Double modifications of TiO2 by doping with WO3 and by dispersing on a SiO2 support were made by the one-pot sol-gel method. Doping with W shifts the TiO2 band gap energy from 3.2 eV to around 3.06 eV. The surface area of the supported W-TiO2/SiO2 material was significantly increased, by approximately 3 times, in comparison to the bare TiO2. The photocatalytic activities of the catalysts were evaluated in the degradation reaction of p-nitrophenol in aqueous solution and basic medium. After 240 min of photodegradation, more than approximately 99% p-nitrophenol could be mineralized with the most active W-TiO2/SiO2 catalyst. Under UV irradiation, p-nitrophenol was initially photodegraded into hydroquinone and benzosemiquinone intermediates, which were further degraded into smaller fragments such as organic carboxylic acids and finally completely mineralized. A proposed photoreaction mechanism was presented based on the key roles of the surface hydroxyl species and superoxide radicals such as O2- and OH, together with W6+/W5+ couples and e-/h+ pairs in the catalysts in the p-nitrophenol photodegradation. The one-pot sol-gel synthesis method was proven to be effective to obtain W-TiO2/SiO2 catalyst with large surface area and high photocatalytic activity, and it can be also used for the preparation of other heterogeneous catalysts.

1. Introduction

Industrial development around the world has greatly increased the volume of contaminated residual water [13]. Nanotechnology for eliminating or reducing environmental pollutants has shown significance in the last decades, particularly for water contaminant control [4]. Heterogeneous photocatalysis represents a part of the solution for addressing environmental concerns due to its significant benefits compared to traditional decontamination methods, allowing the complete mineralization of organic contaminants without producing dangerous intermediate residuals [1, 5]. Photocatalysis involves the use of solar or ultraviolet light and a semiconductor such as TiO2 [3]. However, TiO2 only absorbs light in the UV region due to its relatively large band gap energy (~3.2 eV), which is the main drawback for its application, since only 4~5% of the solar spectrum falls within this range [6]. The photodegradation efficiency of supported TiO2 has improved in comparison with the bare TiO2 solid [7, 8]. Doping TiO2 with transition metal oxides is another effective way to overcome this problem and improve its photoactivity [9, 10].

It is known that tungsten trioxide (WO3) is a semiconductor material with a band gap between 2.5 and 2.8 eV [11, 12]. It is an n-type semiconductor and it may act as visible light catalyst for oxidation reactions [13]. WO3 has several crystalline structures, depending on the synthesis conditions. The most representative tetragonal WO3 structure forms at temperatures above 740°C; the orthorhombic WO3 structure can be obtained between 740 and 330°C; the monoclinic WO3 structure can be synthesized at a temperature interval between 330 and 17°C, and the triclinic structure can be obtained at room temperature below 17°C [14]. Due to its optical properties, WO3 has the capacity to absorb UV and visible light from the electromagnetic spectrum [15], and thus, it has been highlighted as a photocatalyst for the photooxidation of water contaminants [16, 17]. WO3 nanoparticles can be synthesized through the sol-gel method under different conditions [18, 19]. However, considering its high cost, small surface area, and low stability, pure WO3 is rarely used as a catalyst but it is widely used as a photocatalyst electronic promoter or dopant. WO3-doped TiO2 can reduce the band gap energy where WO3 acts as e− sink that would transfer photoexcited e− to the surface grafted nanoclusters, enhancing the indoor air purification photoefficiency with visible light [20]. WO3-doped TiO2 has also been reported to be active for the photodegradation of small organic compounds such as ethanol and tetracycline [21]. When WO3-TiO2 was dispersed on monowalled carbon nanotubes (MWCNT), several factors responsible for the photocatalytic activity enhancement can be found in the WO3-TiO2/MWCNT nanocomposite system. Different synthesis approaches led to different crystal phase compositions thus affecting the photocatalytic activity [22].

p-Nitrophenol, widely used in industry and agriculture, has significantly contaminated water bodies, severely affecting the aquatic ecosystem. This compound has been included in the US Environmental Protection Agency list of the most toxic chemicals [23]. Meijide et al. reported that p-nitrophenol can be effectively degraded by the electro-Fenton process using H2O2 as oxidant. A kinetic model disclosing the p-nitrophenol photodegradation process was developed [24].

In the present work, we attempt to improve the photocatalytic activity of the TiO2 solid by double modifications: (i) doping TiO2 with a small amount of WO3 in order to modify its electron structure and thus its optical properties and (ii) dispersing the W-doped TiO2 nanoparticles onto a support, in order to increase the dispersion and the surface area. Because SiO2 is abounded with large specific area and a good hydrothermal stability, it was chosen as the support material [25]. The TiO2, W-TiO2, and W-TiO2/SiO2 solids were synthesized and later on characterized with X-ray diffraction (XRD), UV-vis spectroscopy, N2 physisorption isotherms, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), and X-ray energy dispersion spectroscopy (EDS). Their photocatalytic activities were comparatively evaluated in the photodegradation reaction of a p-nitrophenol aqueous solution under basic conditions. The photoreaction mechanism and the degradation reaction pathways were explored and discussed by monitoring the reaction route.

2. Experimental

2.1. Synthesis of W-TiO2 Nanoparticles

W-TiO2 materials were synthesized using the sol-gel method consisting of the following stages: (1) 0.2 mol of titanium ethoxide, Ti(OC2H5)4, and 2.4 mol of propanol were mixed; the solution was stirred over the course of 2 h. (2) 10.5 mL of a previously prepared WC16/deionized solution containing 1.3 g WC16 was slowly added to the above solution in order to obtain 5 wt% of tungsten. The pH was adjusted to approximately 3 with 1 M HC1. (3) The mixture was kept under stirring and reflux at 70°C for 72 h in order to obtain a gel. Afterwards, the solvent was evaporated and the formed material was washed with deionized water and dried at 100°C. The material was calcined in air at 400°C for 8 h with a temperature increasing rate of 1°C/min. The obtained solid was noted as W-TiO2.

2.2. Synthesis of W-TiO2/SiO2

The W-TiO2 dispersed on SiO2 catalyst was prepared with the following stages: (1) 4 g of SiO2 was suspended in 3.5 mol of propanol. (2) In a different container, 0.2 mol of titanium ethoxide Ti(OC2H5)4 and 2.4 mol of propanol were mixed; the solution was stirred over the course of 2 h. (2) 10.5 mL of a previously prepared WC16/deionized solution containing 1.3 g WC16 was slowly added to the above solution in order to obtain 5wt% W. The pH was adjusted to approximately 3 with 1 M HC1. The final molar ratio of the precursors used in the synthesis was as follows: . (3) The mixture was kept under stirring and reflux at 70°C for 72 h to obtain a gel. After that, the solvent was evaporated and the formed material was washed with deionized water and dried at 100°C. The material was calcined in air at 400°C for 8 h with a temperature increasing rate of 1°C/min. The obtained solid was noted as W-TiO2/SiO2 catalyst.

2.3. Characterization

The X-ray diffraction (XRD) patterns of the solids were obtained with a Siemens D500 diffractometer coupled to a tube with a copper anode and Cu Kα () radiation. The measurements were realized at room temperature at 35 kV and 20 mA conditions. The X-ray diffraction patterns were recorded at a region between 5° and 75°.

The specific surface area, pore volume, and pore size distribution were determined with a Micromeritics ASAP 2000 instrument, using N2 as an adsorbent. The N2 physisorption data was converted into surface area using the Brunauer-Emmett-Teller (BET) method. The pore size distribution was calculated using the desorption isotherms according to the Barrett-Joyner-Halenda (BJH) model. Before carrying out the adsorption-desorption profile analysis, the samples were degassed in vacuum at 250°C for 4 h.

The synthesized materials were studied by diffuse reflectance UV-vis spectroscopy (DRS-UV-vis); the UV-vis spectra were obtained with a Varian Cary 100 instrument equipped with an integration sphere, using BaSO4 as reference, at a wavelength interval from 200 to 800 nm. The UV-vis spectra were reported in diffuse reflectance mode.

The morphological features of the catalysts were studied employing a JEOL JSM-5900 LV scanning electron microscope (SEM) and a secondary emission vacuum with variable range control. The element distribution on the solid was analyzed with an energy dispersive spectrometer (EDS), model Oxford. Transmission electron microscopy (TEM) was employed in order to closely observe the dispersion of the WO3 and TiO2 nanoparticles on the surface of the SiO2 solid. Micrographs were obtained with a JEM-ARM200CF microscope at an accelerating voltage of 200 kV.

The metal oxidation state and the chemical valence of the solids were determined using X-ray photoelectron spectroscopy (Thermo Scientific K-Alpha) at 150 W and 15 kV voltage. The deconvolution of the XPS peaks and curve fittings were made using the XPS Peak 4.1 software.

2.4. Photocatalytic Activity Measurement

The photocatalytic activity of the TiO2, W-TiO2, and W-TiO2/SO2 nanomaterials was evaluated in the degradation reaction of p-nitrophenol in an aqueous solution inside a glass photoreactor with the following steps: 200 mL of an aqueous solution containing 25 ppm of p-nitrophenol was poured into a container at pH 8, adjusted with a 5.0 M solution of NH4OH, and then, 200 mg catalyst was also added. The resulting mixture was stirred inside a dark box for 30 min in order to reach adsorption-desorption equilibrium. Afterwards, a UV light source of 2.16 W (UVP Pen Ray lamp of 18 mA, 2.5 mW/cm, ) was used to irradiate the photoreactor.

At regular intervals of 30 min of irradiation, aliquots of 3 mL were taken, which were filtered (0.45 microns nylon filters) in order to eliminate suspended solid particles. The p-nitrophenol photodegradation was determined by measuring the characteristic absorption peak that appears at approximately 400 nm using UV-vis spectrophotometry (Varian Cary 100 UV-vis). Formation of different intermediates was monitored by liquid chromatography-mass spectrometry (LC-MS) on time analysis in order to explore the possible degradation mechanisms or pathways.

The catalytic stability test was carried out 5 successive times under the same reaction conditions reported above. After the end of each experiment, the catalyst was filtered and dried without calcination, and then, it was used for the next test.

3. Results and Discussion

3.1. X-Ray Diffraction

Figure 1 shows the XRD patterns of the TiO2, W-TiO2, and W-TiO2/SiO2 catalysts. For the pure TiO2 sample, the diffractograms show several peaks at , 37.5, 48.0, 53.8, 55.9, and 63.5° which correspond to the reflections of the (101), (004), (200), (105), (211), and (204) planes in the crystalline anatase TiO2 (JCPDS 21-1272). For the W-TiO2 solid, several diffraction peaks appear at , 23.6, 28.6, 33.3, 33.7, and 41.5° which correspond to the reflections of the (001), (200), (111), (021), (201), and (221) planes of the monoclinic WO3 [26, 27]. For the W-TiO2/SiO2 sample, the peaks became broader and weaker. The XRD peaks of WO3 are slightly wide, which is attributed to the high dispersion of W-TiO2, since the content of W is rather low (approximately 5 wt% in TiO2) in the nanomaterial [28, 29].

3.2. UV-vis Diffuse Reflectance Spectroscopy (DRS-UV-vis)

Figure 2 shows the DRS-UV-vis spectra of the TiO2, W-TiO2, and W-TiO2/SiO2 solids. The values of the band gap energy () were calculated using the Tauc method [30], which consists of graphing the product (light energy) on the -axis and the (αhν)1/r on the ordinate, where is the absorption coefficient of the material. The values of TiO2, W-TiO2, and W-TiO2/SiO2 are reported in Table 1. After W doping, the value diminishes from 3.2 eV for TiO2 to around 3.06 eV for W-TiO2 and W-TiO2/SiO2. Relative to pure TiO2, the UV-vis spectrum of the W-TiO2 catalyst presents a small displacement towards the visible electromagnetic region. This probably resulted from the electronic interaction within the WO3 and TiO2 material heterojunction [31]. The UV-vis spectrum of the W-TiO2/SiO2 solid exhibited an absorption profile similar to W-TiO2. This behavior is attributed to the fact that SiO2 is transparent to the UV-vis radiation [32].

3.3. N2 Adsorption-Desorption Isotherms

The textural properties of the TiO2, W-TiO2, and W-TiO2/SiO2 samples were measured. Table 1 shows the specific surface area (BET), pore volume, and average pore diameter of TiO2, W-TiO2, and W-TiO2/SiO2. TiO2 and W-doped TiO2 solids present a surface area of 140 and 126 m2/g, respectively. W doping leads to a slight reduction of surface area probably due to the partial blockage of some pores by WOx deposits. After loading the sol-gel W-TiO2 nanoparticles on the SiO2 support, the surface area significantly increases to 432 m2/g. Therefore, the supported W-TiO2 solids show greater dispersion and much higher surface area in comparison with the unsupported W-TiO2 solids, probably as a result from the strong mutual interaction between hydroxyls in W-TiO2 and SiO2, preventing rapid crystallization and particle growth. The stretching out hydroxyls in the SiO2 support may highly connect with the –OH species in Ti(OH)4 and W(OH)6 produced by the hydrolysis of the titanium and tungsten precursors during the synthesis, gaining the dispersion of the W-TiO2 nanoparticles formed on the SiO2 solid after calcination.

3.4. Morphological Features and Element Distribution

Figure 3 shows SEM micrographs and the Si, Ti, and O chemical composition distribution of the TiO2, W-TiO2, and W-TiO2/SiO2 solids. For the TiO2 sample, on the selected area only Ti and O elements were detectable by EDS. For the W-TiO2 solid, W, Ti, and O on the surface of TiO2 were observed on the selected area. For the W-TiO2/SiO2 sample, Si, Ti, W, and O distribution was confirmed. Elementary carbon may result from the titanium ethoxide Ti(OC2H5)4 or tetraethylorthosilicate (TEOS) precursors during the synthesis.

More information on the element distribution along the whole samples was investigated with the EDS mapping technique. Figure 4 shows the EDS mappings of TiO2 and W-TiO2 samples. A very uniform distribution of O and Ti elements in the pure TiO2 was observed. In the W-doped TiO2 sample, we may see that W was homogeneously distributed across the whole TiO2 particles.

For a closer observation of the distribution of WO3 and TiO2 nanoparticles on the SiO2 surface, the transmission electron microscopy technique was employed. Figure 5 shows the TEM images of W-TiO2 and W-TiO2/SiO2 samples. On the W-TiO2 sample (Figure 5(a)), both WO3 and TiO2 nanoparticles are observed. Their particle sizes ranged between 5 and 20 nm. Some WO3 and TiO2 particles contacted each other, forming W-TiO2 mixed oxides. Some crystal terrace and kinks in TiO2 crystals are observed. In Figure 5(b), we observe WO3 and TiO2 particles highly dispersed on the surface of the SiO2 solid. Some large TiO2 anatase phases were surrounded with WO3 small particles. The particle size of TiO2 varied between 5 and 20 nm whereas WO3 particles are relatively small, most ranging from 4 to 10 nm. The TEM observations are consistent with XRD results, where wide XRD peaks corresponding to monoclinic WO3 and anatase TiO2 phase are present in both W-TiO2 and W-TiO2/SiO2 samples.

3.5. X-Ray Photoelectron Spectroscopy (XPS) Analysis

Figure 6 shows the XPS spectra of the TiO2, W-TiO2, and W-TiO2/SiO2 catalysts. In Figure 6(a), the core levels of O 1s XPS spectra can be deconvoluted into two parts: the large peak with binding energy (BE) between 528.0 and 531.4 eV corresponds to the lattice oxygen in oxides; the small peaks with a BE value of 531.5~534.0 eV indicate the surface oxygen species such as hydroxyls and adsorbed water [33]. The Ti2p core levels of the XPS spectra are shown in Figure 6(b). Two well-defined peaks are clearly present, and they correspond to the Ti2p1/2 () and Ti2p3/2 () signals, confirming the Ti4+ ion in anatase phase [34]. When TiO2 was doped with W, the peak positions slightly shifted towards the high-energy region; when the W-TiO2 solid was dispersed on the SiO2 surface, this shift became more significant. Considering that the cationic size of W6+ (0.62 Å) is slightly smaller than that of Ti4+ (0.69 Å), during the synthesis procedure, W ions may insert into the Ti lattice points resulting in the W–O–Ti bond. These peak shifts may indicate both, the formation of the W–O–Ti bond in the structure and the interaction between W-TiO2 and the SiO2 support.

The W4f core levels of the XPS spectra are present in Figure 6(c). Each of them can be deconvoluted into two peaks, indicating the coexistence of W6+ ( for W4f 7/2 and for W4f 5/2) and W5+ ( for W4f 7/2 and for W4f 5/2) [35, 36]. The Si2p core level of the XPS spectrum, Figure 6(d), shows only one symmetrical peak at 101~106 eV [37]. The XPS characterization confirms that a certain number of hydroxyl species, surface-adsorbed oxygen species, lattice oxygen, and W6+/W5+ ions coexist in the W-doped TiO2 solids.

The surface atomic concentrations of the samples obtained from the XPS analysis are reported in Table 2. For the W-TiO2/SiO2 catalyst, the surface oxygen concentration is greater than that for both TiO2 and W-TiO2 materials. This is probably due to more hydroxyls present on the surface of the solid. Some of them result from the SiO2 support. These surface oxygen species are the precursors of active oxygen species such as OH or O2 radicals. The presence of more surface active oxygen species may increase the photocatalytic activity.

3.6. Photocatalytic Activity

Usually, p-nitrophenol may exist in two forms in aqueous solution, depending on the pH value. Figure 7 shows the UV-vis absorption bands of p-nitrophenol at . Two bands at 220 nm and 399 nm are shown, corresponding to the p-nitrophenol protonated form and to the nonprotonated p-nitrophenol ion form, respectively. The former usually appears in acidic conditions and the latter can be formed under basic conditions. In the present experiments, the p-nitrophenol photodegradation reaction was initially performed at . Therefore, only the intensity variation of the band corresponding to the nonprotonated p-nitrophenol was monitored during the whole reaction procedure.

Figure 8 shows the absorption spectra of p-nitrophenol in function of the UV irradiation time for the most active catalyst, W-TiO2/SiO2. During the degradation, the intensity of the absorption band gradually diminishes as the pH value gradually increases from the initial to 10 at the end of a 240 min reaction. The pH value variation must be related to the products formed during the degradation, relationship which will be further discussed below.

Figure 9 shows the remaining concentration of p-nitrophenol after photodegradation with the TiO2, W-TiO2, and W-TiO2/SiO2 catalysts under UV irradiation. In the initial adsorption period (15 min), the p-nitrophenol concentration was reduced from 25 ppm to 24.6 ppm for TiO2, to 24.4 ppm for W-TiO2, and to 22.8 ppm for the W-TiO2/SiO2 catalyst. These results indicate that the p-nitrophenol adsorption ability of W-TiO2/SiO2 is greater in comparison to TiO2 and W-TiO2. This can be principally attributed to the greater specific area of W-TiO2/SiO2.

After 240 min of irradiation, the concentration of p-nitrophenol significantly diminishes to 13.2 ppm employing TiO2; thereafter, only 52.8% of p-nitrophenol was converted; so the activity of pure TiO2 for p-nitrophenol degradation is relatively low. A better activity was achieved using the W-TiO2 catalyst since the p-nitrophenol conversion reached 88.2% at the end of reaction. The W-TiO2/SiO2 catalyst exhibits the greatest activity, leaving a remnant of 2.6 ppm of p-nitrophenol at 180 min of reaction. After 240 min, approximately 99% p-nitrophenol was photocatalytically degraded, reaching almost complete removal.

The stability of three catalysts was evaluated. The results are presented in Figure 10. The catalytic test was repeated 4 times under the same reaction conditions. At the end of each experiment, the catalysts were filtered and dried and then used in the next test. We found that the TiO2 catalyst could remove around 50-53% of p-nitrophenol in four successive recycles. In the first cycle, approximately 80% and 99% of p-nitrophenol could be degraded by the W-TiO2 and W-TiO2/SiO2 catalysts, respectively. The degradation of p-nitrophenol gradually diminished in the following trials. In the fourth cycle, an approximately 15-20% drop of the catalytic activity was observed that may result from the catalyst mass loss during the filtration procedure. In the fifth cycle, we added a little amount of fresh catalysts for compensating the catalyst loss and keep the catalyst mass unchanged. Then, we found that the p-nitrophenol photodegradation rate became very similar to the results obtained in the first cycle, confirming that the decrease in catalytic activity was caused by the catalyst mass loss in the operation procedure.

4. Discussion

In this work, we dispersed W-TiO2 nanoparticles on SiO2 solids by the sol-gel method. SiO2 may take two roles: (1) it provides a larger surface area for better dispersing the W-TiO2 active phases, as confirmed by the significant increase of approximately 3 times, of the surface area in comparison to that of bare TiO2, and approximately 3.5 times relative to W-TiO2 (see Table 1); and (2) SiO2 also provides additional surface hydroxyls that participate in the formation of OH radicals under UV irradiation.

The XPS characterization confirms that in W6+ and W5+ ions, a certain number of hydroxyl species and surface oxygen species coexist in the W-TiO2 and W-TiO2/SiO2 catalysts. Therefore, some W6+/W5+ couples are formed in these catalysts even before the photoreactions. The formation of W5+ may result from W6+ accepting an electron transferring from surface OH species; and it may also indicate the existence of some oxygen defects around W6+ in the crystalline structure of WO3 nanocrystals. During the W-TiO2 synthesis, some W6+ ions can replace some of the Ti4+ ions in the TiO2 lattices in order to form a W–O–Ti bond. However, because W6+ (0.62 Å) has a smaller ion radium than Ti4+ (0.69 Å), the W–O–Ti bond formation may lead to an unbalance of the local electron cloud distribution that results in a more active local oxygen, favoring the surface oxidation reaction.

It is widely recognized that there are some electron (e-)–hole (h+) pairs in the surface of the activated induced by UV irradiation

In the W-doped TiO2 catalyst under UV irradiation, the charge transfer cycles in the W6+/W5+ couples may occur via W6+ capturing an electron and converting to W5+ (equation (2)) or via W5+ hole (h+) trapping and converting to W6+ (equation (3)):

Therefore, the coexistence of W6+/W5+couples, (e-)–(h+) pairs, and surface oxygen species such as hydroxyls may promote the surface reduction-oxidation cycles.

It has been reported that a heterojunction structure can be formed in the interfaces of the dopant and parent semiconductor nanoparticles [38]. When such a heterojunction is initiated by photons with energy higher than or equal to the bandgap of the semiconductor, herein, W-TiO2, the photoinitiated electron-hole pairs can be quickly separated: the electrons in the TiO2 conduction band (CB) may flow to the WO3 nanoparticles CB because the TiO2 CB is more negative than that of WO3. Simultaneously, the valence bands (VB) of WO3 and TiO2 oxides are mainly constituted by their O2p orbital, and therefore, their VB energy levels are rather close. Considering the charge balance in the semiconductor, the photoexcited holes in the VB would more likely transfer from WO3 to TiO2 [38]. Such charge transferring effectively inhibits the recombination rate of electron-hole pairs. Electron and hole migration driven by the electron field force within the heterojunction further promotes the reduction-oxidation cycles of the W6+/W5+ couple in the W-TiO2/SiO2 catalyst.

In the present experiment, the photodegradation of p-nitrophenol was performed in the presence of air. Under the oxidation reaction condition and the UV irradiation, surface W5+ ions can be oxidized to W6+ ions via W5+ transferring an electron to the molecular O2 adsorbed on the surface of the TiO2 particle, forming O2- (equation (4)); on the other hand, the accumulated electrons in WO3 can be consumed by reducing W6+ ions to W5+ ions via interacting with surface hydroxyl (OH-) species which liberate an electron and convert OH- into OH radicals (equation (5)).

Therefore, in the p-nitrophenol photodegradation process, an oxidation half cycle (equations (3) and (4)) and a reduction half cycle (equations (2) and (5)) coexist, accompanied by electrons transfer from the hydroxyls to the OH radicals, and from O2 to O2-, among the electron (e-)–hole (h+) pairs and W6+/W5+ couples in the catalyst surface. These OH and O2- radicals are the principal active species for the photodegradation of p-nitrophenol. A photodegradation reaction mechanism involving the oxidation-reduction cycles taking place on the surface of the W-TiO2/SiO2 catalyst is proposed in Scheme 1.

When the p-nitrophenol photodegradation was catalyzed with the W-TiO2/SiO2 catalyst, various intermediates were formed involving several reaction steps. At the beginning of the degradation stage under UV irradiation, the stretching –NO2 and –OH groups interacted with the catalyst and were easily attacked by surface active oxygen species such as superoxide radicals O2- and OH, leading to the formation of the primary intermediate fragments such as 2,4-benzoquinone, p-benzoquinone, and 4-nitrocatechol.

The formation of the primary intermediate fragments is related to the special molecular structure of p-nitrophenol. Collado et al. reported that there is an electron withdrawing effect of the p-hydroxybenzoic acid [39]. In this work, the –NO2 group in the p-nitrophenol structure may generate an electron withdrawing effect in the para position of the aromatic ring, which influences the hydroxyl group and thus inhibits the rate of the electrophilic attach reaction for the formation of catechol and hydroquinone. As a result, 2,4-benzoquinone, p-benzoquinone, and 4-nitrocatechol are chiefly formed at the initial stage. Because the –NO2 group is a leaving group which can be eliminated easily, p-nitrophenol could form the hydroquinone and benzosemiquinone intermediates via the rupture of the C–N bond [40].

As the reaction further proceeded, the formed benzoquinones and nitrocatechol species gradually disappeared, and instead, 3-hydroxymuconic acid and other organic acids (acetic acid, oxalic acid, and maleic acid) were detectable. These results indicate that the primary fragments were further degraded into secondary species with a smaller molecular mass through the cleavage of C–C bonds by cycloaddition reactions, leading to the aromatic ring opening followed by the formation of secondary fragments such as various organic carboxylic acids.

After 4 hours of photocatalytic reaction, the secondary intermediates were further degraded into species with smaller molecular masses and finally mineralized into CO2 and NH4+ and water. The pH value increased from 8 to 10 during the reaction, indicating that the intermediates with small molecules or short-chain acids formed in the last step were finally degraded into CO2 and water. Formation of additional OH- species in the solution may lead to the pH increment. These results suggest the complete mineralization of p-nitrophenol. The possible reaction pathways are pictured in Scheme 2.

5. Conclusions

The one-pot sol-gel synthesis method reported herein was proven to be effective to obtain a W-TiO2/SiO2 catalyst with a large surface area and high photocatalytic activity, and it can be also used for the preparation of other heterogeneous catalysts. The XPS and UV-vis spectra show that after W doping, a slight displacement of the electromagnetic spectrum towards the visible region was observed for the W-TiO2 catalyst. When the W-TiO2 nanoparticles were dispersed on SiO2, their surface area was greatly increased by approximately 3 times. The synthesis conditions reported herein favor the formation of the crystalline titania anatase phase and enhance the dispersion of W-TiO2 nanoparticles with high dispersion and a large surface area.

In the photodegradation reaction of p-nitrophenol in basic conditions under UV irradiation and after 4 h reaction, the most active W-TiO2/SiO2 catalyst degraded more than 99% of p-nitrophenol. The activity of the catalyst has been attributed to the high dispersion of W-TiO2 within the silica matrix with the greater surface area and to the formation of a heterojunction at the interface between WO3 and TiO2. The surface hydroxyls and superoxide radicals such as O2- and OH, together with W6+/W5+ reduction-oxidation couples and e-–h+ pairs, took key roles in the complete mineralization of p-nitrophenol.

Data Availability

The data used to support the findings of this work are available from the corresponding authors upon request.

Conflicts of Interest

The authors declare herein that this article currently has no conflict of interests.


We would like to thank the financial support from the projects of Instituto Politécnico Nacional (SIP-20181280 and SIP-20180199) and an international scientific collaboration between the International Joint Research Centre of Clean Energy and Chemical Engineering at the East China University of Science and Technology and the Instituto Politécnico Nacional in Mexico. J.A. Wang thanks the Departamento de Ciencias Básicas, Universidad Autónoma Metropolitana-Azcapotzalco for supporting his investigation during his Sabbatical year.