Stannous Tungstate Semiconductor for Photocatalytic Degradation and Photoelectrochemical Water Splitting: A Review

: The use of photocatalysis and photoelectrocatalysis is expected to achieve the efﬁcient utilization of solar energy to alleviate and even solve the problems of energy depletion and environmental pfollution. At present, stannous tungstate materials have attracted extensive attention in the ﬁelds of photocatalysis and photoelectrocatalysis as favorable candidates for such utilization because of their narrow band gap energy (which is ~1.7 eV for the α phase and ~2.7 eV for the β phase, respectively) and unique band structure (which covers the oxidation and reduction potential of water). However, their practical application is still limited by excessive electron–hole recombination and poor stability. In this review, basic information (crystal and electronic structures) related to photocatalysis and photoelectrocatalysis is presented. Additionally, various strategies to enhance the photocatalytic and photoelectrochemical properties of stannous tungstate materials, such as morphological modiﬁcation, crystal facet engineering, doping modiﬁcation, and multicomponent compositing, are summarized. Furthermore, the achievements and difﬁculties of the relevant studies are discussed. The information presented in this review can provide a reference for subsequent research on the photocatalytic and photoelectrochemical performance of tungstate-based materials.


Introduction
With the rapid development of technology, energy has become a necessity for social operation, and the human consumption of non-renewable traditional energy (from sources such as oil and coal) is decreasing. The energy crisis has become an urgent problem that needs to be solved. Therefore, researchers are enthusiastically looking for alternative renewable clean energy sources. Among the many potential energy sources, solar energy has attracted significant attention due to its large reserves and reproducibility. Although solar energy exhibits a variety of excellent properties, its dispersion, instability and intermittency have led to its low utilization. Therefore, exploring efficient ways to better use solar energy is necessary for the generation of alternative power. Photocatalytic and photoelectrochemical methods show promise for degrading pollutants and generating new kinds of energy (such as hydrogen and chemical) via the use of solar energy. All the relevant reactions are completed by the photo-generated holes and electrons; thus, they follow a similar principle. Since the first report of TiO 2 in 1972 [1], different kinds of materials with a smaller band gap than that of TiO 2 (3.0-3.2 eV) have been explored [2].
One of these materials is stannous tungstate (SnWO 4 ), which is a potential catalyst due to its narrow band gap, suitable band position, and nontoxicity [3]. However, due to its low charge separation efficiency and poor stability [4], its photocatalytic and photoelectrochemical performance has not met theoretical expectations. To solve this problem, many researchers are committed to obtaining a better catalytic performance and higher solar energy efficiency by means of morphology control, doping, and multi-component compositing. In this paper, we describe the synthesis method, physical characteristics, and This difference in structure leads to different properties for the α and β phases. The W-O bond in the WO 4 2− tetrahedron (~1.75 Å) [5] is shorter than that in the WO 6 octahedrons (~1.80-2.15 Å) [6]. It leads to a larger crystal field-splitting energy and a higher orbital energy of W in β-SnWO 4 . The conduction band edge of the α phase is mainly composed of a W 5d orbital, while the β phase contains an Sn 5p orbital, in addition to the W 5d orbital ( Figure 2). The valence bands of both SnWO 4 are formed by hybridizing the O 2p and Sn 5s orbitals, and the Sn-O bonds are almost identical in length (~2.20-2.80 Å) [5,6]. Thus, the positions of the valence bands are similar. These characteristics result in a narrower band gap for α-SnWO 4 (~1.7 eV) than for β-SnWO 4 (~2.7 eV) [3], and α-SnWO 4 has a wider range of light absorption and a higher theoretical photocurrent density (~17 mA cm −2 [9]) than β-SnWO 4 . In addition, the β phase is a direct band gap and the α phase is an indirect band gap [3], which leads to a higher carrier separation efficiency for the α phase than the β phase [10]. Due to the wider band gap, the photo-generated electrons (holes) of β-SnWO 4 show a more significant reduction (oxidation) activity. Thus, β-SnWO 4 was used in photocatalysis, and only α-SnWO 4 has been used in photoelectrochemical water splitting and sulfite oxidation. However, in addition to the excellent optical properties mentioned above, the use of stannous tungstate exhibits problems that urgently need to be solved. Since Sn is divalent and oxidizable in SnWO 4 , it is susceptible to photocorrosion, which is attributed to stannous, and the following reaction occurs: SnWO 4 + H 2 O + 2h + → SnO 2 + WO 3 + 2H + . Therefore, SnWO 4 has poor stability, which limits its performance in practical applications [4]. At the same time, a recombination of carriers occurs in the diffusion process of SnWO 4 . Its diffusion length is short, which limits the number of photo-generated carriers that can arrive at the surface of the crystal, leading to unsatisfactory photocatalytic and photoelectrochemical performance. Various methods have been developed to synthesize SnWO 4 materials, and their differences in thermodynamic stability lead to changes in the synthesis of α-SnWO 4 and β-SnWO 4 . Because the α phase is a kind of crystalline phase at low temperatures, its synthesis conditions are milder. Many methods can be used to synthesize the α phase, e.g., hydrothermal methods, solid-solid reaction methods [3,[11][12][13], and magnetron sputtering methods [10,14,15]. Each synthesis method has its own advantages. The hydrothermal method can be used to obtain crystals with a complete crystal phase, small particle size, and uniform distribution. A magnetron sputtering method can obtain dense and uniform highquality thin films. The solid-solid reaction method possesses a simple process and a large output. However, the synthesis of the high-temperature crystalline β-phase is relatively harsh because it requires higher temperature conditions (which should be obtained by rapid cooling at more than 940 K). It has been reported that most β-SnWO 4 powders are synthesized with a high-temperature solid phase method [11,12,16,17]. β-SnWO 4 can also be obtained by using NaWO 4 as a raw material through kinetic control under more mild conditions [18]. Jan Ungelenk and Claus Feldmann synthesized β-SnWO 4 using a microemulsion method at low temperatures [19]. When W and O combine to form a discrete WO 4 2− tetrahedron in NaWO 4 , the structure is similar to that of β-SnWO 4 , so Na 2 WO 4 is likely to be directly converted to β-SnWO 4 . Additionally, assisted by hydrothermal [20,21] and microwave-assisted hydrothermal methods [22], α-SnWO 4 can also be synthesized using NaWO 4 and SnCl 2 as the precursors.
To synthesize the film for photoelectrochemical water spliting, an indirect or direct coating method can be used. In indirect coating, as-prepared SnWO 4 powders are drop-cast onto conductive substrates. This is an easy-to-perform method, but the performance of the photoanode is weak due to the poor charge transfer caused by bad connections between the particles. For direct coating, Moritz Kölbach et al. employed a pulsed laser-deposited method that used an α-SnWO 4 target formed by annealing WO 3 (99.99%) and SnO (99.99%) powders [23]. Bozheyev et al. reported on the use of a magnetron co-sputtering method in which Sn (99.99%) and W (99.95%) were the targets, and the deposition was performed with the existence of O 2 [24]. Recently, Gottesman used a new method-rapid thermal processing-to synthesize an α-SnWO 4 film. The technique could be used to treat a sample at higher temperatures, without destroying the glass-based F:SnO 2 (FTO), and a desired crystallinity with few defects was obtained [25]. Liu et al. used a chemical vapor deposition to convert WO 3 into α-SnWO 4 , and the reaction is shown in Equation (1) [26].
An in situ hydrothermal conversion method was also reported, in which WO 3 was used as a precursor to react with Sn 2+ in the solution [27]. The possible reactions occurring in this method are illustrated as Equations (2)-(3) [28]. As a result, the α-SnWO 4 film exhibits similar nanostructured array morphology to that of WO 3 films due to the inherited behavior ( Figure 3) [28]. The anion in hydrothermal conversion solutions may also affect the morphology of α-SnWO 4 films [29]. 4WO 3 + 3SnCl 2 3SnWO 4 + WCl 6 (1) Generally speaking, SnWO 4 has good optical properties, such as a narrow band gap and a special band structure, which results in excellent application prospects for the fields of photocatalysis and photoelectrocatalysis (especially in the field of water splitting). However, it is necessary to pay attention to the instability and short carrier diffusion length of the material itself. Furthermore, because of the differences in crystal structure and band structure between the two crystal phases (α phase and β phase), their properties and synthesis methods are different. For instance, the photo-generated electrons (holes) of the β phase have more significant reduction (oxidation) activity than that of the α phase, so the β phase is suitable for photocatalytic degradation. The α phase has a wider light absorption range and higher carrier separation efficiency, so it has potential for use in photoelectrocatalysis.

Research on Modification of SnWO 4
Since Cho et al. first reported its optical properties in 2009 [3], SnWO 4 has been widely used in photocatalytic degradation [30] and photoelectrochemical water splitting [31]. Researchers have been working to modify SnWO 4 via morphological modification, facet regulations, doping, and the construction of multicomponent composites in order to obtain better photocatalytic degradation rates of organic pollutants (Table 1), as well as improved photoelectrochemical water splitting and sulfite oxidation performance (Table 2).

Morphological Modification
By regulating the synthesis technique, which is a universal method for enhancing the catalytic performance of a material, a target material with a specific morphology can be obtained. The morphology of a material is usually designed for the following purposes [57]: (1) to increase the specific surface area and reduce the nanometer size, as a high specific surface area can provide more active sites, and a small nanometer size can provide a shorter charge transfer path that improves carrier separation efficiency and enhances light absorption efficiency; (2) to improve the crystallinity of materials, as a high crystallinity can reduce the recombination center and improve carrier separation efficiency, but it is difficult to simultaneously improve crystallinity and specific surface area; (3) to obtain special structural shapes (such as 0D, 1D, 2D, and 3D materials), as different dimensions have diverse characteristics. Thus, directional design can improve the performance of the materials.
In the study of SnWO 4 for photocatalysis, there have been many attempts to obtain improved properties by adjusting the morphologies of the substrate materials. Zhu et al.  [36]. Alharthi et al. synthesized honeycomb-like β-SnWO 4 with a hydrothermal method and calcined it at 700 • C, which enabled the photodegradation of rose bengal in 150 min [53]. For photoelectrochemical water splitting, it is difficult to control the morphology of the powder present on an electrode. Accordingly, Zhu et al. synthesized a porous α-SnWO 4 film through the hydrothermal conversion of WO 3 films [31], and they obtained a photocurrent density of 0.08 mA cm −2 at a potential of 1.23 V RHE . In contrast, topography optimized nails (Figure 4b) [27,29], nanosheets (Figure 4c) [56], and long-plates [28] have presented better PEC activity than α-SnWO 4 , with a powder stacked morphology because the array structures provide pathways for charge transfer, with less recombination.  [45]; α-SnWO 4 with (b) dense nanorods [27] and (c) nanosheet [56] morphology.

Crystal Facet Engineering
During morphology regulation, the ratio of exposure facets may change as well [58]. It is noted that reaction activities vary for different facets [59,60]. Thus, charge transfer ability can be modified by controlling the anisotropic growth of crystals in a certain direction. The β-SnWO 4 enclosed by (100) and (110) facets obtained the strong photocatalytic degradation of methylene blue [19]. Harb et al. found that the (001) facet of α-SnWO 4 and the (100) facet of β-SnWO 4 showed strong charge separation and transport capabilities via calculations with the HSE06 code [10]. They also discussed the (121), (210), (111), (200), and (040) exposed facets of α-SnWO 4 using DFT, and the (210) and (121) facets were shown to perform oxygen evolution reaction due to the positive position of their valence band (Figure 5a) [61]. They also noted that the (001) facet had a high surface energy, but it may be thermodynamically less stable than the other five facets. Wang et al. discussed the oxygen evolution performance of α-SnWO 4 (010) facets with different kinds of termination, and O-Sn termination showed better OER activity than did the others [62]. These DFT researchers have demonstrated that crystal facet engineering is another viable method for obtaining a good SnWO 4 substrate material. Inspired by these projects, our group synthesized an α-SnWO 4 array and added F ions to its precursor solution to increase the ratio of the (001) facet (Figure 5b). The facet-controlled film had a higher photocurrent than the uncontrolled film due to the better OER performance (Figure 5c) [29], and a~1.9 times greater photocurrent was obtained in an unbiased PEC cell.

Doping Modification
Doping can change the position of an energy band, reduce the width of a band gap, and improve the efficiency of carrier separation and transfer, thus improving the catalytic performance of the materials. For example, Zn 2+ doping was introduced into α-SnWO 4 by Su et al. [44]. Because Zn and Sn have similar electronegativity values and suitable sizes, Zn 2+ is used to replace Sn 2+ in SnWO 4 to form Sn 1−x Zn x WO 4 . This ensures that the Zn 4s and Zn 4p orbitals participate in the hybridization of the valence band and the conduction band, respectively, thus changing the position of the energy band and reducing the band gap. The authors also found that Zn 2+ doping changed the morphology of the material, and the resulting Sn 0.955 Zn 0.045 WO 4 could degrade nearly 90% of MO in 20 min. Zhu et al. introduced defects into α-SnWO 4 through Bi 3+ doping [37]. Unlike Zn 2+ doping, Bi does not participate in the formation of energy band hybridization, but cation vacancies are formed due to changes in the charges. These defects form effective energy levels in the band gap which enhance light absorption, inhibit electron-hole pairing, and improve the photocatalytic properties of the material. Although the mechanisms of Zn 2 and Bi 3 doping are not exactly the same, both of them can improve the performance of SnWO 4 by changing its morphology and band structure. At the same time, because β-SnWO 4 has a wider band and a lower carrier separation efficiency than α-SnWO 4 , the doping of the two ions may lead to better performance improvements for β-SnWO 4 . In addition to these two kinds of hetero ions, there may be more suitable ions for SnWO 4 modification (such as Mo and Co) that need to be further studied. In theory, Azofra et al. replaced Sn with Ge and substituted W with Mo in DFT calculations, and they obtained the enhanced orbital hybridization of the VBM/CBM electronic states, which amplified the amount of generated holes/electrons on top of the (110)/(100) facets [63]. However, these result have not been experimentally confirmed.

Multicomponent Composite
The multicomponent composite strategy includes coating a sedimentary protection layer and building heterojunction. Constructed multicomponent catalysts generally have the following advantages over single catalysts: (1) they expand the light absorption spectra, (2) they promote carrier separation, (3) they inhibit carrier recombination, and (4) they prevent photocorrosion.
In photocatalysis, the compositing of reduced graphene oxide (RGO) and β-SnWO 4 can be achieved via microwave heating [12] and hydrothermal methods [50]. In such a composite, RGO, as a supporting material, provides a larger surface area and more active sites than β-SnWO 4 , thereby improving the specific surface area and catalytic activity of the material. At the same time, the high conductivity of RGO makes the electrons generated by β-SnWO 4 move towards the RGO, while the holes remain inside β-SnWO 4 , which improves the carrier separation efficiency of the materials. The compositing of RGO and α-SnWO 4 was reported by Huang et al. [43], and Wang et al. [41] combined hexagonal boron nitride (2D materials analogous to graphite) with α-SnWO 4 . Both groups improved the photocatalytic activity of α-SnWO 4 for degrading methyl orange (MO) and tetracycline. In a composite of Ag-NPs and α-SnWO 4 [38], the photo-absorption efficiency can be increased through the surface plasmon resonance (SPR) of Ag-NPs. Because Ag-NPs have lower Fermi levels than the bottom of the conduction band of α-SnWO 4 , the photoelectrons can migrate to Ag, increasing the efficiency of carrier separation. Moreover, Ag can more easily transfer electrons to a solution when it is in direct contact with the solution, which accelerates the transfer of interfacial carriers. The construction of heterojunctions is widely studied in SnWO 4 material research. The formation of heterojunctions is mainly based on the combination of different energy band positions of different materials. Accordingly, photogenerated electrons are easier to transfer to materials with lower conduction bands, while photogenerated holes tend to transfer to materials with higher valence bands. It is possible to enhance the carrier separation efficiency of materials by selecting suitable materials for compositing. For example, the heterojunctions of α-SnWO 4 and g-C 3 N 4 were shown to effectively inhibit carrier recombination, thereby increasing the catalytic activity of the material [47]. SnWO 4 /ZnO [51], SnWO 4 /BiOBr [49], SnWO 4 /UiO-66/g-C 3 N 4 ternary heterojunction [54], one dimensional ZnWO 4 @SnWO 4 core-shell heterojunction [48], and SnS/α-SnWO 4 [40] have also been reported. In addition, α-SnWO 4 and SnO 2 [42], along with β-SnWO 4 [52], can form heterojunctions to improve the efficiency of carrier separation.
In photoelectrocatalysis, Kölbach and Schnell et al. deposited NiO x on α-SnWO 4 to protect the substrate from photocorrosion [4], which improved material stability and maintained a stable current for sulfite oxidation at 1.23 V vs. RHE in 30 min. Furthermore, α-SnWO 4 is susceptible to photo-corrosion and oxidation during photoelectrochemical water splitting, which results in a lower photocurrent value (<10 µA cm −2 ). However, the photocurrent of the material is greatly increased after protection by NiO x (~0.75 mA cm −2 ). Another study found that a thin oxide layer (SnO 2 ) might form at the interface of α-SnWO 4 and NiO x , which explains the reason for the decreasing photovoltage after the loading of NiO x [9]. Thus, the exploration of alternative techniques for loading protection layers, such as vacuum evaporation and atomic layer deposition, is recommended.

Other Techniques
In addition to the aforementioned approaches, photocatalysis and photoelectrocatalysis performance may be influenced by the external environment [64], as has been studied in recent years. Liu et al. discussed the PEC performance of SnWO 4 /Sn electrodes at different operating temperatures, and they found that higher photocurrents could be obtained at 70 • C (Figure 6a) [55]. Schnell studied the pH-dependent stability of α-SnWO 4 , and pH = 7 (Figure 6b) was suggested to be used for testing due to the formation of the surface oxide layer as a protection layer [65]. Besides, α-SnWO 4 , with higher crystallinity, was formed by Cottesman, presenting~0.95 mA/cm 2 after loading with NiO x for sulfite oxidation.

Conclusions and Outlook
In general, SnWO 4 is considered a type of potential catalytic material for photocatalysis and photoelectrocatalysis, especially for water splitting, because of its optical properties and special band structure (which covers the oxidation and reduction potential of water). β-SnWO 4 is always used in photocatalytic degradation due to the high reduction (oxidation) activity of its photo-generated electrons (holes), while α-SnWO 4 is employed in both photocatalysis and photoelectrocatalysis. In recent years, many researchers have studied the application of SnWO 4 in photocatalysis and photoelectrocatalysis, and they have improved the properties of materials by means of morphology control, doping, and multicomponent combinations. Compared with photocatalysis, photoelectrocatalysis has the advantage of discrete oxidation and reduction processes. This makes it easier to separate gaseous products, study reaction mechanisms and kinetics, and adjust reaction selectivity. Therefore, the application of α-SnWO 4 in photoelectrocatalysis needs to be further explored. However, it is worth noting that the catalytic performance of α-SnWO 4 and its applications in solar energy have not reached their expected values. For example, the highest reported experimental current for PEC water oxidation is only~0.79 mA cm −2 at 1.23 V RHE , which is far from the theoretical value. Other problems include low efficiency and poor stability. To solve these problems, obtaining a high-performing substrate through morphology control, crystal surface control, and doping may prove to be an effective research route. Then, the PEC performance of α-SnWO 4 can be further upgraded by constructing heterojunctions, depositing protective layers, and depositing cocatalysts. Of these methods, loading cocatalysts is the most promising because the catalysts can also act as protection layers. However, a suitable loading method to avoid oxidizing α-SnWO 4 is required, and pinholes should be excluded to prevent α-SnWO 4 from attaching to the electrolytes.

Conflicts of Interest:
The authors declare no conflict of interest.