Article pubs.acs.org/IC
Construction of α−β Phase Junction on Bi4V2O11 via Electrospinning Retardation Effect and Its Promoted Photocatalytic Performance Chade Lv, Gang Chen,* Jingxue Sun,* and Yansong Zhou MIIT Key Laboratory of Critical Materials Technology for New Energy Conversion and Storage, School of Chemistry and Chemical Engineering, Harbin 150001, P. R. China S Supporting Information *
ABSTRACT: The creation of a phase junction structure in photocatalysts is a wise approach to promote photocatalytic performance, as phase junctions possess the potential to inhibit the recombination of photoinduced charge carriers. Here, Bi4V2O11 nanofibers with an α−β phase junction are fabricated via electrospinning with subsequent calcination. Electrospinning offers the opportunity to keep α-Bi4V2O11 from transforming into β-Bi4V2O11 completely due to an electrospinning retardation effect, leading to the formation of an α−β Bi4V2O11 phase junction. Furthermore, the α−β Bi4V2O11 phase junction realizes a well-established type-II band alignment. Photoelectrochemical measurements and photoluminescence spectroscopic investigations demonstrate that the phase junction structure has a significant impact on the separation and transfer of photogenerated electrons and holes. Thus, the α−β phase junction on Bi4V2O11 holds the key to achieving promoted efficiency in the photocatalysis process. because of its potential application in photocatalysis.21 To date, single-phase Bi4V2O11 has been synthesized and studied generally. Nevertheless, single-phase Bi4V2O11 cannot avoid the intrinsic drawback of rapid recombination of photoinduced carriers. Inspired by the progress in phase junction semiconductor photocatalysts, it is envisaged that construction of a phase junction by combining two different crystalline phases of Bi4V2O11 could lead to enhanced photocatalytic activity. Construction of a phase junction on Bi4V2O11 could address the intrinsic drawbacks of Bi4V2O11 for promoted photocatalytic performance without relying on extra semiconductors. However, to the best of our knowledge, because of the difficulty in maintaining a thermodynamically metastable phase, no attempts have been reported for the construction of a phase junction on Bi4V2O11. Because electrospinning is likely conducive for obtaining a thermodynamically metastable phase for constructing a target phase junction, fabrication of a phase junction on Bi4V2O11 via electrospinning might be a feasible approach.22 Meanwhile, one-dimensional electrospun fibers are ideal matrixes to fabricate junctions with nanosized interfaces for adequate utilization of junction structure.20 Furthermore, junctions on electrospun fibers with one-dimensional nanostructure possess the combined advantages of rapid diffusion of photoinduced carrier transfer along the long direction, large surface area for increased reaction sites, and outstanding light harvesting, which
1. INTRODUCTION Construction of junctions (such as p−n junctions) between two different semiconductors has been proven to be a reliable strategy for promoting the separation and transfer of photoinduced charge in photocatalysis.1 Therefore, there has been a recent upsurge in the number of studies on the fabrication of photocatalysts with junction structure.2−4 Very recently, the strategy of fabrication junctions has been further advanced by constructing phase junctions between two different crystalline phases of a single semiconductor, such as anatase−rutile or anatase−B TiO2,5−7 α−β Ga2O3,8−10 α−β Bi2O3,11 monoclinic−tetragonal BiVO4,12,13 sphalerite−wurtzite ZnS,14 and monazite monoclinic−monoclinic BiPO415 phase junctions. Well-fabricated phase junctions can endow photocatalysts with enhanced activity due to the enhanced feasibility of the formation of favorable band alignments, well-matched interfaces, and improved photo harvesting.7,8,16 As a result, because of the bright prospect for photocatalysis, great effort should be put toward the development of novel photocatalysts with phase junctions to meet social requirements. As a nontypical stoichiometric bismuth vanadate, Bi4V2O11 inherits the unique properties in band structure of BiVO4 photocatalysts, determining its excellent photocatalytic properties.17−19 In addition, this nontypical stoichiometric photocatalyst possesses stronger mobility of charge carriers, more effective capacity for separation of photoinduced electron−hole pairs, and a much narrower bandgap (∼2.2 eV) relative to those of the corresponding typical stoichiometric photocatalyst.20 Therefore, Bi4V2O11 has been investigated in recent years © XXXX American Chemical Society
Received: January 19, 2016
A
DOI: 10.1021/acs.inorgchem.6b00130 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Figure 1. XRD patterns of BVO-440, BVO-500, and BVO-550 (a) from 10° to 70° and (b) magnified from 28.0° to 30.0°. field emission scanning electron microscope (FE-SEM). The operating voltage was set to 20 kV, and the samples were prepared by dropping the preultrasonic-dispersed (10 min) ethanol turbid liquid onto a silicon chip. Transmission electron microscopy (TEM) and highresolution TEM (HRTEM) of the hierarchical structures were carried out on an FEI Tecnai G2 S-Twin operating at 300 kV. The thermal stability was determined by a SETARAM DSC-141 under a stream of air and a heating rate of 10 K min−1. UV−vis diffuse reflectance spectra were acquired by a spectrophotometer (HITACHI UH-4150), and BaSO4 was used as the reflectance standard. X-ray photoelectron spectroscopy (XPS) was accomplished using a Thermo Scientific ESCALAB 250Xi X-ray photoelectron spectrometer with a pass energy of 20.00 eV and an Al Kα excitation source (1486.6 eV). Timeresolved fluorescence decay spectra were obtained on a HORIBA FluoroMax-4 spectrofluorometer monitored under 301 nm laser excitation. 2.2. Photocatalytic and Photoelectrochemical Measurements. The photocatalytic activities of the samples were evaluated by photocatalytic reduction of Cr(VI), degradation of MB, and nitrogen fixation under visible light using a 300 W Xe lamp (Trusttech PLS-SXE 300, Beijing) with a cutoff filter of λ ≥ 400 nm. The reduction of Cr(VI) was irradiated with a visible light source in the presence of citric acid. In a typical process, 0.05 g of the as-prepared sample as photocatalyst was added to 100 mL of Cr(VI) solution (10 mg L−1, which was based on Cr in a dilute K2Cr2O7 solution). After the photocatalyst was dispersed in the solution with an ultrasonic bath for 5 min, the solution was stirred for 55 min in the dark to reach adsorption equilibrium. Before being exposed to visible-light irradiation, 0.05 g of citric acid was added to the solution. The photocatalyst was removed by centrifugation at given time intervals, and the Cr(VI) concentration was determined at 540 nm by the diphenylcarbazide (DPC) method using UV−vis spectroscopy.28,29 The initial concentration of MB was 10 mg/L; 0.05 g of photocatalysts was added to 100 mL of MB solution. Before the photodegradation experiments were initiated, the suspension was magnetically stirred in the dark for 55 min to reach adsorption−desorption equilibrium and sonicated for 5 min. Once the photodegradation experiment started, 4 mL aliquots of solution were sampled at given time intervals and centrifuged to remove the photocatalysts. The filtrates were analyzed by the variations of the absorption-band maximum (664 nm). For photocatalytic nitrogen fixation, 0.05 g of photocatalyst and 2 mL of isopropanol aqueous solution (10 g/L) were added to 100 mL of distilled water in a reactor. Water circulation was used to maintain the reactor at 25 °C. After being sonicated for 5 min, the mixture was continuously stirred in the dark for 20 min with high-purity bubbled N2. Once the N2 fixation experiment began, 10 mL of the solution was removed each 1 h, and the photocatalyst was removed by centrifugation. The concentration of liquid (NH3) was monitored by Nessler’s reagent colorimetry at 420 nm with a HITACHI UH-5300 UV−vis spectrometer.30 The photocurrent transient response measurement was carried out at the open circuit potential. The light source employed was a 300 W
are all favorable for photocatalytic reactions.23−25 Hence, it is of significant interest to fabricate Bi4V2O11 nanofibers via electrospinning for achieving the construction of a phase junction. Herein, we present an α−β phase junction on onedimensional Bi4V2O11 nanofibers fabricated by electrospinning. A reasonable electrospinning retardation effect is proposed, which determines the successful construction of the phase junction on the Bi4V2O11 nanofibers. The key role of the α−β phase junction on Bi4V2O11 is to realize efficient separation and transfer of photoinduced carriers of which the lifetime is prolonged. The efficient separation and transfer of photoinduced carriers is shown to be due to the well-matched band alignment of the α−β phase junction on Bi4V2O11. Thus, the Bi4V2O11 nanofibers with an α−β phase junction exhibit remarkably higher photocatalytic activity than α-Bi4V2O11 or βBi4V2O11.
2. EXPERIMENTAL SECTION 2.1. Preparation and Characterizations. All reagents were received from Aladdin Chemical Co., Ltd., and used without further purification. In a typical synthesis, 0.970 g (2.0 mmol) of Bi(NO3)3· 5H2O was dissolved in 10 mL of N,N-dimethylformamide (DMF) with magnetic stirring at room temperature, and then, 6 mL of acetic acid was added to the mixture. After the Bi(NO3)3·5H2O dissolved, 0.117 g (1.0 mmol) of NH4VO3 was added slowly to the solution. Then, 10 mL of absolute ethanol was added, followed by the slow addition of 2.0 g of PVP (Mw ≈ 1300000). The spinnable precursor sols were finally obtained after continuous stirring for 12 h. All of the precursor sols were transferred to a syringe attached to a needle with an inner diameter of 0.901 mm. The positive voltage applied to the tip was 18 kV, and the distance between the needle tip and the collector was 14 cm. The feeding rate was controlled at 0.5 mL h−1, and the humidity level was maintained around 25% RH. The as-spun fibers were collected from a collector plate (Al foil). As reported, Bi4V2O11 possesses three principal polymorphic forms, α (monoclinic), β (orthorhombic), and γ (tetragonal), and the corresponding phase transformation temperatures are 445 and 567 °C.26,27 Thus, to fabricate Bi4V2O11 nanofibers with an α−β phase junction, the as-spun fibers were calcinated at 500 °C in air for 30 min at a heating rate of 1 °C min−1 (denoted as BVO-500). Pure α and β phase Bi4V2O11 nanofibers were obtained at 440 and 550 °C (denoted as BVO-440 and BVO-550, respectively). In addition, as-spun fibers were also calcinated at 460, 480, 520 °C (denoted as BVO-460, BVO-480, and BVO-520, respectively). The structures of the obtained samples were confirmed by X-ray diffraction (XRD) on a Rigaku D/max-2000 diffractometer with Cu Kα radiation (λ = 0.15406 nm). Diffraction patterns were collected from 10° to 90° at a speed of 4° min−1 with a scan width of 0.02°. The morphologies of the samples were observed by a Camscan MX2600FE B
DOI: 10.1021/acs.inorgchem.6b00130 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
Bi4V2O11 reaction calcinated at 500 °C (without PVP, denoted as BVO-500 SSR) are also provided in Figure S2, where the XRD peaks and Raman bands can be well-assigned to βBi4V2O11, demonstrating that the phase junction cannot be fabricated by a solid-state reaction under the same conditions. The morphology and microstructure of BVO-500 nanofibers are displayed in Figure 3. Figure 3a shows a fairly smooth and
Xe light source. A CHI604C electrochemical working station with a standard three-compartment cell was employed to measure the photoelectrochemical characteristics of the samples. FTO glass coated with photocatalysts was used as the working electrode, and a piece of Pt sheet, Ag/AgCl electrode, and 0.5 M sodium sulfate served as the counter electrode, reference electrode, and electrolyte, respectively.
3. RESULTS AND DISCUSSION XRD analysis is carried out to identify the crystalline phase of Bi4V2O11 nanofibers calcinated at different temperatures. As observed in Figure 1a, all of the samples are assigned to Bi4V2O11 without any impurity. The magnified pattern (Figure 1b) reveals that the strongest peak of Bi4V2O11 nanofibers calcinated at 440 °C (BVO-440) located at 28.66° is wellindexed to monoclinic Bi4V2O11 (α-Bi4V2O11, JCPDS: 821481). However, the strongest peak of Bi4V2O11 nanofibers calcinated at 550 °C (BVO-550) is shifted to 28.57°, which is assigned to orthorhombic Bi4V2O11 (β-Bi4V2O11, JCPDS: 420135). The shift of the diffraction peak identifies the occurrence of a phase transformation. In addition, the corresponding peak of Bi4V2O11 nanofibers calcinated at 500 °C (BVO-500) is located between the two peaks, which furnishes the possibility of introducing an α−β Bi4V2O11 mixed phase in BVO-500. With the aim of further verifying the existence of the α−β Bi4V2O11 mixed phase in BVO-500, Raman spectra were obtained (Figure 2). A Raman band in BVO-440 around 826
Figure 3. SEM images of BVO-500 fibers (a) before and (b) after calcination; TEM (c) and HRTEM (d) images of BVO-500 fibers.
uniform surface of the fibers before calcination. The average diameter of as-spun nanofibers is approximately 250 ± 50 nm, and the length is approximately several micrometers. After annealing at 500 °C, BVO-500 retains the one-dimensional structure (Figure 3b). However, the diameter decreases to 100−200 nm due to the disappearance of organic composition. The shrinkage leads to the coarse surface of fibers. A TEM image confirms that BVO-500 nanofibers are one-dimensional with diameters of approximately 100−200 nm, which agrees well with that revealed by the SEM images (Figure 3c). In the HRTEM image (Figure 3d), the interplanar spacing of 0.304 nm assigns well to the (025) plane of α-Bi4V2O11. Meanwhile, the interplanar spacing of 0.236 nm verifies the (212) plane of β-Bi4V2O11. Obviously, two independent crystal lattices, αBi4V2O11 and β-Bi4V2O11, are observed simultaneously with a perfect interfacial boundary, indicating the successful construction of the phase junction. The SEM and TEM images of BVO-440 and BVO-550 are presented in Figures S3 and S4, respectively. The BVO-440 and BVO-550 nanofibers also maintain one-dimensional structures after heat treatment (Figure S3). HRTEM images shown in Figure S4 illustrate that BVO-440 and BVO-550 are assigned to α-Bi4V2O11 and βBi4V2O11, respectively. Overall, these data imply that the target phase junction could be constructed by the electrospinning method. As shown above, the construction of an α−β phase junction on Bi4V2O11 nanofibers is successfully achieved by electrospinning with subsequent calcination. Therefore, it is speculated that the maintenance of α-Bi4V2O11 could be attributed to the electrospinning method. For validating this conjecture, TG and DTA measurements are carried out for asspun fibers, as shown in Figure 4. There exists approximately 85% weight loss in the range 100−500 °C in the TG curve corresponding to the decomposition of redundant ions and polymer PVP templates. In addition, two exothermic peaks at approximately 320 and 480 °C could be observed. The former indicates the decomposition of NO3−,33 and the latter is assigned to the complete decomposition of the main polymer chain of PVP.33 Interestingly, a small endothermic peak shows up at 490 °C, revealing that the β phase comes into being. In a
Figure 2. Raman spectra of BVO-440, BVO-500, and BVO-550.
cm−1 is assigned to α-Bi4V2O11, and a Raman band in BVO-550 around 852 cm−1 belongs to β-Bi4V2O11, indicating that BVO440 and BVO-550 are α-Bi4V2O11 and β-Bi4V2O11, respectively.31,32 Interestingly, two bands located at 824 and 848 cm−1, corresponding to the V−O band of α-Bi4V2O11 and βBi4V2O11, respectively, are simultaneously displayed in BVO500. Therefore, α-Bi4V2O11 and β-Bi4V2O11 coexist in BVO-500 in accordance with the XRD results. Moreover, the Raman spectra of the samples calcinated at different temperatures are shown in Figure S1. As the calcination temperature increased, the Raman band intensity assigned to β phase was enhanced, whereas that of α phase was weaker (480, 500, and 520 °C). When calcinated at 550 °C, only β phase was obtained. In addition, the ratio of α/β phases in the junction could be roughly calculated on the basis of the Raman band intensity. The ratio was approximately 4:3 at 480 °C, 1:1 at 500 °C, and 3:4 at 520 °C. The XRD and Raman results of the solid-state C
DOI: 10.1021/acs.inorgchem.6b00130 Inorg. Chem. XXXX, XXX, XXX−XXX
Article
Inorganic Chemistry
electrospinning method to fabricate Bi4V2O11 nanofibers can result in the construction of an α−β phase junction. For identifying the interfacial chemical bonding and surface chemical status of the phase junction, XPS measurements are taken,37 and the corresponding results are shown in Figure 6. The XPS survey spectra (Figure 6a) confirm the presence of Bi, V, and O in BVO-440, BVO-500, and BVO-550. The Bi 4f spectra of the samples are displayed in Figure 6b. Two symmetric peaks are displayed in BVO-440, and the energies are indicative of Bi3+ [E (Bi 4f7/2)] = 159.1 eV; E (Bi 4f5/2) = 164.4 eV].38 The Bi3+ peaks in BVO-500 shift approximately 0.2 and 0.4 eV toward lower binding energies relative to those of BVO-440 and BVO-550, respectively. Interestingly, two new peaks appear at 159.7 (159.8) and 165.0 eV (165.1 eV) in BVO-500 (BVO-550), which are assigned to Bi5+.39,40 On the basis of the defect relationship, the formation of Bi5+ could make the intrinsic oxygen vacancies fill with oxygen atoms. Accordingly, coexistence of Bi3+ and Bi5+ is suggested in BVO500 and BVO-550. Additionally, BVO-550 possesses more Bi5+ compared with that of BVO-500. The presence of Bi5+ could be ascribed to the fierce repulsion between oxide ions and Bi 6S2 electrons of Bi3+ ions during the phase transformation process.41,42 This results in the loss of Bi 6S2 electrons and some Bi3+ ions converting to Bi5+ ions.40 Furthermore, the lower binding energy shifts of Bi in BVO-500 and BVO-550 could be attributed to the increased electron concentration caused by the introduction of Bi5+. The V 2p of BVO-440, BVO-500, and BVO-550 is analyzed in Figure 6c and indicates that the V ions are V5+ in each of the samples.20,38 For BVO500, the V 2p peaks shift approximately 0.3 and 0.1 eV toward lower binding energies compared with those of BVO-440 and BVO-550, respectively. Likewise, as observed in Figure 6d, the O 1s peak on the curves of BVO-500 shift approximately 0.3 and 0.1 eV toward lower binding energies versus that of BVO440 and BVO-550. The shift toward lower binding energy could be attributed to the formation a Bi−O bond at the interface of the α−β phase junction.20 The interface bonding might contribute to the increase of electron concentration, enhancing the electron screening effect.43 These consequences demonstrate that interfacial chemical bonding is formed to construct a phase junction between α-Bi4V2O11 and β-Bi4V2O11, which could facilitate the separation and transfer of interfacial photoinduced carriers. For the purpose of investigating the impact of the constructed α−β Bi4V2O11 phase junction on the separation and transfer of photoinduced carriers, photocurrent measurements are carried out on BVO-440, BVO-500, and BVO-550. As shown in Figure 7a, all electrodes are prompt in generating a fast and reproducible photocurrent response to each on/off cycle. Compared with BVO-440 and BVO-550, BVO-500 exhibits significantly enhanced photocurrent intensity, which infers that the construction of the α−β phase junction on Bi4V2O11 could effectively promote the separation and transfer of photoinduced carriers. For further confirming the outstanding capability of charge separation and transfer in an α−β Bi4V2O11 phase junction, time-resolved fluorescence decay spectra are performed. Figure 7b displays the time-resolved fluorescence decay spectra of the samples. All of the fluorescence intensities for BVO-440, BVO-500, and BVO550 decay exponentially. Clearly, in contrast to the decay curves of α-Bi4V2O11 and β-Bi4V2O11, the decay kinetics of the α−β Bi4V2O11 phase junction is relatively slow. It is believed that the increased lifetimes of carriers in BVO-500 are associated with
Figure 4. TG-DTA analysis of as-spun fibers.
DTA curve of BVO-500 SSR (Figure S5), the corresponding endothermic peak exists at ∼450 °C. These results imply that the phase-transition of α → β in BVO-550 electrospun fibers exhibits hysteresis, thus giving rise to the formation of an α−β Bi4V2O11 phase junction. On the basis of the results presented above and electrospinning properties, a electrospinning retardation effect for the construction of the α−β Bi4V2O11 phase junction is proposed (Figure 5). First, because of the unique process of electro-
Figure 5. Formation process of the α−β Bi4V2O11 phase junction.
spinning, Bi and V ions could be surrounded well by polymer template. At the low temperature (