1 Introduction

Volatile organic compounds (VOCs), which mainly contain alkanes, aromatics, alkenes, carboxylic acids, esters and alcohols [1], have been proven to seriously damaged environment and human health owning to their toxic carcinogenesis and environmental destructiveness such as photochemical smog, greenhouse effect and stratospheric ozone depletion. To solve this problem, several effective VOC elimination techniques such as adsorption [2], ozonation [3], chemical combustion [4], biological degradation [5] and photocatalytic oxidation [6,7,8] have been proposed in recent decades. Among the above methods, photocatalytic oxidation technology is a promising method for removing gaseous pollutants with a low concentration, owing to its excellent features of operation at room temperature and high activity towards various pollutants which can react to final products (CO2 and H2O) [9,10,11]. This technology is basically founded on the application of semiconductor materials with ultraviolet (UV) light at ambient temperature [12, 13]. For instance, TiO2 is regarded as the appropriate semiconductor photocatalyst for converting various VOCs into less harmful molecules due to its low cost, high stability and great capability [14].

Despite its advantages for VOC degradation, some challenges with TiO2 cannot be ignored. For instance, intrinsic large band-gap limited its application to high wavelength region under visible light. In addition, high recombination rate of photo-generated charge carriers, which lowers the photocatalytic application on photocatalysis oxidation under visible light or natural solar light irradiation. Furthermore, the photocatalytic oxidation efficiency is reduced by the high recombination of electrons (e−) and holes (h+). Therefore, photocatalysis process should gain considerable attention to seek highly efficient photocatalysts, which can slow down the recombination rate, accelerate the charge separation efficiency and also respond to visible light absorption [15, 16].

Compared with TiO2 and other semiconductor materials, WO3 has become the ideal choice for photocatalytic oxidation of VOCs due to its exhibition of broader wavelength region in solar spectrum, stable physicochemical properties and better electron transport performance than those of TiO2 [17,18,19,20]. Pristine WO3 is a kind of yellow powder solid, in which oxygen atoms enclose the tungsten atoms to form a corner-sharing distorted octahedra. Oxygen atoms are at the corners of an octahedron and a tungsten atom occupies a center position of the octahedron. In fact, the structure of tungsten oxide is influenced by temperature: a triclinic structure exists from − 50 to 17 °C and then a monoclinic structure is stable from 17 to 330 °C. Above 330 °C and until 740 °C, an orthorhombic structure exists, and finally, above that temperature WO3 becomes the tetragonal phase. The polymorphic property of WO3 could form a phase junction to promote the photocatalytic activity for VOC degradation [21]. However, the unmodified WO3 has a low light energy conversion efficiency as the reduction potential of the electrons in WO3 is low, which accelerates the recombination rate of photo-generated electrons and holes. Some significant efforts are proposed to overcome this shortcoming such as doping [22], noble metal deposition [23] and coupling with other semiconductors [24]. Based on the published literature, despite abundant reported articles focusing on modifying WO3-based photocatalysts, there is an absence of a review on the enhanced performance of tungsten-based photocatalysts for degradation of VOCs in the gas phase.

Here, we provide a short review to enable the reader to make connections among the morphology features, surface defects, electronic properties of the photocatalyst and stable photocatalytic activity toward varies VOCs. Therefore, this paper reviews modification techniques for overcoming the inherent WO3 limitations and improving the photocatalytic activity of VOCs. Such techniques include controlling the morphology, introducing defects, coupling with heterojunction, doping ions and using co-catalyst. This paper also describes the challenges and research directions for the further exploration of VOCs degradation with tungsten-based photocatalysts in engineering applications.

2 Issues in photocatalytic oxidation of VOCs with tungsten oxide

Photocatalytic reactions for degradation of VOCs with semiconductor catalyst are a surface chemical oxidation process. This chemical reaction was focused on the critical process: the semiconductor realizes the band-gap energy requirement to generate chemical active species under the absorption of ultraviolet or visible radiation, then the primary oxidant species (hydroxyl radicals), which are formed by oxidizing adsorbed OH− or adsorbed water for oxidation of VOCs [25]. This oxidation of VOCs include some processes such as adsorption of VOCs in gas phase onto the surface of semiconductors, separation of the electrons and holes and transferring to the surface of semiconductors, generation of reactive oxygen species (ROS) to facilitate chemical degradation, and the desorption of products or intermediates [26]. However, the photocatalysis oxidation of VOCs on surface of some semiconductors exist some inherent drawbacks such as the easy recombination of photogenerated electrons and holes on the surface of the semiconductor materials [27], which impeded the production of ROS.

In addition, compared with the photocatalytic degradation of wastewater in liquid phase, the solid semiconductor catalyst is difficult to evenly disperse in the VOCs pollutants, which restricts the full contact between the catalyst and pollutants. Although advanced and precision characterization and calculation methods are used to research the relationship between the catalyst and pollutants [28, 29], the reaction mechanism for VOCs degradation is still complex, which impedes further research to photocatalytic reactions. Furthermore, it is testified that the deactivation of photocatalysts is closely related to the reactive intermediates of VOCs, which is a main barrier that diminishes the value of practical applications for photocatalytic VOCs degradation [30]. Besides, the complex pollutant composition and rigorous reaction conditions lead to a challenge of confirming the surface oxidation mechanisms for VOC degradation.

As a promising photocatalyst, tungsten trioxide has many advantages, such as strong adsorption with gas phase pollutants, substantial utilization of the abundant solar spectrum, exceptional photochemical stability and resilience to reaction condition effects [31]. However, the low conduct band (CB) location of W-containing single oxide semiconductor has an inferior reductive potential, which restricts their capacity for reduction of oxygen and results in an accumulation of free electrons, then followed by an increased incidence to recombine with holes, which leads to a low conversion efficiency for light energy [32,33,34]. Therefore, WO3 with modification is thought to be appropriate for VOCs decomposition.

3 Strategies for improving the VOCs photocatalytic oxidation performance with modified WO3

3.1 Controlling the morphology structure of WO3

In general, WO3 semiconductors with smooth and porous surface areas are beneficial to improve their photocatalytic performance [35,36,37]. The methods of hydrothermal reactions and template-mediated synthesis were used by Sayama et al. [38] for preparing five WO3 samples (WO3 prepared from amorphous peroxo-tungstic acid-WO3 (PA), and WO3 powder was prepared by calcination-WO3 (C), mesoporous WO3, homemade WO3 powder and commercial WO3), acting as various photocatalysts to degradation of hexane. WO3 (PA) showed a higher photocatalytic activity and more efficient absorption than those of other samples in visible light region because of its special morphology. The flat and smooth morphology of WO3 (PA) with a relatively large particle were measured by an optical microscope. In contrast, in all the other investigated samples with large aggregates, very rough surface morphologies were observed. Those sample powders showed large light reflection because of Mie scattering, which showed low photocatalytic activity for hexane degradation [39]. However, the small roughness and large surface porosity of the WO3 (PA) powder reduced light reflection, which would improve the light absorption and enhance the high activity for hexane degradation.

The various morphologies of tungsten trioxide nanosheets were also synthesized by Wicaksana et al. [40] to research the size and crystallinity of the nanocube structure, nanobundle structure and nanoparticle structure of WO3 in Fig. 1, which can be achieved by adjusting the pH of aqueous solution and regulating the concentration of sulfate precursors. The WO3 nanocubes exhibit greater conversion of ethylene than those of the nanobundles and nanoparticles due to the cuboid morphology with more “edged” nature, which reduce the recombination of photogenerated electron–hole. Xie et al. [41] reported that WO3 in a sheet-like structure could increase the reduction potential of the conduction compared to the cuboid structure because the dominance of the exposed crystal facets blue-shifted the band-gap of the WO3 in a sheet-like form, contributing to the enhanced photo activity of photocatalytic water oxidation and pollutants degradation.

Fig. 1
figure 1

Reproduced with permission from Ref. [40] Copyright 2014 MDPI

SEM image for the different sharp distribution of WO3 structure. a WO3 morphology of slabs with columnar clusters; b WO3 morphology of nanobundles; c WO3 morphology of nanocubes; d WO3 morphology of nanoparticles.

Full size image

3.2 Defect modification

Defects in tungsten oxide materials can exhibit strong adsorption capability towards VOCs [42,43,44,45]. Wang et al. [46] reported oxygen vacancies can act as active sites on the VOC degradation of WO3. The results showed that oxygen vacancies from WO3−x can capture oxygen atoms from the formaldehyde and H2O, which boost the production of hydroxyl radicals and lead to the oxidization and degradation of formaldehyde and benzene. Lu et al. [47] reported that tungsten oxide nanowire bundles with high concentration of oxygen vacancies provided an efficient activity for ethanol dehydration. The oxygen vacancies facilitate full utilization for solar energy and serve as active sites for the improvement of ethanol adsorption and degradation. Action mechanism and oxidation process of WO3−x as photocatalyst were also illustrated in Fig. 2. Defect band (DB) can be formed to serve as an intermediate bridge to induce and transfer hot electrons and hot holes by the oxygen vacancy of WO3−x materials, which can generate the plasmonic thermal effect under full-spectrum light irradiation in Fig. 2a, and then ethanol dehydrogenation and dehydration process would be informed. Finally aldehyde is generated, which was discussed in Fig. 2b.

Fig. 2
figure 2

Reproduced with permission from Ref. [47] Copyright 2019 Elsevier

a Photoelectron and hot electron generation in plasmonic WO3−x under different light irradiation. b Dehydrogenation and dehydration of ethanol over WO3−x as photocatalysts.

Full size image

Similar work was carried out by Bai et al. [48], who synthesized self-organized W18O49 nanowires with high oxygen vacancy concentrations via a facile one-pot method. Depending on the defect structure caused by oxygen vacancy, the W18O49 nanowires show unexpected selectivity in photocatalytic dehydration of isopropyl alcohol to propylene.

3.3 Heterojunction constructing

Doping hybrid WO3 with other semiconductor to form a heterojunction is well-investigated to be an effective way for improving the separation efficiency of the photo-generated charges or extending the spectral response to full light region [49, 50].

3.3.1 WO3/TiO2 heterojunction

TiO2 is a practical photocatalyst due to its strong oxidation power and low material cost. It has been reported that the formation of WO3/TiO2 heterojunction semiconductor can be used as an effective photocatalyst for degradation of VOCs [51, 52]. Zhang et al. [53] reported a hydrothermal method of the electrospun TiO2 nanofibers to fabricate heterostructured TiO2/WO3 nanocomposites, which were loaded onto the inner walls of the photoreactor and fixed with nylon meshes to enhance the contact area between photocatalysts and toluene. The TiO2/WO3 nanocomposites showed a higher toluene degradation degree (85.3%) in flowing air than those of the TiO2 nanofibers and the WO3 nanoparticles. Single crystal structure of WO3 nanorods that closely contacts with the TiO2 nanofibers can reduce resistance of electrons transmission between the grain boundaries of TiO2/WO3 nanocomposites and accelerate the separation of the photo-generated carriers to improve the photocatalytic performance. To further research, the Z-scheme-type mechanism and heterojunction-type mechanism were compared to discuss the efficient photocatalytic decomposition of hexane in Fig. 3 [54]. According to the fluorescence spectra experiments, the heterojunction-type mechanism is more probable to expound the photocatalytic process of vaporous hexane degradation. The photogenerated electrons are stably transferred from the N-doped TiO2 conduction band to that of WO3, accelerating the generation of superoxide, while the hydroxyl radicals are formed by the hole reaction with water and hydroxyl ions, and the hazardous hexane molecules then are decomposed into some intermediate products, such as CO2 and H2O.

Fig. 3
figure 3

Reproduced with permission from Ref. [54] Copyright 2016 Elsevier

a Z-scheme- and b heterojunction-type mechanisms involving N-doped TiO2/WO3 composite catalyst for the decomposition of hexane.

Full size image

3.3.2 WO3/carbon based heterojunction

Recently, graphitic carbon nitride is the widely used photocatalyst because of its high reduction ability, chemical stability and visible-light absorption [55]. Although the composite photocatalysts WO3/g-C3N4 are applied in solar cells, splitting of water and carbon dioxide storage [56,57,58], their photocatalytic application in degradation of volatile organic pollutants are not common. g-C3N4 blended with WO3 was successfully synthesized using a planetary mill by Jin et al. [59]. Compared with original samples, WO3 nanosheets loaded with g-C3N4 particles exhibited a high activity for decomposition of acetaldehyde. The more positive valence band (VB) potential of WO3 with high oxidation ability and more negative CB potential of g-C3N4 with superior reduction ability were utilized by the composite photocatalyst to accelerate electron separate and transfer from WO3 to g-C3N4, which enhanced the complete oxidation of acetaldehyde. Similar research with photodegradation of acetaldehyde (HCHO) under visible-light irradiation was reported by Katsumata et al. [60]. Meanwhile, the enhanced visible-light-driven mechanism for g-C3N4/WO3 composites were discussed. Some of the photoinduced electrons in carbon nitride surfaces are separated and transferred to the CB of WO3, and partial photogenerated holes in WO3 surfaces are transferred to the VB of g-C3N4. This cyclic process enhances the formation of free radicals such as superoxide and hydroxyl radicals.

In addition, ternary heterojunctions such as CaFe2O4/WO3 [61] and CuBi2O4/WO3 [62] were also reported for enhanced oxidation of VOCs.

3.4 Doped with WO3

As a photocatalyst with a relatively wide band-gap, WO3 is motivated by the near ultraviolet and blue regions of the solar spectrum. Doping WO3 with different elements such as Mg [63], Cs [64], Zn [65], and Bi [66] to narrow the band-gap and increase the photocatalytic performance of WO3 were reported. To enhance the absorption of visible light and effective transformation of photo-generated electron–holes, Fe ions were introduced, which can narrow the band-gap of WO3 [67]. Sheng et al. [68] reported an interesting Fe-doped WO3 catalytic material strongly adhered on a wood rabbit craft. This rabbit craft sample doped Fe (the content is 4.56%) showed a 98.21% degradation rate of formaldehyde in 6 h. The detailed photocatalytic reaction mechanism of HCHO was also proposed in Fig. 4: Fe3+ as a scavenger can not only trap the electrons and holes of WO3 to prevent the recombination of excited charge carriers, but also participate in the oxidation reactions with hydroxyl ions and oxygen to form hydroxyl radicals and superoxide radicals, and then these energetic free radicals attack formaldehyde to form the final product such as CO2 and H2O. Irie et al. [69] also did the similar work through doped Cu(II) into the interior structure of WO3, which exhibits 16 times higher photocatalytic ability for 2-propanol decomposition than that of N-doped TiO2.

Fig. 4
figure 4

Mechanism of photocatalytic degradation of HCHO by wood covered with Fe-doped WO3. Reproduced with permission from Ref. [68] Copyright 2016 Elsevier

Full size image

3.5 Cocatalyst loading

Loading a cocatalyst with tungsten oxide were extensively studied in utilization of solar energy to remove harmful organic pollutants in gaseous phase [70,71,72,73,74,75].

3.5.1 Noble metal co-catalyst

Noble metal (Pt, Ag and Pd) loaded onto WO3 were confirmed to be a promising photocatalyst attributed to the surface plasmon resonance effect of noble metal nanoparticles, which can accelerate the effective transfer of free charges to the surface of WO3 and facilitate the generation of active free radicals [76,77,78]. Zhao et al. [79] researched that loading appropriate Pt nanoparticles onto WO3 can enhance photocatalytic activity for removal of acetaldehyde. Meanwhile, 0.5 wt% Pt-loaded WO3 photocatalyst showed a complete decomposition of isopropyl alcohol, which was also reported by Abe et al. [80].

In addition, the photocatalytic properties of silver nanoparticles loaded on tungsten oxide nanocrystals in photocatalytic oxidation of cyclohexane were also researched by Xiao et al. [81]. A probable mechanism of Ag-WO3 nanoparticles with high photocatalytic activity was showed in Fig. 5. The Ag nanoparticles loaded on WO3 nanoparticles act as electronic traps to facilitate the electron-hole separation. Simultaneously, the surface plasmon resonance effect of Ag nanoparticles is attributed to the intense absorb ability of visible light, and then the electromagnetic fields near the surface of Ag nanoparticles is enhanced, which can facilitate the formation of main hydroxyl radicals and easily oxidize cyclohexane. A possible convention pathway for the photocatalytic oxidation of cyclohexane was proposed in Fig. 5b. Hydroxyl radicals (OH) and cyclohexyl peroxy radical (CyOO) are generated to trigger the radical oxidation process, which would selectively oxidize cyclohexane into cyclohexanol and cyclohexanone.

Fig. 5
figure 5

Reproduced with permission from Ref. [81] Copyright 2018 Elsevier

Scheme of a the mechanism and b probable pathway for the photocatalytic oxidation of cyclohexane using Ag-loaded WO3 composite as the catalyst.

Full size image

3.5.2 Loading other cocatalysts

Loading cocatalysts on tungsten oxide, such as carbon nanomaterial and base metal nanomaterial were reported to decomposition of VOCs under visible light irradiation [82,83,84]. Kim et al. [85] prepared a carbon nanomaterial with a core–shell structure, which was loaded with WO3 and applied for the degradation of acetaldehyde and toluene under visible light. Such superior activity of nanodiamond/WO3 is attributed to the graphitic carbon shell on the diamond core, which can promote the charge separation and interfacial electron transfer. Recently, Fukumura et al. [86] synthesized a transparent thin layer photocatalyst, which was formed by a simple physical mixture of WO3 nanoparticles and colloidal CeO2. Further research revealed that CeO2 might behave as a provider of reactive oxygen species in the thin layer photocatalyst system, which promote the degradation reaction of acetaldehyde.

A performance comparison of the modified WO3-based photocatalysts, including the photocatalyst, photocatalytic efficiencies, VOC initial concentration et al., is presented in Table 1. It strongly suggests that modified WO3-based photocatalysts are able to efficiently decompose varies of organic compounds. Furthermore, available data from Table 1 shows that coupling heterojunction and loading a noble metal may be effective methods to the degradation of different target VOCs. Although the performance of removing VOCs with modified tungsten-based catalyst were improved in many researches, there is still a problem to evaluate the photocatalytic oxidation performance on an objective basis and confirm the best photocatalytic material in uniform standards, as each relevant photocatalytic study on VOCs was carried out using different target VOCs and various experimental conditions. There is a long way between the laboratory research achievement and requirements for practical applications.

Table 1 Comparison of WO3-based photocatalysts for VOC degradation
Full size table

4 Conclusion and outlooks

This review proposed the issues of photocatalytic degradation of VOCs with tungsten oxide and summarized a series of strategies for improving the performance of WO3-based catalyst in degradation of VOCs. It is indicated that all the enhancement of the photocatalytic efficiency for degradation of VOCs are concluded following four aspects: (1) broaden light absorption; (2) increase the active reactive sites on the surface of the modified WO3-based photocatalyst, (3) extend the reaction surface area between the WO3-based photocatalyst and VOCs; (4) boost the separation of the photogenerated carriers and inhibit the recombination of charge carriers. The relationship between the various strategies and their possible enhancement mechanisms include: (1) controlling the WO3 morphology and doping suitable dopants, which can increase the light absorption; (2) forming WO3 defect structure that can serve as active sites and enhance the adsorptivity with VOCs; (3) introducing a heterojunction with other semiconductor, which can promote the separation of the photogenerated carriers; (4) loading co-catalyst that can also facilitate the effective charge transfer due to the surface plasmon resonance effect or good electrical conductivity of nanophase materials.

Although the performance of removing VOCs with modified tungsten-based catalyst were improved in many researches, there are still a lot of opportunities to enhance the photocatalytic efficiency of WO3-based photocatalysts for removal of VOCs. Here, several research directions are proposed as reference for confirming the best photocatalytic material and proper practical applications.

  1. (1)

    Although a series of strategies are developed for improving the performance of WO3-based catalyst in degradation of VOCs, it is necessary to introduce synergistic effects with different strategies, which may lead to incredible photocatalytic performance than that of single strategies.

  2. (2)

    The deep mechanism of WO3-based photocatalysts for removal of VOCs is still vague. For instance, the charge-transfer dynamics and construction of heterojunctions mechanism still exist much controversy and lack of theoretical guidelines. Besides, advanced and precision characterization and calculation methods should be applied for VOC degradation, such as in situ Fourier transform infrared spectroscopy and density functional theory calculations.

  3. (3)

    A further challenge should be focused on the development of efficient photocatalysts to solve practical problems. In addition, the efficient utilization of solar energy and lamplight should be highlighted in environmental purification. For example, photo-reactors can be designed for treatment of high concentration VOCs in chemical industries under solar energy, and the tungsten-based catalyst can be attached to the furniture to remove VOCs of indoor air under incandescent lamp. Therefore, there are still a lot of challenges and concerns to be addressed for providing prioritization solutions on VOCs degradation in the future.