Introduction

Semiconductor photocatalysis has achieved keen attention in utilizing solar power to solve environments deterioration and energy crisis1,2,3. Hence, numerous photocatalyst materials were developed, including TiO24,5,6, ZnO7, WO38,9,10, CdS11,12, SnS213, Ag based photocatalysts14,15,16, etc. However, the development of photocatalysts with high efficient is still a huge and continuous undertaking.

The Bi-based photocatalyst materials, such as Bi2O317, BiOCl18, BiOBr19, BiOI20, BiVO421,22,23, Bi2WO624, and Bi2MoO625, have obtained great attention. Among these as-prepared photocatalysts, BiVO4 with strong visible light response capacity and good stability has been extensively investigated in environmental remediation and water splitting26,27,28,29,30,31. Nevertheless, a big problem that affects the photocatalytic property of BiVO4 is its unsatisfied charge carrier separated efficiency. To conquer this problem, many researchers had developed some methods for modifying BiVO4. Cao et al. successfully prepared Au-BiVO4 photocatalyst, which could present much higher visible-light photocatalytic performance for wastewater treatment and clean energy product than the individual BiVO432. Except Au modification, Pd, AuPd and CoPd were used to modify BiVO4 to enhance its photocatalytic property33,34,35. However, these noble metals or noble metal alloys were high cost so that this method was difficult to be wide application. Hence, it was necessary to develop an economical and convenient method to modify BiVO4.

Recently, amorphous semiconductor materials have been exploited and exhibited specific photocatalytic property36,37. Compared to crystals, amorphous semiconductor materials exist a most remarkable advantage that it possesses much smaller band gap than their crystalline counterparts, which conduces them to present more expansive light absorption range, which is an important condition for conversing solar energy37. However, amorphous semiconductor materials exhibited the short-range atomic order, and existed a number of defects, which could become the charge recombination centers, causing themselves inactive or weak performance. So could numerous defects existed on amorphous semiconductor materials as the electrons capturer be applied for modifying other photocatalysts? This view is very significative and interesting. Hence, considering the wide light responded property and cost, Fe based semiconductors enter into our view. Among Fe based semiconductors, FeOOH exhibits extensive visible light response capacity, which caused FeOOH coupled with other photocatalysts to modify the visible-light-irradiation photocatalytic property, some FeOOH with certain crystalline phase was used to modify photocatalyst materials, such as β-FeOOH/TiO238, β-FeOOH/g-C3N439, α-FeOOH/AgVO340, etc. Actually, FeOOH contained various crystalline phases, including α-phase, β-phase, δ-phase, γ-phase, and amorphous phase. Among these different crystalline phase FeOOH, amorphous FeOOH could exhibit excellent oxygen evolution rate by photoelectrochemical and superior pseudocapacitive performance41,42. However, it is relative lack of report about amorphous FeOOH as a modifier to be used in photocatalysis. Hence, combined with above proposed view, it is essential to investigate the role of amorphous FeOOH modifier.

Herein, a novel amorphous FeOOH modified BiVO4 was successfully prepared, and photocatalytic performance for producing O2 from water was investigated. It could be found that, after amorphous FeOOH evenly covered the surface of BiVO4, as-prepared photocatalysts exhibited better migration of photogenerated charges, and stronger visible light responded activity. These positive factors promoted A-FeOOH/BiVO4 presenting improved the photocatalytic performance. Hence, this work shows an effective and simple modified method for designing and preparing highly efficient photocatalysis materials.

Experimental

The synthesis of catalysts

To obtain BiVO4 material, in a beaker, Bi(NO3)3·5H2O (5 mmol) dissolved in HNO3 solution (5 mL 3 mol·L−1) and ethylene glycol (20 mL) mixed solution. Then in other beaker, NH4VO3 (5 mmol) and 0.25 g sodium dodecylbenzenesulfonate (SDBS) were dissolved in hot water (20 mL). After stirred for 30 min, above two solution mixed, and the pH of solution was adjusted to 5 using NaOH solution. Stirring for 60 min, obtained suspension solution was placed into high pressure reactor with PTFE liner, maintained at 180 °C for 24 h. After filtration, wash and desiccation, BiVO4 was prepared.

Amorphous FeOOH/BiVO4 was prepared as follows: BiVO4 (400 mg) was mixed into 40 mL deionized water containing FeCl3·6H2O and NH4HCO3 (The molar ratio of FeCl3·6H2O and NH4HCO3 is 1:3). Stirring for 60 min, the solid powder was obtained through centrifugation, wash and desiccation. According to the theoretical content of amorphous FeOOH in amorphous FeOOH/BiVO4, the obtained powder catalysts were marked as A-FeOOH/BiVO4(2 wt%), A-FeOOH/BiVO4(5 wt%), A-FeOOH/BiVO4(8 wt%) and A-FeOOH/BiVO4(10 wt%), respectively.

Characterizations and photocatalytic experiment

Supporting Materials showed their details.

Results and Discussion

Figure 1 demonstrates the XRD of resultant photocatalysts. As can be observed, resultant BiVO4 exhibited highly crystalline, and there were some mainly diffraction peaks at 2θ of 18.6°, 28.9°, 30.5°, 34.4°, 35.3°, 39.4°, 42.3°, 46.1°, 46.6°, 47.3°, 53.3°, 58.3° and 59.9o, which indexed (101), (013), (112), (200), (020), (211), (105), (123), (204), (024), (301), (303) and (224) diffraction planes of monoclinic BiVO443. For amorphous FeOOH, there was no obvious diffraction peak to be found. Hence, in A-FeOOH/BiVO4(8 wt%), XRD diffraction peaks of BiVO4 could only be detected. The XRD results exposed that amorphous FeOOH had little impact the crystal phase of BiVO4. Moreover, no other diffraction peak was found, meaning that resultant samples possessed the high purity.

Figure 1
figure 1

The X-ray diffraction patterns of as-prepared photocatalysts.

Whereafter, the XPS of A-FeOOH/BiVO4(8 wt%) was further investigated. As revealed in Fig. 2A, Bi, V, O and Fe could be found in the survey XPS spectra, according with the composition of material. Bi 4 f XPS spectrum (Fig. 2B) presented 164.3 eV and 159.1 eV two peaks that attributed to Bi 4f5/2 and Bi 4f7/2 in BiVO444,45. V 2p XPS spectrum (Fig. 2C) had two peaks at 516.3 eV and 524.3 eV, matching with V 2p1/2 and V 2p3/2 in BiVO446. Figure 2D (Fe 2p XPS spectrum) shows 724.2 (Fe 2p1/2) and 710.8 eV (Fe 2p3/2) that were consistent with FeOOH47. Hence, the results of XPS further confirmed that sample contained FeOOH and BiVO4, consistent to the XRD.

Figure 2
figure 2

The XPS high resolution (A) survey spectra, (B) Bi 4f, (C) V 2p and (D) Fe 2p spectra of A-FeOOH/BiVO4(8 wt%).

The morphologies of BiVO4 and A-FeOOH/BiVO4(8 wt%) were seen through SEM. As revealed in Fig. 3, BiVO4 and A-FeOOH/BiVO4(8 wt%) exhibited similar star-shaped particles with the size of about 3.5 μm, which demonstrated that introduced amorphous FeOOH did not influence on the feature of BiVO4. In addition, the elemental compositions of the resultant A-FeOOH/BiVO4(8 wt%) were measured through EDS mapping analysis. The related element mapping images were exhibited in Fig. 3C. Clearly, Fe, V, Bi and O only appeared in observed view, meaning the successful preparation of A-FeOOH/BiVO4.

Figure 3
figure 3

The SEM images of (A) BiVO4 and (B) A-FeOOH/BiVO4(8 wt%) and (C) SEM-EDS mapping images of A-FeOOH/BiVO4(8 wt%).

Subsequently, to further observe the microstructure of A-FeOOH/BiVO4, transmission electron microscopy were also analyzed. As revealed in Fig. 4A, A-FeOOH/BiVO4(8 wt%) exhibited the similar star-shaped particles, which was consistent with the SEM. Clearly, amorphous FeOOH could not be observed at low resolution TEM image. As a result, the HRTEM of A-FeOOH/BiVO4 (Fig. 4B) was provided. The lattice fringe spacing of 0.237 nm attributed to BiVO4 (220) was obviously observed. Moreover, the thickness of ultrathin amorphous FeOOH nanolayers was about 2 nm, and evenly adhered on the surface of BiVO4. Amorphous FeOOH nanolayers did not reveal a lattice spacing, demonstrating traditional amorphous structure.

Figure 4
figure 4

The (A) TEM and (B) HRTEM images of A-FeOOH/BiVO4(8 wt%).

The photocatalytic property over resultant photocatalysts was measured through producing O2 from slitting water. Before producing O2, to improve oxygen evolution rate, the sacrificial agent need be added. The previous report indicated that BiVO4 could exhibit better photocatalytic performance for producing O2 in the NaIO4 solution than that in the AgNO3 solution48. Meanwhile, to verify this result, we also measured the photocatalytic performance for producing O2 over A-FeOOH/BiVO4 photocatalyst in the same concentration NaIO4 or AgNO3 solution. As shown in Fig. S1, A-FeOOH/BiVO4 photocatalyst in the NaIO4 solution could present high oxygen evolution rate (OER) than A-FeOOH/BiVO4 photocatalyst in the AgNO3 solution. Hence, in this system, we selected NaIO4 solution as the sacrificial agent. Then the results of oxygen evolution over all of photocatalysts were shown in Fig. 5A. For a series of resultant A-FeOOH/BiVO4 photocatalysts, O2 could be persistently produced with reaction time prolonging. Compared with pure BiVO4, resultant A-FeOOH/BiVO4 photocatalysts could clearly display the improvement of photocatalytic capacity for producing O2. When loading amount of amorphous FeOOH was 8%, the sample had the optimal catalytic rate of O2 evolution. Then Fig. 5B gives their OER. Their OER were 162.3, 691.7, 1077.1, 1206.3 and 962.8 μmol h−1 g−1 for pure BiVO4, A-FeOOH/BiVO4(2 wt%), A-FeOOH/BiVO4(5 wt%), A-FeOOH/BiVO4(8 wt%) and A-FeOOH/BiVO4(10 wt%), respectively. The OER of A-FeOOH/BiVO4(8 wt%) was around 7.4 folds more than pure BiVO4. In addition, the O2 production rates of different materials in previous reports have been in Table 149,50,51,52, and compared with this work. Obviously, as-prepared A-FeOOH/BiVO4(8 wt%) in this work could present excellent advance.

Figure 5
figure 5

(A) Photocatalytic evolution O2 curves and (B) OER over various samples, (a) BiVO4, (b) A-FeOOH/BiVO4(2 wt%), (c) A-FeOOH/BiVO4(5 wt%), (d) A-FeOOH/BiVO4(8 wt%) and (e) A-FeOOH/BiVO4(10 wt%).

Table 1 Comparison of photocatalytic activity in previous reports and this work.

Furthermore, as we known, the stability is a very significant index to appraise its ability. Thus, recycling tests for producing O2 over A-FeOOH/BiVO4(8 wt%) were investigated. As demonstrated in Fig. 6A, the A-FeOOH/BiVO4(8 wt%) presented relative stable OER, after 6 times cycled experiments, as-prepared A-FeOOH/BiVO4(8 wt%) still presented 70% photocatalytic activity of fresh sample. Besides, the XRD of used A-FeOOH/BiVO4(8 wt%) was measured in Fig. 6B. Its XRD did not have obviously change in comparison to fresh A-FeOOH/BiVO4(8 wt%). And the morphology of used sample was also observed in Fig. 6C,D. It can be observed that used sample exhibited a certain aggregate in comparison to fresh sample, which might result in the declined photocatalytic property. Besides, the whole morphology did not present great change.

Figure 6
figure 6

(A) Stability test in O2 photosynthesis for A-FeOOH/BiVO4(8 wt%), (B) the XRD of fresh A-FeOOH/BiVO4(8 wt%) and used A-FeOOH/BiVO4(8 wt%) and (C,D) The SEM images of used A-FeOOH/BiVO4(8 wt%).

Obviously, it was found that covered amorphous FeOOH could effectively modify BiVO4 to improve catalytic activity. Why? To find the reason for promoting effect, some analyzed instruments were tested to investigated light response capacity, the photoinduced charge separated rate and surface area of as-prepared photocatalysts, which are deemed as the main factors to effect on the photocatalytic performance53,54,55.

Figure 7 presents the UV-Vis absorption spectrum of resulted BiVO4 and A-FeOOH/BiVO4(8 wt%). Pure BiVO4 presented remarkable light absorption between 200 and 800 nm. The absorption edge was 525 nm. Band gap was 2.36 eV. After covered by amorphous FeOOH, the light absorption capacity of as-obtained A-FeOOH/BiVO4(8 wt%) gained the obvious enhancement. Hence, the enhanced photocatalytic property might be anticipated.

Figure 7
figure 7

The UV-Vis absorption spectrum of (a) BiVO4 and (b) A-FeOOH/BiVO4 (8%).

Photoluminescence (PL) property could analyze the separation and transfer efficiency of photoinduced charges56. Therefore, the PL of BiVO4 and A-FeOOH/BiVO4 (8%) were measured in Fig. 8. The emission band intensity of A-FeOOH/BiVO4 (8%) was clearly declined in comparison of pure BiVO4. This result implied that covered amorphous FeOOH could effectively reduce the recombined efficiency of photoinduced electrons and holes, conducing to improve photocatalytic property57.

Figure 8
figure 8

The photoluminescence spectra of (a) BiVO4 and (b) A-FeOOH/BiVO4 (8%).

Subsequently, electrochemistry measurements (photocurrent and EIS measurement) were used to assess the separated and transfer efficiency of photoinduced charges. Currently, the stronger photocurrent manifests the more effective separation and transfer rate of photo-charges58,59. As illustrated in Fig. 9A, pure BiVO4 and A-FeOOH/BiVO4(8 wt%) could produce certain intensity photocurrent signals. And the order of photocurrent intensity from strong to weak was A-FeOOH/BiVO4(8 wt%) > pure BiVO4. Evidently, resultant A-FeOOH/BiVO4(8 wt%) exhibited the better photocurrent intensity, revealing that the separated rate of photogenerated charges for A-FeOOH/BiVO4(8 wt%) photocatalyst was significantly improved by covering amorphous FeOOH.

Figure 9
figure 9

(A) Transient photocurrent property and (B) EIS of (a) BiVO4 and (b) A-FeOOH/BiVO4 (8%).

Then EIS techniques were further used to estimate charge separation property. Figure 9B gives the EIS of pure BiVO4 and A-FeOOH/BiVO4(8 wt%). Small frequency semicircle radius exposes a better charge transfer rate. As demonstrated in Fig. 9B, the semicircle radius of A-FeOOH/BiVO4(8 wt%) was shorter than that of pure BiVO4, meaning that A-FeOOH/BiVO4(8 wt%) possessed a higher separation and transport rate of photogenerated charges.

Besides, the BET surface area of pure BiVO4 and A-FeOOH/BiVO4(8 wt%) were 4.9 and 5.2 m2 g−1, respectively. The addition of amorphous FeOOH had little effect on the surface area of photocatalyst. Additionally, the surface hydrophilic property of photocatalyst was performed to measure interact with the water. As presented in Fig. 10A,B, water contact angle (CA) of BiVO4 (69.55°) and A-FeOOH/BiVO4(8 wt%) (48.35°) was measured. This result meant that the covered amorphous FeOOH made BiVO4 possess water favorable wetting capacity, providing a good chance to oxidate H2O in aqueous environment60.

Figure 10
figure 10

Water contact angle of (A) BiVO4 and (B) A-FeOOH/BiVO4(8 wt%).

Hence, combined with above analyzed results, it was found that, the main reason that as-prepare A-FeOOH/BiVO4 possessed the better photocatalytic performance than BiVO4 could be explained that the former exhibited higher efficiency for separation of photogenerated charges, and stronger strong visible responded activity compared with the latter. Hence, the remarkable improvement of photocatalytic capacity for producing O2 was obtained.

Whereafter, to speculate the photocatalytic mechanism, the energy structure of BiVO4 might be calculated61

$${\rm{Eg}}={{\rm{E}}}_{{\rm{VB}}}-{{\rm{E}}}_{{\rm{CB}}}$$
(1)
$${{\rm{E}}}_{{\rm{VB}}}={\rm{X}}-{{\rm{E}}}_{{\rm{e}}}+1/2{\rm{Eg}}$$
(2)

Herein Eg, EVB, ECB, X, Ee are band gap of photocatalyst, the valence band potential, conduction band potential, electro negativity of component atoms, hydrogen scale (4.5 eV), respectively. Here, for BiVO4, X is 6.15 eV62, Eg is 2.36 eV (Fig. 7). Hence, the CB and VB potentials of BiVO4 were respective 0.47 and 2.83 eV.

Finally, in Fig. 11, a probable photocatalytic mechanism for A-FeOOH/BiVO4 was presented. BiVO4 could generate electrons and holes under light irradiation. Due to quantum-tunneling effect (QTE)63,64, formed charges from the conduction band of BiVO4 could transfer through amorphous FeOOH, or cationic vacancy network in amorphous FeOOH phase65. Then NaIO4 as the electrons sacrificial agent consumed electrons. Because the cationic vacancy might be also activated by trapping hole66, then holes remained on the VB of BiVO4 would migrate to amorphous FeOOH surface to produce O2. Hence, in this process, the existence of amorphous FeOOH could boost the separation of photoinduced charges for A-FeOOH/BiVO4 system, obtaining the enhancement of photocatalytic performance.

Figure 11
figure 11

The proposed mechanism of photocatalytic water oxidation over A-FeOOH/BiVO4 photocatalyst.

Conclusions

We successfully produced a novel amorphous FeOOH modified BiVO4, and investigated it photocatalytic performance for producing O2 from water. It could be found that, amorphous FeOOH modified BiVO4 exhibited higher migration rate of photogenerated charges, and strong visible responded capacity than BiVO4, which resulted that amorphous FeOOH modified BiVO4 could present better photocatalytic property than BiVO4, and kept excellent performance and structure stability. Hence, this work provides a simple and inexpensive modified method for design and synthesis of effective photocatalysts.