Nanostructured MoO3 for Efficient Energy and Environmental Catalysis

This paper mainly focuses on the application of nanostructured MoO3 materials in both energy and environmental catalysis fields. MoO3 has wide tunability in bandgap, a unique semiconducting structure, and multiple valence states. Due to the natural advantage, it can be used as a high-activity metal oxide catalyst, can serve as an excellent support material, and provide opportunities to replace noble metal catalysts, thus having broad application prospects in catalysis. Herein, we comprehensively summarize the crystal structure and properties of nanostructured MoO3 and highlight the recent significant research advancements in energy and environmental catalysis. Several current challenges and perspective research directions based on nanostructured MoO3 are also discussed.


Introduction
In the 21st century, the impending depletion of fossil fuels and urgent environmental concerns are among the most challenging issues. Exploration of renewable and sustainable energy and the development of new and improved materials are the solutions to keep environmental sustainability [1][2][3][4]. Catalysts are increasingly important in the development of innovative clean energy, environmental protection, and energy conversion [5][6][7]. Accordingly, the search for developing superior nanostructured catalysts is a sustainable alternative. Noble metal catalysts such as Pt, Pd, Rh, and Au exhibit high activity, to tackle the energy and environmental challenges. However, the noble metal catalysts are impacted by their scarcity, high cost, and relatively low stability, which impede their general use in large scale applications. As a result, catalysts with higher activity and lower cost are urgently required for large-scale practical applications [8]. Transitional metal oxides dominate widespread The monoclinic structure of β-MoO3 is markedly different from the crystal structure of α-MoO3, which possesses a ReO3-related structure. The MoO6 octahedral units share corner oxygen atoms in the direction of the c axis, and edge sharing occurs in the direction of the c axis (see Figure 1b). A transformation from the β to α phase took place spontaneously at the temperature ranging from 387 to 450 °C, according to the reported result. Moreover, β-MoO3 exhibited higher catalytic properties than α-MoO3 in some catalysis reactions [32][33][34].
Hexagonal h-MoO3 is built up by zigzag chains of MoO6 octahedra linked to each other by corner sharing along the c axis. A very salient crystalline structure for h-MoO3 is the presence of a tunnel (~ 3.0 Å in diameter) running along the c direction. It can serve as a conduit and intercalation site for mobile ions (see Figure 1c). The h-MoO3 phase can generally be formulated as (A2O)x·MoO3·(H2O)y, where A represents an alkali-metal ion or ammonium ion, and the exact values of x and y depend on the details of the preparation and subsequent treatment [6,26,32,35]. The tunnel structure in h-MoO3 exhibits considerably higher sensitivity, coloration efficiency, and faster response in comparison with the orthorhombic α-MoO3. It is mainly attributed to the h-MoO3 of tunnel structure with a higher structure openness degree, giving rise to an accelerated electron-hole separation in photochromism and a facile Li + ion insertion/extraction in electrochromism [36].
The above four different crystal structures have their unique properties, which accordingly influence their applications. The crystal structures of α-MoO3 and MoO3-II have distinctive double The monoclinic structure of β-MoO 3 is markedly different from the crystal structure of α-MoO 3 , which possesses a ReO 3 -related structure. The MoO 6 octahedral units share corner oxygen atoms in the direction of the c axis, and edge sharing occurs in the direction of the c axis (see Figure 1b). A transformation from the β to α phase took place spontaneously at the temperature ranging from 387 to 450 • C, according to the reported result. Moreover, β-MoO 3 exhibited higher catalytic properties than α-MoO 3 in some catalysis reactions [32][33][34].
Hexagonal h-MoO 3 is built up by zigzag chains of MoO 6 octahedra linked to each other by corner sharing along the c axis. A very salient crystalline structure for h-MoO 3 is the presence of a tunnel (~3.0 Å in diameter) running along the c direction. It can serve as a conduit and intercalation site for mobile ions (see Figure 1c). The h-MoO 3 phase can generally be formulated as (A 2 O) x ·MoO 3 ·(H 2 O) y , where A represents an alkali-metal ion or ammonium ion, and the exact values of x and y depend on the details of the preparation and subsequent treatment [6,26,32,35]. The tunnel structure in h-MoO 3 exhibits considerably higher sensitivity, coloration efficiency, and faster response in comparison with the orthorhombic α-MoO 3 . It is mainly attributed to the h-MoO 3 of tunnel structure with a higher structure openness degree, giving rise to an accelerated electron-hole separation in photochromism and a facile Li + ion insertion/extraction in electrochromism [36].
The above four different crystal structures have their unique properties, which accordingly influence their applications. The crystal structures of α-MoO 3 and MoO 3 -II have distinctive double layers. Some positive ions can easily inject into the layer structure, which can be applied in electrochromism, catalysis, and as an electrode material of lithium ion batteries. However, MoO 3 -II is metastable, the crystal structure is easily converted to stable α-MoO 3 , which limits its application. The monoclinic structure of β-MoO 3 is a monoclinic ReO 3 -related structure, in which each MoO 6 octahedral shares all the corners with adjacent MoO 6 octahedral. Breaking the Mo-O bonds can produce more unsaturated Mo atoms on the surface compared with α-MoO 3 , which would behave as active centers for oxidation of small organic molecules, such as partial oxidation of methanol [34]. Unlike α-MoO 3 , metastable h-MoO 3 easily permits the ready intercalation and migration of some monovalent cations because of the open structure tunnels. The unique structure would play an important role in enhancing the charge transfer characteristics and displaying an ionic conductive nature. The h-MoO 3 has the potential to exhibit excellent photochromic, electrochromic, and electrochemical properties. Crystal structure is the intrinsic property, but all the stoichiometric MoO 3 phases have wide bandgaps, which seems to impact some applications. Creating oxygen vacancies, reducing crystal dimensions, introducing dopants, or transforming into quantum dots can be used to manipulate the band structure and improve the performance of MoO 3 .

The Morphology of Nanostructured MoO 3
A large number of architectures with controllable sizes or morphologies have been reported to optimize the structure and composition, such as zero-dimensional (0D) quantum dots (QDs), and one-dimensional (1D), two-dimensional (2D), and three-dimensional (3D) architectures, which is significant for revealing the relationship between structure and performance [6].
QDs represent small-size nanocrystals with all three dimensions in several nanometers. The properties were changed significantly from those bulk semiconductors due to their small sizes and the quantum confinement effect [37]. Lu et al. [38] prepared MoO 3 QDs through combining intercalation and thermal exfoliation and studied the optical properties of MoO 3 QDs. The dispersion of MoO 3 QDs showed a tunable strong localized surface plasmon resonance (LSPR) after simulated solar light illumination and the plasmon peak red shifted with an extension of illumination time, which was different from the MoO 3 nanosheets. Lu and co-workers [39] also observed the morphology changes in combination with different photochromic phenomena through preparation of quantum dots from MoO 3 nanosheets by UV irradiation (see Figure 2a). It suggested that the morphology changes were mainly influenced by consumption of photoexcited holes.
One-dimensional structure is mainly used to describe structures of which the growth is along one direction and the one dimension is less than 100 nm. One-dimensional architectures can be divided into nanowires, nanorods, nanotubes, nanobelts, and so on [26,[40][41][42][43][44][45], which are widely used in many fields due to their unique structural traits. For instance, Meduri et al. [40] reported a MoO 3−x nanowire array (see Figure 2b). The 1D MoO 3 nanowires could provide good conduction pathways for electronic conductivity, along with a shorter path for lithium diffusion, which showed lower lithium intercalation voltages and flat voltage plateau. One dimensional α-MoO 3 nanorods were prepared with a size of about 50 nm in diameter and few microns long by solution combustion method; the TEM image of α-MoO 3 nanorods is shown in Figure 2c. By comparison and analysis, the thermodynamically stable α-MoO 3 single phase exhibited high specific capacitance and good cycle stability for supercapacitor applications [41]. Single-walled MoO 3 nanotubes were synthesized by Wang and co-workers, and the MoO 3 SWNTs (see Figure 2d) were synthesized by a thiol-assisted hydrothermal method through an interface controlled self-rolling mechanism [42]. Enyashin et al. [43] proposed the atomic models of MoO 3 nanotubes and studied the electronic structure and chemical-bonding indices. The results showed that the holey wall structures of MoO 3 nanotubes made it possible to dope the materials with various atoms. Moreover, α-MoO 3 nanobelts can display excellent performance for supercapacitors. It was reported that the α-MoO 3 nanobelts' electrode exhibited a higher specific capacitance of 369 F/g and a good cycle stability, with more than 95% of the initial specific capacitance maintained after 500 cycles, which suggested the α-MoO 3 nanobelts could be a potential electrode material [44]. Zheng et al. [26] prepared h-MoO 3 nanobelts and studied the photochromic and electrochromic properties; the TEM image of as-prepared h-MoO 3 nanobelts is shown in Figure 2e. Higher structure openness degree in the tunnel structure of h-MoO 3 could promote electron-hole separation and allow cation insertion/extraction and diffusion which exhibited a better performance on photochromic and electrochromic response than that of α-MoO 3 .
Two-dimensional nanostructured MoO 3 attracted extensive attention due to its tunable bandgap and high charge carrier mobility. Balendhran and co-workers presented a 2D MoO 3 based biosensing platform. The nanostructured film made of 2D α-MoO 3 nanoflakes was used as the conduction channel, which significantly reduced response time due to the high permittivity of the 2D α-MoO 3 nanoflakes [46]. In addition, Balendhran et al. [47] also demonstrated that the energy bandgap of 2D high-dielectric α-MoO 3 . The α-MoO 3 forming MoO 3−x can reduce the bandgap and enhance charge carrier mobility. Compared with the bulk MoO 3 , 2D-MoO 3 nanosheets had a better chemical sensor performance due to the 2D structure with a large surface area and more reactive sites [48] (see Figure 2f). Cheng et al. [49] reported that MoO 3−x nanosheets can display strong LSPR from the visible to the near-infrared region, owing to the layered crystal structure, which could be used as highly efficient catalysts for the hydrogen generation from ammonia borane with visible light irradiation.
Three-dimensional architectures can increase the surface area and provide more facets with active sites. Three-dimensional architectures mainly include nanoflowers, nanospheres, and mesoporous structures. Nanoflowers can be described as compositions of low-dimensional nano building blocks for certain structures, such as nanorods, nanosheets, and so forth. Sui et al. [50] reported a flower-like α-MoO 3 with hierarchical structure (see Figure 2g). The sensors based on α-MoO 3 flowers exhibited highly sensitive and distinctively selective to trimethylamine. The α-MoO 3 flowers may have less agglomerated configurations, more effective charge transportation, and expose more active sites, as well as specific facets. Another type of promising architecture, nanospheres were widely used in the fields of catalysis, sensing energy storage conversion, and so on [51]. Generally, hollow MoO 3 nanospheres were prepared by drawing support from templates. Triblock copolymer micelles have been used as templates to form hollow MoO 3 nanospheres. The size could be adjusted by choosing a suitable micellar core [52]. Du et al. [53] synthesized highly dispersed MoO 3 nanospheres by ultrasonic irradiation (see Figure 2h) and probed their growth mechanism. Mesoporous structure because of their intrinsic high surface areas and the nano-wall structure were beneficial to diffusing the ions and electrons, which can enhance the electrochemical or photocatalytic properties [6]. Brezesinski and co-workers reported the ordered mesoporous α-MoO 3 nanocrystalline walls for the application of thin-film pseudocapacitors, which exhibited the superior capacitive charge-storage properties compared with both the mesoporous amorphous and nonporous crystalline MoO 3 [54]. Luo et al. also prepared mesoporous MoO 3−x (see Figure 2i) with good activity and stability for HER under both acidic and alkaline conditions [55].
Based on the above overview, the performance of MoO 3 is generally influenced by the morphology, and the essence is that the structure of MoO 3 is different. The controllable preparation of diversified nanostructured MoO 3 is an efficient approach to achieve excellent performance. However, it should be noted that some nanostructured MoO 3 with well-controlled morphologies and sizes were not applied well. Hence, it is better to rationally design and synthesize with well-controlled nanostructured MoO 3 based on advances in nanomaterials science and engineering.

The Application of MoO 3 in Energy Related Catalysis
Due to its low-cost, nontoxicity, multiple oxidation states, and van der Waals gap along the [010] direction, molybdenum oxide (MoO 3 ) has attracted more and more attention in hydrogen evolution, oxygen evolution, and fuel cells in recent years. However, because of its poor inherent electrical conductivity and slow electrochemical kinetic process, the widespread application of MoO 3 is limited, and a great amount of effort has been devoted to enhancing the catalytic performance of MoO 3 .

Application in Hydrogen Evolution
Owing to high calorific value and its renewable and clean properties, hydrogen (H 2 ) is regarded as one of the most promising energy carriers for the future. As a layered n-type semiconductor, MoO 3 has been studied as a catalyst for electrocatalytic hydrogen evolution, photocatalytic hydrogen evolution, and ammonia borane dehydrogenation.

Electrocatalytic Hydrogen Evolution Reactions
Electrocatalytic hydrogen evolution reaction (HER) is an efficient method. It is an environmentally friendly process that does not create any by-products. Although the layered structure of MoO 3 is suitable for insertion/removal of small ions such as H + , MoO 3 always shows poor catalytic performance for HER, due to the lack of active sites [55]. Therefore, MoO 3 always serves as core the substance, as the MoO 3 /MoS 2 core-shell structure [56][57][58]. The MoO 3 core provides high specific surface area and facile charge transport, as depicted in Figure 3a. In order to eliminate the influence of the resistance of the solution and obtain the real kinetic activity of the MoO 3 /MoS 2 , an iR-corrected test was also performed, and the data is shown in Figure 3b. The HER activity of MoO 3 /MoS 2 remained stable even if it was tested for 10,000 cycles ( Figure 3c) [56]. Besides, Pt [59], Pd [60], RuO 2 [19] can also load on MoO 3 , to get excellent electrocatalysts for electrocatalytic hydrogen evolution. Owing to high calorific value and its renewable and clean properties, hydrogen (H2) is regarded as one of the most promising energy carriers for the future. As a layered n-type semiconductor, MoO3 has been studied as a catalyst for electrocatalytic hydrogen evolution, photocatalytic hydrogen evolution, and ammonia borane dehydrogenation.

Electrocatalytic Hydrogen Evolution Reactions
Electrocatalytic hydrogen evolution reaction (HER) is an efficient method. It is an environmentally friendly process that does not create any by-products. Although the layered structure of MoO3 is suitable for insertion/removal of small ions such as H + , MoO3 always shows poor catalytic performance for HER, due to the lack of active sites [55]. Therefore, MoO3 always serves as core the substance, as the MoO3/MoS2 core-shell structure [56][57][58]. The MoO3 core provides high specific surface area and facile charge transport, as depicted in Figure 3a. In order to eliminate the influence of the resistance of the solution and obtain the real kinetic activity of the MoO3/MoS2, an iR-corrected test was also performed, and the data is shown in Figure 3b. The HER activity of MoO3/MoS2 remained stable even if it was tested for 10,000 cycles ( Figure 3c) [56]. Besides, Pt [59], Pd [60], RuO2 [19] can also load on MoO3, to get excellent electrocatalysts for electrocatalytic hydrogen evolution.  [56]. Reprinted with permission from [56]. Copyright 2019 American Chemical Society. (d) Diffuse reflectance ultraviolet-visible spectra (DR UV-vis) and photos (onset images) for comMoO3 and as-synthesized mMoO3; (e) XPS spectrum details for Mo 3d binding energy regions; (f) polarization curves of mMoO3 materials on Ni foam electrode in 0.1 M of KOH [55]. Reprinted with permission from [55]. Copyright 2016 WILEY-VCH.  is presented with its iR-corrected data; (c) the cycling stability of nanowires sulfidized at 200 • C, measured as current density at -0.4 V vs. RHE, normalized to initial current density [56]. Reprinted with permission from [56]. Copyright 2019 American Chemical Society. (d) Diffuse reflectance ultraviolet-visible spectra (DR UV-vis) and photos (onset images) for comMoO 3 and as-synthesized mMoO 3 ; (e) XPS spectrum details for Mo 3d binding energy regions; (f) polarization curves of mMoO 3 materials on Ni foam electrode in 0.1 M of KOH [55]. Reprinted with permission from [55]. Copyright 2016 WILEY-VCH. (g) A photograph of the MoO 3−x product dispersed in ethanol; (h) XRD pattern of MoO 3−x and (inset) crystal structure of the orthorhombic MoO 3 . Time course of H 2 evolution from NH 3 BH 3 aqueous solution at room temperature for different samples: (i) in the dark and (j) under visible light irradiation (λ > 420 nm) [49]. Reprinted with permission from [49]. Copyright 2014 WILEY-VCH.
In order to extend application in electrochemical hydrogen evolution, researchers found that oxygen vacancies can dramatically improve catalytic performance of MoO 3 . Figure 3d showed that MoO 3 with oxygen vacancies possessed a much narrower band gap compared to commercial MoO 3 , and the color changed from greenish to blue (Figure 3d onset). Figure 3e demonstrates the formation of oxygen vacancies in MoO 3 . As shown in Figure 3f, the oxygen vacancies which were close to Mo 5+ can serve as active sites [55]. In this way, 2D α-MoO 3 nanosheets were fabricated and exhibited a considerable HER performance with low overpotential of 142 mV, to achieve 10 mA/cm 2 current density [61]. Zhang et al. also found that MoO 3−y (valence state of V and VI) can be used for efficient electrocatalytic hydrogen evolution [62]. Second, the morphology is another key factor. Mesoporous MoO 3 [55] and MoO 3 nanosheets [61] mentioned above have demonstrated the importance of morphology. Zhang et al. synthesized porous MoO 3 with large specific surface area of 113.8 m 2 /g, and it can work as a bifunctional electrocatalyst for (oxygen evolution reactions) OER and HER [63]. The large specific surface area of as-synthesized porous MoO 3 increased electrochemical active area (0.057 mF/cm 2 ) which was four times bigger than commercial MoO 3 (0.012 mF/cm 2 ). Third, hetero-atoms doping is another efficient strategy to improve electrocatalytic performance of MoO 3 . Li et al. have confirmed that P doped MoO 3−x nanosheet (P-MoO 3−x ) can show the low overpotential of 161 mV to reach 10 mA/cm 2 current density with low Tafel slope of 42 mV per decade [64]. They also found that P doping sites could attract protons to adjacent oxygen vacancies and then form H ads on oxygen vacancies, which meant that P doping sites and oxygen vacancies promoted HER together. Haque et al. also demonstrated that NH 4 + doped MoO 3 (2D Crys-AMO) just need 138 mV to achieve 10 mA/cm 2 current density. After NH 4 + doping, orthorhombic MoO 3 was transformed into hexagonal phase, resulting in the formation of highly ordered intracrystalline pores (serve as active sites) [65]. When the hetero-atoms doping leads to the structure defect, such as oxygen vacancies or transformation of intracrystalline phase, the HER activity of MoO 3 may be promoted dramatically. In addition, when the atomic radius of hetero-atoms is similar to oxygen, it may be easier to be doped into the lattice of MoO 3 . Stability is another crucial factor for catalysts. In general, MoO 3 -based nanocatalysts, which served as substrate or active center, possessed excellent stability in both acid and alkaline media for HER [19,55]. The current density can remain stable at a specific voltage for 12 [55], 24 [19], and even 40 h [65].

Photocatalytic Hydrogen Evolution Reactions
Because solar energy is inexhaustible, photocatalytic hydrogen evolution reaction is an attractive strategy to produce hydrogen from water [66]. MoO 3 is a well-known direct-band-gap semiconductor with high work function and good hole conductivity. However, due to wide band gap energy (about 3.0 eV), the photo-induced electrons/holes (e − /h + ) pairs are easy to recombine, resulting in low conversion efficiency of incident light [67]. To solve this issue, MoO 3 -based nanocomposites were synthesized and studied. Ma et al. fabricated an MoO 3 /polyimide composite, and the growth of MoO 3 increased the light absorption and suppressed the recombination of e − /h + pairs [66]. MoO 3 -TiO 2 nanotubes were studied, and MoO 3 -TiO 2 annealed at 650 • C showed an almost-100-times-higher donor concentration than pure TiO 2 nanotubes. The MoO 3 -TiO 2 nanotubes possessed lower charge transfer resistance and improved separation efficiency because of the appearance of MoO 3 [68]. Esparza et al. fabricated a Mo-coated Pt HER catalyst which was O 2 -insensitive and stable in acidic media. The formed Mo membrane kept H 2 and O 2 far away from Pt, suppressing both oxygen reduction reactions (ORR) and hydrogen oxidation reactions (HOR). Therefore, the HER efficiency of the Mo-coated Pt HER catalyst was promoted [69]. Guo et al. synthesized MoS 2 @MoO 3 core-shell nanowires with a high hydrogen evolution rate of 841.4 µmol/(h·g). The MoO 3 widened the range of light absorption and produced more photo-induced carrier, which accelerated HER rate greatly [70]. Direct bonds formed between CdS NWs and MoO x clusters were achieved. The MoO x clusters induced deep electron-trap states by generating long-lived electrons, to improve the activity. The MoO x clusters could also effectually dissociate adsorbed water molecules, resulting in improved photocatalytic HER activity [71]. Under light irradiation, no obvious diminution in the photocatalytic activity of MoO 3 -based materials for HER was observed for 15-30 h [66][67][68][69][70][71].

Ammonia Borane Dehydrogenation
Due to nontoxicity, room-temperature stability, and high hydrogen storage content (19.6 wt.%), ammonia borane (NH 3 BH 3 ; AB) has been regarded as an attractive hydrogen storage material candidate [49,72]. With suitable catalysts, the hydrolysis of NH 3 BH 3 can be obtained as illustrated in Equation (1): MoO 3 shows strong LSPR signal, a near-field enhancement phenomenon, in the visible light region. Yamashita group firstly reported visible-light-induced hydrogen evolution enhancement from NH 3 BH 3 solution in plasmonic MoO 3−x [49], and the blue color of MoO 3−x (Figure 3g) was consistent with another report [55]. Figure 3h showed that the MoO 3−x is orthorhombic phase. The activity of MoO 3−x was obviously higher than commercial MoO 3 under visible light irradiation (Figure 3i,j). After that, they prepared Pd/MoO 3−x hybrid and it exhibited great plasmon-enhanced hydrolysis of NH 3 BH 3 [72]. The MoO 3−x nanoparticles were fabricated, and the enhanced LSPR property generated by the introduction of oxygen vacancies [73]. Further, they explored the effect of reduction temperature to MoO 3 , and MoO 3−x -200 • C (H 2 reduction temperature) showed a higher dehydrogenation activity [74]. The oxygen vacancies could narrow the band gap of MoO 3 , then increased the absorption of visible light in a wider range. Lu et al. synthesized MoO 3 -doped MnCo 2 O 4 microspheres comprised of nanosheets. The Mo doping provided a small pore diameter leading to an enhanced specific surface area (BET area changed from 13.2 to 62.1 m 2 /g), which is contributed to enhance hydrolysis of NH 3 BH 3 [75]. The stability of existing MoO 3 -based catalyst is not desirable due to relatively short test time (60-70 min) [72][73][74][75].
In summary, the application of MoO 3 in hydrogen evolution is still in preliminary stage. And the further improvement in the activity and stability is still anticipated. Most of works reported that MoO 3 can serve as substance instead of active center in electrocatalytic and photocatalytic hydrogen evolution. As for NH 3 BH 3 dehydrogenation, Yamashita group has been devoted to developing MoO 3 as a promising material. There still exists broad space for the researchers to develop nanostructured MoO 3 as an excellent catalyst for hydrogen evolution. The possible strategy including introduction of oxygen vacancies, hetero-atom doping and hybridization with other materials.

Electrocatalytic Oxygen Evolution Reactions
Electrocatalytic OER is another half-reaction of water splitting. Owing to transfer of four electrons for the evolution of an O 2 molecule, OER is more sluggish compared to HER and becomes the rate-determining step of water splitting [76,77]. There are just a few of research studies about MoO 3 applied in OER, in which MoO 3 is regarded to be catalytically inert for OER [78]. Tariq et al. prepared IrO 2 -MoO 3 composites and 30% mole fraction of iridium contents, which would be more favorable for OER. The mass specific OER activity of iridium active centers was greatly enhanced by seven-fold [76].  Figure 4a showed the CMO with uniformly spherical structure of 500 nm in diameter. Further, on a damaged area (Figure 4b), the core-shell structure can be observed, which means that the CMO possesses a MYLIKE chocolate-like structure. And Figure 4c further demonstrated the uniform distribution of Co, Mo, and O in the CMO. The CMO exhibits the overpotential 340 mV to reach 10 mA/cm 2 current density with a Tafel slope of 49 mV per decade (Figure 4d) [78]. The co-doping, bringing numerous active sites (about 6.550 × 10 −3 mol/g) and amorphous structure of the CMO, was responsible for the improvement of OER activity. The durability of these catalysts was not so good, and the current density could remain stable at 10 mA/cm 2 for only 3-15 h in KOH aqueous solution [76][77][78].

Photoelectrochemical Oxygen Evolution Reactions
Photoelectrochemical (PEC) water splitting has been regarded as a potential avenue for sustainable energy supply. A main obstacle is the need of efficient and stable water oxidation photocatalysts [79,80]. As a high work function (>6.3 eV) and layered semiconductor, MoO 3 always hybridizes with other materials (such as BiVO 4 ) in order to match band potential and obtain an efficient catalyst. Lou  slope of 49 mV per decade (Figure 4d) [78]. The co-doping, bringing numerous active sites (about 6.550 × 10 −3 mol/g) and amorphous structure of the CMO, was responsible for the improvement of OER activity. The durability of these catalysts was not so good, and the current density could remain stable at 10 mA/cm 2 for only 3-15 h in KOH aqueous solution [76][77][78].

Photoelectrochemical Oxygen Evolution Reactions
Photoelectrochemical (PEC) water splitting has been regarded as a potential avenue for sustainable energy supply. A main obstacle is the need of efficient and stable water oxidation photocatalysts [79,80]. As a high work function (>6.3 eV) and layered semiconductor, MoO3 always hybridizes with other materials (such as BiVO4) in order to match band potential and obtain an efficient catalyst. Lou et al. fabricated a Bi2MoO6/MoO3 heterojunction by using anodic oxidation of a molybdenum foil and, subsequently, a hydrothermal method at 160 °C for 24 h. The Bi2MoO6/MoO3 photoanode showed about 100% faradic photocurrent-to-O2 conversion efficiency. They found that, when Bi 3+ was introduced into the MoO3 membrane, the valence band moved upward, which would lengthen the wavelength of light absorbed by Bi2MoO6/MoO3. The photocurrent could remain stable for 8 h by using the Bi2MoO6/MoO3 [79]. He et al. prepared the MoO3/BiVO4 heterojunction film by a drop-casting method, wherein the BiVO4-based precursor was dropped on FTO glass and then dried and annealed. At 0.8 V (vs. SCE), the photocurrent of MoO3/BiVO4 was six times higher than bare BiVO4 film. The improvement can be attributed to band potentials and conductivity differences between MoO3 and BiVO4 [80]. MoO3/Ag/TiO2 nanotube arrays were also successfully synthesized and investigated systemically. The enhanced photoelectrochemical performance is related to the tight contact at the interface. Specifically, the photoinduced electrons could move from valence band of TiO2 to conduction band of MoO3 through a Ti-O-Mo bond [81]. Chen et al. prepared Mo-doped BiVO4/MoOx electrode (1.2% Mo), and Figure 4e shows the formation of the Mo-doped BiVO4/MoOx heterojunction. With the increase of Mo doping amount, MoOx would gradually accumulate and form a film on the surface of BiVO4, and then continue to grow on the film. At 1.23 V (vs. RHE), the photocurrent density of the catalyst (2.67 mA/cm 2 ) was five-fold higher than that of BiVO4 electrode [82]. The valence and conduction band edges of BiVO4 and MoOx are suitable for forming the type II heterojunction at the interface and then accelerate the carrier transfer and separation.
In short summary, the research on MoO3 for the application of OER and PEC water splitting is limited, but the published reports indicate that MoO3 is a promising OER and PEC catalyst through morphology and structure design. MoO3 has a wide band gap (about 3 eV) and easily forms heterojunctions with other materials, so as to obtain photoelectric OER catalysts with higher activity. More effective methods are needed to exploit the potential of MoO3. Meanwhile, the reason for enhanced catalytic behavior also needs to be further investigated, to provide a principle for designing next-generation electrocatalysts. In short summary, the research on MoO 3 for the application of OER and PEC water splitting is limited, but the published reports indicate that MoO 3 is a promising OER and PEC catalyst through morphology and structure design. MoO 3 has a wide band gap (about 3 eV) and easily forms heterojunctions with other materials, so as to obtain photoelectric OER catalysts with higher activity. More effective methods are needed to exploit the potential of MoO 3 . Meanwhile, the reason for enhanced catalytic behavior also needs to be further investigated, to provide a principle for designing next-generation electrocatalysts.

Direct Methanol Fuel Cells
With increasing demand of energy, the direct methanol fuel cells (DMFCs) have attracted more and more attention due to their high energy density (5 kW·h/L). However, the low activity of catalysts and the use of noble metals block its commercial application [83][84][85][86]. Nonstoichiometric MoO 3 (MoO x ) has a rutile-type structure and shows metallic conductivity, which is relatively stable and highly active [83]. Cabrera et al. first reported Pt/MoO x /glassy carbon used for DMFCs by using an electrochemical deposition method. The lower valence of molybdenum and the proton spillover effect from hydrogen molybdenum bronze may account for the improved activity [83,84]. Justin et al. also prepared Pt-MoO 3 /C composite. Insulating MoO 3 was electrochemically reduced to conductive hydrogen molybdenum bronze (HxMoO 3 ), which can keep the Pt surface clean in order to oxidize methanol. MoO 3 plays a crucial role in the transition process of adsorption intermediates to carbon dioxide [85]. Zhang et al. fabricated Pt/MoO 3 nanowires by using an impregnation-calcination method. After loading of Pt nanoparticles, the Pt/MoO 3 maintained its morphology of nanowires, as shown in Figure 5a. Figure 5b,c shows cyclic voltammetry, linear sweep curves of Pt/MoO 3 and Pt/C, respectively. The Pt/MoO 3 showed much higher electrocatalytic activity and stability for the methanol oxidation, compared to Pt/C (the same Pt loading) [86]. The promoted activity may be attributed to more active catalytic sites and three-dimensional structures generating reactant diffusion microchannels. The stability of MoO 3 -based catalysts mentioned above is not satisfactory, and the time of stability test is less than 1 h.

Oxygen-Reduction Reactions
As a type of high efficiency, environmentally friendly energy conversion device, fuel cells have attractive much more attention in recent years [87]. Platinum (Pt) is a benchmark catalyst and always shows excellent performance for ORR. However, it is impossible to use Pt in practical applications that are large in scale because of its scarcity and high cost. Therefore, developing low-content Pt catalysts is a promising way and can be used for real application eventually. Recent researches indicated that MoO 3 can modify a Pt electrode to enhance its catalytic performance and stability [88,89]. After electrodeposition of MoO 3 , the Pt/MoO x /GCE showed a significant increase in the cathodic peak current. The improved performance could be ascribed to the combinations between Pt and MoO 3 , which is a doped effect that always happens between transition elements and leads to the optimal mutual electronic density of states [90]. Karuppasamy et al. fabricated Au-MoO 3 and efficiently decreased the usage amount of Au. The Au/MoO 3 exhibited a one-dimensional morphology of nanorods, and there was no obvious aggregation of Au nanoparticles (Figure 5d). Further, according to particle size of statistics, the mean diameter of Au nanoparticles was 5.9 nm (Figure 5e). After 1000 cycles, no obvious change in the half-wave potential was observed, indicating excellent durability. The HRTEM was employed to study the specific structure of Au/MoO 3 . The MoO 3 made the appearance of low-coordinated stepped Au atoms in the edge, such as (210), (310), and (410), indicating these were main planes of Au nanoparticles (Figure 5f). Then, Figure 5g,h confirmed that the majority of planes of Au nanoparticles was (111) plane. Due to the presence of MoO 3 , the formation of poisonous intermediates, such as OH species, was suppressed on the high index plane of Au nanoparticles; therefore, oxygen was reduced more efficiently [91].
In summary, the glassy carbon electrodes with Pt-modified MoO 3 have been investigated widely for DMFCs and ORR. The MoO 3 can modify the electronic structure of Pt and suppress the formation of poisonous intermediates on high index plane of Au. However, MoO 3 directly used for DMFCs or ORR has not been reported. This phenomenon may be assigned to the relatively fewer research studies and the fact that MoO 3 always serves as a catalyst of partial oxide, suggesting it does not exist at the active sites for breaking O-O bonds. Besides these points, the oxidation mechanism of methanol and oxygen is complex, and there is no clear mechanism explanation about MoO 3 -based materials for DMFCs and ORR. A lot of work still needs to be done for MoO 3 -based materials applied in DMFCs and ORR. In summary, the glassy carbon electrodes with Pt-modified MoO3 have been investigated widely for DMFCs and ORR. The MoO3 can modify the electronic structure of Pt and suppress the formation of poisonous intermediates on high index plane of Au. However, MoO3 directly used for DMFCs or ORR has not been reported. This phenomenon may be assigned to the relatively fewer research studies and the fact that MoO3 always serves as a catalyst of partial oxide, suggesting it does not exist at the active sites for breaking O-O bonds. Besides these points, the oxidation mechanism of methanol and oxygen is complex, and there is no clear mechanism explanation about MoO3-based materials for DMFCs and ORR. A lot of work still needs to be done for MoO3-based materials applied in DMFCs and ORR.

Photodegradation of Organic Pollutants
Photocatalysis is the acceleration of a photoreaction rate in the presence of a catalyst, and it is a green and sustainable catalytic technology that has been widely studied for chemical synthesis, water treatment, environmental cleaning, and self-cleansing processes [92]. The photocatalytic reaction occurs at the interface, based on the absorption of photons with energy larger than the band gap of photocatalyst, with electrons excited from the valence band to the conduction band and  [91]. Reprinted with permission from [91]. Copyright 2017 Elsevier Ltd.

Photodegradation of Organic Pollutants
Photocatalysis is the acceleration of a photoreaction rate in the presence of a catalyst, and it is a green and sustainable catalytic technology that has been widely studied for chemical synthesis, water treatment, environmental cleaning, and self-cleansing processes [92]. The photocatalytic reaction occurs at the interface, based on the absorption of photons with energy larger than the band gap of photocatalyst, with electrons excited from the valence band to the conduction band and producing electron-hole pairs. Photocatalytic efficiency mainly depends on the power and wavelength of the photon source; the properties of the catalyst include its electronic structure, defect density, surface area, and surface-to-volume ratio [93]. Therefore, improving the adsorption capacity of organic pollutants and enhancing the ability of light capture and photo-induced generation performance of electron-hole pairs are important. MoO 3 is one of the most important photocatalysts because of its interesting semiconducting layered structure, rich chemistry associated with multiple valence states and high thermal and chemical stability. Due to the abovementioned outstanding properties, nanostructured MoO 3 has been extensively explored for photocatalytic processes, which are summarized below.
Organic dyes are widely used in textile manufacturing, but they are an environmental threat; hence, the treatment of dyes in wastewater is highly necessary [94]. In general, the degradation rates of rhodamine B (RhB), methylene orange (MO), and methylene blue (MB) are used to evaluate the photocatalytic activity of catalysts. Owing to the distinctive layered structure, MoO 3 exhibited strong adsorption performance and good photocatalytic activity for photodegradation organic pollutants [95][96][97][98]. Inadequately, the large band gap (2.7-3.2 eV) of MoO 3 usually limits the photocatalytic performance of decomposing organic pollutants [99,100]. Based on MoO 3 , various strategies are employed to develop the new efficient photocatalytic materials with the desired properties. These include the doping collaborated with other applicable semiconductors or elements and the fabrication of composites of heterostructures [15].
Hybrid nanomaterials can be designed to help enhance the degradation efficiency. Lots of efforts have been made in the design of MoO 3 -based hybrids such as Eu(Gd)-doped MoO 3 [101,102], rGO/C-MoO 3 [15], and MoO 2 /MoO 3 [100].  Nanostructures/n-TiO 2 nanofiber heterojunctions exhibited a two-times-higher first-order rate constant for the degradation of RhB than that of TiO 2 nanofibers. Significant improvement of the visible-light-driven photodegradation has also been observed, using Bi 2 Mo 3 O 12 /MoO 3 [104], MoS 2 /MoO 3 [105,106], and TiO 2 /MoO 3 heterostructures [107]. It was attributed to combining a feasible semiconducting material with MoO 3 to form heterostructural photocatalysts that can enhance the interfacial charge transfer and minimize the recombination of photogenerated electron-hole pairs, so that the photocatalytic efficiency is greatly improved. Besides, to further shorten degradation period and improve degradation efficiency, the MoO3-based direct solid-state Z-scheme system with a visible-light-driven semiconductor photocatalysts have been studied. He et al. [108] developed Z-scheme type MoO3-g-C3N4 for enhanced photodegradation activity of methylene orange under visible light irradiation. The high photocatalytic activity of MoO3-g-C3N4 is mainly attributed to the synergetic effect of MoO3 and g-C3N4 in electron-hole pair separation via the charge migration between the two semiconductors (see Figure 7a). Figure 7b showed the PL spectra of MoO3, g-C3N4 and 1.5 wt.% MoO3-g-C3N4 samples. The 1.5 wt.% MoO3-g-C3N4 sample had the strongest PL peak, the reason was the combination of high concentrations of the electron on the CB of g-C3N4 and holes on the VB of MoO3, which generated higher concentration ·OH species. Feng et al. [99] reported AgBr/MoO3 monolithic catalyst for degrading RhB under visible light irradiation. The formation of well-defined novel Z-scheme between AgBr and MoO3 effectively accelerated dye-sensitization and charge transfer, resulting in high activity in degrading RhB solution. Figure 7c showed the schematic illustration of photosensitized degradation of the RhB dyes over the AgBr/MoO3 composite before and after contacting for the AgBr/MoO3 system. Further experiments suggested that ultrafast degradation of the RhB on the AgBr/MoO3 nanocomposites was due to both the photocatalytic process and the dye sensitization. The charge-transfer mechanism was shown in Figure 7d. These results opened up a new avenue in surface-and interface-engineering techniques, enhancing the further utilization in the field of energy transformation and environmental improvement.  [100]. Reprinted with permission from [100]. Copyright 2019 Elsevier. (c-e) The energy band alignment of p-MoO 3 nanostructures and n-TiO 2 nanofiber heterojunctions and the postulate mechanism for photodegradation of RhB under UV irradiation [103]. Reprinted with permission from [103]. Copyright 2014 American Chemical Society.
Besides, to further shorten degradation period and improve degradation efficiency, the MoO 3 -based direct solid-state Z-scheme system with a visible-light-driven semiconductor photocatalysts have been studied. He et al. [108] developed Z-scheme type MoO 3 -g-C 3 N 4 for enhanced photodegradation activity of methylene orange under visible light irradiation. The high photocatalytic activity of MoO 3 -g-C 3 N 4 is mainly attributed to the synergetic effect of MoO 3 and g-C 3 N 4 in electron-hole pair separation via the charge migration between the two semiconductors (see Figure 7a). Figure 7b showed the PL spectra of MoO 3 , g-C 3 N 4 and 1.5 wt.% MoO 3 -g-C 3 N 4 samples. The 1.5 wt.% MoO 3 -g-C 3 N 4 sample had the strongest PL peak, the reason was the combination of high concentrations of the electron on the CB of g-C 3 N 4 and holes on the VB of MoO 3 , which generated higher concentration ·OH species. Feng et al. [99] reported AgBr/MoO 3 monolithic catalyst for degrading RhB under visible light irradiation. The formation of well-defined novel Z-scheme between AgBr and MoO 3 effectively accelerated dye-sensitization and charge transfer, resulting in high activity in degrading RhB solution. Figure 7c showed the schematic illustration of photosensitized degradation of the RhB dyes over the AgBr/MoO 3 composite before and after contacting for the AgBr/MoO 3 system. Further experiments suggested that ultrafast degradation of the RhB on the AgBr/MoO 3 nanocomposites was due to both the photocatalytic process and the dye sensitization. The charge-transfer mechanism was shown in Figure 7d. These results opened up a new avenue in surface-and interface-engineering techniques, enhancing the further utilization in the field of energy transformation and environmental improvement. , and Heterojunction-type (d2) charge-transfer mechanisms for the AgBr/MoO3 system [99]. Reprinted with permission from [99]. Copyright 2017 Elsevier B.V.
Apart from the use of decomposing dyes in wastewater, MoO3 is also effective for photoreduction of aqueous Cr (VI). Very recently, Zhang et al. [109] prepared three-dimensional (3D) MoO3@ZIF-8 core-shell nanorod composite photocatalysts and studied the Cr (VI) degradation mechanism, which were shown in Figure 8a,b. Compared with the pure ZIF-8 and MoO3 nanowires, the MoO3@ZIF-8 catalysts exhibited superior photocatalytic activity for Cr (VI) reduction under visible light (see Figure 8c). The reduction of Cr (VI) was up to 100% with 15% of ZIF-8 in 45 min. Moreover, the composite had good recyclability, and the photocatalytic activity remained almost constant after four cycles. Recently, Jing et al. [110] designed and synthesized Mo2C/MoO3 and employed it as catalyst for the photoreduction of Cr (VI) and photodegradation of MO under visible light. The photocatalytic reaction mechanism of photodegradation of MO and photoreduction of Cr (VI) for Mo2C/MoO3 under visible-light irradiation was illustrated (see Figure 8d). The Mo2C/MoO3 heterostructure showed the good photocatalytic activity for wastewater treatment. Apart from the use of decomposing dyes in wastewater, MoO 3 is also effective for photoreduction of aqueous Cr (VI). Very recently, Zhang et al. [109] prepared three-dimensional (3D) MoO 3 @ZIF-8 core-shell nanorod composite photocatalysts and studied the Cr (VI) degradation mechanism, which were shown in Figure 8a,b. Compared with the pure ZIF-8 and MoO 3 nanowires, the MoO 3 @ZIF-8 catalysts exhibited superior photocatalytic activity for Cr (VI) reduction under visible light (see Figure 8c). The reduction of Cr (VI) was up to 100% with 15% of ZIF-8 in 45 min. Moreover, the composite had good recyclability, and the photocatalytic activity remained almost constant after four cycles. Recently, Jing et al. [110] designed and synthesized Mo 2 C/MoO 3 and employed it as catalyst for the photoreduction of Cr (VI) and photodegradation of MO under visible light. The photocatalytic reaction mechanism of photodegradation of MO and photoreduction of Cr (VI) for Mo 2 C/MoO 3 under visible-light irradiation was illustrated (see Figure 8d). The Mo 2 C/MoO 3 heterostructure showed the good photocatalytic activity for wastewater treatment.  [110]. Reprinted with permission from [110]. Copyright 2019 Elsevier.
In brief, MoO3 has shown a good activity in degradation of the organic pollutants due to the design of morphology, hybrid structure, and heterostructure in catalysts. However, most photocatalytic processes were finished to decompose dyes under visible-light irradiation, so it is meaningful to synthesize nano-photocatalytic materials with a wide variety of wavelength bands. Additionally, based on the reported works, the study of stability and recyclability is poor; however, it is crucial for the practical applications of catalysts.

Selective Catalytic Reduction of NOx with NH3
Nitrogen oxides (NOx) cause many environmental pollution problems, such as acid rain, photochemical smog, and ozone depletion. Selective catalytic reduction (SCR) of NOx with NH3 has become the dominant technology for decreasing NOx emission from stationary and mobile sources [111,112]. MoO3 plays an important role in traditional catalysts, for example, with MoO3-based catalysts for SCR exhibiting good catalytic NOx removal performance. The redox behavior of V2O5-MoO3/TiO2 catalysts was investigated by Casagrande and co-workers, who pointed out that the ternary catalysts are more easily reduced and reoxidized than the corresponding binary samples (V2O5/TiO2 and MoO3/TiO2 catalysts) [113]. Recently, based on CeO2/MoO3 binary or ternary catalysts that were favored by researchers and CeO2 that possesses excellent redox property also with plenty of Lewis acid sites, the catalysts with MoO3 loaded on CeO2 have been well studied, with much more acid sites on surface. The synergistic effect between CeO2 and MoO3 contributed to the selective oxidation of NH3 to N2 [114,115]. For instance, Zhu et al. [116] studied the surface  [110]. Reprinted with permission from [110]. Copyright 2019 Elsevier.
In brief, MoO 3 has shown a good activity in degradation of the organic pollutants due to the design of morphology, hybrid structure, and heterostructure in catalysts. However, most photocatalytic processes were finished to decompose dyes under visible-light irradiation, so it is meaningful to synthesize nano-photocatalytic materials with a wide variety of wavelength bands. Additionally, based on the reported works, the study of stability and recyclability is poor; however, it is crucial for the practical applications of catalysts.

Selective Catalytic Reduction of NO x with NH 3
Nitrogen oxides (NO x ) cause many environmental pollution problems, such as acid rain, photochemical smog, and ozone depletion. Selective catalytic reduction (SCR) of NO x with NH 3 has become the dominant technology for decreasing NO x emission from stationary and mobile sources [111,112]. MoO 3 plays an important role in traditional catalysts, for example, with MoO 3 -based catalysts for SCR exhibiting good catalytic NO x removal performance. The redox behavior of V 2 O 5 -MoO 3 /TiO 2 catalysts was investigated by Casagrande and co-workers, who pointed out that the ternary catalysts are more easily reduced and reoxidized than the corresponding binary samples (V 2 O 5 /TiO 2 and MoO 3 /TiO 2 catalysts) [113]. Recently, based on CeO 2 /MoO 3 binary or ternary catalysts that were favored by researchers and CeO 2 that possesses excellent redox property also with plenty of Lewis acid sites, the catalysts with MoO 3 loaded on CeO 2 have been well studied, with much more acid sites on surface. The synergistic effect between CeO 2 and MoO 3 contributed to the selective oxidation of NH 3 to N 2 [114,115]. For instance, Zhu et al. [116] studied the surface structure of M x O y /MoO 3 /CeO 2 system (M = Ni, Cu, Fe) and its influence on SCR of NO by NH 3 . The results revealed that the intensity of the interaction between MoO 3 and other metal oxides was different, which could be listed as follows: NiO/MoO 3 /CeO 2 > CuO/MoO 3 /CeO 2 > Fe 2 O 3 /MoO 3 /CeO 2 . The NO conversion for different catalysts in "NO + NH 3 + O 2 ' reaction is shown in Figure 9a. Figure 9b shows the schematic drawing of ammonia adsorption and decomposition on Brønsted and Lewis acid sites. The reactivity of "NO + NH 3 + O 2 " reaction was strongly associated with acid properties of the catalysts. Additionally, some researchers studied the catalytic behaviors of MoO 3 -based catalysts with respect to resistance to phosphorus, as shown in Figure 9c,d. The 1.3P/Ce-Mo(0.5)-O catalyst showed a higher NH 3 reaction rate than the other two catalysts below 350 • C, which suggested that the resistance of the CeO 2 catalyst to phosphate was improved with the addition of Mo, and phosphorus poisoning significantly affected catalyst activity [115]. Nevertheless, the NO x conversion is still within a relatively narrow temperature window for MoO 3 -based NH 3 -SCR catalysts, and it is necessary to develop catalysts with superior catalytic performance at both high and low temperatures, with excellent resistance to coexisting poisoning pollutants such as SO 2 and phosphate.  Figure 9a. Figure  9b shows the schematic drawing of ammonia adsorption and decomposition on Brønsted and Lewis acid sites. The reactivity of "NO + NH3 + O2" reaction was strongly associated with acid properties of the catalysts. Additionally, some researchers studied the catalytic behaviors of MoO3-based catalysts with respect to resistance to phosphorus, as shown in Figure 9c,d. The 1.3P/Ce-Mo(0.5)-O catalyst showed a higher NH3 reaction rate than the other two catalysts below 350 °C, which suggested that the resistance of the CeO2 catalyst to phosphate was improved with the addition of Mo, and phosphorus poisoning significantly affected catalyst activity [115]. Nevertheless, the NOx conversion is still within a relatively narrow temperature window for MoO3-based NH3-SCR catalysts, and it is necessary to develop catalysts with superior catalytic performance at both high and low temperatures, with excellent resistance to coexisting poisoning pollutants such as SO2 and phosphate.  [116].

Selective Catalytic Oxidation of Propene
Propene has an extensive production and application in numerous industries, and it is a primary contributor to photochemical smog, making it unfriendly to the environment [117]. There are lots of reports focusing on the catalytic oxidation of propene. MoO3 is well-known for its structure sensitivity in selective oxidation of propene. The products of selective oxidation mainly include acrylic acid and acrolein. Schuh et al. [118] studied the influence of the morphology of  [116]. Reprinted with permission from [116]. Copyright 2009 Elsevier B.V. (c) NH 3 reaction rate vs. temperature over the Ce-Mo(0.5)-O and CeO 2 catalysts with different phosphorus content in NH 3 oxidation reaction. (d) Normalized NO oxidation rate vs. temperature in NO oxidation reaction over different catalysts after P poisoning [115]. Reprinted with permission from [115]. Copyright 2013 American Chemical Society.

Selective Catalytic Oxidation of Propene
Propene has an extensive production and application in numerous industries, and it is a primary contributor to photochemical smog, making it unfriendly to the environment [117]. There are lots of reports focusing on the catalytic oxidation of propene. MoO 3 is well-known for its structure sensitivity in selective oxidation of propene. The products of selective oxidation mainly include acrylic acid and acrolein. Schuh et al. [118] studied the influence of the morphology of α-MoO 3 in the selective oxidation of propylene. The results suggested that the morphologies of the samples have a significant effect not only on the selectivity to acrolein, but also on the propylene conversion (see Figure 10a,b). The rod-like structures with (100) facet seemed to be of decisive importance for the catalytic activity, and the same conclusion was drawn by Volta and coworkers [119,120]. Moreover, supported Mo oxides may exhibit different structural and catalytic properties for selective oxidation of propene. Ressler et al. [121] systematically studied the catalytic properties of MoO 3 supported on nanostructured SiO 2 and compared the conversion quantity of h-MoO 3 /SiO 2 with that of α-MoO 3 by mass spectrometric analysis of the gas-phase composition during thermal treatment of in propene and oxygen (see Figure 10c). It exhibited stability and catalytic properties different from other binary bulk oxides and could directly oxidize propene to acrylic acid without additional metal sites. Besides this, MoO 3 was used for the epoxidation of propylene to propylene oxide (PO) by molecular oxygen [122]. The work of Jin et al. [123,124] on Ag-MoO 3 and Ag-MoO 3 -ZrO 2 catalysts showed that MoO 3 resulted in a significant improvement in the efficiency of the catalyst. The addition of MoO 3 increased the selectivity to 34% at a propylene conversion of 4.6%, whereas the pure Ag catalyst had a PO selectivity of only 0.8% at a propylene conversion of 11%. The olefinic carbon of propylene was easily adsorbed by adding MoO 3 into the silver catalyst. MoO 3 was as an electron and structure-type bifunctional promoter in this system. α-MoO3 in the selective oxidation of propylene. The results suggested that the morphologies of the samples have a significant effect not only on the selectivity to acrolein, but also on the propylene conversion (see Figure 10a,b). The rod-like structures with (100) facet seemed to be of decisive importance for the catalytic activity, and the same conclusion was drawn by Volta and coworkers [119,120]. Moreover, supported Mo oxides may exhibit different structural and catalytic properties for selective oxidation of propene. Ressler et al. [121] systematically studied the catalytic properties of MoO3 supported on nanostructured SiO2 and compared the conversion quantity of h-MoO3/SiO2 with that of α-MoO3 by mass spectrometric analysis of the gas-phase composition during thermal treatment of in propene and oxygen (see Figure 10c). It exhibited stability and catalytic properties different from other binary bulk oxides and could directly oxidize propene to acrylic acid without additional metal sites. Besides this, MoO3 was used for the epoxidation of propylene to propylene oxide (PO) by molecular oxygen [122]. The work of Jin et al. [123,124] on Ag-MoO3 and Ag-MoO3-ZrO2 catalysts showed that MoO3 resulted in a significant improvement in the efficiency of the catalyst. The addition of MoO3 increased the selectivity to 34% at a propylene conversion of 4.6%, whereas the pure Ag catalyst had a PO selectivity of only 0.8% at a propylene conversion of 11%. The olefinic carbon of propylene was easily adsorbed by adding MoO3 into the silver catalyst.
MoO3 was as an electron and structure-type bifunctional promoter in this system. The acrolein selectivity as a function of propylene conversion [118]. Reprinted with permission from [118]. Copyright 2015 Elsevier B.V. (c) Evolution of ion currents of CO2, acrolein, and acrylic acid obtained by mass spectrometric analysis of the gas phase composition during thermal treatment of MoO3/SiO2 or α-MoO3 in propene and oxygen [121]. Reprinted with permission from [121]. Copyright 2007 Elsevier B.V.

Selective Catalytic Oxidation of Methane
As one of the greenhouse gases, methane effects on global warming. Hence, its emissions have attracted more and more attention [125]. The conversion of CH4 into chemicals is much more economical and energy-efficient compared to CO2. Many publications in this field have been totally The acrolein selectivity as a function of propylene conversion [118]. Reprinted with permission from [118]. Copyright 2015 Elsevier B.V. (c) Evolution of ion currents of CO 2 , acrolein, and acrylic acid obtained by mass spectrometric analysis of the gas phase composition during thermal treatment of MoO 3 /SiO 2 or α-MoO 3 in propene and oxygen [121]. Reprinted with permission from [121]. Copyright 2007 Elsevier B.V.

Selective Catalytic Oxidation of Methane
As one of the greenhouse gases, methane effects on global warming. Hence, its emissions have attracted more and more attention [125]. The conversion of CH 4 into chemicals is much more economical and energy-efficient compared to CO 2 . Many publications in this field have been totally committed to the oxidation of methane at relatively high temperature, using transition metal oxides catalysts. The catalysts based on MoO 3 were the most widely studied because of their higher activity and selectivity for methane oxidation [126]. Liu et al. [127] prepared MoO 3 /SiO 2 catalysts and studied the kinetics and mechanism of the partial oxidation of methane, using N 2 O as the oxidant. These researchers also found that the selective oxidation reaction was initiated by the formation of O − ions generated from the interaction of N 2 O. The structural effects of MoO 3 on partial oxidation of methane to formaldehyde were investigated by Smith et al. The results indicated that Mo=O sites located on the side plane tend to form formaldehyde [128]. Moreover, Arena et al. [129] studied the working mechanism of MoO 3 /SiO 2 catalysts in the partial oxidation of methane to Formaldehyde. The influence of the oxide loading on the surface structure and compared with the dispersity of MoO 3 /SiO 2 and V 2 O 5 /SiO 2 catalysts. Taylor et al. [130] prepared a series of catalysts based on MoO 3 and WO 3 , the MoO 3 based catalysts were more effective for the production of methanol. The Cu/MoO 3 and Ga 2 O 3 /MoO 3 catalysts showed selectivity and methanol yield advantage, respectively. MoO 3 demonstrated oxygen insertion ability due to its n-type semiconductivity. However, the mechanism of the co-operative effect in catalysis was not well studied. Most studies focused on the catalytic activity and selectivity, but the research of durability is rather poor.

Other Catalysis to Reduce Air Pollutants
There are only a few reports about MoO 3 catalysts applied for oxidation of CO and some volatile organic compounds (VOCs), such as methanol, (CH 3 ) 2 S 2 , benzene, and chlorobenzene [131][132][133]. Mohamed et al. [131] reported the oxidation of CO to CO 2 by MoO 3 /CeO 2 , and the results suggested that the dispersed MoO x species and Ce 3+ /Ce 4+ redox couples had high capacity toward oxygen, which was most likely to be the active species for CO oxidation. Wang et al. [132] studied the catalytic incineration of (CH 3 ) 2 S 2 on CuO-MoO 3 /γ-Al 2 O 3 and the promoter effect. The results revealed that CuO-MoO 3 /γ-Al 2 O 3 has a good activity and durability, and Cr 2 O 3 was the most effective promoter. Moreover, MoO 3 as a dopant in V 2 O 5 -MoO 3 /TiO 2 catalysts improved redox properties and enhanced chlorobenzene oxidation catalytic activity at a low temperature, and this illuminated new applications for pollutants [133]. These researches will lay an early foundation for exploring MoO 3 catalytic properties in the catalysis of other VOCs.

Summary and Outlook
Nanostructured MoO 3 with unique structures and wide availabilities become one of the most promising materials employed for versatile applications. In this review, we have presented the basic structure and main applications of nanostructured MoO 3 . The MoO 3 mainly consists of four phases: orthorhombic (α), monoclinic (β), hexagonal (h), and high pressure monoclinic (ε). All of these phases display unique physical and chemical properties for different performances.
We mainly outlined the applications of nanostructured MoO 3 in the field of energy and environmental catalysis, including water splitting, fuel cells, photocatalytic degradation, and selective thermocatalysis. The catalytic activities of nanostructured MoO 3 will be improved through the design of morphology, hybridization, and hybrid structure. The bandgap structure is optimized, and charge transfer is enhanced by constructing heterojunction nanocomposites. Although progress has been made in catalysis based on nanostructured MoO 3 , the current research is at nascent stage. It is still required to develop new nanostructured MoO 3 materials to meet more demands in practical applications. For instance, selective catalysis to remove air pollutants requires very high temperatures; therefore, the exploitation of new and improved nanostructured MoO 3 applied to mild conditions is needed. The catalytic nature and mechanism in MoO 3 -based catalysts are still challenging and need further studies. Limited research has been done to investigate the stability and recyclability in the selective thermocatalysis. There are numerous research studies about MoO 3 materials used for energy catalysis, especially in the field of hydrogen evolution. All kinds of strategies, including (1) morphology control, (2) phase transition, (3) introduction of oxygen vacancies, (4) hetero-atom doping, and (5) hybridization with other materials, have been employed to improve catalytic performance and durability of MoO 3 -based materials. However, the activity or performance of MoO 3 -based materials still do not completely reach the commercial standard, and the mechanism is not fully understood. The strategies mentioned above have been carried out, but there is no systematic and deep study on morphology, phase, oxygen vacancies, etc. Therefore, many problems, such as 'how do oxygen vacancies influence catalytic performance?' still need to be solved. It is more common to see that MoO 3 serves as a support material instead of active centers as a whole. Although some researchers have devoted themselves to addressing this [55,[61][62][63][64][65], the activity enhancement is still limited. On the other hand, the durability of MoO 3 -based materials still needs to be improved, and the stability test time should be prolonged. Thus, MoO 3 has huge treasure to be excavated for HER, OER, and fuel cells. In short summary, generally speaking, there are two aspects for further studying MoO 3 in the future: (i) the great improvement of catalytic performance and stability; and (ii) insight into structure-activity relationship and mechanism. Novel strategies and new methods must be proposed in order to achieve these long-term goals.