Fabrication of Bi₂O₄/Bi₂WO₆ composite for high photocatalytic performance

ABSTRACT

The Bi₂O₄/Bi₂WO₆ composites were synthesized by one step hydrothermal method, which were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), Fourier transformed infrared spectroscopy (FT-IR), photoluminescence spectra (PL), X-ray photoelectron spectroscopy (XPS) and UV – Vis diffuse reflectance spectroscopy (DRS). The photocatalytic performance of the composite was tested by degrading methyl orange (MO) under visible light irradiation. The results showed that the Bi₂O₄/Bi₂WO₆ composites performed better than pure Bi₂O₄ and Bi₂WO₆, whose cycle stability was also greatly improved. Among them, S3 sample (Bi₂O₄:Bi₂WO₆ = 16:1) could degrade by about 95% methyl orange (MO) within 50 min, and after three cycles the degradation rate could still reach 74%. The enhanced photocatalytic activity and stability was attributed to the formation of heterojunctions. This work provided a promising composite photocatalyst with high performance for organic pollutant degradation.

GRAPHICAL ABSTRACT

KEYWORDS:

• Bi₂O₄/Bi₂WO₆
• heterostructures
• visible light
• photocatalytic activity
• stability

1. Introdution

At present, water pollution resulting from organic pollutants becomes a serious environmental concern. In the treatment of water pollution, photocatalytic technology has received a lot of attention [1,2]. After TiO₂ was found to have photocatalytic behaviour, TiO₂-based materials have been extensively investigated as photocatalysts [3,4]. However, due to its wide band gap (~3.2 eV), TiO₂ can only respond in the ultraviolet region, which accounts for less than 5% of sunlight [5,6]. Therefore, developing new visible light photocatalysts has become the current hot spot.

Recently, some bismuth-containing oxides with narrow band gaps present a wide range of application in photocatalysis, such as BiVO₄ [7], Bi₂MoO₆ [8] and BiOX (X=Cl, Br, I) [9–11]. The bismuth in these photocatalysts all appears as Bi³⁺. It is known that the 6s2 hybrid orbit of Bi³⁺can narrow the band gap energy and increase the valence band width by overlapping O2p, which is beneficial to the improvement of photocatalytic performance [12]. In addition, it is reported the empty 6s orbital of Bi⁵⁺ ion is also able to improve the band gap structure and endow related compounds with excellent visible light photocatalytic activity [13]. For example, it was reported that NaBiO₃ possessed strong absorption under visible light and thus superior activity towards the degradation of organic pollutant [14]. However, few studies have involved bismuth-containing compounds with the mixed valence states so far.

Bi₂O₄ is a mixed valence oxide of bismuth, which contains Bi³⁺and Bi⁵⁺. It has reported to have a good visible light photocatalytic activity [15]. Nonetheless, the stability of Bi₂O₄ is not good, since Bi⁵⁺ in Bi₂O₄ tends to Bi³⁺ under long-term lighting conditions. It is necessary to enhance its stability [16]. It is well known that constructing type-II heterojunctions with other semiconductors is a feasible strategy to improve and photocatalytic performance. Recently, we find that Bi₂WO₆, as a binary bismuth complex oxide, possesses a narrow bandgap (~2.7 eV), which shows excellent photo-absorption and photocatalytic activity [17]. Therefore, we imagine that the stability and catalytic performance of Bi₂O₄ might be improved by constructing Bi₂O₄/Bi₂WO₆ heterostructure.

Experiment

Materials preparation

All the chemicals were purchased from Sinopharm Chemical Reagent Co., Ltd. (China) and used as received without further purification. The Bi₂O₄/Bi₂WO₆ composites were synthesized by one-step hydrothermal method. At room temperature, a certain amount of NaBiO₃ and 2 mmol of Bi(NO₃)₃ and 1 mmol of Na₂WO₆ were placed in a beaker with 70 ml of water and sonicated for 5 min. Then, the ultrasonically suspended suspension was introduced into a 100 ml Teflon-lined stainless steel autoclave and maintained at 180°C for 6 h. The product was collected by centrifugation, washed for three times with distilled water, and then dried at 80°C for 12 h. In this experiment, the amount of NaBiO₃ was 8 mmol, 16 mmol, 32 mmol and 64 mmol, and the corresponding composite samples were called S1, S2, S3 and S4. Correspondingly, pure Bi₂O₄ was produced by NaBiO₃ with hydrothermal, and pure Bi₂WO₆ was made by Bi(NO₃)₃ and Na₂WO₆ with hydrothermal method.

Characterization

The phase structure tests were performed by an X-ray diffractometer (XRD, Bruker D8 Advance) using Cu-Kα radiation. The morphology of the material was characterized by field emission scanning electron microscopy (SEM, Quanta 250) and transmission electron microscopy (TEM, Tecnai G2 F20). Chemical mapping was achieved using energy-dispersive X-ray detector and scanning electron microscopy modules. X-ray photoelectron spectroscopy (XPS: Thermo ESCALAB250, U.S.A) was used to detect the surface properties of the sample. The optical properties of the samples were obtained using UV–vis diffuse reflectance spectroscopy (DRS, Varian Cary 300) and photoluminescence spectroscopy (PL, Varian Cary-Eclipse 500).

Photocatalytic activity tests

The photocatalytic activity of Bi₂O₄/Bi₂WO₆ composites was tested under the irradiation of visible light with methyl orange (MO) aqueous solution. In the experimental measurement, a 150 W xenon lamp with a cut-off filter of 400 nm was used as a light source for visible light, and its light intensity is about 0.25mWcm⁻². The photocatalytic activity of the experimental process is as follows: 0.05 g Bi₂O₄/Bi₂WO₆ composite sample was added 50 mL 20 mg/L MO solution（pH ≈ 6.8）. Before using visible light irradiation, the suspension is placed in a dark environment and stirred for about 10 min to establish an equilibrium of adsorption/desorption. Then, take samples every 20 min for 2 h, 3 ml each and centrifuge. Finally, the sample was tested on a UV–visible spectrophotometer to analyse the change of concentration of MO, wherein the special wavelength of MO was set at 464 nm.

Photoelectrochemical measurement

On the electrochemical workstation (CHI660B, China), photoelectrochemical measurements were carried out using a standard three-electrode where platinum electrode (1 × 1 cm²) is as a counter electrode and a saturated calomel electrode (SCE) is as a reference electrode. A 0.5 M Na₂SO₄ solution was used as electrolyte. The composite electrode of Bi₂O₄/Bi₂WO₆ was uniformly coated on 1 cm × 1 cm tin fluoride oxide glass (FTO) to prepare the working electrode.

Results and discussion

Structural and morphology characterization

The purity and crystallization phase of Bi₂O₄, Bi₂WO₆ and their composites with different proportions were tested by XRD (Figure 1). The diffraction peaks of Bi₂O₄ match well with the monoclinic phase of Bi₂O₄ (JCPDS card number 50–0864), while all the diffraction peaks of Bi₂WO₆ can be indexed to the orthorhombic phase (JCPDS card number 39–0256). This shows that pure samples were successfully synthesized. The XRD patterns of S1, S2, S3 and S4 sample present the superposition of the diffractive peaks of the two components, indicating that the obtained catalyst is indeed composed of Bi₂O₄ and Bi₂WO₆. As the composite ratio increases, the peak intensity of Bi₂O₄ gradually increases, suggesting the increase of the content of Bi₂O₄ in the sample. The relative mass fraction of Bi₂WO₆ in Bi₂O₄/Bi₂WO₆ composites can be estimated according to equation (1)(1) $w_1=\frac{I_1}{I_1+I_2(RI{R}_R/RI{R}_2}$(1) :

Figure 1. XRD pattern of the resultant products.

(1) $w_1=\frac{I_1}{I_1+I_2(RI{R}_R/RI{R}_2}$(1)

where W₁ is the mass fraction of Bi₂WO₆, I₁ and I₂ are the strongest integral intensities of Bi₂WO₆ and Bi₂O₄, respectively. RIR₁ and RIR₂ values are obtained from MDI Jade 5.0. By calculation, the content of Bi₂WO₆ is estimated as 45.6%, 21.4%, 12.7% and 5.9% in the S1, S2, S3 and S4 sample, respectively.

The morphology of the composite samples can be obtained by SEM techniques. As shown in Figure 2a, pure Bi₂O₄ shows an irregular rod-like morphology with a length of about 2 µm and a diameter of about 0.5 µm. In contrast, Figure 2f shows the morphology of pure Bi₂WO₆. It can be seen that pure Bi₂WO₆ generally shows a coherent sheet-like structure. As shown in Figure 2b, there are almost basically short sticks like the morphology of pure Bi₂O₄. However, the difference is that the length of the rod becomes shorter, which is about 500 nm. It is very likely that a small amount of Bi₂WO₆ has an effect on the morphology of the sample. In Figure 2c, it can be seen that S3 appears has less rod morphology, which tends to degenerate into fragmented floccules. In Figure 2d, the rod-shaped Bi₂O₄ crystals further shorten, which tend to stick together. As for Figure 2e, it can be seen that there is a combination of rod-like and small pieces, and a large number of pieces are attached to the surface of fragmented rod.

Figure 2. SEM images of (a) Bi₂O₄ (b) S4 (c) S3 (d) S2 (e) S1 (f) Bi₂WO₆.

In order to study the elemental distribution of Bi₂O₄/Bi₂WO₆ composites, the EDS elemental analysis of S3 was carried out. As shown in Figure 3, three elements of W, O and Bi were detected in the sample, which were evenly distributed, suggesting that Bi₂O₄ and Bi₂WO₆ were uniformly grown together.

Figure 3. Surface scan image of S3 (a) Surface scan image (b) O element (c) W element (d) Bi element.

Sample S3 was further characterized by TEM. As shown in Figure 4a, the shape of the sample is generally rod-shaped, with some small particles attached to the rod. We can clearly see the complex shape of rods and pieces in Figure 4b, in which the pieces are tightly attached to the surfaces of rods. The TEM results are well consistent with SEM.

Figure 4. TEM image of S3.

XPS spectra were used to identify the elemental composition and chemical state of composite sample S3, pure Bi₂O₄ and Bi₂WO₆. All spectra were compared with C1s at 284.8 eV. As shown in Figure 5a, full-scan XPS spectra show that O, Bi and W are indeed present in the sample. Figure 5b shows the Bi4f spectrum of Bi₂O₄, Bi₂WO₆ and sample S3. Among them, the peaks of Bi₂WO₆ at binding energies of 158.7 eV and 164 eV can be attributed to the characteristic spin-orbit splitting of Bi4f_5/2 and Bi4f_7/2 of Bi³⁺ [18]. However, Bi₂O₄ can be decomposed into two doublets at binding energies of 158.2 eV and 158.6 eV (or 163.5 eV and 163.9 eV), and the corresponding sample S3 also has binding energy of 158.4 eV and 158.7 eV (or 163.6 eV and 164 eV  ) are decomposed into two peaks, which corresponds exactly to the two valence states of Bi(III) and Bi(V) [19]. Figure 5c is a high resolution XPS spectrum of W4f of Bi₂WO₆ and sample S3. Among them, the two main peaks at 35 eV and 37.1 eV are W4f_5/2 and W4f_7/2 of W⁶⁺ ions in Bi₂WO₆. Figure 5d. is the XPS spectrum of O1s. For pure Bi₂O₄, the two peaks at 529.1 eV and 531.1 eV (and the two peaks in 529.8 eV and 531.7 eV in pure Bi₂WO₆) can correspond to the lattice oxygen (Bi-O) and hydroxyl oxygen [15]. The binding energy of O in sample S3 synthesized by hydrothermal one step is between two pure samples and should have some coupling effect. Overall, the XPS analysis results can be compared with the XRD analysis results, which proves that we have indeed synthesized Bi₂O₄/Bi₂WO₆ composites.

Figure 5. XPS spectra of Bi2O4, Bi2WO6 and S3 : survey(a), Bi4f(b), W5f(c), O1s(d).

We further tested the optical properties of the composite samples while also testing the optical properties of pure Bi₂WO₆ and pure Bi₂O₄. As shown in Figure 6, the absorption edges of pure Bi₂WO₆ and Bi₂O₄ are approximately 475 nm and 740 nm, which correspond to a band gap of 2.61 eV and 1.68 eV (Eg ≈ 1240/λ), respectively. The absorption edges of the composite samples S1, S2, S3 and S4 are significantly larger than the pure Bi₂WO₆ and less than Bi₂O₄. Among these, the absorption edge of the sample S3 is the largest, which can reach 630 nm, and the forbidden band width is approximately 1.97 eV. UV spectrum clearly shows that with the increase of the composite ratio, the absorption edge first increases and then decreases, and all of them can well response to visible light.

Figure 6. UV-visible diffuse reflectance spectra of samples.

Photocatalytic performance

In order to test the photocatalytic activity of the composite sample, a series of photocatalytic degradation experiments were performed to degrade the typical organic dye MO, as shown in Figure 7. First of all, a blank experiment was carried out, and it was found that the photolysis of pure MO solution is negligible in the absence of photocatalyst. In contrast, pure Bi₂O₄ and Bi₂WO₆ were able to remove 77.0% and 28.8% of the MO after 120 min of visible light irradiation. After Bi₂O₄ and Bi₂WO₆ were recombined, the photocatalytic activity was greatly improved. With the increasing of Bi₂WO₆ content, the photocatalytic activity of the composite sample increases first and then decreases. The MO degradation rates are 77.3%, 89.3%, 95.2% and 91.9% for S1, S2, S3 and S4, respectively, in which S3 sample show the best photocatalytic activity. The MO degradation rates over S3 is comparable to or higher than those of reported Bi-containing compounds [20]. Generally speaking, the specific surface has obvious effect on the photocatalytic activity. However, the specific surface areas are measured as 4.5, 7.3, 5.8, 5.3, 4.9 and 12.5 m²/g for Bi₂O₄, S1, S2, S3, S4 and Bi₂WO₆, respectively. S3 sample possesses moderate surface area but best activity. Therefore, the surface area is not a key factor for the improved activity. The improved photocatalytic activity may be ascribed to the heterojunction effect between Bi₂O₄ and Bi₂WO₆, which helps to restrain the recombination of photo-generated electrons and holes.

Figure 7. Photodegradation of MO.

The stability of photocatalyst is of great significance for practical production applications. The recycling runs of MO degradation over S3 sample and Bi₂O₄ were performed to test their stability. As shown in Figure 8(a) and (b), the degradation efficiency of Bi₂O₄ decreased from 77% to 31% after three cycles, while the degradation efficiency of S3 sample decreased from 95% to 74% after three cycles. Therefore, Bi₂WO₆ modified Bi₂O₄ exhibits excellent stability, which is superior to some reported photocatalysts [21].

Figure 8. Cycling runs for the photocatalytic degradation MO over (a) Bi₂O₄ and（b）S3.

Mechanism

PL and photocurrent

In general, the separation efficiency and migration rate of photogenerated electrons and holes play a very important role which affects the photocatalytic activity of photocatalysts. We can use the photoluminescence spectra (PL) to qualitatively research their recombination processes. The PL signal is mainly derived from the radiative recombination process of self-trapped excitations. The lower the PL intensity, the smaller the probability of the photogenerated electron – hole pairs recombination. As shown in Figure 9a, the PL emission intensity of sample S3 is much lower than that of Bi₂WO₆ and Bi₂O₄, which means that constructing Bi₂O₄/Bi₂WO₆ heterostructure can inhibit the recombination of photogenerated electrons and holes, favouring the enhancement of the photocatalytic efficiency. To further confirm this, the I-t curve was also measured by visible light switching illumination on S3, Bi₂WO₆ and Bi₂O₄ samples. As shown in Figure 9b, all samples exhibited a significant photocurrent response when the lamp was turned on. Compared to Bi₂WO₆ and Bi₂O₄, S3 samples produced higher current densities, indicating that photogenerated electron-holes have a high mobility and a low recombination rate. Therefore, we can confirm that the composite sample S3 effectively inhibits the recombination of photogenerated electrons and holes.

Figure 9. (A) PL spectra of S3, Bi₂O₄ and Bi₂WO₆ samples, (b) transient photocurrent responses of as-prepared samples.

Main active species

In photocatalytic experiments, the three active species that typically cause photocatalytic degradation of organic matter are superoxide radicals (·O₂⁻), hydroxyl radicals (−OH), and active holes (h⁺) [22]. In order to explain the photocatalytic reaction mechanism, it is necessary to find the main active species of MO photodegradation. In this experiment, samples isopropanol (IPA), benzoquinone (BQ) and triethanolamine (TEOA) [23] were used as scavengers for the three active species. We conducted a series of experiments to test the effect of these active species on the photodegradation rate of MO. As shown in Figure 10, using IPA to test the active species -OH, there was no decrease in the photodegradation rate of the MO, indicating that there was no -OH effect during the reaction. The addition of TEOA to the sample solution has a certain decrease in the photodegradation rate of MO, which indicates that there are h⁺ active radicals during photodegradation. After the addition of BQ, the photodegradation rate of MO decreased significantly, indicating that ·O₂⁻ plays a major role in photodegradation. A large number of documents indicate that -OH is not present in Bi-based photocatalysts because the reduction potential of Bi(V)/Bi(III) is more negative than the standard·OH/OH- redox potential - (+2.38 eV). The results of these tests show that the active free radicals of the composite photocatalyst Bi₂O₄/Bi₂WO₆ when photodegraded are ·O₂⁻ and h⁺. The electrons and holes generated by the light, the electrons migrate to the surface of the catalyst to form oxygen radicals with oxygen, and the holes directly interact with the degradants.

Figure 10. Degradation rate of MO in the presence of various radicals scavengers.

XPS valence band spectrum

In order to further analyse the separation of photoelectrons-holes and the photocatalytic process, it is necessary to understand the band arrangement of Bi₂O₄/Bi₂WO₆. As shown in Figure 11, the resulting valence band edges are approximately 0.9 eV and 2.45 eV for Bi₂O₄ and Bi₂WO₆, respectively. Considered that the band gap of Bi₂O₄ and Bi₂WO₆ is about 1.68 eV and 2.61 eV, their conduction bandedges are −0.78 eV and −0.19 eV. Obviously, the band potentials of Bi₂O₄ and Bi₂WO₆ take on a staggered band alignment, which helps to improve the separation and transfer of photogenerated carriers due to the energy bias.

Figure 11. Valence-band XPS spectra of the Bi₂O₄ and Bi₂WO₆.

Photocatalytic mechanism

According to the experimental results and reference energy band theory, we can simply analyse the catalytic mechanism of Bi₂O₄/Bi₂WO₆ composite photocatalyst. The band gap of Bi₂O₄ is relatively narrow, about 1.68 eV, and can be excited to generate electron–hole pairs under visible light irradiation. As shown in Figure 12, since the CB position of Bi₂O₄ is higher than that of Bi₂WO₆, the photo-induced electrons of Bi₂O₄ migrate to the conduction band of Bi₂WO₆ under the irradiation of visible light, while the Bi₂O₄ receives the signal from Bi₂WO₆ at the VB position of Bi₂O₄. The photogenerated holes transferred from VB greatly promote the separation and transfer of photogenerated carriers and extend the lifetime of photogenerated carriers. The electrons accumulated in the CB of Bi₂WO₆ capture dissolved oxygen to generate active species ·O₂⁻ radicals, while the holes on VB of Bi₂O₄ directly degrade MO under irradiation with visible light. Therefore, the catalytic activity of the photocatalyst is greatly improved.

Figure 12. Schematic illustration of the mechanism for the high photocatalytic performance of Bi₂O₄/Bi₂WO₆ composite.

Conclusion

In summary, a Bi₂O₄/Bi₂WO₆ composites were successfully synthesized via a hydrothermal method. The composite exhibited improved photocatalytic activity towards the degradation of methyl orange (MO) under visible light irradiation. Especially, the S3 sample could remove 95.2% of MO in 120 min, suggesting the best activity. The improved activity and stability was attributed to the heterojunction effect between Bi₂O₄ and Bi₂WO₆, which strengthen effective separation of photo-generated charge carriers. Besides, the trapping experiments exhibited that ·O₂⁻ and h⁺ radicals are the main active species in the photocatalytic degradation of MO.

Acknowledgments

This work was supported by Fundamental Research Funds for the Scientific Research Project of Jiangxi Provincial Education Department of China (GJJ2206602, GJJ217002).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The work was supported by the Fundamental Research Funds for the Scientific Research Project of Jiangxi Provincial Education Department of China [GJJ2206602, GJJ217002].