Figure 1 shows schematic illustration of the Fe-Ti/SF composite fabrication process. In a typical experimental procedure, 0.85 mL tetrabutyl titanate and 0.31 g ammonium ferric oxalate were dissolved into 4 mL of H2O2 solution (~ 30%). 0.85 mL tetrabutyl titanate and 0.31 g ammonium ferric oxalate were dissolved into 4 mL of H2O2 solution (~ 30%). Then 4 g spent FCC catalyst was added into this solution. Finally, this mixture was grinded for 2 h and calcined at 800 °C for 4 h. The degradation efficiency of methylene blue can be used to evaluate the photocatalytic performance of Fe-Ti/SF composite. The methylene blue can be degraded into the carbon dioxide (CO2) and water (H2O) at last.
XRD patterns of the spent FCC catalyst, Fe-Ti/SF, Fe/SF and Ti/SF are shown in Fig. 2 (2-Theta range from 5° to 80°). Figure 2a shows XRD patterns of the spent FCC catalyst. There are four diffraction peaks at 6.3°, 10.3°, 12.1° and 15.9°, which can be indexed to (111), (220), (311) and (331) planes of zeolite Y phase (JCPDS No. 77-1551), respectively. Figure 2b,c show the XRD patterns of the Fe/SF and Fe-Ti/SF composite. The intensity of the peak (111) in zeolite Y phase decreases after being loaded Fe2O3 or TiO2. In the range of 12°–25°, the peak intensity is higher than Fig. 2a. In Fig. 2d, there are three peaks at 26.4°, 35.3° and 40.1°. These peaks are corresponded to the (101), (004) and (200) planes of the TiO2 phase (JCPDS card No. 73-1764), respectively. The size of the loaded TiO2 is small, which causes the low diffraction peak. Compared with the zeolite Y (111), the diffraction peaks intensity of Fe2O3 is lower. Therefore, there no obvious diffraction peaks of Fe2O3.
Figure 3 presents scanning electron microcopy (SEM) images of the spent FCC catalyst, Fe/SF, Fe-Ti/SF and Ti/SF samples. In Fig. 3a–c, there are some cracks on the surface. In Fig. 3d, the surface of this composite is covered with Fe2O3. In Fig. 3e–f, the peripheral surface of the Fe/SF is coarse. Figure 3g–i show the Fe-Ti/SF composite, this composite is covered with Fe2O3 and TiO2. In Fig. 3i, because the TiO2 deposits in clusters of irregular shapes, the surface of Fe-Ti/SF composite is rough and jagged. Figure 3j–l show the morphologies of the Ti/SF composite. The TiO2 particles are loaded on the surface of supporter. As shown in Fig. 3l, the TiO2 particles with diameter raging from 50 to 200 nm. There is an aggregation phenomenon between TiO2. EDS mappings of the Fe-Ti/SF composite are shown in Supplementary Figure S1. It shows the corresponding mappings of Fe-Ti/SF composite, which clearly indicate the homogeneous of Si, O, Al, Fe and Ti elements. The ingredient of zeolite Y are Al2O3 and SiO2. The Al, Si and O elements originate from zeolite Y, the Fe and Ti are loaded on the zeolite Y.
XPS spectra are used to identify the valence states of the elements of the Fe-Ti/SF composite. Supplementary Figure S2 shows the XPS survey spectrum, which suggests the presence of Fe, Ti, Al, Si and O. In Supplementary Figure S2a, the peaks of Fe 2p3/2 and Fe 2p1/2 are overlap, the three binding energy peaks at 712.0, 721.6 and 725.7 eV. In Supplementary Figure S2b, two banding energy peaks at 458.7 and 464.3 eV correspond to the characteristic of the Ti 2p3/2 and Ti 2p1/2. In Supplementary Figure S2c, the peak of Al 2p is overlap, the banding energy peak at 74.6 eV. In Supplementary Figure S2d, the peak of Si 2p is overlap, the banding energy peak at 102.9 eV. In Supplementary Figure S2e, the binding energy peak at 531.8 eV correspond to the characteristic of the O 1 s.
Figure 4 shows the TEM images of the Fe-Ti/SF composite. In Fig. 4a–c, the Fe2O3 and TiO2 particles are attached to the surface of the spent FCC catalyst. The crystallography of the Fe-Ti/SF composite is investigated with high-resolution TEM (HRTEM). In the Fig. 4d, the nanostructured heterojunction is formed between Fe2O3 and TiO2. The intimately contacted interface is important for accelerating the separation of the electrons and holes. The HRTEM image in Fig. 4d shows two kinds of lattice fringes, the interplanar crystal spacing of 0.352 and 0.252 nm are corresponded TiO2 (101) and Fe2O3 (110), respectively. More TEM images of the Fe-Ti/SF composite are shown in Supplementary Figure S3a–c. The dark color regions under TEM indicate thickness of the zeolite Y. The TEM images further show that Fe2O3 and TiO2 are loaded on the surface of supporter.
In Fig. 5, the macropores properties of the spent FCC catalyst, Fe/SF, Fe-Ti/SF and Ti/SF are analyzed. These insets show the pore diameter of these samples. All the samples show the type-II isotherms, which are corresponded of macropores materials (IUPAC classification). The isotherms of these samples exhibit H3 hysteresis loops associated with the presence of macropores. The pore diameter of the spent FCC catalyst, Fe/SF, Fe-Ti/SF and Ti/SF are 5.6, 5.8, 5.9 and 5.8 nm, respectively. It shows that these samples are macroporous. The surface areas of these composites (spent FCC catalyst, Fe/SF, Fe-Ti/SF and Ti/SF) are calculated to be 226.3, 253.7, 264.5 and 235.2 m2 g−1, respectively. Due to the large surface area, these samples can absorb more methylene blue.
As shown in Supplementary Figure S4, the UV–Vis absorbance spectra of the Fe-Ti/SF, Fe/SF and Ti/SF composites were evaluated. The adsorption edges of the Fe-Ti/SF, Fe/SF and Ti/SF composites are 510, 540 and 400 nm, respectively. The Kubelka–Munk function is applied to determine the band gaps of Fe-Ti/SF, Fe/SF and Ti/SF composites. As shown in inset of the Supplementary Figure S2, the band gaps of the Fe-Ti/SF, Fe/SF and Ti/SF composites from the energy versus [F(R∞)hv]n can be estimated. R∞ and F(R∞) are the limiting reflectance and Kubelka–Munk function, respectively. The slope of the Tauc plot fit well at n = 1/2, which indicated an indirect transition. The band gaps of the Fe-Ti/SF, Fe/SF and Ti/SF composites are 2.23, 1.98 and 3.0 eV, respectively30,31. The partial absorption in the visible range is determined by the band gap energy value, which shows that these samples have potential photocatalytic activity.
The schematic diagram of the band position of Fe-Ti/SF is shown in Fig. 6. The interparticle electrons intensity of Fe-Ti/SF is higher than the single component Fe2O3 or TiO2, which can promote the separation of electrons and holes. The Fermi levels of the Fe2O3 and TiO2 are lower than their conduction band (CB). In the heterostructure, the Fermi levels of Fe2O3 and TiO2 reach a new equalization state. There is a new electric field between the Fe2O3 and TiO2. The interparticle electrons rapidly transfer in the heterostructure between the Fe2O3 and TiO2. The electrons of a higher CB transfer to the lower one, the holes move in the opposite way. The electrons of Fe2O3 (CB) transfer to TiO2 (CB). The holes of TiO2 (VB) transfer to Fe2O3 (VB). The electrons are trapped by O2 (dissolved in the methylene blue) to form ⋅O2, the ⋅O2 oxidized the methylene blue. Finally, the methylene blue is degraded in CO2 and H2O.
The electron transfer efficiency can be measured by the electrochemical impedance spectroscopy (EIS). As shown in Fig. 7, the radius (Nyquist plot) of the Fe-Ti/SF is much smaller than those of Fe/SF and Ti/SF, which shows that Fe-Ti/SF has lower electron transfer resistance. The heterostructure of the Fe2O3 and TiO2 enhances the separation of the electrons and holes.
In Fig. 8, the Fourier transform infrared spectroscopy (FT-IR) spectra of the spent FCC catalyst, Fe/SF, Fe-Ti/SF and Ti/SF are measured. The peak at 3,425 cm−1 corresponds to the absorption the stretching vibration of O‒H group stretching. The peak at 1645 cm−1 corresponds to the bending vibration absorption of the O‒H32. In Fig. 8a–d, the peak at 1,082 cm−1, 848 cm−1 and 456 cm−1 correspond to the stretching or bending vibration of the Si–O–Si33,34. In the samples of the (b) Fe/SF and (c) Fe-Ti/SF, the absorption peaks at 560, 610, 1,100, and 3,500 cm−1 correspond to the stretching vibration of Fe–O. In Fig. 8d, the peaks at 910 cm−1 to 960 cm−1 corresponds to the stretching vibration of the Ti‒O‒Si.
In Fig. 9, the photocatalytic performance of the Fe-Ti/SF is evaluated by the methylene blue degradation experiment. With increase of the reaction time, the methylene blue characteristic peak is gradually declined, which indicates concentration of methylene blue is gradually declined. The photocatalytic degradation efficiency of the methylene blue is ~ 94.2%. The inset of Fig. 9 shows the color change of the methylene blue. Figure. S5 shows the degradation of methylene blue with the Fe-Ti/SF, Fe/SF and Ti/SF composites. The methylene blue degradation efficiencies (with Fe-Ti/SF, Fe/SF and Ti/SF composites) are ~ 94.2%, ~ 22.3% and ~ 54.0% in 120 min, respectively. The results show that the Fe-Ti/SF composite has the highest photocatalytic activity. Supplementary Figure S6 shows the photocatalytic degradation efficiency of methylene blue with Fe-Ti/SF, Fe2O3-TiO2, Fe2O3 and TiO2.
As shown in Fig. 10, the recycling experiments of the Fe-Ti/SF composite are implemented by the methylene blue degradation, which evaluates the stability of this Fe-Ti/SF composite. After the fourth cycle reaction, the photocatalytic activity of the Fe-Ti/SF composite is not significant loss. The degradation efficiency of the methylene blue with Fe-Ti/SF composite decreases to 84.0% from the pristine degradation efficiency (94.2%). These results indicate that the Fe-Ti/SF exhibit a relatively stable photocatalytic performance.