Photocatalytic and Photo-Fenton Catalytic Degradation Activities of Z-Scheme Ag2S/BiFeO3 Heterojunction Composites under Visible-Light Irradiation

Z-scheme Ag2S/BiFeO3 heterojunction composites were successfully prepared through a precipitation method. The morphology and microstructure characterization demonstrate that Ag2S nanoparticles (30–50 nm) are well-decorated on the surfaces of polyhedral BiFeO3 particles (500–800 nm) to form Ag2S/BiFeO3 heterojunctions. The photocatalytic and photo-Fenton catalytic activities of the as-derived Ag2S/BiFeO3 heterojunction composites were evaluated by the degradation of methyl orange (MO) under visible-light irradiation. The photocatalytic result indicates that the Ag2S/BiFeO3 composites exhibit much improved photocatalytic activities when compared with bare Ag2S and BiFeO3. The optimum composite sample was observed to be 15% Ag2S/BiFeO3 with an Ag2S mass fraction of 15%. Furthermore, the addition of H2O2 can further enhance the dye degradation efficiency, which is due to the synergistic effects of photo- and Fenton catalysis. The results of photoelectrochemical and photoluminescence measurements suggest a greater separation of the photoexcited electron/hole pairs in the Ag2S/BiFeO3 composites. According to the active species trapping experiments, the photocatalytic and photo-Fenton catalytic mechanisms of the Ag2S/BiFeO3 composites were proposed and discussed.


Introduction
Wastewater containing organic dyes or pigments has caused serious damage to environment and human health in recent times. Moreover, most of the organic pollutants are toxic, non-biodegradable, and are difficult to mineralize under natural conditions [1]. Several wastewater treatment routes have been used to eliminate organic pollutants from wastewater [2][3][4][5][6][7][8][9][10]. Compared with other elimination methods, advanced oxidation processes (AOPs) are demonstrated to be attractive methods for the efficient elimination of organic contaminants. In these processes, strong oxidizing radical species such as hydroxyl (•OH) radicals can be generated; they degrade most organic pollutants into innocuous byproducts [8,9]. Among various AOPs, photocatalytic and photo-Fenton-like catalytic processes have attracted remarkable attention for the decomposition of dyes, owing to their low cost, mild reaction deionized water, and the resulting Na 2 S solution was then added into the AgNO 3 solution (above). The mixture was subjected to vigorous magnetic stirring for 5 h. During this process, a black suspension was obtained. The black product was collected by centrifugation. After washing several times using deionized water and then drying at 60 • C for 6 h in a vacuum oven, the final Ag 2 S nanoparticles were obtained.

Preparation of BiFeO 3 Polyhedra
For the synthesis of BiFeO 3 , 5 mmol of Bi(NO 3 ) 3 ·5H 2 O and 5 mmol of Fe(NO 3 ) 3 ·9H 2 O were added into 20 mL of 25% (v/v) dilute nitric acid solution under magnetic stirring. Exactly 60 mL of 4.5 mol·L −1 KOH solution was dropped into the above solution. After magnetic stirring and ultrasonic treatment, a yellow suspension was obtained. Then, the suspension was transferred into a Teflon-lined stainless steel autoclave (capacity: 100 mL), and heated at 200 • C for 9 h. After that, the product was collected and washed with deionized water. After drying at 80 • C for 12 h, the final product of the BiFeO 3 polyhedra was obtained.

Fabrication of the Ag 2 S/BiFeO 3 Composites
For the fabrication of the Ag 2 S/BiFeO 3 composites, 0.1 g of BiFeO 3 particles and a certain amount of AgNO 3 were added into 30 mL of deionized water under vigorous stirring, to achieve a uniform mixture. Exactly 20 mL Na 2 S solution (possessing a certain concentration) was slowly added into the above suspension, which was then subjected to magnetic stirring for 5 h. In this process, the color of the mixture changed from brownish into black, indicative of the successful assembly of Ag 2 S nanoparticles on the surface of the BiFeO 3 polyhedral particles. After washing and drying, as in the procedure described above, the final Ag 2 S/BiFeO 3 composite was obtained. By adjusting the amounts of AgNO 3 and Na 2 S, a series of Ag 2 S/BiFeO 3 composites with different Ag 2 S mass fractions (5%, 10%, 15%, and 20%) in the composites were fabricated. The corresponding composites were named 5% Ag 2 S/BiFeO 3 , 10% Ag 2 S/BiFeO 3 , 15% Ag 2 S/BiFeO 3 , and 20% Ag 2 S/BiFeO 3 .

The Photo-Fenton Catalytic Activity Test
To investigate the photo-Fenton catalytic activities of the samples, MO was used as the model pollutant, and the visible light (300 W xenon lamp with a 420 nm cut-off filter) was used as the light source. A certain amount of the catalyst was dispersed into the MO solution (100 mL) in the dark under magnetic stirring for 30 min. During this process, the absorption/desorption equilibrium between the catalyst and the dye was achieved. After that, an amount of H 2 O 2 was added into the reaction solution, and then irradiated by visible light. During the reaction, 2.5 mL of the reaction solution was sampled at a given time interval, and centrifuged to remove the catalyst. The residual MO concentration was obtained by measuring the absorbance of the reaction solution at λ MO = 464 nm. The effects of catalyst dosage, H 2 O 2 concentration, and MO concentration on the degradation efficiency of MO were investigated, based on the above experimental process. The photocatalytic activity of the samples was evaluated under the same conditions without the addition of H 2 O 2 . Ethanol (10% (v/v)), ammonium oxalate (AO, 2 mmol·L −1 ) and benzoquinone (BQ, 1 mmol·L −1 ) were separately added into the reaction solution, with the aim of examining the active species in the catalytic process.

Photoelectrochemical Measurement
An electrochemical workstation (CHI 660C, Shanghai Chenhua Instrument Co. Ltd, Shanghai, China) with a three-electrode system was used for the photoelectrochemical measurements [63]. In the three-electrode system, a platinum foil acted as the counter-electrode, and a standard calomel electrode was used as the reference electrode. For the preparation of the working electrode, stoichiometric amounts of photocatalyst (15 mg), carbon-black (0.75 mg), and polyvinylidene fluoride (PVDF, 0.75 mg) were uniformly dispersed in 1 mL 1-methyl-2-pyrrolidione (NMP) to obtain a slurry mixture. The slurry mixture was coated onto a fluoride-doped tin oxide glass electrode (effective area: 1.0 × 1.0 cm 2 ), and dried at 60 • C for 5 h. During the photoelectrochemical measurements, a 300 W xenon lamp with a 420 nm cut-off filter was used as the light source, and 1 mol L −1 Na 2 SO 4 aqueous solution was used as the electrolyte. A 0.2 V bias potential was applied for the testing of photocurrent-time (I-t) curves. The electrochemical impedance spectroscopy (EIS) spectra were recorded by applying a sinusoidal voltage pulse (amplitude: 5 mV; frequency range: 10 −2 -10 5 Hz).

Hydroxyl Radical Measurement
The •OH radicals generated on the irradiated samples were measured by fluorimetry. To achieve this aim, terephthalic acid (TA) as an •OH scavenger [64]. The measurement procedure was as follows: TA was added into an NaOH solution with a concentration of 1.0 mmol·L −1 , to obtain a 0.25 mmol·L −1 TA solution. Exactly 60 mg of the catalyst was added into 100 mL of TA solution. The mixture was magnetically stirred for several minutes to ensure the uniform dispersion of the catalyst. Subsequently, H 2 O 2 was added into the reaction solution (2.0 mmol·L −1 ) and then irradiated by a 300 W xenon lamp with a 420 nm cut-off filter. At a given time interval, 2.5 mL of the reaction solution was sampled to measure its photoluminescence (PL) spectrum (the catalyst was removed by centrifugation). A fluorescence spectrophotometer was used to measure the PL spectrum of the reaction solution (excitation wavelength: 315 nm). The •OH radicals produced during the photocatalytic process were detected under the same conditions without the introduction of H 2 O 2 .

Characterization
X-ray powder diffraction (XRD) was employed to investigate the phase purity of the samples on a D8 Advance X-ray diffractometer (Bruker AXS, Karlsruhe, Germany). The morphologies and microstructures of the products were observed by field-emission scanning electron microscopy (SEM) and field-emission transmission electron microscopy (TEM). The SEM investigation was performed on a JSM-6701F scanning electron microscope (JEOL Ltd., Tokyo, Japan), and the TEM observation was carried out on a JEM-1200EX transmission electron microscope (JEOL Ltd., Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) was used to record the chemical states of the elements on a PHI-5702 multi-functional X-ray photoelectron spectrometer (Physical Electronics, Chanhassen, MN, USA). The ultraviolet-visible (UV-Vis) diffuse reflectance spectra of the samples were tested by using a TU-1901 double beam UV-Vis spectrophotometer (Beijing Purkinje General Instrument Co. Ltd., Beijing, China) with BaSO 4 as a reference. An RF-6000 fluorescence spectrophotometer (Shimadzu, Kyoto, Japan) was available to record the PL spectra of the samples (excitation wavelength:~350 nm). Figure 1 shows the XRD patterns of the BiFeO 3 , Ag 2 S and Ag 2 S/BiFeO 3 composites. For the bare BiFeO 3 and Ag 2 S samples, their diffraction peaks could be completely indexed in terms of the rhombohedral structure of BiFeO 3 (PDF card no.  and the monoclinic structure of Ag 2 S (PDF card no. 14-0072), respectively. When Ag 2 S is decorated on BiFeO 3 , the diffraction peaks of BiFeO 3 undergo no obvious change, suggesting that BiFeO 3 maintains a rhombohedral structure. Meanwhile, weak characteristic diffraction peaks of Ag 2 S are observed in the composites, indicating the decoration of Ag 2 S on BiFeO 3 . In addition, no other impurity phases are detected in the composites.  Figure 2a shows the absorption spectra of the BiFeO 3 and Ag 2 S/BiFeO 3 composites transformed from the UV-Vis spectra, using the Kubelka-Munk (K-M) theory. It is worth noting that, with the increase of Ag 2 S content, the composites exhibit a gradually increasing light absorption over the whole wavelength range, which is mainly due to the strong light absorption of Ag 2 S in the UV-Vis light region. The digital images of the samples inserted in Figure 2a show that the color of bare BiFeO 3 is brown, and the 15% Ag 2 S/BiFeO 3 composite manifests a black color. The deepening of the apparent color for the composite further confirms its enhanced visible-light absorption. To obtain the energy bandgap of the samples, the plots of (α hν) 2 vs hν using the Tauc relation are carried out, where α is defined as the K-M absorption coefficient, and hν is the incident photon energy. As depicted in Figure 2b, the bandgap (E g ) of bare BiFeO 3 is estimated to be 2.15 eV by extrapolating the linear portion of the plot to the hν axis. A similar E g of BiFeO 3 in the 20% Ag 2 S/BiFeO 3 composite is obtained, suggesting that the band structure of BiFeO 3 undergoes no detectable change after the decoration of Ag 2 S. In addition, based on its bandgap energy, the valence band (VB) and the conduction band (CB) potentials of BiFeO 3 can be obtained by using the following relations (Equations (1) and (2)): In the above relations, X is the absolute electronegativity of BiFeO 3 , and it is estimated to be 5.93 eV, based on the data reported in the literature [65]. E e is defined as the energy of free electrons on the hydrogen scale (~4.5 eV). The VB and CB potentials of BiFeO 3 were found to be +2.5 and +0.35 V vs normal hydrogen electrode (NHE), respectively.

XPS Analysis
To investigate the surface chemical states of 15% Ag 2 S/BiFeO 3 , XPS detection was carried out. Figure 3a-e show the high-resolution XPS spectra of Bi 4f, Fe 2p, O 1s, Ag 3d, and S 2p, respectively. From Figure 3a, the Bi 4f 5/2 and Bi 4f 7/2 binding energy peaks were found at 163.9 and 158.7 eV, respectively, indicating the existence of Bi 3+ [21]. Figure 3b depicts the Fe 2p spectrum, which displays two separate peaks at 724.2 (Fe 2p 1/2 ) and 710.1 eV (Fe 2p 3/2 ), respectively. The Fe 2p 3/2 binding energy can be divided into two peaks at 711.3 and 709.8 eV, corresponding to Fe 3+ and Fe 2+ , respectively [21,66,67]. The weak peak detected at 718.6 eV is characterized as the satellite peak for Fe 3+ . Figure 3c presents the O 1s spectrum. By fitting the O 1s binding energy, two peaks at 529.3 and 531.1 eV are found. The signal at 529.3 eV is assigned to the lattice oxygen, while the peak at 531.1 eV is mainly attributed to the surface vacancies and the chemisorbed oxygen [21,62]. On the Ag 3d spectrum (Figure 3d), the peak at 368.4 eV is assigned to Ag 3d 5/2 , and the peak at 374.3 eV is attributed to Ag 3d 3/2 of Ag + [68]. In the spectrum of S 2p (Figure 3e), the signal of S 2p can be deconvoluted into two peaks at 162.7 and 161.5 eV, which are caused by the S 2p 1/2 and S 2p 3/2 , respectively [68].

Morphology Observations
Figure 4a-d shows the SEM images of Ag 2 S, BiFeO 3 , and 15% Ag 2 S/BiFeO 3 . It is seen that Ag 2 S exhibits spherical-like shapes with diameters of 30-50 nm. The BiFeO 3 sample is mainly composed of polyhedral-like particles with edge lengths of 500-800 nm, and the particles have smooth surfaces. From the SEM image of the 15% Ag 2 S/BiFeO 3 composite, small Ag 2 S nanoparticles are clearly observed to be attached onto the surface of the BiFeO 3 polyhedral-like particles. Figure 4e presents the TEM image of the 15% Ag 2 S/BiFeO 3 composite, further confirming the decoration of the Ag 2 S nanoparticles on the polyhedral BiFeO 3 particles. The high-resolution TEM (HRTEM) image obtained from the interface of the two phases in the 15% Ag 2 S/BiFeO 3 composite is shown in Figure 4f, revealing two distinct sets of lattice fringes. The interplanar spacing of~0.39 nm corresponds to the (012) plane of BiFeO 3 , and the interplanar spacing of~0.27 nm corresponds to the (122) plane of Ag 2 S. This indicates the intimate contact between Ag 2 S and BiFeO 3 to form a Ag 2 S/BiFeO 3 heterojunction structure. To further investigate the distribution of Ag 2 S on the surface of BiFeO 3 , the elemental mapping observation of 15% Ag 2 S/BiFeO 3 was performed, as shown in Figure 5.

Photo-Fenton Catalytic and Photocatalytic Performances
MO was used as the model pollutant to evaluate the visible-light-driven photocatalytic activities of the samples, as shown in Figure 6a. The direct photolysis of the dye without the photocatalyst, and the dye adsorption on the photocatalyst were also investigated, which reveals that the dye MO has a good stability after a short period of visible-light irradiation; the absorption of the dye on the photocatalyst is very small. After 4 h of photocatalysis, the introduction of BiFeO 3 and Ag 2 S leads to MO degradation percentages of~33% and~30%, respectively. This implies a relatively weak photocatalytic activity from bare BiFeO 3 and Ag 2 S. When Ag 2 S is decorated on BiFeO 3 , the formed Ag 2 S/BiFeO 3 composites exhibit obviously improved photocatalytic activities. With an increase in the Ag 2 S content, an optimal composite sample is observed for 15% Ag 2 S/BiFeO 3 . Further increasing the Ag 2 S content cannot result in an enhanced photocatalytic activity for the composites. This suggests that the construction of the Ag 2 S/BiFeO 3 heterojunction is critical for the effective enhancement of photocatalytic activity. However, the excessive Ag 2 S loading may result in full coverage by Ag 2 S on the BiFeO 3 surface, thus decreasing the photon absorption of BiFeO 3 . To further compare the photocatalytic activities of the samples, kinetic analysis of the photocatalytic degradation of MO was carried out. As illustrated in Figure 6c, the plots of Ln(C t /C 0 ) vs reaction time t presented a good linear behavior, and they can be modeled well by using the first-order kinetic equation Ln(C t /C 0 ) = k app t, where k app is the apparent first-order reaction rate constant. In every case, the standard deviation (SD) was smaller than 0.07. The obtained values of k app and SD are inserted in Figure 6c. From the reaction rate constants, it is concluded that the optimal composite sample, 15% Ag 2 S/BiFeO 3 , has a photocatalytic activity ca. 4.0 and 4.1 times higher than those of bare BiFeO 3 and Ag 2 S, respectively. To further examine the photo-Fenton catalytic behavior of the samples, the degradation of MO in the photo-Fenton processes was performed, as shown in Figure 6b. In the absence of the photocatalyst, the degradation of MO is observed to be very small under visible light irradiation and in the presence of H 2 O 2 , indicating minor self-degradation for MO. With the addition of catalysts and H 2 O 2 under visible light irradiation, dye degradation is significantly enhanced, to be much higher than that in the bare photocatalytic process. The photo-Fenton catalytic activities of the samples are in the same orders as their photocatalytic activities. The kinetic plots for the photo-Fenton catalytic degradation of MO are shown in Figure 6d. The derived reaction rate constants reveal that the 15% Ag 2 S/BiFeO 3 composite exhibits a photo-Fenton catalytic activity ca. 3.5 times higher than that of bare BiFeO 3 . Figure 7 shows the degradation percentage of MO over the 15% Ag 2 S/BiFeO 3 composite during three successive photocatalytic or photo-Fenton catalytic processes. It is seen that the catalytic activity of the composite does not undergo a significant decrease. After three successive recycles, the photocatalytic and photo-Fenton catalytic degradation percentages of MO still reach~70% and 84%, respectively. Choosing 15% Ag 2 S/BiFeO 3 as the photocatalyst, further photo-Fenton degradation experiments were carried out, aimed at investigating the effects of photocatalyst dosage, H 2 O 2 content, and dye concentration on dye degradation. Figure 8a shows the effect of catalyst dosage on the degradation of MO. The corresponding kinetic plots are presented in Figure 8b. The degradation percentage of MO increases gradually with an increase in the catalyst loading from 0.1 to 0.6 g·L −1 . However, when the catalyst loading exceeds the optical value (0.6 g·L −1 ), the catalytic efficiency starts to decrease. Generally, increasing the catalyst loading can provide more active sites for H 2 O 2 activation to produce more •OH radicals, thus accelerating the degradation of the dye. Nevertheless, excessive addition of the catalyst will enhance the light screening effect, and reduce light absorption, thus resulting in a decrease in the catalytic efficiency.  The effect of H 2 O 2 concentration on the degradation of MO is shown in Figure 8c. As observed, the degradation percentage of MO increases from~81% to~97%, with an increase in the H 2 O 2 concentration from 0.5 to 2 mmol·L −1 , which is due to the fact that more •OH radicals can be produced by increasing the H 2 O 2 concentration. However, when the concentration of H 2 O 2 reaches 2.5 mmol·L −1 , the MO degradation percentage exhibits a decreasing trend. This phenomenon is mainly attributed to the consumption of •OH radicals by the reaction of additional H 2 O 2 with •OH [15]. The corresponding kinetic plots and derived reaction rate constants (Figure 8d) further confirm the effect of H 2 O 2 concentration on the degradation of MO.
The effect of MO concentration on its degradation is shown in Figure 8e. The corresponding kinetic plots are presented in Figure 8f. It is seen that the degradation percentage of MO decreases with increasing the initial MO concentration, which is due to the decreased light transparency. In spite of this, the 15% Ag 2 S/BiFeO 3 composite still photo-Fenton-catalyzes a 50% rate of removal of the dye at a high MO concentration of 40 mg·L −1 . This implies that the present Ag 2 S/BiFeO 3 composite photocatalyst could offer a practical application in the treatment of high-concentration dye wastewater. From an application viewpoint, the stability of the catalyst is another important factor to be considered besides its catalytic activity.

Photogenerated Charge Behavior
It is known that photocatalytic and photo-Fenton catalytic activities are related to the performance of the photogenerated charges. Photoelectrochemical and PL measurements were performed to investigate the photoinduced charge behaviors of the samples [70,71]. Figure 9a shows the photocurrent response profiles of BiFeO 3 and 15% Ag 2 S/BiFeO 3 . Compared with bare BiFeO 3 , the 15% Ag 2 S/BiFeO 3 composite exhibits an enhanced photocurrent density, suggesting that the decoration of Ag 2 S nanoparticles results in an enhanced separation of photogenerated e − -h + pairs. The EIS spectra of BiFeO 3 and 15% Ag 2 S/BiFeO 3 are presented in Figure 9b. In the EIS spectra, the arc radius for 15% Ag 2 S/BiFeO 3 is smaller than that for bare BiFeO 3 , which indicates the low interfacial resistance of the composite. Figure 9c displays the PL spectra of BiFeO 3 and 15% Ag 2 S/BiFeO 3 . A steady emission peak is detected at~520 nm, which is mainly due to the recombination of the photogenerated charges. The 15% Ag 2 S/BiFeO 3 composite possesses a relatively weak PL emission peak in comparison to bare BiFeO 3 , indicating that the recombination of photoinduced charges is suppressed in the composite. The above results reveal that the photogenerated electrons and holes of BiFeO 3 are successfully separated and migrated by the decorations of the Ag 2 S nanoparticles.

Active Species Detection
To unveil the active species that are responsible for the photocatalytic and photo-Fenton catalytic degradation of MO over the Ag 2 S/BiFeO 3 composites, the active species trapping experiments were performed. To achieve this aim, ethanol was used as the •OH scavenger, AO as the h + scavenger, and BQ as the •O 2 − scavenger [72], and the results are shown in Figure 10. During the photocatalytic process, the introduction of AO leads to an obvious reduction of dye degradation, and a similar suppression effect is detected after the addition of BQ. When ethanol is added, the degradation efficiency of the dye is slightly decreased.   Figure 11a,b shows the time-dependent PL spectra of the TPA solution over the 15% Ag 2 S/BiFeO 3 composite in the photocatalytic and the photo-Fenton catalytic processes, respectively. During the photocatalytic and photo-Fenton catalytic processes, an obvious PL signal is detected at around 429 nm, and its intensity exhibits an increasing trend with the irradiation time, suggesting that the production of the •OH radicals during the photocatalytic and photo-Fenton catalytic processes. However, the PL signal becomes more intense in the photo-Fenton process, which indicates the generation of more •OH radicals. This result is consistent with the active species-trapping experiments.

Photocatalytic and Photo-Fenton Mechanisms
The photocatalytic mechanism of the Ag 2 S/BiFeO 3 composites are schematically shown in Figure 12a. Based on the previous report, the CB/VB edge potentials of Ag 2 S are estimated to be −0.3/+0.7 V vs NHE [73]. The CB/VB edge potentials of BiFeO 3 are obtained as +0.35/+2.5 V vs NHE, respectively. This suggests that BiFeO 3 and Ag 2 S can be coupled together to form a promising Z-scheme composite photocatalysts, due to their well-matched overlapping band-structures. Under illumination from visible light, both BiFeO 3 and Ag 2 S are excited to generate electrons in the CB, and holes in the VB. For bare BiFeO 3 and Ag 2 S, most of the photogenerated charges tend to be recombined, and only a few of them take part in the photocatalytic reaction. In contrast, in the Ag 2 S/BiFeO 3 heterojunction composites, the photoexcited electrons in the CB of BiFeO 3 will migrate to Ag 2 S and combine with the photoexcited holes in the VB of Ag 2 S. This charge migration and combination process remarkably promote the separation of the photogenerated electrons, and holes in BiFeO 3 and Ag 2 S. As a result, more photoinduced holes in BiFeO 3 and photoinduced electrons in Ag 2 S are able to take part in the photocatalytic reaction. This is the dominant mechanism, resulting in enhanced photocatalytic activity for the Ag 2 S/BiFeO 3 composites. Compared to the redox potential of O 2 /•O 2 − (−0.13 V vs NHE) [21], the sufficiently negative CB potential of Ag 2 S indicates that the photoexcited electrons in Ag 2 S can reduce O 2 to generate •O 2 − . On the other hand, the photogenerated holes in the VB of BiFeO 3 possess enough oxidation ability to generate •OH (E 0 (OH − /•OH) = +1.99 V vs NHE) [74]. Under visible-light irradiation, and in the presence of H 2 O 2 , photocatalytic and Fenton reactions simultaneously take place, and moreover, the two catalytic processes exhibit an important synergistic effect, as schematically shown in Figure 12b. In the photo-Fenton catalytic process, Fe 2+ on the surface of BiFeO 3 can react with H 2 O 2 to generate •OH and Fe 3+ . Then, Fe 3+ will be reduced into Fe 2+ by the photogenerated electrons. In this cyclic reaction, more •OH radicals are being produced, but the consumption of the photogenerated electrons leads to the decreased generation of •O 2 − radicals.
This could be the reason that •OH plays an important role in the degradation of the dye during the photo-Fenton catalytic process. The main photo-Fenton reaction process can be briefly described by Equations (3) Fe 3+ + e − (Ag 2 S) → Fe 2+ (8) •OH, h + , •O 2 − + MO → degradation products (9)

Conclusions
A series of Z-scheme Ag 2 S/BiFeO 3 heterojunction composites have been constructed by assembling Ag 2 S nanoparticles on the surface of the BiFeO 3 polyhedra. It is demonstrated that the Ag 2 S/BiFeO 3 composites can be employed as a promising catalyst for the photocatalytic and photo-Fenton catalytic decomposition of MO under visible-light irradiation. Compared to the bare photocatalytic process, the synergistic effect of photo-Fenton catalysis in the presence of H 2 O 2 leads to a higher degradation efficiency of the dye. The optimum composite sample is observed for 15% Ag 2 S/BiFeO 3 , which photo-Fenton-catalyzes~97% degradation of MO after 4 h of reaction under the optimal conditions. The enhanced catalytic activity of the Z-scheme Ag 2 S/BiFeO 3 heterojunction composites can be explained as the result of efficient separation of photoexcited e − -h + pairs, resulting from the Z-scheme electron transfer.