Investigation on Pt-WO₃ Catalytic Interface for the Hydrodeoxygenation of Anisole

Abstract

As a model compound for lignin derivatives, anisole and its conversion are crucial for the upgrading of biomass resources. Anisole molecule contains a characteristic aryl ether bond (C_aryl-O-CH₃); therefore, the selective cleavage of the C-O bond to efficiently produce high-value chemicals poses a significant challenge. Constructing bimetallic synergistic active sites through tuning the metal-support interface is considered an effective strategy. In this work, the WO₃-promoted Pt/SiO₂ catalysts were investigated to enhance the performance of anisole hydrodeoxygenation (HDO) to hydrocarbons. Experimental results demonstrate that WO₃ significantly promotes HDO selectivity, increasing from 37.8% to 86.8% at 250 °C. Moreover, moderate doping improves low-temperature (<250 °C) HDO activity, confirming the presence of synergistic effects. In contrast, excessive WO₃ suppresses anisole conversion. Characterization results reveal that WO₃ stabilizes metallic Pt and facilitates H₂ dissociation. Concurrently, strong hydrogen spillover between Pt and WO₃ promotes oxygen vacancy formation on WO₃. This transforms disordered adsorption of anisole on SiO₂ into directed adsorption of the anisole’s oxygen species onto WO₃. This work achieves high anisole HDO selectivity through the Pt-WO₃ interface tuning, offering novel insights for efficient lignin conversion.

1. Introduction

Driven by environmental and energy challenges, such as diminishing fossil fuel reserves and escalating environmental pollution, biomass energy has emerged as a promising alternative to alleviate energy crises and achieve carbon neutrality, leveraging its renewability, carbon-neutral nature, and resource abundance [1,2]. Biomass primarily encompasses lignocellulose, starches, saccharides, and lipids. As a major component of lignocellulosic biomass, the efficient conversion of lignin into fuels and high-value chemicals represents a pivotal research objective [3]. However, bio-oil derived from lignin depolymerization contains significant amounts of oxygen-containing functional groups, resulting in lower heating value and stability. Consequently, chemical treatments such as deoxygenation are necessary to improve its energy utilization efficiency [4,5]. Lignin-based model compounds are commonly employed for catalytic performance testing to investigate the reaction mechanisms and underlying catalyst principles. Anisole is widely selected as a lignin model compound due to its typical structure containing the representative C_aryl-O-CH₃ bond type characteristic of lignin [6,7,8,9,10]. The selective cleavage of the C_aryl-O bond is the key step of hydrodeoxygenation (HDO), which yields benzene and cyclohexane.

Anisole hydrodeoxygenation proceeds primarily through three pathways (Figure 1) [6,11,12,13]. (1) Demethylation (DME) pathway: Anisole undergoes hydrogenolysis to form a phenol intermediate, followed by hydrogenation saturation of the aromatic ring, yielding cyclohexanone. Partial hydrogenation of cyclohexanone generates cyclohexanol. (2) Hydrogenation-Deoxygenation (HYD) pathway: Initial hydrogenation saturation of the aromatic ring produces methoxycyclohexane. Subsequent demethylation forms cyclohexanol, which undergoes dehydration to yield cyclohexane. (3) Direct Deoxygenation (DDO) pathway: Anisole undergoes direct demethoxylation to produce benzene and methanol. Benzene is then fully hydrogenated to cyclohexane over metal active sites.

The reaction primarily involves two key steps: hydrogenation occurring on metal sites and cleavage of the C-O bond on oxophilic sites [13]. Among these, the deoxygenation reaction is more difficult to occur than hydrogenation due to the higher bond energy of the C-O bond. Metals with low oxophilicity tend to favor hydrogenation over deoxygenation [14]. This leads to a mismatch between the deoxygenation and hydrogenation rates on the catalyst surface, resulting in lower yields of deoxygenation products. Bimetallic catalysts combining hydrogenation metals (Pt, Pd, Ni, Ru, Rh) with oxophilic metals (W, Mo, Al, Co, Zr, Nb) are considered highly promising [15]. Consequently, further research is needed on how to precisely control the reaction pathway of anisole hydrogenation and deoxygenation using bimetallic materials.

Platinum (Pt) catalysts exhibit good catalytic activity in the HDO reaction of lignin model compounds [16,17,18]. Studies have found that Pt-based catalysts typically serve as active centers for the adsorption and activation of aromatic rings, hydrogenation, as well as H₂ adsorption and dissociation of H species in the hydrodeoxygenation of lignin model compounds [18,19]. Studies have shown that on Pt-supported acidic Al-SBA-15 molecular sieves, the synergy of metal sites and acidic sites can also efficiently drive the anisomethyl ether (HDO) reaction, which will increase the yield of cyclohexane and significantly reduce the usage of precious metals [20]. Oxophilic sites promote the adsorption of oxygen-containing molecules and activate C-O bonds, thereby enhancing the catalyst’s deoxygenation performance. Metal oxides, such as WO₃ [21,22,23], MoO₃ [24,25], and Nb₂O₅, possess the ability to preferentially bond with oxygen atoms in anisole, providing additional deoxygenation active sites in catalysts. Modifying hydrogenation metal catalysts by interfacing with oxophilic metal oxides is a crucial strategy for achieving efficient hydrodeoxygenation of anisole. The oxophilic metal oxides influence the structure of hydrogenation metal active sites by improving dispersion via geometric effects and modifying the electronic environment of the metal surface, ultimately achieving high HDO selectivity.

Furthermore, research indicates that oxygen vacancies are among the primary choices for deoxygenation sites [26,27]. Unsaturated coordination structures on metal oxide surfaces often correspond to the formation of oxygen vacancy defects. In the HDO reaction of lignin model compounds, the selective adsorption mechanism at defect sites preferentially coordinates with oxygen-containing functional groups in the substrate molecule, facilitating electron density redistribution in the C-O bond. Hydrogen species generated on adjacent hydrogenation metal sites migrate near the adsorption site, enabling efficient hydrogenolysis of the C-O bond. WO₃, characterized by variable valence states, is a key component for constructing oxygen vacancies [22,28,29]. Its critical role in the formation of intrinsic oxygen defects and redox cycling processes has been widely applied in heterogeneous catalytic systems.

In this work, WO₃ is dispersed onto a SiO₂ support via the impregnation method to prepare Pt-xWO₃/SiO₂ catalysts (where x represents the molar ratio of W/Pt; for example, Pt-3WO₃/SiO₂ indicates a W/Pt ratio of 3:1), where WO₃ is introduced to modulate both the geometric and electronic structures of Pt. The catalytic performance is systematically evaluated in gas-phase anisole hydrodeoxygenation (HDO), with particular focus on elucidating the activation mechanism of C-O bonds promoted by WO₃. The study found that the introduction of WO₃ forms Pt/WO₃ interfacial sites, which not only enhances the catalyst’s deoxygenation capability but also improves the dispersion of Pt nanoparticles on the support while simultaneously suppressing the sintering of Pt species during the catalytic reaction.

2. Results and Discussion

In the HDO reaction of anisole, the reaction temperature has a significant impact on the catalyst activity and selectivity. By regulating the reaction temperature, a balance can be achieved between conversion and selectivity. Figure 2a,b show the HDO performance of WO₃-promoted Pt/SiO₂ catalysts under temperature gradients. Overall, the conversion of anisole on WO₃-promoted Pt/SiO₂ catalysts increases with temperature, especially Pt/SiO₂ (black line) and Pt-WO₃/SiO₂ (red line). Below 250 °C, Pt-WO₃/SiO₂ exhibits the highest activity, while above 250 °C, Pt/SiO₂ shows slightly higher activity than Pt-WO₃/SiO₂. It can be seen that the selectivity of Pt/SiO₂ and Pt-WO₃/SiO₂ catalysts is similar at low temperatures (100–150 °C), and their main products are methoxycyclohexane (hydrogenation product). When the temperature reaches 200 °C, the hydrogenation and deoxygenation selectivity (50.6%) of the Pt-WO₃/SiO₂ catalyst begins to significantly increase. At 250 °C, the difference in HDO selectivity between Pt/SiO₂ and Pt-WO₃/SiO₂ catalysts is predominant. The HDO selectivity of Pt-WO₃/SiO₂ catalyst is 85.8% with benzene at 60.7% and cyclohexane at 25.1%, respectively, but the HDO selectivity of Pt/SiO₂ is only 37.8%. Therefore, the appropriate addition of WO₃ is beneficial for improving the selectivity of benzene and cyclohexane without significantly altering the activity of anisole HDO. However, with the further increase in W load, the activity of Pt-xWO₃/SiO₂ catalysts significantly decreased, which may be due to excessive WO₃ hindering the contact between anisole and active sites. Linking HDO selectivity and conversion with W loading, Pt-WO₃/SiO₂ is the optimal catalyst, as shown in Figure S1. It can be observed that the selectivity of HDO shows a “volcano type” trend, first increasing and then decreasing, and the temperature corresponding to the maximum value decreases with the increase in W content, with Pt-3WO₃/SiO₂ at 300 °C, Pt-6WO₃/SiO₂ at 250 °C, and Pt-12WO₃/SiO₂ at 200 °C, as well as Pt/WO₃ at 200 °C, indicating that temperature is a key factor affecting HDO selectivity. In addition, it also indicates that the structure of the catalyst is another key factor. Specifically, the W content regulates the reaction pathway on Pt-xWO₃/SiO₂ catalyst, gradually transitioning from the reaction pathway on Pt/SiO₂ to Pt/WO₃. Table S1 compares the performance of reported catalysts, highlighting the advantages of this series of catalysts in HDO selectivity.

In order to investigate the stability and selectivity of the catalyst within the thermodynamic control region, performance tests were conducted by increasing the amount of catalyst used to achieve a conversion of over 20% for anisole at 250 °C. As shown in Figure 2c,d, the conversion of anisole on WO₃-promoted Pt/SiO₂ catalysts remained stable for a period of 300 min, indicating that the Pt-WO₃ interface structure is relatively stable. However, as the W content increased, the conversion decreased, which is consistent with the trend in the temperature gradients (Figure 2a), and the HDO selectivity of the W-promoted catalyst all approached 91%. In addition, we reduced the conversion to below 20% at 250 °C to investigate the intrinsic performance of the catalyst in the kinetic control region. As shown in Figure 2e,f, the selectivity of demethylation products (cyclohexanone and cyclohexanol) for Pt/SiO₂ is 47.6%, the selectivity of hydrodeoxygenation (HDO) products (benzene and cyclohexane) is 37.5%, and the selectivity of hydrogenation products (methoxycyclohexane) is 14.8%, respectively. The selectivity to the HDO products of the Pt-WO₃/SiO₂ catalyst is 87.6%, and the selectivity of the hydrogenation product is 12.4%. Moreover, the selectivity of HDO products on Pt/WO₃ is almost 100%, but compared to Pt-WO₃/SiO₂, the selectivity of cyclohexane is significantly reduced (Figure 2f), indicating that the latter has stronger hydrogenation ability than the former. Results of the kinetic test are similar to those of the thermodynamic control region activity test, indicating that the addition of WO₃ to Pt/SiO₂ is beneficial for the cleavage of C-O bonds and the generation of deoxygenated products. In addition, the presence of SiO₂ increases the hydrogenation ability of Pt/WO₃, promoting further hydrogenation of benzene to cyclohexane. The consistency between kinetic and activity measurements underscores the robustness of the Pt and WO₃ synergistic effect in driving selective HDO pathways.

In order to detect the crystal phase of the catalyst support and the state of the active metal, XRD tests were conducted on Pt/SiO₂, Pt-xWO₃/SiO₂, and Pt/WO₃, and the results are shown in Figure 3. In the XRD patterns of all Pt-xWO₃/SiO₂ samples, a broad diffraction peak appears at 2θ = 22.5°, which belongs to SiO₂ (JCPDS No.77-1315) [8]. In the XRD patterns of fresh samples, a weak peak appears at approximately 39.8° on the Pt/SiO₂ curve (Figure 2b), which is assigned to the metal Pt [30], indicating weak interaction between Pt and SiO₂, and Pt is prone to agglomeration during calcination. However, this characteristic diffraction peak is not observed after doping with WO₃, indicating that WO₃ doping can promote uniform dispersion of Pt species and suppress their aggregation. No additional diffraction peaks are observed in the Pt-(1, 3, 6) WO₃/SiO₂ fresh samples except for SiO₂ (22.5°), which may be due to the low loading of WO₃. As the WO₃ loading further increases, clear diffraction peaks appear at positions of 23.1°, 23.7°, 24.1°, and 34.0°, etc. on Pt-12WO₃/SiO₂, which are attributed to the (001), (020), (200), and (220), etc. crystal planes of WO₃ (JCPDS No.20-1324). In order to compare the changes in catalyst structure during the reaction process, XRD tests were conducted on the samples after the reaction. As shown in Figure 2c,d, the peak of Pt can only be observed at a position of 39° on the Pt/SiO₂ and Pt-WO₃/SiO₂ curves, which reflects the stability of the Pt-WO₃ interface structure. Furthermore, the WO₃ related characteristic diffraction peaks of the Pt-12WO₃/SiO₂ and Pt/WO₃ catalysts after the reaction showed disappearance, broadening, and intensity reduction, indicating that the WO₃ on the catalyst surface was partially reduced to low valence WO_3-x during the reaction process, while forming a certain amount of oxygen vacancies (O_v), which is consistent with the decrease in stability of WO₃ under a certain temperature reduction conditions in the literature [31].

The physical structural properties, such as specific surface area and pore size, of WO₃-promoted Pt/SiO₂ catalysts are shown in Figure S2 and

Table 1. N₂ physical absorption and desorption analysis for catalysts.

Sample	SBET [a] (m2·g−1)	Vtotal [b] (cm3·g−1)	dmeso [a] (nm)
Pt/SiO2	65	0.26	17
Pt-WO3/SiO2	67	0.39	27
Pt-3WO3/SiO2	61	0.27	20
Pt-6WO3/SiO2	66	0.28	18
Pt-12WO3/SiO2	65	0.30	19
Pt/WO3	0.96	0.005	19

^[a] obtained by the BET method. ^[b] obtained by a single plot method.

. The isotherms of each catalyst have inflection points and H4 hysteresis loops, belonging to type IV isotherms, which are inherent characteristics of mesoporous catalysts. Generally, the larger the specific surface area and pore volume, the more favorable it is for the reaction substrate to diffuse and be adsorbed onto the support surface, which can enhance the mass transfer process of HDO reaction. From Table 1, it can be seen that the specific surface area, pore volume, and pore size of Pt/SiO₂ and Pt-xWO₃/SiO₂ catalysts are very similar, which may be due to the minor amount of Pt loading. The specific surface area and pore volume of the Pt/WO₃ catalyst are much smaller than those of the Pt/SiO₂ and Pt-xWO₃/SiO₂ catalysts. Research has shown that Pt/WO₃ catalysts have acidic sites and oxygen vacancy (O_v) required for HDO reactions, but their specific surface area and pore volume are too small, resulting in less substrate adsorption and thus reducing their conversion. Our experiment also confirmed that due to the low specific surface area of WO₃, the activity of Pt/WO₃ is much lower than that of Pt/SiO₂ (Figure 2a). It is worth noting that the pore size of the Pt/WO₃ is similar to that of the Pt/SiO₂ and Pt-xWO₃/SiO₂, which is likely due to the accumulation of mesopores caused by particle accumulation during the testing process. The actual loading amounts of Pt and W components on the catalysts were determined using ICP-OES, and the test results are shown in Table S2. The actual loading of all catalysts is slightly lower than the nominal loading amount, while the measured W/Pt ratio of the catalysts is also close to the nominal ratio.

Further analysis of the phase structure of Pt/SiO₂ and Pt-WO₃/SiO₂ catalysts after reaction was conducted using TEM. Figure 4(a1–j1) and Figure 4(a2–j2) show the HR-TEM, STEM, and element mapping images of Pt/SiO₂ and Pt-WO₃/SiO₂ catalysts after reaction, respectively. Overall, the morphology of the SiO₂ support in the two samples remains consistent and is not affected by the introduction of WO₃, both of which have a nanoparticle structure (Figure 4(a1,a2)). From the yellow dashed circle in Figure 4(c1), it can be seen that Pt particles agglomerate on the Pt/SiO₂ surface after the reaction. This phenomenon is further verified in the element mapping images (Figure 4(g1,i1)). After the reaction, a small portion of Pt in the Pt-WO₃/SiO₂ underwent agglomeration (Figure 4(d2)), while the majority of Pt did not show significant agglomeration. The element distribution mapping (Figure 4(f2–j2)) shows that Pt and W atoms are uniformly dispersed on SiO₂, and some Pt and W signals overlap, indicating the possibility of WO₃ covering some Pt particles. There is evidence to suggest the presence of oxygen defects in WO₃ [29,32], which facilitate strong interactions (SMSI) with the loaded Pt nanoparticles, thereby promoting high dispersion of Pt species. This is consistent with the XRD characterization conclusion, further indicating that WO₃ restricts the migration of Pt nanoparticles by forming physical barriers or chemical anchoring. After doping Pt/SiO₂ catalyst with WO₃, the aggregation of Pt nanoparticles can be effectively suppressed, and the dispersibility of active metal Pt can be improved. Especially, the higher the WO₃ doping content, the more favorable it is for Pt dispersion, verified by the XRD results (Figure 3).

Raman spectroscopy was used to study the functional group features of W and O on the surface of catalysts. The test results of fresh and reacted catalysts are shown in Figure 5a,b. No Raman characteristic peaks are detected on the fresh Pt/SiO₂ sample. Three peaks appeared on the Pt-WO₃/SiO₂ spectrum at 264, 322, 710, and 800 cm⁻¹, respectively. Peaks in the lower wavenumber region (264 and 322 cm⁻¹) are attributed to the bending vibrations of W-O-W, and peaks in the higher wavenumber region (710 and 800 cm⁻¹) are attributed to the asymmetric stretching and symmetric vibrations of O-W-O [22,33]. As the WO₃ loading increased, the intensity of the characteristic peaks of W-O-W and O-W-O gradually increased. Compared to fresh catalysts, both the peak intensity of the bending and stretching modes of W-O-W and O-W-O bonds in the Raman spectra of Pt-xWO₃/SiO₂ catalysts after reaction are significantly reduced (Figure 5b), indicating the formation of oxygen vacancies on the catalyst surface. In addition, the peak positions are all shifted toward a higher wavenumber region, changing from 264 to 277 cm⁻¹, 322 to 368 cm⁻¹, 710 to 711 cm⁻¹, 800 to 804 cm⁻¹, and partial enlarged view of the Pt-xWO₃/SiO₂ series catalyst is shown in Figure S3, indicating that electron transfer occurred between Pt and W and the formation of oxygen vacancies on WO₃ during the reaction process.

To verify the interaction between Pt and WO₃, UV-Vis spectroscopy was used to investigate the coordination environment of WO₃-promoted Pt/SiO₂ catalysts. UV-Vis spectra of fresh catalysts are shown in Figure 5c. A small peak appeared at 250 nm in the Pt/SiO₂ spectrum, which may be attributed to the absorption peak of Pt particles on SiO₂ [34]. With the loading of WO₃, the peak gradually shifted toward higher wavenumber, indicating a strong interaction between Pt and WO₃. Notably, an absorption peak appeared near 204 and 265 nm, belonging to tetrahedral monotungstate and octahedral polytungstate [35,36,37,38,39]. In addition, a strong absorption peak appeared at 387 nm on the Pt/WO₃ spectrum, which is related to the WO₃ crystal. According to the literature [40], the absorption peaks of the three types of pure phase WO₃ appear at 208, 270, and 400 nm, respectively. Compared to pure WO₃, the absorption peaks of Pt-xWO₃/SiO₂ and Pt/WO₃ catalysts loaded with Pt shifted toward the lower band, further indicating a strong interaction between W species and Pt nanoparticles. The absorption peaks of the three types of pure phase WO₃ appear at 208, 270, and 400 nm, respectively.

The H₂-TPR experiment was used to test the reduction degree of Pt-based catalysts. Figure 5d shows no obvious reduction peak on the Pt/SiO₂ curve, which may be due to the easy reduction of Pt on SiO₂ at low temperatures (<50 °C). The curves of Pt-WO₃/SiO₂ and Pt-3WO₃/SiO₂ are similar, indicating that H₂ dissociated from Pt sites at low temperatures can easily reduce nearby WO₃, which also confirms the strong hydrogen spillover effect between Pt-WO₃. As the amount of WO₃ continues to increase, a peak will appear near 348 °C on Pt-6WO₃/SiO₂ and Pt-12WO₃/SiO₂, attributed to the reduction peak of WO₃ converted to WO_2.9 [41,42,43], which also indicates excessive doping of WO₃.

XPS spectroscopy was used to determine the surface elemental composition and chemical valence states of the catalyst’s surface. Analyze the role of introducing WO₃ on the state of the Pt element in the HDO reaction process. XPS tests were conducted on Pt/SiO₂ and Pt-WO₃/SiO₂ catalysts after the reaction. C 1s spectra are shown in Figure 6a, with three peaks appearing after peak fitting. The characteristic peaks around 284.8, 286.1, and 288.6 eV are attributed to C-C, C=O, and O-C=O, respectively (

Table 2. C 1s XPS peaks fitting results of the used catalysts.

Sample	C-C (eV)	C=O (eV)	O-C=O (eV)
Pt/SiO2	284.8	286.1	288.6
Pt-WO3/SiO2	284.8	286.1	288.9

) [34,44]. It is interesting that there is a significant difference in the Pt 4f spectra of the two samples after reaction (Figure 6b). The Pt/SiO₂ sample was fitted with four peaks, with binding energies of 72.8 and 76.4 eV, 74.8, and 78.2 eV, assigned to Pt²⁺ 4f7/2 and 4f5/2, as well as Pt⁴⁺ 4f7/2 and 4f5/2 [45,46], indicating that Pt loaded on SiO₂ is easily oxidized to PtO_x (

Table 3. Pt 4f XPS peaks fitting results of the used catalysts.

Sample	Pt0 4f(7/2) (eV)	Pt0 4f(5/2) (eV)	Pt2+ 4f(7/2) (eV)	Pt2+ 4f(5/2) (eV)	Pt4+ 4f(7/2) (eV)	Pt4+ 4f(5/2) (eV)	Pt0/(Pt0 + Ptσ+) 4f(5/2) (%)
Pt/SiO2	/	/	72.8	76.4	74.8	78.2	0
Pt-WO3/SiO2	71.3	74.7	73.0	76.6	/	/	77.6

). The XPS spectrum of Pt-WO₃/SiO₂ was also fitted with four peaks, but they appeared at 71.3 and 74.6 eV, 73.0 and 76.6 eV, respectively, assigned to 4f7/2 and 4f5/2 for Pt⁰ and Pt²⁺, respectively, indicating that the addition of WO₃ can effectively stabilize Pt⁰, prevent oxidation during the HDO process of anisole, and promote H₂ cleavage. This is attributed to the electronic interaction between Pt and WO₃. Furthermore, compared to the Pt²⁺ 4f7/2 binding energy of Pt/SiO₂ (72.8 and 76.4 eV), the binding energy of Pt-WO₃/SiO₂ (73.0 and 76.6 eV) shifts toward a higher direction by 0.2 eV, demonstrating the strong interaction between Pt and W, which promotes electron transfer.

Figure S4 shows the W 4f XPS spectrum of the used Pt-WO_x/SiO₂ catalyst. After peak fitting, the graph can be resolved into two main peaks, W 4f5/2 (38.0 eV) and W 4f7/2 (35.9 eV), as well as two weak peaks, W 4f5/2 (37.3 eV) and W 4f7/2 (34.9 eV), corresponding to the W⁶⁺ and W⁵⁺ species, respectively. Therefore, WO_x in the used Pt WO_x/SiO₂ catalyst mainly exists in the form of W⁶⁺. As a variable valence metal, W is prone to generate oxygen vacancies on WO₃ in both oxidizing and reducing atmospheres, and tends to stabilize. Therefore, compared to SiO₂, WO₃/SiO₂ has more oxygen vacancies, which can provide specific adsorption for benzyl ether.

Compared to pure Pt/SiO₂, the disordered adsorption of anisole on SiO₂ is replaced by the ordered adsorption of oxygen vacancies on WO₃/SiO₂ after the introduction of WO₃, resulting in a significant increase in HDO selectivity. The evaluation results of Pt-WO₃/SiO₂ catalyst indicate that the main products of its anisole HDO are benzene and cyclohexane. The reaction mechanism of the Pt-WO₃/SiO₂ catalyst for anisole HDO is shown in Scheme 1. The XPS results demonstrate that the introduction of WO₃ support can inhibit the oxidation of Pt species. The Raman results showed a significant decrease in the peak intensity of the bending and stretching modes of the O-W-O bond after the Pt-WO₃/SiO₂ catalyst reaction, indicating the formation of oxygen vacancies on the WO₃ surface. Oxygen vacancies can adsorb lone pair electrons on the oxygen of reaction substrates and intermediates, polarize C-O bonds, and thus obtain more hydrogenation and deoxygenation products. There is electron transfer between Pt nanoparticles and WO₃, indicating that the addition of WO₃ has not only geometric and physical effects on Pt species but also electronic effects. There is a synergistic effect between Pt species and WO₃.

3. Materials and Methods

Catalyst preparation. Pt-xWO₃/SiO₂ series catalysts were prepared using a simple stepwise incipient wetness impregnation method. The specific experimental procedures were as follows:

(1)

Pretreatment of silica (SiO₂) particles: crush the silica (SiO₂, Q50, ASONE International, Osaka, Japan, AR, 99%) particles to 80–100 mesh (particle size range of 147–177 µm) as a support for subsequent catalyst preparation.

(2)

Preparation of xWO₃/SiO₂ series support: Taking the synthesis of 0.3 wt.% WO₃/SiO₂, for example, weigh 1 g of SiO₂ support into a beaker. Then, dissolve 0.0042 g of (NH₄) 6H₂W₁₂O₄₀·xH₂O (Aladdin Chemicals, Shanghai, China, AR, 99.5%) in 1 mL of deionized water and sonicate to prepare a homogeneous solution. Add the (NH₄) 6H₂W₁₂O₄₀ solution dropwise onto the SiO₂ support while stirring continuously to ensure a uniform dispersion of the solution. After adding all the solutions dropwise, let it stand and age for 2 h, then dry the sample in an 80 °C oven for 12 h. The dried sample was then placed in a muffle furnace and heated to 500 °C at a rate of 2 °C/min for 4 h. Other xWO₃/SiO₂ supports with different W loading amounts (0.9, 1.8, and 3.6 wt.%) were prepared using the above method, except for the different (NH₄) 6H₂W₁₂O₄₀·xH₂O values measured.

(3)

Preparation of pure WO₃ support by (NH₄) 6H₂W₁₂O₄₀·xH₂O thermal decomposition method. Specifically, a certain amount of (NH₄) 6H₂W₁₂O₄₀·xH₂O in a muffle furnace, raise the temperature to 600 °C at a rate of 2 °C/min and maintain it for 4 h.

(4)

The preparation process of Pt-based catalysts was the same as that of xWO₃/SiO₂ supports, except that 1g of SiO₂, WO₃, and xWO₃/SiO₂ supports were separately weighed into beakers. Then, 0.0060 g of H₁₂N₆O₆Pt (Aladdin Chemicals, Shanghai, China, Pt ≥ 50%) was added to 1 mL of deionized water. Finally, the samples were calcined at 350 °C for 5 h in a muffle furnace. Specifically, the loading amount of Pt was 0.3wt.%, and the loading amounts of W were 0.9, 1.8, and 3.6 wt.%. Pt/W ratio was taken here to simplify the sample naming, for example, 0.3 wt.% Pt-0.3 wt.% WO₃/SiO₂ was named Pt-WO₃/SiO₂, 0.3 wt.% Pt-0.9 wt.% WO₃/SiO₂ was named Pt-3WO₃/SiO₂, and so on.

Catalyst characterization. X-ray powder diffraction (XRD) patterns were obtained on a Rigaku Smartlab diffractometer (Tokyo, Japan) with Cu Kα radiation (λ = 1.5406 Å), tube voltage of 35 kV, and operating current of 25 mA. Data collection at 5 °/min from 5° to 90°, or at 1 °/min from 35° to 45°. The diffraction patterns are referenced to Joint Committee on Powder Diffraction Standards (JCPDS) database. Textural properties of the catalyst were characterized by N₂ physisorption using a Micromeritics ASAP 2020 plus analyzer (Norcross, GA, USA) at 77K. Calculate the specific surface area and pore size distribution using the BET and BJH methods. Inductively coupled plasma optical emission spectrometry (ICP-OES) was tested using the iCAP-7000 Plus spectrometer (Thermo Fisher, Waltham, MA, USA). X-ray photoelectron spectroscopy (XPS) was measured on an ESCALAB 250XI spectrometer (Thermo Fisher, USA) with Al Kα radiation (HV = 1486.6 eV). The binding energy of all samples was referenced to C 1s at 284.8 eV.

High-resolution transmission electron microscopy (HR-TEM) images and high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) images were obtained using a Talos F200s G2 field-emission perspective electron microscope (FEI, Hillsboro, OR, USA) with an acceleration voltage of 200 kV. The chemical composition of the catalyst surface was tested using the FEI Super-X energy dispersive spectroscopy (EDX) system.

Raman spectra were measured using a confocal Raman micro-spectrometer (Thermo Fisher Scientific DXR2xi, USA) equipped with a 432 nm diode laser and a CCD detector. During testing, a high signal-to-noise ratio spectrum was obtained by adjusting the exposure time, scanning frequency, and laser intensity. Select different locations for multiple tests on each sample, and take the one with the best repeatability as the final Raman spectrum. Diffuse reflectance UV-Vis spectra (DRS) were recorded on a UV-Vis spectrophotometer (Agilent Cary 5000, Santa Clara, CA, USA), using BaSO₄ as a reference.

Hydrogen temperature-programmed reduction (H₂-TPR) was conducted on a chemisorption analyzer (Micromeritics Auto Chem II 2920, USA) equipped with a thermal conductivity detector. For H₂-TPR, weigh 100 mg of the sample into a U-shaped quartz reaction tube. Pre-treat at 300 °C for 60 min under pure He gas, then cool down to 30 °C, wait for the detection signal to stabilize, switch to 10% H₂/Ar, and reach 800 °C at a heating rate of 10 °C/min. Record the signal change curve of the thermal conductivity detector continuously through the data acquisition system.

Catalytic performance evaluation. The activity test of Pt-based catalyst for HDO reaction was conducted on a continuous fixed bed at atmospheric pressure, equipped with a quartz glass reaction tube with an inner diameter of 6 mm and a length of 63 cm, and an online gas chromatography (Shimadzu GC-2014, Kyoto, Japan) with dual FID detectors. In a typical test, mix a certain amount of catalyst powder evenly with 300 mg of quartz sand (20–30 mesh) and place it in the middle of the quartz reaction tube. Before the reaction, 20 mL/min of 5 vol% H₂/Ar was introduced and pre-reduced at 350 °C for 1 h. Then, the temperature was lowered to the corresponding level and stabilized for 1 h before switching the gas path, allowing 5 vol% H₂/Ar to pass through a conical bubble flask containing liquid anisole and enter the quartz reaction tube together with anisole vapor for reaction. Gas chromatography was used for online analysis of reaction products.

Anisole conversion (${\mathrm{X}}_{\mathrm{Anisole}}$) was calculated using the following formula:

$${\mathrm{X}}_{\mathrm{Anisole}}\left(\%\right)={\displaystyle \frac{{ [\mathrm{Anisole}]}_{\mathrm{in}}-{ [\mathrm{Anisole}]}_{\mathrm{out}}}{{ [\mathrm{Anisole}]}_{\mathrm{in}}}} \times 100\%$$

(1)

The selectivity (${\mathrm{S}}_{\mathrm{i}}$) of each product was calculated using the following formula:

$${\mathrm{S}}_{\mathrm{i}} \left(\%\right)={\displaystyle \frac{{ [{\mathrm{Mol}}_{\mathrm{i}}]}_{\mathrm{out}}}{{ [\mathrm{Anisole}]}_{\mathrm{in}}-{ [\mathrm{Anisole}]}_{\mathrm{out}}}} \times 100\%$$

(2)

Anisole selectivity (${\mathrm{S}}_{\mathrm{HDO}}$) was calculated using the following formula:

$${\mathrm{S}}_{\mathrm{HDO}} \left(\%\right)={\displaystyle \frac{{\sum \mathrm{Mol}}_{\mathrm{HDO}}}{{ [\mathrm{Anisole}]}_{\mathrm{in}}-{ [\mathrm{Anisole}]}_{\mathrm{out}}}} \times 100\%$$

(3)

where ${ [\mathrm{Anisole}]}_{\mathrm{in}}$ is the molar concentration of reaction anisole steam at the inlet, ${ [\mathrm{Anisole}]}_{\mathrm{out}}$, ${ [{\mathrm{Mol}}_{\mathrm{i}}]}_{\mathrm{out}}$, and ${\sum \mathrm{Mol}}_{\mathrm{HDO}}$ are the molar concentration of anisole, each product of HDO reaction, and the total product steam at the outlet. The important products in the HDO reaction are benzene derivatives, followed by cyclohexane.

Antoine equation:

lgP = A − B/(T + C)

(4)

where P is the saturated vapor pressure of the gas (mmHg), T is the absolute temperature (K), A, B, and C are Antoine constants, which can be obtained from the literature or databases. The Antoine equation was used to calculate the relationship between the saturated vapor pressure of the gas and temperature. Taking the calculation of the saturated vapor pressure of 20 °C anisole as an example: According to the database (NIST Chemistry WebBook), the Antoine constants of 20 °C anisole are A = 6.989, B = 1453, and C = 200. By substituting into the Antoine equation, the partial pressure of 20 °C anisole is 2.410 mmHg.

4. Conclusions

In summary, Pt-xWO₃/SiO₂ and Pt/WO₃ series catalysts were prepared using a stepwise impregnation method, and the effect of WO₃ doping on the HDO performance of Pt/SiO₂ catalysts was investigated using anisole HDO as a probe reaction. The experimental results showed that the main products of the reaction are benzene, cyclohexane, cyclohexanol, cyclohexanone, and methoxycyclohexane. The HDO selectivity on Pt/WO₃ is high, but the conversion of anisole is low. The conversion rate of anisole is high, but the HDO selectivity is low on the Pt/SiO₂ catalyst. The appropriate addition of WO₃ increased the HDO selectivity (91%) of Pt-xWO₃/SiO₂ while hardly changing the conversion of anisole, but the excessive addition of WO₃ led to a decrease in conversion rate. Further characterization of Pt-based catalysts revealed that the larger specific surface area of SiO₂ provides a favorable platform for the adsorption and further reaction of anisole, and it is precisely this disordered adsorption that results in lower HDO selectivity. After doping with WO₃, the strong interaction between Pt and WO₃ can improve the dispersion of Pt nanoparticles on SiO₂ support and suppress the sintering of Pt species during the catalytic reaction process. It is precisely this strong interaction that enables WO₃ to stabilize Pt⁰ and undergo electron transfer during the reaction process, accelerating the dissociation of H₂ and reducing WO₃ sites through the hydrogen overflow effect, forming oxygen vacancies. Oxygen vacancies, as hydrophilic sites, are prone to adsorb O in anisole. The spillover of H on WO₃ accelerates the cleavage of C-O bonds and enhances reaction activity. In addition, compared to Pt/SiO₂, the Pt-xWO₃/SiO₂ catalyst with added WO₃ has new Pt and WO₃ heterojunction active sites. The synergistic effect of Pt and WO₃ accelerates the cleavage of the C-O bond in the reaction substrate, which also enhances the catalytic activity.