Powder X-ray diffraction patterns of TiO2, α-Fe2O3 and ternary Ag@TiO2/α-Fe2O3 photocatalyst are shown in Fig. 1. As can be observed, in the diffractogram of the ternary nanocomposite, all diffractions peaks characteristic of titanium dioxide, silver and iron oxide are detected. No other peaks or impurities can be detected which prove the purity of the obtained sample. Based on the XRD data, we can assume the successful fabrication of the ternary Ag@TiO2/α-Fe2O3 heterojunction. Figure 2 displays the morphology of the ternary photocatalyst. As shown, we can detect the deposition of both AgNPs and α-Fe2O3 NPs on the TiO2 surface. Furthermore, the HRTEM image confirms the ternary heterojunction by the presence of (111), (003) and (101) lattice spacing characteristic of Ag, α-Fe2O3 and TiO2 NPs, respectively. The SAED pattern further confirms the presence of the three system by the detection of all characteristic electron diffraction responses. To further confirm the successful synthesis of Ag@TiO2/α-Fe2O3 nanocomposites, XPS mesearement was performed to study the chemical composition and their corresponding valence states (Fig. 3). The XPS spectra of the asprepared sample are shown in Fig. 3. In the Ti 2p spectrum, two peaks at 458 eV, and 465 eV were assigned to Ti 2p3/2 et Ti 2p1/2, respectively. Values agree well with the Ti4+ state in TiO2. The spectrum of Fe which exhibites also two peaks at 711 eV and 725 eV can be ascribed to the Fe 2p element of hematite α-Fe2O3. In the case of O 1s, it can be detected the presence of asymmetrical peak located at 529.9 eV which can be attributed to lattice oxygen and surface oxygen. For Ag element, two characteristic peaks are located at 273.6 and 367.8 eV which can be assigned to Ag 3d3/2 et Ag 3d5/2 respectively. Values agree well with Ag0 metal. The deconvolution of the XPS peaks of Ag (Figure S1) revealed the presence of peaks characteristic of Ag-O bonds. As mentioned above, HRTEM results showed the deposotion of AgNPs on both TiO2 and α-Fe2O3 surfaces. The XPS data agree well with the HRTEM observation. However, the presence of XPS peaks characteristic of Ag-O bonds confirmed the the adsorption processes of AgNPs on TiO2 and α-Fe2O3 surfaces, through chemical oxygen bonds. This chemisorption between AgNPs and metals oxide ensures the formation of .he ternary heterojunction and avoiding metal desorption [12, 13]. The optical properties of Ag@TiO2/α-Fe2O3 nanocomposite, hybrid Ag@TiO2 and as well as bare TiO2 and α-Fe2O3 were investigated using UV-visible absorption spectroscopy (Fig. 4-a). Regarding the TiO2, Ag@TiO2 and Ag@TiO2/α-Fe2O3 spectra, each spectrum shows an absorption edge at 346 nm that corresponds to a band gab energy of 3.85 eV. Result agree well agreement with the band gap energy reported in the literature. On the other hand, for Ag@TiO2 and Ag@TiO2/α-Fe2O3, an additional absorption band located between 435 and 480 nm which can be attributed to the surface plasmon resonance of silver nanoparticles. On the other side, iron oxide is generally reported to absorb strongly in the ultraviolet (UV) region. However, regarding the α-Fe2O3 and Ag@TiO2/α-Fe2O3 spectra, effectively an absorption edge located at 245 nm can clearly detected and can be assigned to the electronic transition in the hematite 𝛼-Fe2O3 structure. The change of the absorption behaviour of Ag@TiO2/α-Fe2O3 compared to bare TiO2 and α-Fe2O3, implies that the charge-transfer transition between the three materials occurs while loading Ag to TiO2 and α-Fe2O3. This observation was further supported by PL measurements (Fig. 4-b). Indeed, as can be seen, the PL spectrum of all samples exhibited blue emission located at 445 nm. It has been reported that the visible luminescence, related to deep level emissions, mainly results from defects such as interstitials and oxygen vacancies. On the other hand, as can be seen in Fig. 4-b, a considerable PL emission quenching of Ag@TiO2/α-Fe2O3 nanocomposite was observed which indicated that a lower recombination rate of the photogenerated carrier could be efficiently achieved resulting from the synergistic effects between Ag, TiO2 and α-Fe2O3. This result implying that the intimate contact between Ag, TiO2 and α-Fe2O3 could make for the vectorial migrate of charge carriers among the nanocomposite, enhancing the photogenerated carrier’s separation and therefore improving the photocatalytic efficiency.
The photocatalytic efficiencies of Ag@TiO2/α-Fe2O3 nanocomposites were evaluated using MB dyes as a model pollutant. Photocatalytic activities of hybrid Ag@TiO2, bare TiO2 and α-Fe2O3 were also measured for comparison. As shown in Fig. 5-a, the photodegradation rate of MB was found to be the highest using the ternary Ag@TiO2/α-Fe2O3 photocatalyst. On the other side, the photodegradation rate using pristine α-Fe2O3 photocatalyst is the lowest. However, the photocatalytic performance of hematite α-Fe2O3 is limited due to the charge carrier recombination. On the other hand, the hybrid Ag@TiO2 photocatalyst exhibited interesting photodegradation rate due to the presence of plasmonic AgNPs which can generate more electrons and therefore boost the photocatalytic activity of TiO2. It can be seen that ternary Ag@TiO2/α-Fe2O3 photocatalyst exhibits much higher photodegradation activities than that of hybrid Ag@TiO2 may be due to the support given by α-Fe2O3 NPs which increases the surface are of the photocatalyst and also increases the light absorption which generates more electron-hole pairs for dye photodegradation and consequently enhances the photocatalytic activity of the ternary nanocomposite. To examine the reaction kinetics of photocatalysts, experimental data were fitted by a first-order kinetic equation (Ln(C0/C) = kapt) using the Langmuir–Hinshelwood model. It can be seen from the curves displayed in Fig. 5-b that the photodegradation process followed first order kinetics. A proposed possible photocatalytic mechanism is illustrated in Fig. 5-c. After excitation the VB electrons of both TiO2 and α-Fe2O3 were excited to the CB, creating holes in the VB. These photogenerated electrons-holes recombined rapidly, leading to a low photocatalytic performance of the photocatalyst. However, after loading the AgNPs, the photogenerated electrons could be continuously transported from TiO2 and α-Fe2O3 to AgNPs. Those vector transfers led to spatial separation of the photogenerated carries with an electron transferred to AgNPs while the holes remain trapped at the TiO2 and α-Fe2O3 surface. Subsequently, the adsorbed oxygen molecule (O2) can after that react with electrons produce therefore reactive oxygen species (such asO2.−, .O2−) that could oxidize and destruct MB dyes.