistent with the XRD results. The above results indicate the existence of binary composite ZFO/BOI. At the same time, there is a close relationship between ZFO and BOI, forming a Z-scheme heterojunction, which is advantageous for separating photogenerated carriers.
Figure 3 shows the result of using the EDS test to determine the element types in the composite ZFO/BOI. Fig. 3b-f can be seen the five elements of O, Zn, Fe, Bi and I in the composite. It is not difficult to find from the EDS element distribution diagram that the distribution of each element is consistent with the shape of the composite material, which further proves that the composite material ZFO/BOI was successfully prepared without other impurities.
3.2. XPS and N2 adsorption–desorption
We use XPS to study the elemental composition and chemical valence of ZFO/BOI-2 nanocomposites. Fig. 4 detects the peak signals of the six elements C, Zn, Fe, Bi, O and I in the ZFO/BOI-2 nanocomposite. The C1s spectrum (Fig. 4a) fits three peaks at 585.3, 586.8 and 589.2 eV, respectively, which indicates that the carbon species in this sample has three different chemical environments. The binding energy (BE) of 285.3 eV coincides with -C-C and is identified as graphite or adventitious carbon. The peak at 286.8 eV is designated to be embedded in the C-O bond in the interlayer compound. The peak close to 289.2 eV indicates the presence of carbonate species (C=O).(Zeng et al. 2019) The peaks at 1045.2 and 1021.9 eV in Fig. 4b can be designated as the binding energies of Zn 2p1/2 and Zn 2p3/2, which are similar to the standard data of Zn2+.(Li et al. 2011) In the Fe 2p spectrum from 740 to 705 eV (Fig. 4c), the two prominent peaks belonging to Fe 2p1/2 and Fe 2p3/2 appear at 725.8 and 711.9 eV, while the satellite peak of Fe 2p3/2 is at 720 eV. The binding energy of Fe 2p2/3 is 710.9 eV, which matches the binding energy of Fe 2p2/3 in ZFO very well. Meanwhile, Fe 2p3/2 peaks can be deconvolved into two peaks located at 711.9 and 710.9 eV, allocated to octahedral and tetrahedral Fe3+, respectively. (Cai et al. 2016, Li et al. 2018b) According to Fig. 4d, the two typical peaks of Bi 4f are located at 159.6 and 164.9 eV, and respectively, this finding confirmed that Bi3+ cations in ZFO/BOI composites correspond to Bi 4f5/2 and Bi 4f7/2, respectively.(Qiu et al. 2017, Yang et al. 2018) The O1s map (Fig. 4e) fitted three peaks at 530.2, 531.5 and 532.8 eV respectively. The characteristic peak at 530.2 eV was derived from the Bi-O bond of {Bi2O2}2+ layer in BOI, and the peak at 531.9 eV was corresponding to the Fe-O bond in ZFO. The peak at 532.8 eV corresponds to the characteristic peak of free oxygen or hydrated oxides on the sample surface.(Bai et al. 2019, Tian et al. 2020) Two peaks were observed at about 619.5 and 631 eV corresponding to I 3d5/2 and I 3d3/2, respectively (Fig. 4f), which were attributable to the I− ion of BOI.(Li et al. 2018a) In addition, the above analysis indicated the presence of BOI and ZFO in the composite sample.
The specific surface area and pore size distribution of pure ZFO and ZFO/BOI-2 composites were studied by speed ratio surface and porosity analyzer. As shown in Fig. 5a, the nitrogen adsorption-desorption isotherms of pure ZFO and ZFO/BOI-2 photocatalysts are in the range of 0-1.0 P/P0, and there is an evident IV type H3 hysteric ring. The IV type H3 hysteric ring indicates that the sample has a mesoporous structure and is composed of flake nanoparticles,(Raza &Faraz 2020) which is consistent with SEM and HRTEM. In addition, the specific surface areas of pure ZFO and ZFO/BOI-2 were 88.988 and 42.538 m²/g, respectively, calculated by Brunauer-Emmett-Teller(BET) model. The corresponding pore size distribution of the samples was determined by the BJH method, and the average adsorption pore size of ZFO and ZFO/BOI-2 was 151.06 and 218.39Å.
3.3. Optical analysis
The light absorption capacity of semiconductor materials has an important influence on the photocatalytic activity. We analyzed the spectral absorption range of the photocatalyst by UV-Vis absorption spectrum. Fig. 6 shows the UV-Vis absorption spectra of the nanomaterials in the 250-800 nm range. Compared with pure BOI samples, the absorption intensity of ZFO/BOI samples was significantly improved in the range of 600-800 nm, and the absorption edge was redshifted. This indicates that the binary composite photocatalyst can enhance the response in the visible region. According to Fig. 6b, the Eg values of ZFO and BOI were 1.71 and 2.02 eV, respectively. The red shift of the Z-scheme heterojunction ZFO/BOI may be due to the synergy between the flake BOI and ZFO nanoparticles.
PL spectroscopy is an effective method to study the composite behaviour of photo charges generated by photocatalyst. The PL spectra of pure ZFO and ZFO/BOI nanocomposites are shown in Fig. 7. Pure ZFO and ZFO/BOI nanocomposites show a luminescence peak at 505 nm. The luminescence intensity of ZFO/BOI nanocomposites is lower than that of pure ZFO, and the luminescence peak intensity of ZFO/BOI-2 is the lowest. The recombination rate of photogenerated carriers is the lowest. ZFO nanoparticles were modified on the surface of the flake BOI nanostructures to construct the ZFO/BOI heterojunction and the recombination rate of photogenerated carriers could be effectively reduced. However, when the content of BOI in ZFO/BOI is too high, many electron-hole pairs will recombine on the surface of BOI.
As shown in Fig. 8, the characteristic peaks located at 3445 cm−1 and 1626 cm−1 are respectively the O-H bond stretching vibration peaks and H-O-H.(Yosefi et al. 2017) In addition, the vibration mode corresponding to CH3-bending was detected at 1383 cm−1.(Tamaddon et al. 2020) For pure BOI, the absorption peak located at 500 cm−1 is derived from the stretching vibration of the Bi-O bond.(Zhou et al. 2017) In addition, for the spinel ZFO sample, the A position is mainly occupied by Zn2+, and Fe3+ mainly occupies the B position. The characteristic peaks located at 571cm−1 and 418 cm−1 correspond to the stretching vibrations of the Zn-O and Fe-O bonds in the tetrahedral and octahedral positions, respectively.(Khasevani &Gholami 2019)
3.4. photo electrochemical properties analysis
The CB and VB potentials of ZFO and BOI were determined by Mott-Schottky (M-S) plots. By extrapolating the linear part of the M-S plots to the horizontal axis. The slope of the M-S plot of ZFO is positive, indicating that it is an n-type semiconductor. That of BOI is negative, showing as a p-type semiconductor. According to V(NHE)=V(SCE)+0.059pH+0.242(pH=7),(Huang et al. 2021) the relative standard calomel electrode (vs SCE) potential is converted to the standard hydrogen electrode (vs NHE) potential. It can be seen from Fig. 9a that the V(SCE) of ZFO is -1.21 V, the V(SCE) of BOI is 1.07 V. It is generally believed that the VB edge potential of p-type semiconductors and the CB edge potential of n-type semiconductors can be approximately equal to the VFB of semiconductors. Since ZFO is an n-type semiconductor, the conduction band potential of ZFO is VCB=V(NHE)= -0.55 V. The valence band potential of BOI is VVB=V(NHE)=1.72 V because BOI is a p-type semiconductor.
EIS impedance spectroscopy and transient photocurrent test research were carried out to understand the photo-Fenton mechanism in-depth. Fig. 10a shows that the EIS of nanomaterials is approximately semicircle in the high-frequency region. The semicircle diameter of ZFO/BOI composite in the high-frequency region is smaller than that of pure ZFO and BOI, which indicates that it has a low charge transfer resistance. The results show that the ZFO/BOI heterojunction can significantly improve photogenerated electron-hole pairs' separation and migration efficiency. Fig. 10b shows the transient photocurrent response of the sample in five cycles of on/off under simulated sunlight. The photocurrent densities of ZFO, BOI, ZFO/BOI-1, ZFO/BOI-2 and ZFO/BOI-3 photocatalysts were 5.82×10−8, 1.34×10−7, 1.27×10−6, 1.71×10−6 and 6.5×10−7 mA/cm2,respectively. After the light is turned off, the photocurrent is reduced to the initial value. Compared with pure ZFO and BOI, the photocurrent density of ZFO/BOI-2 heterojunction is significantly increased, indicating that the photogenerated carriers of ZFO/BOI-2 heterojunction can be separated and migrated more effectively. By comparing the EIS and transient photocurrent responses of pure and composite photocatalysts, the ZFO/BOI nanocomposites showed better-photogenerated charge separation and migration ability, which indicated that the ZFO/BOI heterojunction composite photocatalyst might have better Fenton activity.
3.5. Photo-Fenton analysisis
The photo-Fenton activity of ZFO/BOI heterojunction was tested by dissolving RhB in simulated solar degradation. Fig. 11a shows the degradation rates of photo-Fenton degradation of RhB by pure ZFO, BOI and ZFO/BOI nanocomposites with different composite ratios. Under the action of photocatalytic degradation, the concentration of RhB dye decreases with the increase of the illumination time. Under the illumination condition of 90 min, ZFO, BOI, ZFO/BOI-1, ZFO/BOI-2 and ZFO/BOI-3 nanocomposites had degradation rates of 27.5, 58.9, 84.1, 88.6 and 78.1%, respectively. In Fig. 11c, the degradation rates of ZFO, BOI, ZFO/BOI-1, ZFO/BOI-2 and ZFO/BOI-3 nanocomposites were 24.3, 51.9, 96.9, 100 and 94.6%, respectively, under the light condition of only 20 min. The photo-Fenton degradation rate of ZFO/BOI heterojunction was significantly higher than that of pure ZFO and BOI, and the photo-Fenton activity of ZFO/BOI-2 was the strongest. The excellent photonic Fenton properties of ZFO/BOI nanocomposites are since highly dispersed ZFO particles are uniformly distributed on the surface of the sheets of BOI and fully contact to form a ZFO/BOI heterojunction. ZFO/BOI heterojunction can enhance the visible light absorption capacity. Moreover, this novel ZFO/BOI heterojunction can promote the migration and separation of photogenerated carriers and inhibit their recombination, thus improving the photo-Fenton degradation ability. Through the first-order kinetic model equation: ln(Ct/C0) = -Kappt, the kinetics characteristics of the degradation of RhB by ZFO/BOI heterojunction photo-Fenton can be quantitatively studied, where Kapp is the first-order kinetic reaction rate, and C is the concentration of RhB dye. Fig. 11b is ZFO/BOI photocatalytic degradation of corresponding first-order kinetics curves of ZFO, BOI, ZFO/BOI-1, and ZFO/BOI-2, ZFO/BOI-3 samples of the reaction rate constant Kapp were 0.0025, 0.0083, 0.016, 0.021 and 0.015 min−1. Fig. 11d is Fenton ZFO/BOI light degradation of corresponding first-order kinetics curve, ZFO, BOI, ZFO/BOI-1, ZFO/BOI-2, and ZFO/BOI-3 samples of the reaction rate constant Kapp were 0.0147, 0.0857, 0.1518, 0.4008 and 0.1404 min−1. The Kapp value of ZFO/BOI-2 was the highest in the composite sample, indicating that the photo-Fenton degradation rate of ZFO/BOI-2 was the highest, and the photo-Fenton degradation rate was about 19 times that of the photocatalytic degradation rate. As the amount of BOI increases, Kapp first increases and then decreases. A part of photogenerated carriers will recombine on the BOI surface because when the BOI content in ZFO/BOI is too high.
Test the active species of ZFO/BOI-2 photo-Fenton to degrade RhB through the dynamic species capture experiment, using ethylenediaminetetraacetic acid (EDTA), p-benzoquinone (BQ) and isopropyl alcohol (IPA) to capture h+, ·O2- and ·OH. The illustration in Fig. 12 shows the degradation of RhB over time in the ZFO/BOI-2 sample after adding different capture agents. The degradation rate decreased when EDTA or BQ was added. The photocatalytic degradation rate of ZFO/BOI-2 samples decreased significantly after adding IPA, indicating that shows that ·OH is the main active substance. The role of ·O2- and h+ in the degradation of photo-Fenton is relatively weak.
Repeated cyclic tests investigated the stability of ZFO/BOI-2 nanomaterials under the same environment. In Fig. 13, the first degradation rate of the catalyst was 100% after 25 min. Although the degradation rate decreases slightly with the number of cycles, the degradation rate can still reach 81.6% after three cycles, indicating that the cycling stability of ZFO/BOI-2 composite material is good. The ZFO/BOI photocatalyst can be recovered by an external magnetic field due to the ferromagnetic nature of ZFO. ZFO/BOI photocatalyst shows good stability and reproducibility in RhB degradation, which can be used in the actual dye wastewater purification.
3.6. Possible mechanism
It is proposed that the carrier migration mode of Z-scheme heterojunction ZFO/BOI (Fig. 14) is generated because BOI and ZFO absorb enough energy photons to excite electron transition from the valence band to the conduction band. The valence band and conduction band of BOI are lower than that of ZFO, and the photoelectrons in the conduction band of BOI can migrate to the valence band of ZFO to recombine with h+. The remaining carriers undergo redox reactions in the valence band of ZFO and the conduction band of BOI, respectively. In the process of photo-Fenton degradation, e− of ZFO conduction band reacts with O2 in water to generate ·O2-. In addition, because of the strong oxidation of h+, part of it can directly degrade RhB, a small part h+ can react with OH- to form ·OH. The simplified possible reaction of generating hydroxyl radicals in the process of adding hydrogen peroxide to the acidic solution is determined by r1 and r2, as shown below:
Fe3+ + hν + H2O → Fe2+ + ·OH + H+ (r1)
Fe2+ + H2O2→ Fe3+ + ·OH + OH- (r2)
·O2-, h+ and ·OH radicals produced in the photo-Fenton reaction degrades RhB dye into H2O, CO2 and small molecule substances. This is consistent with the above capture experiment results. Therefore, Z-scheme heterojunction ZFO/BOI construction has a noticeable effect on inhibiting the recombination of photo-generated charges and can also improve carriers' transport and separation efficiency. The photo-Fenton process helps to strengthen its economic and environmental sustainability.