Electrochemical Synthesis of Photoelectrocatalytic Thin Films of Hexagonal BiPO 4 Nanorods

HexagonalBiPO 4 nanorodsthinﬁlmswithremarkablephotoelectrocatalytic(PEC)performanceweresuccessfullyfabricatedonFTO substrates via cathodic electrodeposition. Chronoamperometry, linear sweep voltammertry (LSV) and electrochemical impedance spectroscopy (EIS) were used to study the photoelectrochemical properties of these materials. The photoelectrocatalytic capability was evaluated by the degradation of methyl blue (MB). The ﬁlm deposited in 40 min has the best catalytic activity and good stability with the 30 h degrading experiment. The photocurrent was increasing because the concentration of MB was being degrading. The CV analysis which reveals a potential shift and the increasing anodic and cathodic peak currents with increasing scan rate proves a diffusion-controlled process. The hydroxyl radicals are the main active species that can oxidize the adsorbed organic pollutants. Mott–Schottky (MS) measurements match the typical for n-type semiconductors.

BiPO 4 has received increasing attention in photocatalytic degradation of organic dyes since the report by Pan et al. [1][2][3][4][5] Bismuth salt photocatalysts have a narrow bandgap because of the high valence band which is caused by Bi 6s and O 2p antibonding states. 6,7 For example, BiOBr exhibits superior photocatalytic activity for the degradation of rhodamine B (RhB) 8,9 and cylindrospermopsin 10 under UV and visible light irradiation. In addition, PO 4 3− , possessing a large negative charge, is postulated to help the e − /h + separation, which can effectively improve its photocatalytic activity. 11,12 So far, various methods have been developed for synthesizing BiPO 4 , such as chemical vapor deposition (CVD), 13,14 hydrothermal, [14][15][16][17] microwave synthesis 18 and sonochemical synthesis, 19 etc. Cathodic electrodeposition has shown a power ability to control the growth of the crystal, and it presents a simple, quick, controllable and economical method for the preparation of large area thin films. [20][21][22][23][24][25] The growth rate and surface morphology can be well controlled by deposition potentials, current densities, deposited time and salt concentrations. Recently, considerable research has been dedicated to the use of cathodic electrodepostion to form self-organized semiconductor structures of materials such as ZnO, 26,27 BiVO 4 , 28 La(OH) 3 29-32 and BiOCl. 33 Up to date, powder photocatalysts have several inevitable weaknesses among the researches of photocatalysis in aqueous solution, including the separation of powder from system after reaction, high electron-hole recombination rate of catalyst which reduces the photocatalystic activity, and the limitation of the photoefficiency due to the slow interfacial electron transfer. Since the report of the Honda-Fujishima effect on TiO 2 electrodes under UV light, 34 photoelectrocatalytic (PEC) oxidation has been regarded as an ideal means of curbing environmental organic dyes pollution in aqueous solution using different semiconductor electrodes. [35][36][37][38][39] Moreover, the applied potential can enhance charge separation and facilitate the transfer of photogenerated electron. In our previous work, the BiPO 4 film prepared by hydrothermal process has performed remarkable photoelectrocatalytic activity. 40 However, a more facile way to synthesize more stable film still needs to be explored.
In this paper, the BiPO 4 nanorods were first synthesized by a one-step cathodic electrodepostion approach on FTO substrates from precursor solution. Structure, morphology and the electrochemical performance were characterized as well. The photoelectrocatalytic activity was evaluated by the degradation of MB.

Experimental
Electrochemical deposition.-All chemical reagents were of analytical purity without further purification. All electrochemical dez E-mail: tzhangym@jnu.edu.cn position experiments were performed with a HDV-7C transistor potentiostatic apparatus connected to a three-electrode cell. A graphite rod was used as the auxiliary electrode. Fluorine doped SnO 2 (FTO) coated glass (5.0×4.0 cm 2 ) with a sheet resistance of 10 was used as the working electrode which was connected to the cell with a double salt bridge system. An Hg/Hg 2 Cl 2 electrode was used as the reference electrode. The FTO substrate was cleaned ultrasonically in ethanol, distilled water and 10% HNO 3 , and then rinsed in distilled water again before electrodeposition. Solution for electrodepositing the BiPO 4 nanorods thin films were prepared by dissolving 10 mM Bi(NO 3 ) 3 . 5H 2 O in a solution of 10 mM ethylene diamine tetraacetic acid at < pH 1 with HNO 3 . Then 100 mM Na 3 PO 4 · 12H 2 O was added, which was adjusted to pH 1 with concentrated HNO 3 . Finally a few milliliters 30% H 2 O 2 were added. After cathodic electrodeposition, BiPO 4 nanorods on the working electrode were obtained with a current density of 0.3 mA cm −2 at 70 • C. Then the FTO substrates with BiPO 4 nanorods electrodes were rinsed with distilled water several times for PEC.
Characterization.-The obtained products were analyzed by Xray diffraction (XRD; Bruker D8 Advance) with Cu-Kα radiation (λ = 0.15406 nm) in scan range of 10 • -80 • to determine the phases and structures. The views of the materials were investigated by using a field emission scanning electron microscopy (FESEM) using a Zeiss ULTRA 55 at an acceleration voltage of 5 KV. High resolution transmission electron microscopy (HRTEM) and selected area electron diffraction (SAED) observations were performed on a JEOL 2100 F field emission electron microscope. The photoelectrochemical experiment was measured on an electrochemical system (SP-150, France). UV-vis diffuse reflectance spectra (DRS) of the samples were conducted by using Hitachi U-3010 UV-vis spectrophotometer. Raman spectroscopy was characterized by Renishaw inVia.

Photoelectrochemical
measurements.-Photoelectrochemical measurements were carried out in a conventional three-electrode system. The counter and the reference electrodes were platinum and saturated calomel electrode (SCE), respectively. The UV light was obtained by an 11 W germicidal lamp (Institute of Electric Light Source, Beijing) which was used as the excitation light source for ultraviolet irradiation and the average light intensity was 1.0 mW cm −2 . The electrolyte solution of 0.1 M NaCl was used. Photo response was measured in the mode of chronoamperometry. Potentio electrochemical impedance spectra (EIS) and Mott-Schottky (MS) measurements were recorded in the potentiostatic mode. The amplitude of the sinusoidal wave was 10 mV, and the frequency range of the sinusoidal was 100 kHz to 0.05 Hz. Cyclic voltammetry studies were performed at different scan rates of 2 ∼ 50 mV/s in a potential range between 0.6 and −0.6 V vs. SCE.
The photocatalytic activities were evaluated by the decomposition of MB under UV light (λ = 254 nm). The quartz electrolytic cell was filled with 10 −5 M MB solution. The filtrates were analyzed by recording variations of the maximum absorption band (663 nm for MB) using a TU -1810 UV-vis spectrophotometer. The BiPO 4 thin films served as the working electrode. Prior to the irradiation, the prepared composite electrodes were magnetically stirred in the dark for 20 min to reach an absorption-desorption equilibrium.

Results and Discussion
Structure and morphologies of the BiPO 4  The morphologies of the products synthesized under different times were viewed by SEM. Figs. 1b-1d show the typical lowmagnification SEM images of the as-prepared BiPO 4 thin films, and it is seen that a large quantity of nanorods were successfully synthesized on the FTO substrates. It also shows the amount of the BiPO 4 nanorods in the surface morphology of FTO increases with the increasing of deposition time. The inset figures in Figs. 1b-1d are the high-magnification SEM images, showing that these nanorods have a diameter of 400-800 nm, a length of 1000∼2000 nm and a well defined hexagonal morphology, which is successfully obtained via a template-free cathodic electrodeposition.
Significantly, HRTEM images were further demonstrated to observe the fine structure of BiPO 4 nanorods composite. The results in  Fig. 3a shows the Raman pattern of the BiPO 4 nanorods film. The Raman analysis demonstrates that the BiPO 4 nanorods are well crystallized with hexagonal structure. To further investigate the surface elemental composition and oxidation state of the product, X-ray photoelectron spectroscopy (XPS) analysis was performed and the surveys are shown in Fig. 3b. In the high resolution XPS spectra of the resulting film, the double broad peaks with higher binding energy of 159.4 eV and 164.7 eV are consistent with the characteristic Bi 4f 7/2 and Bi 4f 5/2, respectively, with a peak separation of 5.3 eV (Fig. 3c). This indicates the existence of a trivalent oxidation state for bismuth. The peak in Fig. 3d of P 2p locats at 132.9 eV suggests that the P in the sample exists in the oxidation state of P 5+ . As shown in Fig. 3e, the O may be fitted to two types of chemical states: crystal lattice oxygen and adsorbed oxygen. The peak at 532.1 eV is assigned to the adsorbed oxygen, while the peak at 530.6 eV is related to crystal lattice oxygen, and the results are in line with previous reports on BiPO 4 . The UV-vis diffuse reflectance spectra of BiPO 4 nanorods coated on FTO glass showed a bandgap at around 2.94 eV (Fig. S1).
Electrochemical characterization. -Fig. 4a displays the chronoamperographs of the BiPO 4 thin film, showing an effective charge transfer process under UV light illumination and 1 V bias. The BiPO 4 thin film depositing in 40 min presents the highest photocurrent, indicating that the 40 min film has the highest separation efficiency of photogenerated electron-hole pairs. The value of the photocurrent of the linear sweep voltammogram for the BiPO 4 thin film electrodeposited in 40 min is larger than the others (Fig. 4b) at −1.5 V bias potential. These results confirm that the film deposited in 40 min has optimal PEC performance. LSVs both in the dark and under 1 mW/cm 2 illumination are also performed on the BiPO 4 thin film in 0.1 M phosphate buffer at pH 7 with and without the addition of 0.1 M sodium sulfite (Na 2 SO 3 ) (Fig. 4c). As expected for an n-type semiconductor, the BiPO 4 thin film serves as a photocathode and generats anodic photocurrent through the consumption of photoinduced holes for water at the semiconductor/electrolyte interface. The photo-assisted BiPO 4 thin film yields the greatest cathodic shift of the onset potential for PEC water oxidation, which is ∼600 mV at 4.60 V vs. SCE. When sulfite was added, the anodic photocurrent density increased. It increases to 3.51 mA/cm 2 . This is because of the sulfite, and the anodic photocurrent generated is exclusively ascribed to the oxidation of sulfite, which is kinetically easier than water oxidation. 41,42 Thus, the substantial enhancement in photocurrent in the presence of sulfite demonstrates that the water oxidation photocurrent on the BiPO 4 thin film is mainly limited by poor kinetics for water oxidation on its surface. In fact, the photocurrent generated by the BiPO 4 thin film for sulfite oxidation (Fig. 4c(2) or (2 )) is very exciting in that few oxide-based photocathodes known to date can generate photocurrent at that level in a pH 7 medium. This result suggests that it may be possible for BiPO 4 to generate a similar level of photocurrent for water oxidation when it is coupled with a proper oxygen evolution catalyst.
The radius of the arc on the EIS Nynquist plot reflects the reaction rate occurring at the surface of electrode. The arc radius on the EIS Nynquist plot of the BiPO 4 thin film is smaller than that of film under UV irradiation (Fig. S2(A)). The sizes of the arc radius are reduced by combination of photoirradiation and applied bias potential of 0.5 and 1.0 V (Fig. S2(B)). As shown in Fig. S2(C), it is the smallest at the 2.0 V bias potential combined with photoirradiation, which suggested that a more effective separation of photogenerated electron-hole pairs and faster interfacial charge transfer occurred on the BiPO 4 thin film. 43,44 At the bias potential of 1.0 V, the size of the arc radius of the film deposited in 40 min is the smallest. These photoelectric characteristics prove that the combination of photoirradiation and the applied bias potential is an effective way to improve photocatalytic efficiency. The applied bias potential not only can separate the holes and electrons but also can directly electrolyze MB. 45,46 Thus, MB can be effectively degraded.   degraded using the BiPO 4 nanorods film, and it can also be degraded via the electro-oxidation process at the bias potential of 1.5 or 3 V. However, only 3% MB is degraded under UV light without BiPO 4 . Clearly, the largest degradation rate of MB is attained when both applied potential (EC) and UV-light irradiation (PC) are introduced. It is consistent with the conclusions of the Chronoamperographs, LSVs and EISs. Fig. 5b shows the PEC degradation efficiencies of MB by different BiPO 4 thin films with 3 V bias potential (and 3.5 V see Supporting Information Fig. S4). According to the experimental results, the asprepared BiPO 4 thin films at 3.0 V reveal the significant degradation of MB as a function of UV illumination time. After 5 h under UV irradiation, the relative concentrations of MB remain 35.0%, 19.6% and 59.9%, respectively corresponding to depositing time of 30 min, 40 min and 60 min. On the other hand, when applying 3.5 V bias potential, the relative concentrations are 55.3%, 30.2% and 33.7% (Fig. S3). It clearly demonstrates that the 40 min depositing electrode have best photoelectrocatalytic performance at bias potential of 3 V and 3.5 V.
Also, after five recycles for the photodegradation of MB by the combination of UV light irradiation and electro-oxidation with 3.0 V, the film do not exhibit any significant loss of activity, as shown in Fig. 5c, confirming that the BiPO 4 nanorods are not photocorroded during the photocatalytic oxidation of the pollutant molecules. To test the stability of the sample during the photocatalytic process, we compare the XRD patterns before and after 5 recycles of the reaction (Fig.  S5). After reaction, no obvious change has been observed, indicating its good stability.
The oxidative species in the photocatalytic process can be detected through the trapping experiments of radicals and holes. As shown in Fig. 5d, under UV light irradiation the photodegradation of MB is scarcely restrained by the addition of a hole scavenger EDTA while it is obviously inhibited when tBuOH (hydroxyl radicals scavenger) is added. 4 This indicates that hydroxyl radicals are the main active species that can oxidize the adsorbed organic pollutants. A possible mechanism for the photoelectrocatalytic degradation of MB over BiPO 4 film was illustrated in Fig. S7. As pattern showed, BiPO 4 particles generate and separate electron/hole pairs under UV irradiation and applied bias. Then, photogenerated holes in the valence band combine with hydroxide (OH − ) to produce hydroxyl radical ( . OH), which can directly break the ring structure of MB molecules and convert them into CO 2 and H 2 O. Fig. 5e shows photoelectrochemical catalysing MB using a BiPO 4 thin film (20 cm 2 ) with 3.0 V of an applied potential vs. SCE. The photocurrentis decreasing slightly by the reason of removing 3 ml electrolyte to analyze the changing of the concentration per hour, which correspond the Fig. 5b. Overall, the photocurrent is increasing because of the concentration of MB is being degrading.
The CV analysis is carried out for the BiPO 4 sample which exhibits two anodic peaks and two cathodic ones corresponding to the reverse reactions. The reduction peaks around −0.8 V vs. Hg/HgO are corresponding to the reduction of Bi(III) to Bi metal and the oxidation peaks around −0.2 V vs. Hg/HgO are corresponding to the oxidation of Bi metal to Bi(III) at 50 mV s −1 , respectively. Moreover, there is a potential shift and an increasing current with increasing scan rate in the range of 2-50 mVs −1 , which shows further insight into electrochemical performance as shown in Fig. 5f. The cathodic and anodic peak current is linearly proportional to the square root of scan rates (inset of Fig. 5f). This is a characteristic of a diffusion controlled process limited by the diffusion of either NaCl or the reaction intermediates. 47 Mott-Schottky (MS) analysis was carried out in darkness by using the SP-150 BIO-LOGIC science workstation. MS plots of the as-prepared BiPO 4 film shows reversed sigmoidal plots with an overall shape (Fig. S6). All prepared samples ranges from about −0.6 V to 0.6 V vs. SCE, which are identified with that typical for n-type semiconductors. 48

Conclusions
In summary, a novel hexagonal BiPO 4 nanorods with space group P3121(152) has been synthesized on FTO for the first time via cathodic electrodeposition technique. Simultaneously, the amount of the nanorods increases with the deposited time. The d-spacing of 0.606 nm, 0.349 nm and 0.307 nm in HRTEM and SAED are well matched to the (100), (110) and (111) crystallographic plane of hexagonal BiPO 4 . The BiPO 4 thin film shows a notable photo response of current under UV illustration, while EIS arc radius of the 40 min depositing sample appears to be the smallest. The PEC degradation of MB is much more efficient than the sum of the electrochemical degradation and photocatalytic process. Particularly, the 40 min depositing film exhibits the best photoelectrocatalytic performance at bias potential of 3 V and 3.5 V. Moreover, the film maintains high photoelectrocatalytic activity in 5 recycles of PEC degradation. The photocurrent is increasing during per degradation reaction. In addition, the main oxidative species are determined to be hydroxyl radicals. A potential shift and the increasing anodic and cathodic currents with increasing scan rate in the range of 2-50 mVs −1 prove a diffusion-controlled process. Moreover, the as-prepared BiPO 4 is a typical n-type semiconductor. These results also indicate that it is possible to synthesize other nonmetal oxy-acid salts materials with superior photoelectrocatalytic activity by cathodic electrodeposition.