Template-Free Synthesis of Monoclinic BiVO4 with Porous Structure and Its High Photocatalytic Activity

Monoclinic BiVO4 photocatalysts with porous structures were synthesized by a two-step approach without assistance of any templates. The as-prepared samples were characterized by X-ray diffraction pattern (XRD), scanning electron microscopy (SEM), Brunauer–Emmett–Teller (BET), ultraviolet–visible (UV–vis) diffuse reflectance spectroscopy (DRS), photocurrent responses, and electrochemical impedance spectra (EIS). It is found that the as-prepared BiVO4 samples had a porous structure with aperture diameter of 50–300 nm. Moreover, the BET specific surface area of the porous BiVO4-200 °C sample reaches up to 5.69 m2/g, which is much higher than that of the sample of BiVO4 particles without porous structure. Furthermore, a possible formation mechanism of BiVO4 with porous structure was proposed. With methylene blue (MB) as a model compound, the photocatalytic oxidation of organic contaminants in aqueous solution was investigated under visible light irradiation. It is found that the porous BiVO4-200 °C sample exhibits the best photocatalytic activity, and the photocatalytic rate constant is about three times of that of the sample of BiVO4 particles without porous structure. In addition, the photocurrent responses and electrochemical impedance spectra strongly support this conclusion.


Introduction
It is widely accepted that photocatalysis is a promising application technology in the fields of air cleaning and water purification [1]. However, the traditional TiO 2 photocatalyst could not make use of visible light which accounts for 45% of solar spectrum due to a large band gap (3.2 eV). As a consequence, the development of new and efficient visible-light-driven photocatalysts attracts much attention.
BiVO 4 is one of the typical complex oxides with narrow band-gap [2,3]. As an n-type semiconductor with a direct band gap of 2.4 eV [4], BiVO 4 could absorb ample visible light and is stable, nontoxic, and relatively cheap [5]. It is generally considered that BiVO 4 has three crystal systems of zircon-tetragonal, scheelite tetragonal, and scheelite-monoclinic phases [6]. However, only the monoclinic phase possesses an excellent visible-light photocatalytic property [7][8][9], which has been applied for degradation of various pollutants, e.g., organic dye [8,10], phenol [11], benzene [12], 4-n-alkylphenols [13,14], and so on. In past years, monoclinic BiVO 4 with various morphologies, such as nanoparticles [15][16][17], nanorods [18], nanotubes [19], nanosheets [4,20], hierarchical nanostructure [21], has been successfully synthesized. Apart from these nanostructure morphologies, porous structures always show an improvement of the photocatalytic activity due to the large surface area and multiple scattering effects [22][23][24]. In recent years, mesoporous BiVO 4 has been prepared by using templates or surfactants, such as silica (KIT-6) [25], silica aerogel [26], colloidal carbon spheres [27] as hard templates, and dodecylamine (DA), oleylamine (OL) or oleic acid (OA) as a surfactant [28], triblock copolymer P123 as a surfactant [11], and so on. Although these templates and surfactants could be removed by NaOH treatment or calcination process, it should be pointed out that the residual templates and surfactants may introduce impurities into the BiVO 4 materials and will in any case complicate the separation process of photo-induced electrons and holes. Moreover, the additive of templates and surfactants may increase the cost. However, realization of controlled one-step template-free or surfactant-free fabrication of porous BiVO 4 nanostructures remains a great challenge. Therefore, the development of a cheap and template-free method suitable for synthesis of high-purity porous BiVO 4 structure is a significant research objective. Recently, Ying et al. reported a template-free approach to fabricate monoclinic BiVO 4 hierarchical microtubes self-assembled by several nanowires [29]. Ma et al. prepared monoclinic BiVO 4 hollow nanospheres via a template-free method using citric acid as chelating agent [30]. Cheng et al. fabricated monoclinic BiVO 4 hollow microspheres by a one-pot template-free hydrothermal method [31]. Li et al. reported a hydrothermal method for synthesizing of monoclinic BiVO 4 hollow microspheres from a novel vanadium source K 6 V 10 O 28¨9 H 2 O without any additives but just through adjusting the reaction pH [32].
In this work, monoclinic BiVO 4 with porous structure was synthesized through a novel two-step method without assistance of any templates. Compared with other template-free methods for preparing BiVO 4 with hollow structure [29][30][31][32], our method is possibly not the best, but it provides a novel approach for the fabrication of monoclinic BiVO 4 with porous structure by using the template-free approach. At the same time, the BiVO 4 with porous structure prepared by our method exhibited a larger BET surface area compared with the previously reported BiVO 4 hierarchical microtubes [29] and BiVO 4 hollow microspheres [31]. Moreover, a possible formation mechanism of BiVO 4 with porous structure was proposed. Furthermore, the photocatalytic performance of as-prepared monoclinic BiVO 4 products with porous structure was evaluated by examining the degradation of methylene blue (MB) under visible light irradiation.

Experimental Materials
The following analytically pure chemicals were used: bismuth nitrate (Bi(NO 3

Synthesis
Firstly, the precursor was synthesized by a modified homogeneous precipitation method. Typically for the precursor, 0.002 mol of Bi(NO 3 ) 3¨5 H 2 O was dissolved into 30 mL of deionized water to form solution A, and 0.05 mol of urea was dissolved into 30 mL of deionized water to form solution B. Subsequently, solution B was added into solution A under stirring. After stirring for half an hour, the mixed solution was heated to 90˝C for 2 h in the water bath. The resulting suspension was separated by centrifugation and collected after washing with deionized water three times to obtain the precursor. Secondly, BiVO 4 with porous structure was synthesized as follows: the as-obtained precursor was dispersed into 10 mL of deionized water by ultrasonication. Then, 0.002 mol of NH 4 VO 3 was dissolved in deionized water (10 mL) with stirring and heating, and dripped into the above precursor suspension followed by further stirring at 70˝C for half an hour in the water bath. After cooling to room temperature, the final pH value was adjusted to be about 7.5 by ammonia solution. Afterwards, the suspension was transferred into a Teflon-lined autoclave of 50 mL capacity, and kept at various temperatures (i.e., 180, 200, 220˝C) for 12 h followed by cooling to ambient temperature naturally, respectively. The precipitate was separated from the reaction media by centrifugation, and then washed with deionized water and ethanol several times, and finally dried overnight at 80˝C to obtain the BiVO 4 product with porous structure. For comparison, BiVO 4 particles without porous structure were prepared as follows: 0.002 mol of Bi(NO 3 ) 3¨5 H 2 O was dissolved into 30 mL of deionized water to form solution A. Then, 0.002 mol of NH 4 VO 3 was dissolved in deionized water (10 mL) with stirring and heating, and dripped into the solution A followed by further stirring at 70˝C for half an hour in the water bath, and the final pH value of the suspension was adjusted to be about 7.5 after cooling to room temperature. Subsequently, the suspension was transferred into a Teflon-lined autoclave and kept at 200˝C for 12 h followed by cooling to ambient temperature naturally. The precipitate was separated by centrifugation, and then washed with deionized water and ethanol several times, and finally dried overnight at 80˝C to obtain the sample of BiVO 4 particles.

Characterization
X-ray diffraction (XRD) measurement was carried out with a D/max-2400 diffractometer (Rigaku, Tokyo, Japan) using Cu-Ka radiation. The morphologies of samples were examined by scanning electron microscopy (SEM, Hitachi S-4800, Hitachi High-Technologies Corp., Tokyo, Japan). The Brunauer-Emmett-Teller (BET) specific surface area and Barrett-Joyner-Halenda (BJH) pore distribution of the samples were measured using a Micromeritics ASAP 2020-M instrument (Micromeritics Instrument Corp., Norcross, GA, USA). Ultraviolet-visible diffuse reflectance spectra (UV-vis DRS) were measured using a Perkin Elmer 950 spectrometer (Perkin Elmer, Norwalk, CT, USA), while BaSO 4 was used as a reference. The photocurrent response and electrochemical impedance spectra of as-prepared photocatalysts was measured on an electrochemical analyzer (CHI660E) (Shanghai Chenhua Instruments Co., Ltd., Shanghai, China) in a standard three-compartment cell using 0.5 M Na 2 SO 4 (pH = 6.8) solution as the electrolyte. For the preparation of working electrodes for electrochemical measurements, a homogeneous catalyst ink was first prepared by dispersing 4 mg of catalyst material and 80 µL of a 5 wt % Nafion solution in 2 mL of H 2 O by ultrasonication, and then 400 µL of catalyst ink dispersions was drop-coated directly onto the precleaned indium tin oxide (ITO) glass surface by microsyringe and placed on a hot plate to speed drying. The surface of working electrode exposed to the electrolyte was a circular film with the geometrical surface areas of 4 cm 2 . Platinum foil was used as counter electrode and Ag/AgCl electrode as the reference electrode. A 350 W Xe lamp with a cutoff filter of 420 nm was used for excitation. The photo-responses of the photocatalysts as light on and off were measured at 0.0 V, and the electrochemical impedance spectra were measured at 0.0 V. A sinusoidal ac perturbation of 5 mV was applied to the electrode over the frequency range of 10-10 4 Hz.

Evaluation of Photocatalytic Activity
The photocatalytic performance of as-prepared samples was evaluated by examining the degradation of MB under visible light irradiation. In a typical photocatalytic degradation process, 20 mg of photocatalyst was suspended in the MB solution (10 mg/L, 90 mL). Before irradiation, the suspensions were stirred magnetically in the dark for 60 min to ensure the establishment of adsorption-desorption equilibrium. A 350 W Xe lamp with a cutoff filter of 420 nm was employed for the visible-light irradiation source and positioned 17 cm away from the reactor to start the photocatalytic reaction under continuous stirring conditions. A certain volume of suspension was withdrawn at selected times for analysis. After recovering the photocatalyst by centrifugation (at 15,000 rpm for 10 min), the concentration of MB solution at some points was analyzed by measuring the light absorption of clear MB solution at 664 nm using a spectrophotometer (UV-2450; Shimadzu, Kyoto, Japan). The percentage of degradation was calculated by C/C 0 . Here, C is the concentration of remaining MB solution at each irradiated time interval, while C 0 is the initial concentration.

XRD Analysis
The phase characteristics of the precursor were studied by XRD, as shown in Figure Figure 1, it can be seen that all of the XRD patterns of BiVO 4 products prepared at hydrothermal temperatures of 180, 200, and 220˝C are in good agreement with the standard data of monoclinic BiVO 4 (JCPDS no. 14-0688), with no other characteristic peaks observed. This indicates that monoclinic BiVO 4 products with well-crystallized high-purity were successfully obtained via the two-step processes. Moreover, it is found that monoclinic phase was achieved in the sample of BiVO 4 particles without porous structure.

XRD Analysis
The phase characteristics of the precursor were studied by XRD, as shown in Figure S1 in the Supplementary Materials. The diffraction peaks of the precursor can be indexed as [Bi6O5(OH)3](NO3)5·3H2O, agreeing well with the JCPD Standard card no. 48-0575. Based on the [Bi6O5(OH)3](NO3)5·3H2O precursor, BiVO4 products was synthesized. From Figure 1, it can be seen that all of the XRD patterns of BiVO4 products prepared at hydrothermal temperatures of 180, 200, and 220 °C are in good agreement with the standard data of monoclinic BiVO4 (JCPDS no. 14-0688), with no other characteristic peaks observed. This indicates that monoclinic BiVO4 products with wellcrystallized high-purity were successfully obtained via the two-step processes. Moreover, it is found that monoclinic phase was achieved in the sample of BiVO4 particles without porous structure.

SEM Analysis
From Figure 2a,b, it is observed that the precursor displays a plate-like morphology. Moreover, the as-prepared sample of BiVO4 particles shows particle morphology with particle size of 200-500 nm ( Figure 2c). Furthermore, the low magnification SEM image of as-obtained BiVO4 products prepared at the hydrothermal temperatures of 180 °C is presented in Figure S2 in the Supplementary Materials. It could be seen that the obtained BiVO4 product shows a porous structure. Large magnification SEM images of BiVO4 products prepared at the hydrothermal temperatures of 180 °C (Figure 2d Figure 2f) reveal that the all of the BiVO4 products exhibit a porous structure composed of nanoflakes with thicknesses of 50-100 nm, which means that BiVO4 with porous structure was successfully achieved by our two-step method. Further observation indicates that the aperture diameter of the BiVO4 with porous structure is about 50-300 nm, which is near that of previously reported BiVO4 hollow nanospheres (the cavity diameter of 40 nm ) [30] and much smaller than that of BiVO4 hierarchical microtubes (the outer diameter of 1.2-1.5 μm) [29] and BiVO4 hollow microspheres (the diameter of 2-4 μm). [31].

SEM Analysis
From Figure 2a,b, it is observed that the precursor displays a plate-like morphology. Moreover, the as-prepared sample of BiVO 4 particles shows particle morphology with particle size of 200-500 nm ( Figure 2c). Furthermore, the low magnification SEM image of as-obtained BiVO 4 products prepared at the hydrothermal temperatures of 180˝C is presented in Figure S2 Figure 2f) reveal that the all of the BiVO 4 products exhibit a porous structure composed of nanoflakes with thicknesses of 50-100 nm, which means that BiVO 4 with porous structure was successfully achieved by our two-step method. Further observation indicates that the aperture diameter of the BiVO 4 with porous structure is about 50-300 nm, which is near that of previously reported BiVO 4 hollow nanospheres (the cavity diameter of 40 nm ) [30] and much smaller than that of BiVO 4 hierarchical microtubes (the outer diameter of 1.2-1.5 µm) [29] and BiVO 4 hollow microspheres (the diameter of 2-4 µm). [31].

Formation Mechanism
A possible formation mechanism of as-obtained BiVO4 with porous structure was proposed, as presented in Scheme 1. Firstly, the [Bi6O5(OH)3](NO3)5·3H2O precursor with a plate-like morphology was synthesized in the water bath of 90 °C by a modified homogeneous precipitation method using Bi(NO3)3·5H2O as the raw material, deionized water as the solvent, and urea as the precipitating agent. Then, NH4VO3 was dissolved in deionized water and the pH value was adjusted to be about 7.5 by ammonia solution. In this process, VO3 − might react with OH − resulting from ammonia solution to produce VO4 3− and H + according to Equation (1). Subsequently, the H + started to erode the plate-like [Bi6O5(OH)3](NO3)5·3H2O precursor from the surface gradually, giving rise to a large amount of Bi 3+ according to Equation (2), which then reacted with the surrounding VO4 3− to produce a thin layer of BiVO4 on the surface of the plate-like [Bi6O5(OH)3](NO3)5·3H2O precursor (Equation (3)). Finally, in the hydrothermal process, the reaction with a fast diffusion rate of Bi 3+ inside this BiVO4 layer and the slow diffusion rate of VO4 3− outside the BiVO4 layer led to the formation of porous structure. This explanation was called the Kirkendall effect, which deals with the movement of the interface between diffusion couples [33]. A similar Kirkendall phenomenon was reported in the preparation of YVO4 product [34].
Scheme 1. Schematic illustration for the possible formation mechanism of BiVO4 with porous structure.

Formation Mechanism
A possible formation mechanism of as-obtained BiVO 4 with porous structure was proposed, as presented in Scheme 1. Firstly, the [Bi 6  (1)

Formation Mechanism
A possible formation mechanism of as-obtained BiVO4 with porous structure was proposed, as presented in Scheme 1. Firstly, the [Bi6O5(OH)3](NO3)5·3H2O precursor with a plate-like morphology was synthesized in the water bath of 90 °C by a modified homogeneous precipitation method using Bi(NO3)3·5H2O as the raw material, deionized water as the solvent, and urea as the precipitating agent. Then, NH4VO3 was dissolved in deionized water and the pH value was adjusted to be about 7.5 by ammonia solution. In this process, VO3 − might react with OH − resulting from ammonia solution to produce VO4 3− and H + according to Equation (1). Subsequently, the H + started to erode the plate-like [Bi6O5(OH)3](NO3)5·3H2O precursor from the surface gradually, giving rise to a large amount of Bi 3+ according to Equation (2), which then reacted with the surrounding VO4 3− to produce a thin layer of BiVO4 on the surface of the plate-like [Bi6O5(OH)3](NO3)5·3H2O precursor (Equation (3)). Finally, in the hydrothermal process, the reaction with a fast diffusion rate of Bi 3+ inside this BiVO4 layer and the slow diffusion rate of VO4 3− outside the BiVO4 layer led to the formation of porous structure. This explanation was called the Kirkendall effect, which deals with the movement of the interface between diffusion couples [33]. A similar Kirkendall phenomenon was reported in the preparation of YVO4 product [34].
Scheme 1. Schematic illustration for the possible formation mechanism of BiVO4 with porous structure. Scheme 1. Schematic illustration for the possible formation mechanism of BiVO 4 with porous structure.

Nitrogen Adsorption-Desorption
The porous structure was further confirmed and characterized as follows. Figure 3 shows the nitrogen adsorption-desorption isotherms of BiVO 4 particles and BiVO 4 samples prepared at various hydrothermal temperatures. It can be seen that all the as-prepared samples show a type H3 hysteresis loop according to IUPAC classification [35], indicating the presence of slit-shaped pores or voids in the aggregates of plate like particles, which is consistent with the SEM results in Figure 2. Furthermore, the observed hysteresis loops of the as-prepared samples approach P/P 0 = 1, suggesting the presence of macropores (>50 nm) [36]. These results suggest that pores are presented in the samples. The inset of Figure 3 shows the corresponding BJH pore size distributions of these samples. It is observed that the porous BiVO 4 samples display two peaks around 4 and 60 nm, which could be attributed to the mesopores and macropores, respectively. On the contrary, there is no peak of pore size distribution for the sample of BiVO 4 particles, suggesting there is no apparent pore. Table 1 shows the BET specific surface area and BJH pore volume of these samples. It can be seen that the porous BiVO 4 -200˝C sample shows the largest BJH pore volume and BET specific surface area (5.69 m 2 /g), which is much higher than that of the sample of BiVO 4 particles without porous structure (1.85 m 2 /g). Moreover, it is found that the BET specific surface area of porous BiVO 4 -200˝C sample is much higher than that of the previously reported BiVO 4 hierarchical microtubes (0.3 m 2 /g) [29] as well as BiVO 4 hollow microspheres (1.05 m 2 /g) [31] and is near that of BiVO 4 hollow spheres (5.85 m 2 /g) prepared by employing colloidal carbon spheres (CCSs) as hard templates [27]. The higher BET specific surface area is beneficial for the photocatalytic performance. In addition, it is noted that the porous BiVO 4 -220˝C sample exhibits smaller BET specific surface area compared with porous BiVO 4 -200˝C sample, which may be due to the higher hydrothermal temperature (i.e., 220˝C) inducing the collapse of porous structures.

Nitrogen Adsorption-Desorption
The porous structure was further confirmed and characterized as follows. Figure 3 shows the nitrogen adsorption-desorption isotherms of BiVO4 particles and BiVO4 samples prepared at various hydrothermal temperatures. It can be seen that all the as-prepared samples show a type H3 hysteresis loop according to IUPAC classification [35], indicating the presence of slit-shaped pores or voids in the aggregates of plate like particles, which is consistent with the SEM results in Figure 2. Furthermore, the observed hysteresis loops of the as-prepared samples approach P/P0 = 1, suggesting the presence of macropores (>50 nm) [36]. These results suggest that pores are presented in the samples. The inset of Figure 3 shows the corresponding BJH pore size distributions of these samples. It is observed that the porous BiVO4 samples display two peaks around 4 and 60 nm, which could be attributed to the mesopores and macropores, respectively. On the contrary, there is no peak of pore size distribution for the sample of BiVO4 particles, suggesting there is no apparent pore. Table 1 shows the BET specific surface area and BJH pore volume of these samples. It can be seen that the porous BiVO4-200 °C sample shows the largest BJH pore volume and BET specific surface area (5.69 m 2 /g), which is much higher than that of the sample of BiVO4 particles without porous structure (1.85 m 2 /g). Moreover, it is found that the BET specific surface area of porous BiVO4-200 °C sample is much higher than that of the previously reported BiVO4 hierarchical microtubes (0.3 m 2 /g) [29] as well as BiVO4 hollow microspheres (1.05 m 2 /g) [31] and is near that of BiVO4 hollow spheres (5.85 m 2 /g) prepared by employing colloidal carbon spheres (CCSs) as hard templates [27]. The higher BET specific surface area is beneficial for the photocatalytic performance. In addition, it is noted that the porous BiVO4-220 °C sample exhibits smaller BET specific surface area compared with porous BiVO4-200 °C sample, which may be due to the higher hydrothermal temperature (i.e., 220 °C) inducing the collapse of porous structures.

UV-Vis Diffuse Reflectance Spectroscopy
The optical absorption property is a key factor controlling the photocatalytic activity of a catalyst. Figure 4a shows the comparison of UV-vis absorption spectra of the sample of BiVO 4 particles and porous BiVO 4 -200˝C sample. It is obvious that both of these samples exhibit strong absorption in the visible light range. The band gap energy (Eg) of these two samples was calculated from the absorption data using Equation (4) [37]: αhv " Aphv´Egq n (4) where α is absorption coefficient; A is the absorption constant and n is a constant which depends on the probability of transition; it takes the values 1/2 and 2 for direct allowed and indirect allowed transitions, respectively. Because BiVO 4 has an direct band-gap [38], the energy gap (Eg) of the samples was estimated from the intercept of the tangent in the plots of (αhv) 2 versus photon energy hv.
The calculated Eg values of both samples of BiVO 4 particles and porous BiVO 4 are 2.41 eV, as shown in Figure 4b. This result indicates that both BiVO 4 particles and porous BiVO 4 sample possess the same band gap, suggesting the morphology has little or no effect on the band gap of as-prepared samples.

UV-vis Diffuse Reflectance Spectroscopy
The optical absorption property is a key factor controlling the photocatalytic activity of a catalyst. Figure 4a shows the comparison of UV-vis absorption spectra of the sample of BiVO4 particles and porous BiVO4-200 °C sample. It is obvious that both of these samples exhibit strong absorption in the visible light range. The band gap energy (Eg) of these two samples was calculated from the absorption data using Equation (4) [37]: where α is absorption coefficient; A is the absorption constant and n is a constant which depends on the probability of transition; it takes the values 1/2 and 2 for direct allowed and indirect allowed transitions, respectively. Because BiVO4 has an direct band-gap [38], the energy gap (Eg) of the samples was estimated from the intercept of the tangent in the plots of (αhv) 2 versus photon energy hv. The calculated Eg values of both samples of BiVO4 particles and porous BiVO4 are 2.41 eV, as shown in Figure 4b. This result indicates that both BiVO4 particles and porous BiVO4 sample possess the same band gap, suggesting the morphology has little or no effect on the band gap of as-prepared samples.

Photocatalytic Activity and Electrochemical Analysis
Before the photocatalytic reaction, the sample suspension was stirred for 1 h in the dark to reach the adsorption-desorption equilibrium. The adsorption isotherms of MB solutions for as-prepared samples are shown in Figure S3. It is obvious that the porous BiVO4-200 °C sample exhibits higher adsorption ability than other samples, which is in good agreement with the BET specific surface area. The photocatalytic behaviors of these as-prepared samples were then studied for degradation of MB solutions under visible light irradiation at room temperature. As shown in Figure 5a, the photocatalytic degradation efficiency of MB solutions follows the order: porous BiVO4-200 °C > porous BiVO4-220 °C > porous BiVO4-180 °C > BiVO4 particles. The variation of absorption intensity of MB solutions over these samples at different irradiation times is recorded ( Figure S4), which strongly supports the above result. Additionally, it is known that the conduction band (ECB) edge and valence band (EVB) edge of BiVO4 were about 0.42 eV and 2.6 eV, respectively, which results in a weak reduction ability and high oxidation ability [31]. Therefore, it is generally considered that the mineralization and degradation of MB were ascribed to the high oxidation ability of BiVO4 [11,30,39]. One of the reasons for higher photocatalytic activity of porous BiVO4 samples is the higher BET specific surface area. It is generally considered that the photocatalytic reactions are typically surfacebased processes and the photocatalytic efficiency is closely related to the adsorption property of dyes on the surfaces of a photocatalyst, which largely depends on the BET specific surface area. So, the higher the BET specific surface area, the stronger the adsorption of dyes on the photocatalyst surface,

Photocatalytic Activity and Electrochemical Analysis
Before the photocatalytic reaction, the sample suspension was stirred for 1 h in the dark to reach the adsorption-desorption equilibrium. The adsorption isotherms of MB solutions for as-prepared samples are shown in Figure S3. It is obvious that the porous BiVO 4 -200˝C sample exhibits higher adsorption ability than other samples, which is in good agreement with the BET specific surface area. The photocatalytic behaviors of these as-prepared samples were then studied for degradation of MB solutions under visible light irradiation at room temperature. As shown in Figure 5a, the photocatalytic degradation efficiency of MB solutions follows the order: porous BiVO 4 -200˝C > porous BiVO 4 -220˝C > porous BiVO 4 -180˝C > BiVO 4 particles. The variation of absorption intensity of MB solutions over these samples at different irradiation times is recorded ( Figure S4), which strongly supports the above result. Additionally, it is known that the conduction band (E CB ) edge and valence band (E VB ) edge of BiVO 4 were about 0.42 eV and 2.6 eV, respectively, which results in a weak reduction ability and high oxidation ability [31]. Therefore, it is generally considered that the mineralization and degradation of MB were ascribed to the high oxidation ability of BiVO 4 [11,30,39]. One of the reasons for higher photocatalytic activity of porous BiVO 4 samples is the higher BET specific surface area. It is generally considered that the photocatalytic reactions are typically surface-based processes and the photocatalytic efficiency is closely related to the adsorption property of dyes on the surfaces of a photocatalyst, which largely depends on the BET specific surface area. So, the higher the BET specific surface area, the stronger the adsorption of dyes on the photocatalyst surface, and thus the easier and faster the photocatalytic process [40]. Furthermore, the kinetics of photocatalytic reactions could be described using a first order reaction for low concentrations of MB solutions. The apparent rate constants (k, min´1) are displayed in Figure 5b, which is determined from the slopes of ln(C 0 /C) versus irradiation time. Remarkably, the photocatalytic rate constant of porous BiVO 4 -200˝C is about three times of that of the sample of BiVO 4 particles for degradation of MB dye solutions. In addition, the correlation coefficient (R 2 ) represents the correlation degree between practical photocatalytic reaction and linear fit. It is found that all of the R 2 values are above 0.9, suggesting the data fit the first kinetic model well.
Materials 2016, 9, 685 8 of 11 and thus the easier and faster the photocatalytic process [40]. Furthermore, the kinetics of photocatalytic reactions could be described using a first order reaction for low concentrations of MB solutions. The apparent rate constants (k, min −1 ) are displayed in Figure 5b, which is determined from the slopes of ln(C0/C) versus irradiation time. Remarkably, the photocatalytic rate constant of porous BiVO4-200 °C is about three times of that of the sample of BiVO4 particles for degradation of MB dye solutions. In addition, the correlation coefficient (R 2 ) represents the correlation degree between practical photocatalytic reaction and linear fit. It is found that all of the R 2 values are above 0.9, suggesting the data fit the first kinetic model well.  Figure 6a shows the transient photocurrent responses via three ON−OFF cycles that the electrodes deposited under visible light irradiation, which are correlated with the recombination efficiency of the photogenerated carriers [41][42][43]. It is found that the photocurrent efficiency is consistent with the photocatalytic activities, suggesting the separation and transfer of photoinduced electron−hole pairs are more efficient in the case of BiVO4-200 °C. The electrochemical impedance spectra provide further evidence, as shown in Figure 6b. It can been seen that the diameter of the arc radius for the BiVO4-200 °C electrode is smaller than that for the other electrodes, demonstrating it shows an enhanced separation and transfer efficiency of photogenerated e-h pairs because the smaller arc radius implies a higher efficiency of charge transfer.
Therefore, it can be concluded that the high BET specific surface area and effective separation transfer efficiency of photogenerated e-h pairs are the main reasons for the enhanced photocatalytic activity of as-prepared porous BiVO4 samples.   Figure 6a shows the transient photocurrent responses via three ON´OFF cycles that the electrodes deposited under visible light irradiation, which are correlated with the recombination efficiency of the photogenerated carriers [41][42][43]. It is found that the photocurrent efficiency is consistent with the photocatalytic activities, suggesting the separation and transfer of photoinduced electron´hole pairs are more efficient in the case of BiVO 4 -200˝C. The electrochemical impedance spectra provide further evidence, as shown in Figure 6b. It can been seen that the diameter of the arc radius for the BiVO 4 -200˝C electrode is smaller than that for the other electrodes, demonstrating it shows an enhanced separation and transfer efficiency of photogenerated e-h pairs because the smaller arc radius implies a higher efficiency of charge transfer. and thus the easier and faster the photocatalytic process [40]. Furthermore, the kinetics of photocatalytic reactions could be described using a first order reaction for low concentrations of MB solutions. The apparent rate constants (k, min −1 ) are displayed in Figure 5b, which is determined from the slopes of ln(C0/C) versus irradiation time. Remarkably, the photocatalytic rate constant of porous BiVO4-200 °C is about three times of that of the sample of BiVO4 particles for degradation of MB dye solutions. In addition, the correlation coefficient (R 2 ) represents the correlation degree between practical photocatalytic reaction and linear fit. It is found that all of the R 2 values are above 0.9, suggesting the data fit the first kinetic model well.  Figure 6a shows the transient photocurrent responses via three ON−OFF cycles that the electrodes deposited under visible light irradiation, which are correlated with the recombination efficiency of the photogenerated carriers [41][42][43]. It is found that the photocurrent efficiency is consistent with the photocatalytic activities, suggesting the separation and transfer of photoinduced electron−hole pairs are more efficient in the case of BiVO4-200 °C. The electrochemical impedance spectra provide further evidence, as shown in Figure 6b. It can been seen that the diameter of the arc radius for the BiVO4-200 °C electrode is smaller than that for the other electrodes, demonstrating it shows an enhanced separation and transfer efficiency of photogenerated e-h pairs because the smaller arc radius implies a higher efficiency of charge transfer.
Therefore, it can be concluded that the high BET specific surface area and effective separation transfer efficiency of photogenerated e-h pairs are the main reasons for the enhanced photocatalytic activity of as-prepared porous BiVO4 samples.  Therefore, it can be concluded that the high BET specific surface area and effective separation transfer efficiency of photogenerated e-h pairs are the main reasons for the enhanced photocatalytic activity of as-prepared porous BiVO 4 samples.

Conclusions
In summary, monoclinic BiVO 4 photocatalysts with porous structures were successfully prepared by a two-step approach without assistance of any templates. Kirkendall effect was used to explain the formation mechanism of as-obtained BiVO 4 with porous structure. Compared with the sample of BiVO 4 particles without porous structure, the porous BiVO 4 samples exhibit enhanced photocatalytic activity. In particular, the photocatalytic rate constant of porous BiVO 4 -200˝C sample could reach up to three times of that of the sample of BiVO 4 particles. Furthermore, the high BET specific surface area and effective separation transfer efficiency of photogenerated e-h pairs are considered the main reason for the enhanced photocatalytic activity of as-prepared porous BiVO 4 samples. Finally, it is anticipated that this synthesis method might be used to prepare other vanadate materials.
Supplementary Materials: The following are available online at www.mdpi.com/1996-1944/9/8/685/s1. Figure S1: XRD pattern of the precursor, Figure S2: Low magnification SEM image of as-obtained BiVO 4 products prepared at the hydrothermal temperatures of 180˝C, Figure S3: The adsorption isotherms of MB in the presence of as-prepared samples under dark conditions, Figure S4: UV-vis absorption spectra of MB solutions separated from the photocatalysts of BiVO 4 particles (a); and as-obtained BiVO 4 products prepared at the hydrothermal temperatures of 180˝C (b); 200˝C (c); and 220˝C (d) suspensions during illumination.