Simple Fabrication of Visible Light-Responsive Bi-BiOBr/BiPO 4 Heterostructure with Enhanced Photocatalytic Activity

A Bi-BiOBr/BiPO 4 heterojunction structure was successfully synthesized via a two-step solvothermal method with ethylene glycol as a reducer. Little BiPO 4 irregular polyhedrons and little metal Bi spherical nanoparticles were uniformly dispersed on the surface of BiOBr nanosheets with intimate contact and formed a heterojunction structure between BiPO 4 and BiOBr. It was found that Bi-BiOBr/BiPO 4 had a signi ﬁ cant improvement in photocatalytic performance for RhB degradation compared to bare BiOBr and BiPO 4 . The photocatalytic degradation rate constant of 0.2-Bi/BiOBr/BiPO 4 was 1.44h -1 , which was 3.8 times and 14.2 times more than that of bare BiOBr and BiPO 4 , respectively. This is attributed to the formation of a ternary heterojunction, which bene ﬁ ts the separation of photogenerated electron-hole pairs. Furthermore, with the introduction of metal Bi, the SPR e ﬀ ect of metal Bi can e ﬀ ectively improve the absorption ability of Bi-BiOBr/BiPO 4 photocatalyst, resulting in enhanced photoactivity. In this work, the mechanism of photocatalytic degradation was studied by using the photochemical technique and the capture experiment of active species, and it was revealed that h + and ⋅ O 2- played a major role in the photocatalytic process.


Introduction
With the development of industry, such as textile, plastic, and paper, the varieties and quantities of dyes are rapidly increasing. At present, the discharge of toxic and harmful organic solvents from various industries leads to environmental pollution problems. Recently, it has been reported that more than 80,000 tons of reactive dyes are produced and consumed annually, and more than 15% dyestuff does not bind to the target products and then enters the wastewater stream [1,2]. Dye wastewater is difficult to be biodegraded because of its high content of organic pollutants, complex composition, and good chemical stability, and most dyes are toxic, which has seriously threatened the ecological environment and human health [1,3]. Therefore, it is very important to remove dye wastewater. In order to solve this problem, a variety of strategies, such as chemical oxidation, photocatalytic degradation, filtration, and adsorption, have been employed to treat wastewater containing dyes [4,5]. Among these methods, visible light-driven semiconductor photocatalysis is regarded as an ideal "green" treatment for organic pollutants [6].
Due to their unique energy band structure, low cost, and superlative physical and chemical properties, Bi-based photocatalysts have attracted extensive attention in recent years [7,8]. However, bare Bi-based photocatalysts still have some drawbacks, such as photoetching, narrow photoresponse range, and high recombination rate of the photogenerated electron-hole pairs. BiPO 4 , as a nonmetallic oxyacid type semiconductor, has attracted great interests from many scholars all over the world due to its nontoxicity, low cost, high photoactivity, and positive valence band position [9,10]. However, BiPO 4 has a wide band gap, which inhibits its further application as an efficient visible light-driven photocatalyst [11]. In this regard, coupling BiPO 4 with other semiconductors with a narrow band gap is necessary, which can broaden the absorption range of light and improve the lifetime of photogenerated carriers. It was reported that by combining BiPO 4 with another semiconductor, such as CdS [12], AgBr [13], BiOI [14,15], and BiOCl [16], its photocatalytic performance could be effectively enhanced.
BiOBr, as a member of the bismuth oxyhalide photocatalyst, has been hugely studied owing to its appropriate band gap and the excellent photocatalytic activity under visible light irradiation. It is well known that BiOBr has an internal electrostatic field between the layered structure that is composed of [Bi 2 O 2 ] 2+ slabs and double halogen atom slabs, which can effectively separate the photogenerated electronhole pairs [17][18][19]. Therefore, various synthesis strategies have been developed successfully to prepare BiOBr/BiPO 4 p-n heterojunctions [20][21][22][23]. These works strongly suggested that the fabrication of BiOBr/BiPO 4 p-n heterojunctions can effectively improve the photogenerated electronhole pair separation efficiency and the migration rate of photogenerated carriers.
In recent years, the metal-semiconductor composite structure has been widely used as an effective strategy for improving photocatalytic performance, because it has the surface plasmon resonance (SPR) effect of the metal that can effectively broaden the light absorption range. Bismuth (Bi) has been intensively used as a cocatalyst, due to its unique characteristics, such as low cost, low charge carrier density and electron effective mass, and SPR properties similar to the noble metals [24]. It has been reported that Bi combined with Bi 2 WO 6 [25], BiOI [26], g-C 3 N 4 [27], and Bi 3 O 4 Cl [28] showed stable and significantly enhanced photocatalytic performance.
To the best of our knowledge, there are no existing reports on the Bi-BiOBr/BiPO 4 system. In this paper, a novel Bi-BiOBr/BiPO 4 multicomposite semiconductor photocatalyst was successfully prepared by a simple two-step method. These p-n Bi-BiOBr/BiPO 4 systems exhibit much higher visible photoactivity for RhB than bare BiOBr and bare BiPO 4 , in which metal Bi does act not only as an effective electron acceptor but also as a promoter that enhances the range of light absorption. 2.3. Synthesis of Bi-BiOBr/BiPO 4 . The procedure for preparing Bi-BiOBr/BiPO 4 was as follows: first, the as-prepared BiOBr/-BiPO 4 (0.42 g) was added into 80 mL ethylene glycol with a certain amount of Bi(NO 3 ) 3 ·5H 2 O (0.1 mmol, 0.2 mmol, and 0.3 mmol) under ultrasonic dispersion conditions. After stirring for 2 h, the homogeneous solution was transferred into a 100 mL Teflon-lined stainless steel autoclave and maintained at 180°C for 10 h. The resulting precipitates were centrifugated and washed with distilled water and absolute ethanol for three times, respectively, then dried at 60°C for 12 h.

2.4.
Characterization. The X-ray powder diffraction (XRD) of the products was analyzed using a Bruker D8 Focus Xray diffractometer (Bruker, Germany) with Cu-Kα radiation (λ = 0:154178 nm). The product morphology was analyzed by field emission scanning electron microscopy (FESEM, JEOL JSMT300, Japan) and field emission transmission electronic micrograph (FETEM, JEOL SM-6330F, Japan). UVvis diffuse reflectance spectra of the samples were obtained on an UV-vis spectrophotometer (Cary 5000, Shimadzu, Japan). X-ray photoelectron spectroscopy (XPS) was acquired using the ESCALAB 250 spectrometer (Thermo, USA) with Al Kα (1486.6 eV) line to get the elements and chemical state. The photoluminescence (PL) emission spectra of the products were carried out on a RF-5301PC fluorescence spectrophotometer (Shimadzu, Japan). The photocurrent response and electrochemical impedance spectra (EIS) measurements were carried out in a conventional three-electrode system with a 0.1 M Na 2 SO 4 electrolyte solution by using an electrochemical system (SP-150, France).

Photocatalytic
Activity. The photocatalytic activity of the samples was investigated by degrading rhodamine B (RhB) in an aqueous solution under visible light irradiation. Visible light was provided by a 350 W Xe lamp with a cutoff filter (λ > 420 nm). A total of 0.05 g of the photocatalyst was added to a glass reactor containing 100 mL of RhB aqueous solution (10 mg·L -1 ). Prior to irradiation, the suspensions were magnetically stirred in the dark for 30 min to ensure the establishment of an adsorption-desorption equilibrium.
After equilibrium, the mixture was continuously stirred under visible light irradiation. At a predetermined time interval, 4 mL of suspension was sampled and centrifuged (6000 rpm, 5 min) to remove the catalyst. The concentration of RhB was analyzed using an UV-visible spectrophotometer (TU-1901). In order to evaluate the stability of the photocatalyst, after every 1.5 h of photodegradation, the photocatalysts were separated, washed with deionized water, and dried at 333 K.

Results and Discussion
3.1. XRD. XRD characterization was employed to identify the structural detail of the as-prepared samples. The XRD pattern of the as-prepared BiOBr, BiPO 4 , BiOBr/BiPO 4 , and Bi-BiOBr/BiPO 4 composites with different molar ratios of P to Br is shown in Figure 1.

XPS.
To investigate the surface chemical composition and elemental states, X-ray photoelectron spectroscopy was taken on 0.2 Bi-BiOBr/BiPO 4 -3 sample. The survey spectrum of 0.2 Bi-BiOBr/BiPO 4 -3 is shown in Figure 2(a). It can be seen that the 0.2 Bi-BiOBr-BiPO 4 -3 composite consisted of Bi, O, Br, and P elements. Figure 2(b) shows the high-resolution XPS spectra of Bi 4f. Two strong peaks centered at 158.9 eV and 164.2 eV can be observed, which attribute to Bi 4f 7/2 and Bi 4f 5/2 [33], respectively. Furthermore, these peaks can be subdivided into three different peaks. Among them, the peaks centered at 159.4 eV and 164.6 eV are attributed to Bi 3+ in BiPO 4 [12], while those peaks centered at 158.7 eV and 164.0 eV are ascribed to Bi 3+ in BiOBr [34]. The rest peaks centered at 157.5 eV and 162.8 eV belong to metallic Bi 0 [35]. As shown in Figure 2(c), there are two peaks located at 68.0 eV and 69.0 eV, which are associated with Br 3d 5/2 and Br 3d 3/2 , respectively [36]. It can be seen that a broad characteristic peak is located at 133.1 eV (Figure 2(e)), indicating the presence of PO 4 3in 0.2 Bi-BiOBr/BiPO 4 -3 composite [37]. The XPS spectrum of O 1s can be subdivided into two peaks at about 530.0 eV and 531.3 eV (Figure 2(d)), which are assigned to the lattice oxygen and the hydroxyl groups, respectively [36,38]. Based on XPS spectra and XRD results, it can be concluded that Bi-BiOBr/BiPO 4 composites are successfully fabricated.

Morphology.
The surface morphology and microstructure of the as-prepared samples were measured by SEM, as presented in Figure 3. It is obvious that the morphology of the bare BiOBr sample is a sheet-shaped structure with a smooth clear surface (Figure 3(a)), while that of bare BiPO 4 is an irregular polyhedron structure with a length range of about 0.5-1 μm (Figure 3 Figure 5 displays the UVvis diffuse reflectance spectra of the as-prepared samples.

Optical Absorption Properties.
It can be seen that the absorption edge of BiPO 4 is located at 302 nm, indicating its ineffective visible light absorption ( Figure 5(a)). While in the case of bare BiOBr nanosheets, the absorption edge is lower than 450 nm, implying the possibility of visible light photocatalytic activity. After combining BiPO 4 with BiOBr, compared with bare BiPO 4 , the absorption band edges of the BiOBr/BiPO 4 heterostructure are gradually red-shifted with increasing the molar ratio of Br to P from 1/5 to 1/1, which indicates that all as-synthesized BiOBr/BiPO 4 heterostructures are visible   Journal of Nanomaterials light effective. As shown in Figure 5(b), by introducing metallic Bi, the absorption edge of Bi-BiOBr/BiPO 4 is redshifted and the absorption density around 500 nm becomes stronger, which is caused by the SPR effect of metal Bi [40]. The band gap energies of bare BiOBr and BiPO 4 can be calculated as the following equation [41].
where α, υ, h, A, and E g represent the absorption coefficient, The results are shown in Figure 6(a).
The degradation ratio of RhB over bare BiOBr is 48.6% under visible light irradiation for 90 min. It can be seen that only 18.1% RhB degradation can be reached in the presence of bare BiPO 4 due to its large band gap energy. However, by coupling BiPO 4 with BiOBr, the photocatalytic activity of the BiOBr/BiPO 4 composites is significantly enhanced. For BiOBr/BiPO 4 -3, 79.8% RhB is degraded in 90 min. Comparatively, with the introduction of metallic Bi, the photoactiv-ity of the Bi-BiOBr/BiPO 4 composite sharply increases, and the degradation ratio of RhB over 0.2 Bi-BiOBr/BiPO 4 -3 can reach 89.7%. The degradation of RhB molecules can be regarded as a pseudo-first-order reaction. The photocatalytic reaction kinetics conforms to the formula: ln ðC/C 0 Þ = −kt, where t is the reaction time, k is the degradation rate constant, and C and C 0 are the concentration of RhB at time 0 and t, respectively [44]. The results are shown in Figures 6(b) and 6(c). The decolorization rate constant of RhB over the 0.2 Bi-BiOBr/BiPO 4 -3 is the highest, which is  Journal of Nanomaterials about 3.8, 14.2, and 1.4 times higher than bare BiOBr, bare BiPO 4 , and BiOBr/BiPO 4 -3, respectively. These results strongly indicate that the introduction of metallic Bi can effectively improve the photocatalytic activity of BiOBr/-BiPO 4 composites. The reusability of the 0.2 Bi-BiOBr/-BiPO 4 -3 was investigated under visible light irradiation, and results are shown in Figure 6(d). After four cycling tests, a slight decrease in photoactivity is observed, indicating that the 0.2 Bi-BiOBr/BiPO 4 -3 composite is highly stable during the photocatalysis process.
3.6. PL, Photocurrent, and EIS. PL emission has been widely employed to study the trapping, migration, and separation efficiency of photogenerated charge carriers. It is known that the PL emission is caused by the recombination of excited electrons and holes, and the lower PL intensity may represent the lower efficiency of photogenerated charge carrier recombination under light irradiation [45,46]. Figure 7(a) displays the comparison of PL spectra of BiOBr, BiPO 4 , BiOBr/BiPO 4 -3, and 0.2 Bi-BiOBr/BiPO 4 -3 in the emission peak range of 420-480 nm. The bare BiOBr has the highest emission peak centered at 445 nm, which can be ascribed to the recombination of electron-hole pairs. It is clearly observed that the Bi-BiOBr/BiPO 4 -3 heterojunction shows significantly weaker PL intensity as compared with both bare BiOBr and BiPO 4 , which implies that the recombination rate of electron-hole pairs can be effectively suppressed through the formation of heterojunction among Bi, BiOBr, and 9 Journal of Nanomaterials BiPO 4 . However, the PL intensity of 0.3 Bi-BiOBr/BiPO 4 -3 is higher than that of 0.2 Bi-BiOBr/BiPO 4 -3, demonstrating that loading a proper amount of metallic Bi is helpful for efficient charge transfer and separation. The possible reason is that the excess metallic Bi will work as a recombination center [27]. The photocurrent can directly display the mobility capability of the photoinduced charge carrier. The photoresponses of bare BiOBr, bare BiPO 4 , BiOBr/BiPO 4 -3, and Bi-BiOBr/BiPO 4 -3 heterojunction performed under several on/off sunlight irradiation cycles are shown in Figure 7(b).
Photocurrent characterization has been widely employed to display the charge separation in heterojunction photocatalysts. The higher peak density implies a higher separation efficiency of the photogenerated carriers, which usually plays a positive role in photoactivity promotion [38]. It can be seen that the photocurrent density of 0.2 Bi-BiOBr/BiPO 4 -3 composite increases sharply upon light irradiation and then rapidly decreases to zero when light is turned off. The photocurrent density of 0.2 Bi-BiOBr/BiPO 4 -3 composite is much higher than that of bare BiOBr, bare BiPO 4 , and BiOBr/BiPO 4 composite, indicating a longer lifetime of electron-hole pair photogenerated over 0.2 Bi-BiOBr/-BiPO 4 -3 composite [47,48].
In addition, the effect of the heterostructure on the charge transfer resistance was also investigated, and EIS measurements were carried out to characterize the asprepared samples. It is well-accepted that the smaller arc diameter in Nyquist plots implies a lower resistance of the interfacial charge transfer in the electrode-electrolyte interface region on the surface of the electrode [49]. Figure 7(c) shows EIS Nyquist plots for the BiOBr/BiPO 4 -3 and the Bi-BiOBr/BiPO 4 -3 with different Bi contents. Among these photocatalyst composites, the 0.2 Bi-BiOBr/BiPO 4 -3 sample displays the smallest arc radius, which means a faster interfacial charge transfer process [49,50].

Photocatalytic Mechanism.
In order to explain the mechanism of improved photocatalytic activity, it is necessary to confirm the valence band (VB) and conduction band (CB) potentials of BiPO 4 and BiOBr. As shown in Figure 8, the valence band edges of BiPO 4 and BiOBr were obtained by VB-XPS spectra. It is observed that the VB potentials of the pure BiPO 4 and BiOBr are at 2.8 and 2.2 eV, respectively. Based on the results of DRS, the band edge positions of the conduction band (CB) of bare BiOBr and BiPO 4 photocatalysts can be evaluated using the following equation [51].
According to the results of DRS, the band gap energies (E g ) of BiOBr and BiPO 4 were 2.66 eV and 3.84 eV, respectively. Therefore, E CB of BiPO 4 and BiOBr are calculated to be about -1.04 eV and -0.46 eV, respectively. Figure 9(a) presents the band edge positions of BiOBr and BiPO 4 . As we all know, BiPO 4 is an n-type semiconductor [20,23], whose Fermi energy level lies close to E CB , while BiOBr is a p-type semiconductor, whose Fermi level is located close to E VB [21]. When a p-n junction structure is formed between BiPO 4 and BiOBr, the Fermi levels of BiOBr tend to descend, whereas that of BiPO 4 tend to rise up until an equilibrium state of Fermi levels of BiOBr and BiPO 4 is obtained. Because of the wide band gap of BiPO 4 (E g = 3:84 eV), it is not excited under visible light irradiation. For the Bi-BiOBr/BiPO 4 -3 heterojunctions, under visible light, BiOBr can only be excited to produce electronhole pairs. Then, the photoinduced electrons will transfer from BiOBr to the positive position (the CB of BiPO 4 ), and the photoinduced holes will be left on the VB of BiOBr.
Based on the results of structure characterizations and photocatalytic tests of the samples, a possible photocatalytic mechanism of 0.2 Bi-BiOBr/BiPO 4 -3 is proposed in Figure 9(b). It is reported that the Fermi level of Bi is about -0.17 eV [32,52]. When 0.2 Bi-BiOBr/BiPO 4 -3 is irradiated by visible light (λ > 420 nm), BiOBr can be excited to yield photogenerated electrons and holes. Subsequently, the excited state electrons in the BiOBr conduction band (CB) can readily migrate into the conduction band (CB) of BiPO 4 conduction band and metal Bi and then will contact with O 2 to form ⋅ O 2 -, while the photogenerated holes were left on the VB of BiOBr, reducing the recombination probability of the photogenerated electrons and holes. In this way, the separation of photogenerated electrons and holes can be greatly improved, which results in markedly enhanced photoactivity.
It is vitally important to identify the main active oxidant in the photocatalytic reaction process for a better understanding of the mechanism. In this study, we use triethanolamine (TEOA), benzoquinone (BQ), and isopropanol (IPA) as the scavengers of holes (h + ), superoxide radicals ( ⋅ O 2 -), and hydroxyl radicals ( ⋅ OH), respectively. Figure 10 displays the trapping experiment of active species during the photocatalytic degradation of RhB over 0.2 Bi-BiOBr/BiPO 4 -3 It can be seen that with IPA added to the solution, there is no obvious change for the degradation efficiency of RhB, suggesting that ⋅ OH plays a minor role in this photocatalytic system. When TEOA is used, the degradation efficiency of RhB is significantly depressed, indicating the h + plays a crucial role in the photodegradation process. When BQ is used, the degradation efficiency of RhB is slightly depressed. It suggests that ⋅ O 2 contributes to the degradation of RhB to some extent. Therefore, it can be concluded that h + and ⋅ O 2 -, especially h + , are the main active species responsible for the photocatalytic degradation of RhB over Bi-BiOBr/-BiPO 4 composites.
As the above results, a possible pathway for the photocatalytic degradation of RhB can be proposed as follows [25].

Conclusion
Bi-BiOBr/BiPO 4 heterojunctions were prepared via an in situ solvothermal reduction method. These photocatalysts were tested for RhB degradation under visible light irradiation. Among these heterojunctions, the 0.2 Bi-BiOBr/-BiPO 4 -3 presents the highest photoactivity for RhB degradation. The markedly enhanced photoactivity can be attributed to the SPR effect of metal Bi and the formation of BiOBr/BiPO 4 heterojunction, which can effectively suppress the recombination rate of electron-hole pairs. The main active species in the degradation of RhB are photogenerated holes and superoxide radicals. Cyclic experiments showed that 0.2 Bi-BiOBr/BiPO 4 -3 composites exhibited good stability. This work highlights the important role of metal Bi and p-n heterojunction in improving photocatalytic activity and provides a new idea for the design of new photocatalysts in practical wastewater treatment.

Data Availability
The dataset supporting the conclusions of this article is included within the article.