With the widespread use of fossil fuels and the rapid development of modern industry, gas pollution is becoming a serious issue that we face today. Nitrogen oxides [consisting of mainly nitric oxide (NO) and nitrogen dioxide (NO₂)] are common air pollutants that can lead to not only acid rain, tropospheric ozone depletion, and the greenhouse effect, but also respiratory disease and skin allergies [1-6]. Therefore, various techniques have been developed for NO removal, including physical adsorption [7-9], biofiltration [10, 11], and thermal catalytic reduction [12, 13]. However, the above-mentioned methods are usually complex, expensive, and unsuitable to remove low concentrations of NO (ppb level). Recently, photocatalysis has been considered as a promising technique to achieve low-concentration NO removal. Various photocatalysts such as metals and metal oxides (Bi, TiO₂, BiOX: X = Cl, Br, I, or CO₃, ) [14-20], perovskite-type oxides (ATiO₃, where A = Ba, Sr or Pb) [21-23], and metal-free compounds (C₃N₄) with high NO photodegradation activity have been reported [24, 25]. However, most reported photocatalysts generate a high fraction of NO₂, which is more poisonous than NO, leading to secondary pollution. Therefore, the development of photocatalysts with high selectivity for NO removal is still a great challenge.

Tungsten oxide (WO₃) is a typical n-type semiconductor that has attracted great attention because of its nontoxicity, stable physicochemical properties, and good resistance to photocorrosion [26-34]. Importantly, both theory and experiments have proved that WO₃ possesses a relatively low valence-band edge, which endows it with strong oxidation ability and makes it a promising candidate for NO removal [35, 36]. However, the inherent deficiencies of WO₃, including its low specific surface area, wide bandgap, and high recombination rate of photogenerated electron-hole pairs, severely lower its photocatalytic performance. In addition, most of the reported WO₃ photocatalysts are in powder form, so they have a strong tendency to agglomerate and are difficult to recycle.

To overcome these typical limitations of photocatalysts, a three-dimensional (3D) macroscopic graphene aerogel and its derivatives (reduced graphene oxide (RGO) aerogel and graphene oxide (GO) aerogel) with large surface area, low density, and high porosity have been developed. For example, Li and co-workers synthesized BiOBr/RGO, TiO₂/RGO, and Cu₂O/RGO nanocomposite aerogels via a freeze-drying method [37-40]. These nanocomposite aerogels exhibited high activity for the degradation of pollutants and were easily separated from the aqueous reaction systems to allow easy recycling, opening a new avenue for the development of integrated photocatalysts. In these systems, the graphene aerogel acts as not only the substrate but also has a positive effect on photocatalytic performance because it increases light absorption and slows the recombination of electron-hole pairs [41-44]. Moreover, the monolithic structure of photocatalyst aerogels makes them easy to recycle.

In this work, we report the fabrication of a monolithic WO₃/GO aerogel photocatalyst via a cost-effective freeze-drying process. GO acts as a scaffold to support WO₃ and also enhances the photocatalytic NO degradation activity of WO₃. Moreover, the obtained WO₃/GO aerogel exhibits excellent selectivity, with NO₂ generation fraction as low as 0.5%. Our results demonstrate that WO₃/GO aerogels show great potential for use in indoor air purification.

All chemical reagents were purchased from Chengdu Kelong Co. Ltd. in analytical grade and used without any further purification treatment.

Hexagonal WO₃ powder was prepared via a hydrothermal route according to our previous work [45, 46]. In a typical procedure, ammonium metatungstate (248 mg) and acetic acid (2 mL, 25 vol %) were placed in a 25-mL Teflon-lined stainless-steel autoclave. The autoclave was heated at 180 ℃ for 48 h and then cooled to room temperature. The solid residue was washed with deionized water and then dried at 70 ℃ to obtain WO₃ powder.

GO was synthesized through a modified Hummers method [40]. The final GO solution was diluted to a concentration of 4 mg/mL for use in further experiments.

The prepared WO₃ powder (360 mg) was dispersed in deionized water (90 mL) by ultrasonication at 300 W and 40 kHz for 1 h. GO solution (10 mL, 4 mg/mL) was added dropwise to the above suspension. The mixture was stirred for 24 h to achieve a uniform suspension and then frozen at -80 ℃ for 2 h. After freeze-drying for 48 h, 90 wt% WO₃/GO aerogel was obtained. For comparison, WO₃/GO aerogels with different contents of WO₃ (70 and 80 wt%) were prepared via a similar synthesis process, and are denoted as 70 wt% WO₃/GO and 80 wt% WO₃/GO, respectively. Unless otherwise stated, the WO₃/GO aerogel referred to is the 90 wt% WO₃/GO sample.

Powder X-ray diffraction (XRD) was performed with a PANalytical X'pert diffractometer operated at 40 kV and 40 mA using Cu K_α radiation. Transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) images were recorded on an FEI Tecnai G2 20 microscope operated at 200 kV. Ultraviolet-visible (UV-Vis) diffuse reflection spectra (DRS) were measured on a Shimadzu UV-2600 spectrophotometer. Fourier transform infrared (FTIR) spectra were collected on a Nicolet 6700 spectrometer. X-ray photoelectron spectroscopy (XPS) was performed on a Thermo ESCALAB 250Xi spectrometer with Al K_α radiation at 1486.6 eV. Brunauer-Emmett-Teller (BET) surface area measurements were performed using the nitrogen adsorption-desorption technique on a Quantachrome ASAP 2020 HD88 surface analyzer. Photoluminescence spectra (PL) were measured using Hitachi F-7000 with the light source of MVL-210 (Mejiro Genossen Inc.).

Photocatalytic removal of NO at ppb level was measured using a custom-made testing system [19]. In this experiment, photocatalyst (400 mg) was split equally between two petri dishes that were used in the activity testing. The flow rates of air and NO were 2.0 and 12.0 mL/min, respectively. The removal ratio (η) of NO was calculated by Eq. (1):

(1)

where C is the concentration of NO during the reaction and C₀ is the initial concentration of NO. In addition, the generation ratio (ω) of NO₂ during the photocatalytic reaction was calculated by:

(2)

where C(NO_x) is the total concentration of nitrogen oxides (NO₂ and NO; C(NO_x) = C(NO)+ C(NO₂)) and C₀(NO₂) is the initial concentration of NO₂ before the photocatalytic reaction.

Fig. 1 shows a schematic illustration of the fabrication process of the WO₃/GO aerogel. Firstly, the green WO₃ powder was dispersed in deionized water by ultrasonication for 1 h (Fig. 1(b)) to obtain a homogeneous suspension. Secondly, GO suspension was added dropwise under continuous stirring to achieve a homogenous mixture in which WO₃ powder was uniformly attached on the surface of the GO nanosheets (Fig. 1(a), (c), and (d)). The obtained brown mixture was then freeze-dried to remove water to provide the low-density WO₃/GO aerogel, which can stably rest on a leaf (Fig. 1(e)).

Fig. 2 presents the XRD patterns of GO, WO₃, and WO₃/GO aerogel. GO exhibited one peak at 9.9°, whereas WO₃ displayed four peaks located at 13.95°, 22.71°, 28.17°, and 37.63° that were assigned to the (100), (001), (200), and (210) facets of hexagonal WO₃ (JCPDS 33-1387), respectively [40-44]. The XRD pattern of the WO₃/GO aerogel contained all the peaks from WO₃, indicating that GO did not changed the structure of WO₃. The (001)/(200) and (100)/(200) peak intensity ratios of 0.61 and 0.35, respectively, for the WO₃/GO aerogel were lower than those for WO₃ (0.95 and 0.63, respectively). This influence of the GO nanosheets confirmed their successful introduction to WO₃.

The microstructure of WO₃ and the WO₃/GO aerogel was studied by TEM. As shown in Fig. 3(a) and (b), the WO₃ powder has a plate-like morphology that did not change after the formation of the WO₃/GO aerogel. In addition, two crystal facets with lattice spacings of 0.312 and 0.382 nm consistent with the (200) and (001) facets of WO₃, respectively, with a crossing angle of 90° were observed in the HRTEM images of both WO₃ and the WO₃/GO aerogel (Fig. 3(b) and (d)). The introduction of GO markedly increased the dispersity of WO₃, which is beneficial for enlarging surface area [46].

The BET analysis of WO₃ and the WO₃/GO aerogel was performed. Fig. 4 shows the N₂ adsorption-desorption isotherms for WO₃ and the WO₃/GO aerogel. WO₃ exhibited a type-Ⅳ isotherm pattern with H₁ hysteresis loop, which is associated with the straight pore structure caused by the random accumulation of WO₃ plates. In contrast, the WO₃/GO aerogel displayed H₃ hysteresis loop, which is attributed to a slit-like pore structure, indicating the favorable dispersion of WO₃ on the GO surface, in accordance with the TEM analysis. The increased dispersity of WO₃ in the aerogel enlarges the surface area from 46 to 57 m²/g, which should improve photocatalytic performance [47].

XPS measurements were used to investigate the chemical states, binding energies, and interactions in the aerogel. Fig. 5 depicts the high-resolution W 4f, O 1s, C 1s, and N 1s spectra. Binding energies were referenced to C 1s at 284.8 eV. The peaks for WO₃ at 35.5 and 37.6 eV corresponding to W 4f_5/2 and W 4f_7/2, respectively, can be ascribed to the chemical state of W⁶⁺ in WO₃. These peaks shifted to higher binding energies of 35.7 and 37.8 eV after the introduction of GO [48]. The two peaks in the O 1s spectrum of WO₃ at 530.3 and 531.3 eV are related to W–O and C=O bonds, respectively. After coupling with GO, two new peaks appear at 532.1 and 532.6 eV that are attributed to the C–O–C and –COOH bonds of GO, respectively. In the C 1s spectra, the peak at 286.0 eV is assigned to the C–O bonds on the surface of WO₃, while a new peak at 286.8 eV corresponding to the C=O bond of GO is observed after the introduction of GO [49, 50]. In the N 1s spectra, the peaks located at 399.8 and 401.7 eV are ascribed to the N–H bond in NH₂^– and nitrogen atom of NH₄⁺, respectively. After the formation of the WO₃/GO aerogel, the intensities of the N 1s peaks weaken because of the decrease of the surface content of nitrogen atoms. In addition, the binding energy shifts from 399.8 to 400.3 eV, which indicates the decreased electron density on the N atoms in the aerogel. Fig. 6 shows FTIR spectrum of GO, WO₃ and WO₃/GO aerogel. The characteristic peaks located at 784 cm^-1, 1041 cm^-1, 1637 cm^-1 and 3230 cm^-1 can be ascribed to the O–W–O, C–O, C=O and N–H bonds [46, 51]. Importantly, the peak located at 3230 cm^-1 from N–H bonds disappeared after coupling with GO, indicating the release of NH₄⁺ ions or NH₃ molecules from the channels of WO₃ during the ultrasonication treatment, consistent with the XPS analysis results.

Fig. 7(a) shows the UV-Vis DRS results for WO₃, GO, and the WO₃/GO aerogel. The absorption edge of WO₃ was around 360 nm, indicating a bandgap of 3.44 eV. After coupling with GO, a red shift of the absorption edge was observed, resulting in a narrow bandgap of 3.16 eV, which is attributed to the black-body and surface sensitization behavior of GO and demonstrates the broadened light response of the WO₃/GO aerogel compared with that of WO₃ [18].

Photooxidation of NO over WO₃ and the WO₃/GO aerogel was carried out under visible-light illumination. As illustrated in Fig. 7(b), all the aerogel samples exhibited enhanced NO removal compared with that of WO₃. The highest removal ratio of 51% was obtained for the 90 wt% WO₃/GO aerogel, which is 3.3 times higher than that of WO₃. It should be noted that the integrated structure of the aerogel collapsed when the mass ratio of WO₃ exceeded 90% and thus the 90 wt% WO₃/GO sample was chosen for the recycling test. Fig. 7(c) clearly reveals that the NO removal ratio decreased in first two cycles because the oxidation products (NO₂^- and NO₃^-) blocked the active sites. After the desorption of these products through UV illumination, the catalyst reverted back to its original activity [19, 52-54]. In addition, NO₂ is a photodegradation product of NO that is a poisonous gas and leads to secondary pollution [18]. The as-prepared WO₃ possesses a low NO₂ generation fraction of only 1.8% and after the introduction of GO, the NO₂ generation fraction further decreases as low as 0.5%, indicating the excellent selectivity of WO₃/GO aerogel. Compared with the previously reported works, the as-prepared WO₃/GO aerogel exhibited not only favorable photocatalytic performance but also outstanding selectivity with the lowest NO₂ generation fraction of 0.5%, which is more advantageous in practical application [14-21, 24].

Based on the above XPS valence-band (Fig. 8(a)) and UV-Vis DRS (Fig. 7(a)) analyses, a possible band structure and photocatalytic mechanism for the WO₃/GO aerogel are illustrated in Fig. 9. Generally, a photocatalytic process includes three main steps [18]. Firstly, the pollutant is adsorbed on the surface of the photocatalyst. The WO₃/GO aerogel possesses a large surface area, which endows it with a higher adsorption capacity for Rhodamine B (RhB) of 32 mg/g than that of 2 mg/g for WO₃. Secondly, the photocatalyst absorbs photons and excites electron-hole pairs. The WO₃/GO aerogel exhibited broadened light absorption compared with that of WO₃. Finally, photogenerated carriers are transferred to the photocatalyst surface, where they oxidize NO and RhB [43, 51]. The GO aerogel provides multidimensional electron pathways that hinder the recombination of electrons and holes. Fig. 8(b) shows the PL spectra of WO₃ and the WO₃/GO aerogel. After the introduction of GO, the PL intensity obviously decreases, indicating the decreased recombination of photogenerated carriers. All in all, the large surface area, broadened light absorption, and fast electron transportation work together to result in the WO₃/GO aerogel displaying excellent photocatalytic performance and selectivity.

In conclusion, a WO₃/GO aerogel was prepared by a cost-effective freeze-drying process. The introduced GO not only acted as a support material for the formation of the aerogel but also increased the surface area, light absorption, and charge separation efficiency of the photocatalyst. The WO₃/GO aerogel showed a high NO removal ratio of 51%, which was 3.3 times that of WO₃ powder, and excellent selectivity, with the fraction of NO₂ generated as low as 0.5%. The outstanding photocatalytic NO degradation activity and excellent selectivity of the WO₃/GO aerogel mean it has great potential for use in indoor air purification.