Fabrication and characterization of high efficiency and stable Ag/AgFeO2/Ag3PO4 ternary heterostructures nanocatalyst

An Ag/AgFeO2/Ag3PO4 ternary was synthesized by hydrothermal method with polyvinylpyrrolidone (PVP). The composite materials were characterized by XRD, SEM, TEM, DRS and XPS. XRD, SEM and TEM results are used to characterize the structure and morphology of Ag/AgFeO2/Ag3PO4 samples, DRS results are mainly used to characterize the light absorption capacity of Ag/AgFeO2/Ag3PO4 samples. Photocatalytic results showed that the photocatalytic performance of Ag/AgFeO2/Ag3PO4 photocatalyst was significantly improved, which was 10.6 times than that of pure AgFeO2. The optimal photocatalyst can degrade MO (Methyl Orange) up to 98% in 1 h. Simultaneously, the cyclic experiments showed that it had good stability, from 80% for Ag/AgFeO2/AgPO4 to 58% for pure Ag3PO4 after five cycles. To obtain further insight into the high photooxidative activity of AgFeO2, ab initio density functional theory (DFT) calculations have also been carried out. The mechanism study shows that the synergistic effect of heterojunction and strong SPR of silver nanorods make the catalyst have higher photocatalytic performance and better stability.


Introduction
Semiconductor photocatalysis has received extensive attention and rapid development, which provides an ideal method for energy utilization and environmental pollution control. In view of the efficient utilization of solar energy, a large number of active photocatalysts for visible light have been studied in recent years [1,2]. For example, Ag 2 S-Au, AgBr@rGO, Ag 3 PO 4 @PANI [3][4][5] and other photocatalysts show good photocatalytic effects. [6] Silver orthophosphate (Ag 3 PO 4 ) is an excellent basis for visible light-driven photocatalyst. Previous studies have shown that Ag 3 PO 4 has an indirect bandgap of 2.45 eV [7,8], which can absorb visible light and generate photoexcited holes. Ag 3 PO 4 has good oxidation activity due to high positive potential in the valence band (VB), and the electron mobility is significantly higher than that of holes. [9] 552 Page 2 of 12 However, there are still some problems that limit its development. One is photo corrosion of photoexcited electrons under light irradiation, and the other is greater solubility in water [10]. The above factors seriously reduced the stability of Ag 3 PO 4 . So far, coupling of Ag 3 PO 4 with semiconductors such as Fe 2 O 3 [11], AgI [12], ZnO [13], NiFe 2 O 4 [14], CeO 2 [15], BiOI [16], graphene [17], TiO 2 [18], AgX (X = Cl, Br, I) [19] and g-C 3 N 4 [20] can alleviate these adverse effects and accelerate photoinduced charge separation [21]. To sum up, the Ag/AgFeO 2 /Ag 3 PO 4 composite nano-heterojunction photocatalyst was constructed by adjusting the experimental conditions, which enabled it to have excellent photocatalytic REDOX ability, non-toxicity and low cost.
As a new type of photocatalyst, compound oxide AgFeO 2 has a relatively narrow bandgap of 1.31 eV [22,23], which can expand the range of visible light absorption and has good photocatalytic performance in theory. However, the narrow bandgap leads to a high electron-hole recombination rate, which limits the development of AgFeO 2 as a photocatalyst. To improve the photocatalytic performance of AgFeO 2 , it is attempted to couple AgFeO 2 with other semiconductors, such as AgFeO 2 /g-C 3 N 4 [24], AgCl/Ag/AgFeO 2 [25] and Ag/AgFeO 2 [26], all of which show good photocatalytic performance. An Ag/AgFeO 2 /Ag 3 PO 4 composite semiconductor nanomaterial was prepared by hydrothermal synthesis. In Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructure photocatalyst, elemental silver protects Ag 3 PO 4 from photocorrosion in photocatalytic reaction, thus improving the stability of Ag 3 PO 4 photocatalyst. In addition, the heterostructure of Ag 3 PO 4 /AgFeO 2 can promote the transport of photogenerated carriers and effectively separate electron-hole. The mechanism of photocorrosion inhibition and photocatalytic activity enhancement was discussed.

Materials
All reagents utilized in the work are of analytical purity. Reagents polyvinylpyrrolidone (PVP, Mw = 58000), silver nitrate (AgNO 3

Synthesis of Ag/AgFeO 2 composite
In a typical synthesis process, Ag/AgFeO 2 was prepared by the flowing steps. First, 8 g PVP (Mw = 58000) was solved in 50 ml distilled water. Then, 0.85 g AgNO 3 and 2.02 g Fe(NO 3 ) 3 ·9H 2 O were first added to the solution followed by fierce stirring for 30 min. Sodium hydroxide (NaOH) solution (1.5 g, 10 ml) was added drop by drop to the above solution to form an alkaline system, and the resulting mixture was then stirred for another 20 min. The mixed solution was transferred to an autoclave for hydrothermal reaction at 100 °C for 12 h. The precipitates were collected by centrifuging at a speed of 5000 rpm and washed with anhydrous ethanol and distilled water repeatedly 3 times. Finally, it was dried in an oven at 60 °C for 12 h. The acquired Ag/AgFeO 2 samples were sealed and stored before use.

Preparation of different molar ratio of Ag/AgFeO 2 / Ag 3 PO 4 sample
The Ag 3 PO 4 nanoparticles were loaded onto the crystallized Ag/AgFeO 2 samples by a deposition-precipitation method. Preparation of Ag/AgFeO 2 /Ag 3 PO 4 (mole ratio is 2:1 of Ag 3 PO 4 to Ag/AgFeO 2 ) sample: AgNO 3 (1.02 g, 6 mmol) was dissolved in 25 ml ultrapure water. Then ammonia aqueous solution (1.0 M) was dropwise added to the above solution to form a transparent solution. Afterward, 1 mmol of as-prepared Ag/AgFeO 2 (0.196 g) was added and stirred drastically for 30 min. 0.312 g of NaH 2 PO 4 dissolved in 10 ml deionized water and on the electromagnetic stirrer forms NaH 2 PO 4 solution, after that the solution was added dropwise in the preceding suspension within 20 min of stirring. The product was collected from the solution by centrifugation and washed with anhydrous ethanol and distilled water repeated for 3 times. Finally, it was dried in an oven at 60 °C for 12 h. Different Ag/AgFeO 2 /Ag 3 PO 4 samples were synthesized through the same method as the preparation of Ag/AgFeO 2 /Ag 3 PO 4 (mole ratio is 2:1) sample, just adjusted the mole ratio of AgNO 3 and NaH 2 PO 4 . Corresponding names are AFP-0.5 (mole ratio is 0.5:1), AFP-1 (mole ratio is 1:1), AFP-2 (mole ratio is 2:1), AFP-3 (mole ratio is 3:1) and AFP-4 (mole ratio is 4:1), respectively, AFP means Ag/ AgFeO 2 /Ag 3 PO 4 ternary compounds. The synthesis process of pure Ag 3 PO 4 was similar without adding Ag/AgFeO 2 .

Photocatalytic experiments
The photocatalytic activity of AFP was investigated using an aqueous solution of MO (10 mg/L) at 298 K via a thermostatic bath. Experimental process was as follows: 100 ml of MO aqueous solution and 0.05 g of as-prepared AFP samples were placed in a glass vessel to form a suspension under stirring. After that suspensions were magnetically stirred in dark for 20 min to establish adsorption/desorption equilibrium between the photocatalyst and dye aqueous solution.
Then, a 300 W xenon lamp was used as a light source to provide simulated solar light irradiation. During the illumination, 3 ml of suspension was drawn from the reaction vessel at regular time intervals and centrifuged at 5000 rpm for 10 min to get the supernatant. The degradation of organic dyes was analyzed by measuring the absorbance of samples at 466 nm (MO) with UV-Vis absorption spectra.

Electronic property calculation of AgFeO 2
The spatial structure of AgFeO 2 is edge-connected O-Ag + -O octahedra and edge-shared Fe 3+ O 6 octahedra ( Fig. 1a). So far, AgFeO 2 crystals have two varieties of shapes and morphologies such as 3R structure and 2H structure. Based on the GGA and GGA+U calculations by Ong and co-workers, it was demonstrated they have the same bandgap for 3R and 2H, which was calculated to be 1.15 eV and 1.7 eV, respectively [22,23]. Furthermore, Shuxin Ouyang et al. described the crystal structure and found that (AgMO 2 (M) Al, Ga, In) has a narrower bandgap at 3R than at 2H, which favors the Photoproduction hole migration and thus a stronger oxidation capacity. Therefore, we mainly study the structures of 3R-type AgFeO 2 . [9] As the key photocatalyst, the electronic structures of 3R-type oxide AgFeO 2 were examined by using the ab initio density functional theory (DFT) calculations. The corresponding results are as follows: band structures are shown in Fig. 1(b), and it had a direct bandgap of about 1.31 eV, which agreed well with the previously calculated result [22,23]. It's easy noting that valence bands and conduction bands are so close that could cause harm to the transport of photoexcited electrons and holes. This in turn may result in a high electron-hole recombination rate, thus account for the low photocatalytic activity [7]. Furthermore, the density of state (DOS) and the project density of state (PDOS) of AgFeO 2 are depicted in Fig. 1c−f, and there are some points that should be explained. One is the existence of strong hybridization at energy from − 7 to -2 eV, and another one is the top of the VB next to Fermi level are dominating contributed from O 2p orbital. Therefore, an improved photocatalyst will be based on AgFeO 2 , supplemented with Ag, Ag 3 PO 4 , to form a ternary heterojunction, solving the problem of the high recombination rate of AgFeO 2 , and achieve the best catalytic effect.

XRD analyses
X-ray diffraction (XRD) was explored to determine the composition and crystallographic structure of obtained samples, and the results are shown in Fig. 2. And they could be illustrated in two main steps. First, Fig. 2a shows the XRD patterns of Ag/AgFeO 2 photocatalysts with different masses of PVP. There are two characteristic peaks that are marked and belong to the Ag and AgFeO 2 structures, respectively. Moreover, it could be observed the characteristic diffraction peaks of Ag structures gradually strong up to 8 g when the masses of PVP increase, indicating the decoration of more Ag nanorods onto AgFeO 2 surfaces. And second, in the case of AFP samples, there are three groups of standard diffraction peaks that can be clearly assigned to corresponding samples, as presented in Fig. 2b [27][28][29] Besides, for the pattern of the AFP-2 sample, no other peaks of any impurities are observed, indicating that Ag 3 PO 4 have been deposited on Ag/AgFeO 2 composite with success.

Morphology analysis
Further characterization of the morphologies and microstructures of obtained samples was studied by scanning electron microscopy (SEM) and high-resolution TEM (HR-TEM). As shown in Fig. 3, the uniform dispersion and structures of Ag/AgFeO 2 were observed in the larger scale image of Fig. 3a, and Ag nanorods were formed over the entire surface of the AgFeO 2 cakes. The high-magnification image Fig. 3b showed clearly that Ag nanorods were uniform in size with the estimated average diameter ranging from 68 to 125 nm, which may favor the electron transfer in ternary heterostructures system. And Fig. 3c shows the SEM image of the pure Ag 3 PO 4 , they were found to consist of agglomerated grains composed of irregular particles (2-5 μm). As for Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructures (Fig. 3d), by adjusting the molar ratio of Ag 3 PO 4 , Ag 3 PO 4 particles grown to the surface of Ag/AgFeO 2 composites were obtained, which indicates an intimate contact between Ag 3 PO 4 and Ag/AgFeO 2 . Because Ag 3 PO 4 and AgFeO 2 are too similar to be distinguished, the AFP-2 composite was further investigated by HRTEM to ascertain the decoration of Ag 3 PO 4 on the surface of Ag/AgFeO 2 . As shown in its TEM image ( Fig. 4), these Ag 3 PO 4 particles were in intimate contact with the AgFeO 2 , and the lattice fringes of AgFeO 2 could be clearly identified, which have a spacing of 0.306 nm (012 planes, JCPDS file No. 75-2147).

XPS analysis
The elemental composition and chemical states of AFP-2 composite were investigated by the XPS technique, to reveal more detailed information about the Ag/AgFeO 2 /Ag 3 PO 4  Fig. 5. First, the full XPS spectrum (Fig. 5a) shows that the AFP-2 sample consists of major elements Fe, Ag, P and O. Then in Fig. 5b, the P 2p peak is located at 133.4 eV. The bands at 710.3 and 723.9 eV in Fig. 5c were assigned to Fe 2p 3/2 and Fe 2p 1/2 peaks. And for O 1s XPS spectrum (Fig. 5d), the peak at 529.3 eV could be attributed to the lattice oxygen in Ag 3 PO 4 and AgFeO 2 . Finally, and most importantly, it should be noticed the analysis results of the XPS spectrum of Ag 3d. In Fig. 5e, the stronger peaks at 368.8 and 374.8 eV could be confirmed as Ag 0 on the surface of AgFeO 2 , in consistent with the SEM and HRTEM result. And the rest peaks at 367.6 and 373.5 eV corresponding to Ag 3d 5/2 and Ag 3d 3/2 are Ag + (Fig. 5f). [30,31]

UV-vis DRS
UV-vis diffuse reflectance spectra (DRS) were used to explore the optical absorption abilities of obtained samples, and the results were displayed in Fig. 6. From Fig. 6a, as compared to pure Ag 3 PO 4 , the Ag/AgFeO 2 and AFP-2 photocatalysts both showed better performance in the visible light region of 400 to 800 nm, and the enhanced visible light absorption capability of AFP-2 suggests that it might have higher photocatalytic activity for degradation. [32,33] On the other hand, the bandgap energy (Eg) of these two samples was calculated based on the DRS results to evaluate the optical absorption performance. As can be seen in Fig. 6b, the bandgap energy of Ag/AgFeO 2 and pure Ag 3 PO 4 is about 1.14 and 2.61 eV, respectively. The calculation of semiconductor bandgap width is mainly based on the formula proposed by Tauc, Davis, Mott et al., commonly known as Tauc Plot. The calculation formula is as follows: (ahv) 1/ n =A(hv-E g ). [34][35][36] Comparing with the theoretical calculation result above (1.35 eV), the Eg of Ag/AgFeO 2 is slightly lower which could be assigned to the SPR effect of Ag nanorods formed on the AgFeO 2 surface. When Ag/ AgFeO 2 and Ag 3 PO 4 are doped, Ag 3 PO 4 increases the density of oxygen vacancy on the Ag/AgFeO 2 surface could act as an electron trap that could hinder the electron-hole recombination and in turn increase the photo-catalytic degradation, the lower bandgap values also indicated it may have the better photocatalytic capability. [37,38] Thus, the photocatalytic performance potentiality of the obtained samples was further demonstrated by the results above.

Photocatalytic activity of the photocatalyst
The photocatalytic behaviors of obtained catalysts for the MO degradation under visible light are shown in Fig. 7. As can be seen in Fig. 7a, compared with the photocatalytic activities of Ag/AgFeO 2 samples (degradation efficiency of MO was only approximately 25%), approach to 98% of MO was degraded with 60 min irradiation in the presence of AFP-2 photocatalyst, which proved its superior photocatalytic activity. Moreover, the proper molar ratio of Ag 3 PO 4 to Ag/AgFeO 2 can greatly improve the photocatalytic activity, indicating the heterojunction was formed well, which was also demonstrated above. Meanwhile, Fig. 7b illuminated the MO structure (a model pollutant with a major absorption band at 466 nm) was successfully destructed by the AFP-2 photocatalyst. On the other hand (As shown in Fig. 7c), the kinetics of these photocatalytic reactions can be described using a pseudo-first-order kinetic model ln (C 0 /C) = kt (where C 0 is the premier MO solution and C is the MO concentration at time t, the slope k is the first-order kinetic rate constant) for MO solutions. In comparison, the apparent rate constants for pure Ag/AgFeO 2 and AFP-2 photocatalyst are 0.0053 min -1 and 0.0521 min -1 , respectively (see Fig. 7d), it is 10.6 times higher than that by pure Ag/AgFeO 2 (Table 1).

Photocatalytic stabilities of photocatalyst
Besides the improvement of photocatalytic activities, the enhancement stability for Ag 3 PO 4 is also one of the key points of the article, via the fabrication of Ag/AgFeO 2 / Ag 3 PO 4 ternary heterostructures system. As is known, the stability of a photocatalyst is very important for a photocatalyst to be useful. To study the stability and reusability of Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructures system, used AFP-2 composite powder was collected and reused in five successive MO degradation experiments, this result was showed in Fig. 8. Although some photocatalytic activity losses are observed for the AFP-2 photocatalyst after five times reutilization (but still retained over 80% of its original activity), the rate of MO degradation over pure Ag 3 PO 4 decreased more significantly in the similar conditions (decreasing from 98 to 58%). These results demonstrate Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructures system is very stable.

Effects of reactive species
Radical-trapping experiments with different scavengers were employed to clarify the main active species worked during the photocatalytic degradation of MO and thus determine the reaction mechanism. Figure 9 clearly shows that HO, O 2 and h + were detected using isopropanol (IPA), benzoquinone (BQ), and ammonium oxalate (AO) as trapping agents, respectively, which all were 1 mmol/L. As depicted in Fig. 8, h + was not the active species in the reaction, since the addition of AO did not significantly inhibit the photocatalytic reaction. In contrast, the addition of IPA and BQ showed a significant effect on the photocatalytic degradation (decreased obviously from 90 to 20%), which suggested the HO·and O 2 radicals can play a significant role in the photodegradation under visible light irradiation.
On the basis of the above analysis, a possible mechanism for the degradation of MO of Ag/AgFeO 2 /Ag 3 PO 4 composite under visible irradiation is illustrated in Fig. 10. As is known, both Ag 3 PO 4 and AgFeO 2 can be excited to generate electrons and holes when irradiated by visible light [26,43]. On the one hand, all holes generated in the VB of Ag 3 PO 4 could easily transfer to AgFeO 2 for the valence band of AgFeO 2 has a higher value than that of Ag 3 PO 4 , then the holes in the VB of AgFeO 2 could oxidize H 2 O to form HO· radicals, which are then involved in the photodegradation reaction of MO [44]. They determine the efficiency of the whole photocatalytic reaction. On the other hand, as shown in Fig. 9, the direct coupling of Ag nanowires and AgFeO 2 c6auses the Fermi level to equilibrate which could attain an energy level close to the conduction band of the AgFeO 2 [45,46], and the conduction band of Ag 3 PO 4 is very close but lower than that of AgFeO 2 , so the delicately matched band structures of Ag 3 PO 4 , AgFeO 2 and Ag, are very favorable for the excited electrons to transfer from the conduction bands of Ag 3 PO 4 and AgFeO 2 to the newly rebuilt surface energy band of Ag. Further, the excited electrons (Ag) could be trapped by molecular oxygen in solution to form ·O 2 and other oxidative species [47]. Simultaneously, the effective migration of interfacial electrons could prevent the photocorrosion of Ag 3 PO 4 (the photo-corrosion of Ag 3 PO 4 in the absence of electron acceptors), which guarantees stability of the whole photocatalytic system. Hence, the photocatalytic activity of AgFeO 2 and the photocatalytic stability of Ag 3 PO 4 were greatly improved by the SPR effect of Ag in two ways, more amounts of electrons and holes generation as well as superior charge carrier separation rate [48][49][50].

Conclusions
In this work, novel Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructures were successfully prepared by simple hydrothermal and solution co-precipitation method for the first time. As a novel ternary heterostructured Ag/AgFeO 2 /Ag 3 PO 4 photocatalyst, it showed excellent photocatalytic activity and stability for organic pollutants degradation, the degradation of MO over AFP-2 composite photocatalystreached 98% within 1 h and remained 80% after five cycling runs. On the basis of the above analysis, the enhanced photocatalytic activity and stability could be attributed to a synergistic effect, which includes the effective separation of change carriers under the SPR effect, the energy band structure match of the selected components (Ag, AgFeO 2 and Ag 3 PO 4 ) and high structural stability of AgFeO 2 . More attractively, this kind of Ag/AgFeO 2 /Ag 3 PO 4 ternary heterostructures photocatalysts could provide new insight for the design and fabrication of narrow bandgap photocatalysts like AgFeO 2 .   Data availability Derived data supporting the findings of this study are available from the corresponding author on request.

Conflict of interest
The authors declare that they have no conflict of interest.
Open access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third-party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.