Synthesis and Characterization of Bi₂W_xMo_1−xO₆ Solid Solutions and Their Application in Photocatalytic Desulfurization under Visible Light

Abstract

Photocatalytic oxidative desulfurization has attracted much attention in recent years due to the continuous tightening of the sulfur content requirements in motor fuels and the disadvantages of the industrial hydrodesulfurization process. This work is devoted to the investigation of the photocatalytic activity of Bi₂W_xMo_1−xO₆ solid solutions (x = 1, 0.75, 0.5, 0.25, 0) in the oxidative desulfurization of hydrocarbons under visible light irradiation using hydrogen peroxide as an oxidant. The synthesized photocatalysts were characterized in detail using XRD, SEM, EDS, low-temperature nitrogen adsorption–desorption, and DRS. It was shown that the use of solid solutions Bi₂W_xMo_1−xO₆ with x = 0.5–0.75 leads to the complete oxidation of organosulfur compounds to CO₂ and H₂O within 120 min. The high photocatalytic activity of solid solutions (x = 0.5–0.75) is attributed to their ability to absorb more visible light, the presence of the corner-shared [Mo/WO₆] octahedral layers, which may promote the generation and separation of photogenerated charges, and the hierarchical 3D flower-like structure. The reaction mechanism of the desulfurization was also analyzed in this work.

1. Introduction

Nowadays, an important direction is the development of highly active heterogeneous photocatalysts for industrially important processes, such as water and air purification, hydrogen evolution, the oxidation of organic substrates, and the reduction of carbon dioxide [1]. In recent years, there has also been a growing interest in the photocatalytic oxidative desulfurization (PODS) of liquid hydrocarbons [2,3]. This fact can be explained by the continuous tightening of the sulfur content requirements in motor fuels and the disadvantages of the industrial hydrodesulfurization process, namely harsh reaction conditions, the rapid deactivation of catalysts, and their low activity towards large sulfur-containing molecules, e.g., dibenzothiophene (DBT).

The most common types of semiconductors used in various photocatalytic reactions are inexpensive and naturally abundant transition metal oxides, especially titanium dioxide (TiO₂). The extensive use of this material is also associated with its low toxicity, high efficiency, and stability [4,5]. The most significant limitation of TiO₂ is its large band gap (E_g = 3.0–3.4 eV), which makes it possible to use only the ultraviolet (UV) region to initiate photocatalytic processes [6]. In the field of the development of photocatalysts for PODS, the shifting sensitivity of TiO₂ to the visible region by its modification remains the most common approach. The main strategies are photosensitization using metal nanoparticles, for example, Ag [7], the creation of heterojunctions [8,9], and band gap engineering, namely the doping of TiO₂ with nitrogen or sulfur into anionic positions [10,11].

In the wide range of numerous visible light-responsive metal oxide semiconductors, bismuth tungstate (Bi₂WO₆), which belongs to the Aurivillius family of layered perovskites with the general formula Bi₂A_n−1B_nO_3n+3, has received significant attention in recent years for the photocatalytic removal of organic pollutants from wastewater [12,13], water splitting [14], and CO₂ reduction [15]. The main advantages of Bi-containing layered perovskites are their intense absorption of visible light, high photocatalytic activity, ease of preparation, and ability to control properties by varying their composition and synthesis conditions [16,17].

It has recently been demonstrated that composite materials based on bismuth tungstate Bi₂WO₆ exhibit a relatively high activity in PODS under visible light irradiation [18,19,20]. Wu et al. [18] found that the Nb₂O₅/Bi₂WO₆ exhibited a considerably higher activity than pure Nb₂O₅ and Bi₂WO₆ and the desulfurization rate achieved 99% after visible light irradiation for 120 min. These results could be explained by the formation of a heterojunction and, accordingly, the effective separation of photogenerated electron–hole pairs on the Nb₂O₅/Bi₂WO₆ hybrid. A similar effect was obtained on the NiO-Bi₂WO₆ heterojunction in the photocatalytic oxidation of benzothiophene [20]. Li and co-workers [21] investigated the photocatalytic activity of Bi₂W_1−xMo_xO₆/montmorillonite in the oxidation of DBT using H₂O₂ as an oxidant. The authors demonstrated that the incorporation of Mo atoms into the Bi₂WO₆ structure significantly improved the photocatalytic activity of composites. On the other hand, the effect of physicochemical properties (specific surface area, morphology, optical properties, etc.) of unmodified Bi₂W_1−xMo_xO₆ photocatalysts on their performances in PODS remains unknown.

Herein, we report the preparation of Bi₂W_xMo_1−xO₆ (x = 1, 0.75, 0.5, 0.25, 0) layered perovskites by a hydrothermal route and their photocatalytic activity in PODS under visible light irradiation using H₂O₂ as an oxidant. All the synthesized samples were characterized by X-ray diffraction analysis (XRD), scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), low-temperature nitrogen adsorption–desorption, and UV-Vis diffuse reflectance spectroscopy (DRS). The analysis of the effect of the physicochemical properties of Bi₂W_xMo_1−xO₆ photocatalysts with different Mo and W content on their activity in PODS was performed in detail.

2. Materials and Methods

2.1. Preparation of Bi₂W_xMo_1−xO₆ Solid Solutions

The synthesis of Bi₂W_xMo_1−xO₆ (x = 0, 0.25, 0.50, 0.75, 1) layered perovskites was conducted by a hydrothermal route using Bi(NO₃)₃·5H₂O (99%), Na₂WO₄·2H₂O (99%), and Na₂MoO₄·2H₂O (99%) as starting materials. The starting materials for the synthesis of perovskites were preliminarily stirred in 30 mL of distilled water for 1 h, then transferred into a 50 mL Teflon-lined stainless-steel autoclave and treated at 160 °C for 12 h. The preparation conditions for Bi₂W_xMo_1−xO₆ (x = 0, 0.25, 0.50, 0.75, 1) layered perovskites are listed in

Table 1. Hydrothermal synthesis of Bi₂W_xMo_1−xO₆ perovskites.

x	Perovskite	Molar Ratio of Starting Materials	Temperature and Holding Time
1	Bi2WO6	Na2WO4:Bi(NO3)3 = 1:2	160 °C, 12 h
0.75	Bi2W0.75Mo0.25O6	Na2WO4:Na2MoO4:Bi(NO3)3 = 1.5:0.5:4	160 °C, 12 h
0.5	Bi2W0.5Mo0.5O6	Na2WO4:Na2MoO4:Bi(NO3)3 = 1:1:4	160 °C, 12 h
0.25	Bi2W0.25Mo0.75O6	Na2WO4:Na2MoO4:Bi(NO3)3 = 0.5:1.5:4	160 °C, 12 h
0	Bi2MoO6	Na2MoO4:Bi(NO3)3 = 1:2	160 °C, 12 h

.

2.2. Characterization

XRD patterns were recorded on a Shimadzu XRD-6100 diffractometer (CuKα radiation λ = 1.5418 Å) with a Ni filter (2θ = 10–80°, 2°·min⁻¹). The mean crystallite size D (nm) of Bi₂W_xMo_1−xO₆ layered perovskites was determined from the full width at half-maximum (FWMH) of the corresponding diffraction peaks by using the Scherrer equation:

$$D=\frac{K\mathsf{\lambda }}{\mathsf{\beta }\cos \mathsf{\theta }}$$

(1)

where K is the shape factor (0.94), λ is the X-ray wavelength (1.5418 nm in our case), β is the line broadening at FWHM in radians, and θ is the Bragg angle.

The morphology of the prepared photocatalysts was studied by SEM on a JSM-IT300LV microscope (JEOL, Welwyn Garden City, UK). The elemental composition was proved by energy dispersive X-ray spectroscopy (EDS) equipped with an X-max^N 20 detector (Oxford Instruments, Shanghai, China).

The specific surface areas of Bi₂W_xMo_1−xO₆ layered perovskites were determined from the low-temperature N₂ adsorption–desorption using an Autosorb iQ-C apparatus (Quantachrome Instruments, Boynton Beach, FL 33426, USA). Before measurements, the samples were degassed under a dynamic vacuum condition (base pressure = 1.33 × 10⁻⁴ Pa) at 120 °C for 3 h. Nitrogen (N₂) was used as an adsorptive gas for the adsorption–desorption isotherm measurements. The specific surface area of the photocatalysts was estimated by the Brunauer–Emmett–Teller (BET) method using data taken in the range of 0.05 < p/p₀ < 0.35.

UV-Vis diffuse reflectance spectra were obtained on a Cary 5000 spectrophotometer (Varian, New York, NY, USA). The optical band gaps (E_g) of the as-prepared materials were determined using the Kubelka–Munk theory by the following equation:

$$F\left(R_{\infty }\right)=\frac{{\left(1-R_{\infty }\right)}^2}{2R_{\infty }}=\frac{\mathsf{\alpha }}{s}$$

(2)

where R_∞ is the relative diffuse reflectance, α is the absorption coefficient, and s is the scattering coefficient.

The Kubelka–Munk function F(R_∞) is proportionate to the absorption coefficient α if the scattering coefficient is independent of the wavelength. The Tauc method [22] was used for the determination of the E_g of the semiconductors:

$${\left(\mathrm{F}\left(R_{\infty }\right)h\mathsf{\nu }\right)}^2=C\left(h\mathsf{\nu }-E_g\right)$$

(3)

where ν is the frequency, h is the Plank constant, and C is the proportional coefficient.

2.3. Photocatalytic Oxidative Desulfurization

Model fuel with a sulfur concentration of 200 ppm was prepared by dissolving 0.0863 g of DBT in 100 mL of n-dodecane. PODS experiments were operated in a beaker with magnetic stirring at ambient temperature. A 30 W LED lamp (6500 K, λ = 420–600 nm) was used as the visible light source. The distance from the light source to the model fuel was about 10 cm.

Typically, 10 mL of model fuel was firstly added into a 100 mL beaker, and then photocatalyst (0.1 g) and H₂O₂ (30 wt%, H₂O₂/DBT molar ratio of 10) were also introduced in order. The beaker was placed in the dark with constant magnetic stirring for 30 min to establish an adsorption–desorption equilibrium. Then, the PODS experiments were carried out under visible light irradiation for 120 min. Samples were taken every 20 min. The photocatalyst powder was separated from the reaction mixture by centrifugation.

The analysis of the reaction mixture before and after photocatalysis was carried out on a gas chromatograph Shimadzu GC-2010 Plus equipped with a flame ionization detector and a ZB-FFAP capillary column (length of 50 m; inner diameter of 0.32 mm; film thickness of 0.50 µm). Helium was used as a carrier gas. The desulfurization degree was calculated by the following equation:

X = ((C₀ − C)/C₀) × 100%

(4)

where C₀ (ppm) is the initial sulfur concentration and C (ppm) is the sulfur concentration within a certain period of time.

The main oxidation products were detected using a gas chromatography-mass spectrometer (GC-MS) Shimadzu QP-2010 equipped with an Equity-5 capillary column (length of 30 m; inner diameter of 0.32 mm; film thickness of 0.25 µm). Helium was used as a carrier gas.

2.4. Active Species Trapping Experiments

In order to establish the active species formed during the reaction and the mechanism of PODS, active radical trapping experiments were carried out. In these experiments, isopropanol (IPA), benzoquinone (BQ), and EDTA-2Na were used as radical scavengers for hydroxyl radicals (^•OH), superoxide (O₂^•−), and holes (h⁺), respectively.

2.5. Stability Test

For the investigation of stability, the photocatalyst after PODS was separated from the reaction mixture by the centrifugation, washed with ethanol and distilled water, dried at 80 °C for 12 h, and then directly used for the next run.

3. Results and Discussion

3.1. XRD Analysis

Figure 1 represents the XRD patterns of the Bi₂W_xMo_1−xO₆ layered perovskites. As can be seen, the characteristic peaks are detected at 28.4°, 32.9°, 47.2°, 55.9°, 58.6°, and 76.0°, corresponding to the (113), (200), (220), (313), (226), and (139) crystal planes of Bi₂WO₆ with x = 1 (JCPDS No. 39-0256) [23,24], respectively. No other diffraction peaks were detected, indicating the high purity of the sample. For x = 0, high-purity orthorhombic Bi₂MoO₆, the so-called γ-Bi₂MoO₆ phase (JCPDS No. 21-0102) [25], was obtained. It can also be found from Figure 1 that with an increase of the Mo content in the photocatalyst, the intensity of the (002) diffraction peak, which is characteristic for Bi₂MoO₆, increases compared to the most intense peak (113). At the same time, the narrowing of the (113) diffraction peak was observed, which indicates an increase of the average crystallite size (

Table 2. Lattice parameters and physicochemical properties of Bi₂W_xMo_1−xO₆ layered perovskites.

x	a (Å)	b (Å)	c (Å)	V (Å3)	D113 (nm)	D002 (nm)	SBET (m2·g−1)	Eg (eV)
1	5.446 (2)	5.458 (9)	16.36 (1)	486.7 (7)	13.02	-	25.8	2.90
0.75	5.468 (4)	5.446 (2)	16.39 (5)	487.9 (7)	17.18	-	19.3	2.63
0.5	5.455 (3)	5.461 (1)	16.37 (6)	487.6 (3)	17.96	13.28	16.4	2.60
0.25	5.449 (6)	5.468 (7)	16.34 (1)	486.8 (5)	24.28	20.70	14.8	2.65
0	5.471 (3)	5.480 (6)	16.25 (1)	487.3 (3)	28.38	21.81	15.3	2.68

) from 13.02 nm for Bi₂WO₆ (x = 1) to 28.38 nm for Bi₂MoO₆ (x = 0). The higher crystallinity of the compounds with a high content of Mo could be explained by the lower crystallization temperature of γ-Bi₂MoO₆ compared to Bi₂WO₆ [26].

3.2. Morphology

The SEM images of the as-prepared photocatalysts Bi₂W_xMo_1−xO₆ are shown in Figure 2. Layered perovskite Bi₂WO₆ (x = 1) has a 3D flower-like structure, consisting of nanosheets with widths of 50–80 nm, while Bi₂MoO₆ (x = 0) is composed from irregularly shaped nanosheets with widths of 100–300 nm, which form agglomerates up to 5 µm. Interestingly, at a high W content in the Bi₂W_xMo_1−xO₆ structure, the 3D flower-like morphology is retained, but at x < 0.5 the destruction of the flowers with the formation of individual nanosheets occurs.

The study of the elemental composition of the obtained samples also confirmed the formation of solid solutions with a given composition within the sensitivity of the method (~1–3 at%). Elemental mapping showed a uniform distribution of Bi (Mα), W (Lα), Mo (Lα), and O (Kα) throughout the samples (

Table 3. Elemental composition of layered perovskites Bi₂W_xMo_1−xO₆ (at%) obtained from EDS.

x	Perovskite	Bi	W	Mo	O
1	Bi2WO6	20.96	11.94	-	67.10
0.75	Bi2W0.75Mo0.25O6	22.13	8.82	2.59	65.42
0.5	Bi2W0.5Mo0.5O6	20.93	6.11	5.55	67.22
0.25	Bi2W0.25Mo0.75O6	21.56	2.79	7.75	65.90
0	Bi2MoO6	21.15	-	11.35	67.50

). According to EDS, the actual chemical formulas of the solid solutions, where x = 0.75, 0.5, and 0.25, should be written as Bi₂W_0.78Mo_0.22O₆, Bi₂W_0.53Mo_0.47O₆, and Bi₂W_0.24Mo_0.76O₆, respectively.

3.3. UV-Vis Analysis and Electronic Structure

The UV-Vis spectra of the Bi₂W_xMo_1−xO₆ compounds (Figure 3a) showed that these materials have an intense absorption ability towards visible light (λ ≥ 420 nm). We detected a significant red shift when the Mo atoms were introduced into the crystal lattice of Bi₂WO₆, which indicates a decrease of E_g for the compounds with x ≤ 0.75. For instance, if Bi₂WO₆ has an E_g of 2.90 eV, then Mo-containing perovskites have band gaps in the range of 2.60–2.70 eV (Table 2, Figure 3b).

It is well-known that the conduction band (CB) and valence band (VB) edge potentials of a solid material play a crucial role in the photocatalysis. The edge potentials of semiconductors can be estimated according to the following equations [27]:

E_VB = χ − E_C + 0.5E_g

(5)

E_CB = E_VB − E_g

(6)

where E_CB and E_VB are the conduction and valence band potentials, respectively, χ is the Mulliken electronegativity of a semiconductor [28,29], E_C is the free electron energy with respect to a normal hydrogen electrode (4.5 eV), and E_g is the band gap energy.

The Mulliken electronegativity of compounds A_aB_bC_c can be calculated by the following equation [30,31]:

χ(A_aB_bC_c) = (χ(A)^aχ(B)^bχ(C)^c)^1/(a+b+c)

(7)

where χ(A), χ(B), and χ(C) are the absolute electronegativity of the A atoms, B atoms, and C atoms, respectively, and a, b, and c are the number of A atoms, B atoms, and C atoms in an A_aB_bC_c compound, respectively.

In turn, the absolute electronegativity of atoms is calculated by Equation (8) [32]:

χ = 0.5(I + A)

(8)

where I is the ionization potential and A is the electron affinity (

Table 4. The ionization potential (I), electron affinity (A), and absolute electronegativity (χ) of the Bi, W, Mo, and O atoms in the unit of “eV”.

Atom	I	A	χ
Bi	7.29	0.94	4.12
W	7.98	0.82	4.40
Mo	7.10	0.75	3.93
O	13.61	1.46	7.54

).

Thus, the χ values for the perovskites with x = 1, 0.75, 0.5, 0.25, and 0 are 6.21, 6.19, 6.17, 6.15, and 6.13 eV, respectively. Figure 4 represents the calculated E_CB and E_VB values for the Bi₂W_xMo_1−xO₆ solid solutions using Equations (5) and (6).

The energy levels of the VB maximum (VBM) and CB minimum (CBM) are the main factors in determining the redox power of photogenerated charge carriers in photocatalytic reactions. As can be seen from Figure 4, the incorporation of Mo atoms into the Bi₂WO₆ (x = 1) structure led to a shifting of the CB and VB edge potentials. The solid solutions with x = 0.25–0.5 showed a VBM upshift of 0.18–0.19 eV, while the CBM upshift was not extreme. The observed dependences are in good agreement with earlier studies [33,34,35], where it was demonstrated that the introduction of Mo atoms into the structure of complex oxides leads to a shifting of the CB and VB edge potentials and a narrowing of the band gap.

3.4. Photocatalytic Oxidative Desulfurization Performances

All the prepared layered perovskites Bi₂W_xMo_1−xO₆ exhibited activity in the PODS of a model mixture (200 ppm of sulfur in n-dodecane) under visible light irradiation (λ ≥ 420 nm). To evaluate the photocatalytic activity of the materials, the conversion of DBT (or desulfurization degree) after 120 min of photocatalysis was used. As can be seen from Figure 5a, layered perovskites with x = 0.5–0.75 are characterized by the highest activity in PODS. It should be noted that the prepared materials can catalyze the oxidation of DBT with H₂O₂ without the light—the desulfurization degree in the dark was about 38% after 120 min in the presence of Bi₂WO₆. At the same time, the conversion of DBT was only 6% under visible light without any photocatalyst.

The use of visible light irradiation and Bi₂W_xMo_1−xO₆ solid solutions can significantly promote the formation of ^•OH radicals. As a result, after 120 min of photocatalysis, the desulfurization degree was 68%, 71%, 100%, 94%, and 21% in the presence of Bi₂WO₆, Bi₂W_0.25Mo_0.75O₆, Bi₂W_0.5Mo_0.5O₆, Bi₂W_0.75Mo_0.25O₆, and Bi₂MoO₆, respectively. The obtained results could be considered as one of the most promising in this direction in recent years (Table 5).

Taking into account the two extreme points in the series of solid solutions, the higher photocatalytic activity of Bi₂WO₆ (x = 1) compared to Bi₂MoO₆ (x = 0) could be explained by the higher BET surface area (Table 2) and the hierarchical 3D flower-like structure of bismuth tungstate (Figure 2). On the other hand, the solid solutions with x = 0.25–0.75 with a lower BET surface area compared to Bi₂WO₆ are characterized by improved photocatalytic activity. This suggests that not only does the specific surface area play a significant role in the desulfurization, but also other physicochemical properties of the materials. The enhanced photocatalytic activity of the Bi₂W_0.25Mo_0.75O₆, Bi₂W_0.5Mo_0.5O₆, and Bi₂W_0.75Mo_0.25O₆ solid solutions is primarily attributed to their ability to absorb more visible light than Bi₂WO₆ and Bi₂MoO₆ due to a narrower band gap (Table 2). Moreover, as has been shown previously, the layered structure of perovskites [36] as well as the presence of the corner-shared [Mo/WO₆] octahedral layers may promote the generation and separation of photogenerated charges [37]. In addition, the use of the solid solutions with x = 0.5–0.75, which partially retain a hierarchical 3D flower-like structure, allows the complete conversion of DBT to be achieved after 120 min.

Table 5. Activity of different materials in PODS under visible light irradiation.

No. 1	Photocatalyst	C (ppm) 2	Oxidant	Visible Light Source	Time (min)	X (%) 3	Main Products 4	Ref.
1	Nb2O5/Bi2WO6	200	H2O2	Two 5 W LED lamps	120	99	DBTO2	[18]
2	Fe3O4@SiO2/Bi2WO6/Bi2S3	500	Air	400 W halogen lamp	120	100	Not specified	[19]
3	NiO-Bi2WO6	200	O2	Not specified	180	95	BTO2	[20]
4	CuO–Fe3O4	200	H2O2	350 W Hg lamp	120	80	Not specified	[38]
5	Na–g-C3N4	200	O2	300 W Xe lamp	180	35	CO2, H2O	[39]
6	TiO2@SBA-15	50	H2O2	300 W Xe lamp	90	96	DBTO2	[40]
7	N–CeO2–TiO2	100	H2O2	300 W Xe lamp	180	94	DBTO2	[41]
8	ZnO/TiO2–SiO2	200	None	300 W Xe lamp	240	97	DBTO2	[42]
9	CeO2/MIL-101(Fe)	500	H2O2	500 W Xe lamp	120	90	DBTO2	[43]
10	BiOBr-C3N4/MCM-41	200	H2O2	150 W halogen lamp	120	92	DBTO2	[44]
11	Bi2W0.5Mo0.5O6	200	H2O2	30 W LED lamp	120	100	CO2, H2O	This study

¹ Methanol (No. 1) or acetonitrile (No. 6 and 9) were additionally added as extractants into the reaction mixture; ² the sulfur concentration in the model fuel (DBT was used as a model organosulfur compound except in Nos. 2, 5 and 3, where thiophen and benzothiophene were used as organosulfur compounds, respectively); ³ desulfurization degree; ⁴ DBTO₂—dibenzothiophene sulfone, BTO₂—benzothiophene sulfone.

Table 5. Activity of different materials in PODS under visible light irradiation.

No. 1	Photocatalyst	C (ppm) 2	Oxidant	Visible Light Source	Time (min)	X (%) 3	Main Products 4	Ref.
1	Nb2O5/Bi2WO6	200	H2O2	Two 5 W LED lamps	120	99	DBTO2	[18]
2	Fe3O4@SiO2/Bi2WO6/Bi2S3	500	Air	400 W halogen lamp	120	100	Not specified	[19]
3	NiO-Bi2WO6	200	O2	Not specified	180	95	BTO2	[20]
4	CuO–Fe3O4	200	H2O2	350 W Hg lamp	120	80	Not specified	[38]
5	Na–g-C3N4	200	O2	300 W Xe lamp	180	35	CO2, H2O	[39]
6	TiO2@SBA-15	50	H2O2	300 W Xe lamp	90	96	DBTO2	[40]
7	N–CeO2–TiO2	100	H2O2	300 W Xe lamp	180	94	DBTO2	[41]
8	ZnO/TiO2–SiO2	200	None	300 W Xe lamp	240	97	DBTO2	[42]
9	CeO2/MIL-101(Fe)	500	H2O2	500 W Xe lamp	120	90	DBTO2	[43]
10	BiOBr-C3N4/MCM-41	200	H2O2	150 W halogen lamp	120	92	DBTO2	[44]
11	Bi2W0.5Mo0.5O6	200	H2O2	30 W LED lamp	120	100	CO2, H2O	This study

¹ Methanol (No. 1) or acetonitrile (No. 6 and 9) were additionally added as extractants into the reaction mixture; ² the sulfur concentration in the model fuel (DBT was used as a model organosulfur compound except in Nos. 2, 5 and 3, where thiophen and benzothiophene were used as organosulfur compounds, respectively); ³ desulfurization degree; ⁴ DBTO₂—dibenzothiophene sulfone, BTO₂—benzothiophene sulfone.

Thus, our results showed that effective PODS depends on several physicochemical properties of photocatalysts, such as absorption ability towards visible light, morphology, and BET surface area, which should always be considered together.

It should be noted that the products of incomplete oxidation of DBT, namely dibenzothiophene sulfoxide (DBTO) and dibenzothiophene sulfone (DBTO₂), were not detected in the reaction mass according to the GC-MS analysis. Thus, DBT was completely mineralized to CO₂ and H₂O during the reaction. At the same time, in many previous studies, it has been noted that the main oxidation product is DBTO₂ (Table 5). The implementation of incomplete oxidation will lead to the need for an additional of separation of the oxidation product from the hydrocarbon mixture by extraction methods [45,46].

3.5. Kinetics of PODS

The pseudo-first-order reaction was chosen in order to describe the kinetics of PODS, and the sulfur content in the model mixture should be ascribed by the following equation:

−ln(C/C₀) = k × t

(9)

where C (ppm) is the sulfur concentration within a certain period of time t, C₀ (ppm) is the initial sulfur concentration, and k is the pseudo-first-order reaction rate constant (min⁻¹).

Figure 5b represents the correlations of the time-course variation of −ln(C/C₀) versus reaction time t. It was detected that a linear increasing of −ln(C/C₀) values with the reaction time was observed for all photocatalysts, indicating the pseudo-first order. Overall, the reaction rate constants (k, min⁻¹) decrease in the following order: Bi₂W_0.5Mo_0.5O₆ (0.0533) > Bi₂W_0.75Mo_0.25O₆ (0.0289) > Bi₂W_0.25Mo_0.75O₆ (0.0107) > Bi₂WO₆ (0.0098) > Bi₂WO₆ without light (0.0042) > Bi₂MoO₆ (0.0022) > without any photocatalyst (0.0006). Based on the obtained results, the Bi₂W_0.5Mo_0.5O₆ photocatalyst was chosen for further investigations, namely for the determination of the active species role in PODS.

3.6. Active Radical Trapping

The trapping experiments were carried out by adding to the reaction mixture of IPA, BQ, and EDTA-2Na, which are scavengers of hydroxyl radicals (^•OH), superoxide (O₂^•−), and holes (h⁺), respectively. As can be seen from Figure 6a, ^•OH radicals play the main role in the complete oxidation of DBT since the presence of IPA led to a decrease in the desulfurization degree (from 100% to 34% after 120 min over the Bi₂W_0.5Mo_0.5O₆ photocatalyst). The addition of other scavengers did not significantly change the DBT conversion. The crucial role of ^•OH radicals has also been observed in the previous studies [47,48,49]. These conclusions are in a strong agreement with the energy diagram of the Bi₂W_xMo_1−xO₆ solid solutions (Figure 4). The formation of the superoxide radical is thermodynamically unfavorable (CBM of Bi₂W_xMo_1−xO₆ in the range of 0.26–0.38 eV, E⁰(O₂/O₂^•−) = −0.33 eV). On the other hand, ^•OH can be generated because the CB position of Bi₂W_xMo_1−xO₆ is higher than E⁰(H₂O₂/^•OH) = 0.87 eV [50,51]. According to these results, the photocatalytic mechanism was put forward (Figure 6b).

During the photocatalytic process, the Bi₂W_0.5Mo_0.5O₆ photocatalyst can be activated under visible light irradiation (Equation (10)). The photogenerated electrons (e⁻) react with H₂O₂ to form ^•OH radicals (Equation (11)). In addition, the photogenerated holes (h⁺) can oxidize OH⁻ with the formation of ^•OH radicals (Equation (12)). The oxidation of the DBT molecule occurs due to ^•OH radicals, which are characterized by a strong oxidation ability (Equation (13)).

Bi₂W_0.5Mo_0.5O₆ + hν → e⁻ + h⁺

(10)

e⁻ + H₂O₂ → ^•OH + OH⁻

(11)

h⁺ + OH⁻ → ^•OH

(12)

DBT + ^•OH → CO₂ + H₂O + SO₄²⁻

(13)

3.7. Stability and Recycling Performance of the Photocatalyst

The stability of the photocatalyst is an important factor which determines the possibility of the practical application of the developed photocatalytic materials. Accordingly, the stability of the Bi₂W_0.5Mo_0.5O₆ photocatalyst was studied through reusability experiments. As shown in Figure 7, the desulfurization degree was high and exceeded 90% after five cycles of PODS. A slight decrease in the conversion of DBT could be explained by agglomeration of the photocatalyst particles during PODS.

3.8. Future Research Directions

We also assume that Bi₂W_xMo_1−xO₆@Bi₂MoO₆ or Bi₂W_xMo_1−xO₆@Bi₂WO₆ heterojunctions may be formed in Bi₂W_xMo_1−xO₆ solid solutions. The formation of heterojunctions in materials promotes an efficient spatial separation of charge carriers. However, this assumption requires careful investigation in order to establish the real mechanism of reactions during PODS in the presence of the Bi₂W_xMo_1−xO₆ solid solutions. Thus, the identification of heterojunctions formed during the preparation of Bi₂W_xMo_1−xO₆ could be the subject of further research.

We suppose that further increasing of the photocatalytic activity of these types of materials in PODS could be achieved by the optimization of the synthesis conditions in order to find the most appropriate morphology. It is known that a well-designed structure allows the separation of photogenerated charges and efficient light adsorption to be achieved, taking full advantage of the reflection, refraction, and scattering of photons [52]. It has been demonstrated that modulating reaction temperature, reaction time, and pH value, and changing the surfactant and template during hydrothermal treatment can be efficient for the preparation of Bi₂WO₆ with different morphologies, such as erythrocyte-like [53], flower-like [54], single-crystal-like [55], and 3D hollow [56]. Also, varying the preparation conditions affects the physicochemical properties of the obtained materials (e.g., specific surface area and band gap) [57], which significantly influences the photocatalytic performances. Thus, the second subject of further research in this direction could be the search for the optimal preparation conditions of the solid solutions.

4. Conclusions

In summary, nanosized Bi₂W_xMo_1−xO₆ solid solutions with various compositions (x = 1, 0.75, 0.5, 0.25, 0) were prepared by a hydrothermal method. The incorporation of Mo into the Bi₂WO₆ lattice leads to the Bi₂W_xMo_1−xO₆ solid solutions (x = 0.75, 0.5, 0.25), which are characterized by a narrowed band gap and enhanced visible light-harvesting ability. The solid solutions with x = 0.5–0.75 partially retain a hierarchical 3D flower-like structure, and their use in PODS allows the complete conversion of DBT to be achieved after 120 min. The obtained results confirmed that the physicochemical properties of photocatalysts, such as visible light absorption ability, morphology, and BET surface area, have a significant influence on the reaction and should be considered together for the efficient realization of PODS. In addition, the trapping experiments showed that the sulfur removal proceeds due to ^•OH radicals. We also suggested directions for further research, namely optimization of the preparation conditions of the solid solutions and the identification of heterojunctions formed during the preparation of Bi₂W_xMo_1−xO₆.