Influence of Nanoscale Intimacy in Bi-Functional Catalysts for CO₂-Assisted Dehydrogenation of C₅-Paraffins

Abstract

In this study, Pt₁Sn₁ intermetallic nanoparticles (NPs) on SiO₂/CeO₂@SiO₂ composites were located either on SiO₂ or on CeO₂@SiO₂, thereby varying the average distance (intimacy) between metal sites and CeO_x sites from “closest” to “nanoscale”. The catalytic performance of these catalysts was compared to dual-bed mixtures of Pt₁Sn₁@SiO₂ and CeO₂@SiO₂ powders, which provided a “milliscale” distance between sites. Several beneficial effects on the catalytic performance of CO₂-assisted oxidative dehydrogenation of C₅-paraffins were observed when Pt₁Sn₁ nanoparticles were located on SiO₂ in nanoscale proximity to the CeO₂ sites, as opposed to Pt and Sn species located on CeO₂@SiO₂ with the closest proximity and milliscale intimacy between Pt₁Sn₁ and CeO₂. The former catalysts exhibited the highest C₅-paraffin conversion of 32.8%, with a C₅ total olefin selectivity of 68.7%, while the closest-proximity sample had a lower conversion of 17.4%, with a C₅ total olefin selectivity of 20.9%. The FT-IR (Fourier transform infrared spectroscopy) spectroscopic study of the CO adsorption and X-ray photoelectron spectroscopy results revealed that the closest proximity between Pt and Ce inhibited PtSn alloy formation due to their strong interaction. However, for the nanoscale-proximity sample, neighboring CeO₂@SiO₂ did not disturb Pt₁Sn₁ intermetallic formation. This strategy can be applied to other CO₂ activation catalysts, instead of CeO₂@SiO₂. This paper aims to provide insights into the influence of metal–CeO_x intimacy in bi-functional catalysts.

1. Introduction

CO₂-assisted oxidative dehydrogenation (CO₂-ODH) of alkane to the corresponding olefins (ethylene, propylene, and styrene) involves the use of CO₂ as a soft oxidant, which serves as one of the strategies for CO₂ utilization [1,2,3,4,5]. The CO₂-ODH process can offer an alternative to the catalytic non-oxidative dehydrogenation process used in various industries. This method can lead to an enhanced thermodynamic yield of olefins by removing H₂ through reverse water–gas shift reactions (i.e., H₂ + CO₂ → H₂O + CO) [2,5]. Furthermore, CO₂ can improve the catalyst lifetime by facilitating coke removal via the reverse Boudouard reaction [2,6].

Naphtha steam cracking is a process that involves heating naphtha in the presence of steam at high temperatures (750–950 °C) to produce light olefins, which serve as feedstocks for polyethylene and polypropylene. Typically, ethylene constitutes around 57% of the cracked products, while propylene accounts for approximately 39%, although these percentages may vary based on the specific process conditions and naphtha composition [7]. In addition to these light olefins, naphtha cracking also generates a mixture of saturated hydrocarbons, including pentane and hexane. Typically, these heavier hydrocarbons (mostly C₅ and C₆) are either recycled back into the naphtha cracking process or added to the gasoline pool, both of which have relatively low value. In this regard, there is a high demand for converting these heavier hydrocarbons into value-added products.

C₅ olefins (n-pentene and its isomers) are of immense importance in the petrochemical industry as raw materials in manufacturing a vast array of products comprising rubber, elastomers (polyisoprene, butyl rubber, etc.), fine chemicals, hydrocarbons, and resins [8]. ODH of C₅-paraffin to olefin is a cost-effective process that can overcome the thermodynamic limitations of endothermic dehydrogenation. However, only a limited number of articles have been reported on C₅-paraffin ODH. This is due to the observation that the presence of pure oxygen negatively affects the selectivity and product distribution [8,9,10]. To address the limitations of using oxygen as an oxidant in ODH, CO₂ has been identified as a more viable alternative. However, CO₂ activation remains a concern. This is because dehydrogenation of C₅-paraffin requires a reaction temperature of approximately 500 °C to attain high selectivity, which is significantly lower than the typical ODH reaction temperature (~600 °C) when using ethane and propane as reactants. Consequently, under these conditions, the extent of CO₂ activation is remarkably reduced compared to the light-alkane dehydrogenation process.

Over the past two decades, Pt-based bimetallic catalysts have been extensively studied in the dehydrogenation of lower alkanes. Pt-based bimetallic catalysts featuring rare-earth elements (REEs) as secondary metals have also been utilized very recently by our groups [2,11]. Among them, supported Pt-Sn alloys are extensively reported catalysts in ethane and propane dehydrogenation, exhibiting high catalytic activity, olefin selectivity, and improved catalyst durability [12,13,14,15,16,17,18]. Based on these observations, we explored the combination of CeO₂ and silica-supported Pt-Sn (Pt-Sn@SiO₂) as a bi-functional catalyst at lower temperatures (450–500 °C) for C₅ ODH with CO₂. We chose cerium oxide as the CO₂ activator because various studies have reported that CeO₂, having lattice oxygen and oxygen defects, can facilitate CO₂ reduction during the dehydrogenation reaction [2,14]. However, at lower reaction temperatures, the significantly reduced activation of CO₂ presents a challenge. To balance CO₂ activation with alkane dehydrogenation, the cerium content should be increased, as its surface lattice oxygen can facilitate CO₂ activation. Nonetheless, an excess of cerium can also lead to CO₂ dry reforming reactions, which favor C–C bond cleavage. To address this limitation, the proximity between CeO₂ and Pt-Sn@SiO₂ was investigated. Scheme 1 presents the various integration methods used, including physical mixing, co-impregnation, ball milling, and a dual-bed catalyst (details in the experimental section), in order to achieve superior catalytic activity and high C₅ olefin selectivity.

2. Results and Discussion

2.1. Texture and Structural Properties of 1Pt-0.6Sn@SiO₂ Catalyst

Figure S1a shows the XRD (X-ray diffraction) pattern for the SiO₂ support (CARiACT Grade G-6), which exhibited a broad peak at around a 2θ of 22.6°, indicating the amorphous nature of SiO₂ [13]. The N₂ adsorption–desorption isotherm result (Figure S1b) shows that the SiO₂ support exhibited an increase in N₂ adsorption in the region of 0.45 < P/P₀ < 0.8, which demonstrates the mesoporous nature of SiO₂, resulting in a high specific surface area of 500 m² g⁻¹, with a pore volume of 0.77 cm³ g⁻¹. On this silica support, 1 wt% Pt and 0.6 wt% Sn were supported using the incipient wetness method and subsequently activated in O₂ at 350 °C, followed by H₂ reduction at 550 °C, according to the previous literature [12]. The resulting material was designated 1Pt-0.6Sn@SiO₂ (details in experimental Section 3.2.1). N₂ adsorption–desorption isotherm analysis was also employed to verify the textural characteristics of the 1Pt-0.6Sn@SiO₂ catalyst. As shown in Figure 1a, similar to pristine SiO₂, the region of 0.45 < P/P₀ < 0.8 showed an increase in N₂ adsorption, resulting from the available mesoporous architecture, which exhibited a typical type IV pattern. 1Pt-0.6Sn@SiO₂ had a large surface area of 486 m² g⁻¹, with a pore volume of 0.77 cm³ g⁻¹. For the reduced 1Pt-0.6Sn@SiO₂ catalyst (Figure 1b), the XRD results with a deep scan in the range of a 2θ of 35–47° show the characteristic peaks centered at a 2θ of ~41.8° and 44.1°, which represent the intermetallic compounds of Pt₁Sn₁ [12]. Except for this, no XRD peaks corresponding to Pt and Sn were detected. The particle size calculated from the peak centered at a 2θ of 41.8° using the Scherrer equation (D = Kλ/β.cosθ) was ~2 nm, reflecting the presence of intermetallic NPs composed of Pt and Sn [12]. Additionally, STEM (scanning transmission electron spectroscopy) analysis was conducted to investigate the dispersion of the Pt₁Sn₁ intermetallic NPs. Figure 1c shows uniformly dispersed NPs throughout the SiO₂ support. The particle size distribution was obtained by selecting more than 50 particles (Figure S2), and the average particle size was ~2 nm in diameter, which is well-matched to the XRD result. The reduction properties of the catalysts were measured using H₂-TPR (H₂ temperature-programmed reduction) analysis (Figure S3). The peak observed at 176 °C is related to the co-reduction of Pt and Sn species supported on SiO₂. According to the previous literature, when Pt and Sn were loaded on MCM-41 silica with a Sn/Pt weight ratio of less than 0.6, a hydrogen consumption peak was observed at approximately ~180 °C, attributed to the co-reduction of Pt and Sn [19]. This co-reduction process is facilitated by the reduction of Sn species, aided by the spillover of hydrogen from Pt metal nanoparticles. Furthermore, the XPS (X-ray photoelectron spectroscopy) analysis results were obtained to understand the oxidation state of Pt and Sn species. The binding energies around 71.4 eV for 4f_7/2 and 74.7 eV for 4f_5/2 for the 1Pt-0.6Sn@SiO₂ catalyst revealed the presence of metallic Pt⁰ (Figure 1d) [13,14,20]. On the other hand, the Sn 3d XPS spectra were deconvoluted to study the oxidation state of Sn. An XPS peak centered at 485.7 eV with a small peak at 486.7 eV was observed, corresponding to Sn⁰ and Sn^2+/4+ species, respectively (Figure 1d) [13]. The quantitative ratio between Sn⁰ and Sn^2+/4+ was calculated as 2.6.

The XRD analysis of the 75CeO₂@SiO₂ sample returned characteristic peaks at a 2θ of 28.9°, 33.1°, 47.8°, and 56.3°, which were assigned to the (111), (199), (220), and (311) crystal planes of the cubic fluorite structure of CeO₂, respectively (Figure 2a) [13]. The CeO₂ particle size calculated from the XRD peak 2θ = 28.9° was ~4 nm. The N₂ adsorption–desorption isotherm of 75CeO₂@SiO₂ showed an increase in N₂ adsorption in the 0.45 < P/P₀ < 0.8 region, indicating the presence of mesopores (Figure 2b). The catalyst had a comparatively lower surface area of 113.3 m² g⁻¹, with a pore volume of 0.1 cm³ g⁻¹. Moreover, H₂-TPR analysis was performed (Figure S3), where the reduction peak (appearing as an edge) at 410 °C was attributed to the removal of surface oxygen from small-sized ceria NPs situated on the SiO₂ surface [21]. The broad peaks of H₂ consumption at ~530 °C and above 760 °C indicated the reduction of lattice oxygen from the surface and bulk ceria.

2.2. Effect of Proximity between Ceo_x and Pt/Sn Species on Intermetallic NP Formation

We studied the effect of the proximity between Pt/Sn and CeO₂ in the bi-functional catalyst on the formation of intermetallic compounds between Pt and Sn. This study is based on the assumption that Pt species are known to strongly interact with CeO_2−x via strong metal–support interactions [13,14]. Given this premise, when 1 wt% PtO_x, 0.6 wt% SnO_x, and 75 wt% CeO₂ are simultaneously supported on SiO₂, PtO_x might preferentially be located on CeO₂ without direct contact with SnO_x. This could potentially result in the absence of intermetallic compound formation. To this end, we prepared two catalysts: one involving the co-impregnation of Pt, Sn, and Ce precursors together on a silica support, and another consisting of a physical mixture of equal quantities of 1Pt-0.6Sn@SiO₂ and 75CeO₂@SiO₂ (Scheme 1a–c). Accordingly, the resulting catalysts were designated co-impregnated catalysts, and physically mixed catalysts, respectively.

The co-impregnated and physically mixed catalysts were characterized using various techniques. The XRD patterns obtained for both catalysts showed characteristic peaks related to cerium oxide (Figure S4). From the XRD results, the CeO₂ particle size was calculated for the representative peak (2θ = 28.9°), where the co-impregnated catalyst had a particle size of 3.01 nm, and the physically mixed sample had a particle size of 4 nm. Based on the STEM analysis images (Figure S5), the regions of high intensity correspond to areas with a high concentration of CeO₂. Distinguishing Pt and Sn from CeO₂ particles is challenging due to the dominance of CeO₂ particles. For the co-impregnated catalyst, the CeO₂ particles appeared to be homogeneously dispersed; however, it is difficult to determine the particle size of Pt-based nanoparticles owing to the high content of cerium and its random arrangement. A similar situation was observed for the physically mixed catalysts.

The electronic properties of both the physically mixed and co-impregnated catalysts were studied using XPS analysis, as shown in Figure 3a,b. Figure 3a displays the XPS spectra for Pt 4f, indicating that Pt species were present in metallic form in both catalyst types. However, in the case of Sn 3d, we observed a notable difference between the two catalysts. As compared to 1Pt-0.6Sn@SiO₂, for the co-impregnated catalyst, the Sn 3d XPS spectra after deconvolution showed that the intensity of the peak related to Sn⁰ decreased, and that for Sn^2+/4+ at a higher binding energy increased significantly (Figure 3b). The quantitative ratio between Sn⁰ and Sn^2+/4+ was 0.432, which was about 6 times lower than that of the 1Pt-0.6Sn@SiO₂ catalyst. This indicates that the reduction of SnO_x was prevented by the addition of CeO_x. This observation could be explained by the stabilization effect of CeO₂ on Pt or Sn species. Pt NPs could strongly interact with CeO₂ through a strong metal–support interaction, preventing direct contact between Pt NPs and SnO_x. Under this situation, facilitation of the reduction of SnO_x by spillover hydrogen might also be inhibited [13]. In addition to this, tin oxides might be effectively stabilized by CeO₂ via mixed metal oxide formation. In contrast to the co-impregnated catalysts, the XPS result obtained for the physically mixed sample was similar to that of 1Pt-0.6Sn@SiO₂. The calculated Sn⁰-to-Sn^2+/4+ ratio was 3.5, indicating that microscale intimacy between Pt/Sn and CeO₂ does not interfere with the formation of intermetallic compounds.

To investigate the surface arrangement of Pt atoms on metal nanoparticles for the two catalysts, in situ CO-FTIR analysis was performed. For the co-impregnated catalyst, the peak observed at 2066 cm⁻¹ was attributed to highly coordinated, medium-sized ensembles of Pt atoms (Figure 3c) [22,23]. Additionally, the small peak at ~2080 cm⁻¹ pertains to a collection of Pt atoms arranged in a similar manner to unobstructed, large ensembles [23,24]. This can be explained as follows: With the addition of Ce in the co-impregnation method, the strong interaction between Pt and Ce hindered the formation of the Pt-Sn alloy. As a result, pure Pt nanoparticles were formed. On the other hand, the physically mixed catalyst exhibited a peak centered at 2046 cm⁻¹ (related to Pt-Sn intermetallic NPs) [13,14]. The Pt-Sn NPs are expected to exhibit electron transfer from Sn to Pt due to their electronegativity difference, which leads to a higher degree of Pt electron back-donation into the 2π* antibonding orbital of the CO molecules. This phenomenon accounts for the peak shift towards a lower wavenumber, compared to the pure Pt NPs. In the physically mixed catalyst, in contrast to the co-impregnated catalyst, CeO₂ could not impede the formation of Pt-Sn NPs due to the long distance between CeO₂ and PtSn. Consequently, the physically mixed catalyst enabled the formation of Pt-Sn NPs.

2.3. C₅ Oxidative Dehydrogenation with CO₂: Effect of Nanoscale Intimacy between Ce and PtSn Catalysts

Before studying the effect of the special arrangement of the bi-functional catalysts, different parameters including the Ce content (Figure S6), CO₂ partial pressure (Figure S7), and reaction temperature (Figure S8) effects were inspected to optimize the reaction conditions. These studies were performed using granules of both catalysts, physically mixed. It was observed that mixed 1Pt-0.6Sn@SiO₂ and 75CeO₂@SiO₂, with a feed flow of Ar:CO₂:C₅ = 4:4:1 at 500 °C, had better activity with high C₅ olefin (mono-olefins + dienes) selectivity. These conditions were applied in the following study.

The effect of the proximity of the silica-supported Pt-Sn and CeO₂ catalysts was investigated in the C₅ ODH reaction. For the catalytic activity, the conditions were fixed as follows: T = 500 °C; Ar: CO₂: C₅ molar ratio = 4:4:1 with a total flow rate of 18 mL/min. In the case of the 1Pt-0.6Sn@SiO₂ catalyst, C₅ conversion of 22.9% (1 h time on stream) with 71.4% C₅ total olefin selectivity and 16.3% yield was obtained. However, after the addition of CeO₂ in a physically mixed manner, excellent catalytic activity with 32.8% C₅ conversion, 68.7% C₅ total olefin selectivity, and 22.5% yield was obtained (Figure 4). The introduction of cerium enhanced the C₅ ODH, but the selectivity was slightly decreased. The increase in C₅ conversion can be attributed to the increase in CO₂ conversion (Figure 4d), with CO₂ performing as a soft oxidant activated by CeO₂. On the other hand, the co-impregnated catalyst had a minimal conversion of 17.4% and C₅ total olefin selectivity of 20.9%, with a 3.6% yield, which was even lower than that of 1Pt-0.6Sn@SiO₂. For all studied catalysts, the by-products formed included CH₄, C_nH_2n+2, and C_nH_2n (n = 2, 3, 4) with a collective selectivity of ~7%, as well as other unknown products. Considering the overall catalytic activity, an increasing trend of co-impregnated << 1Pt-0.6Sn@SiO₂ < physically mixed was observed.

To determine the reason for the significant difference in the catalytic performance, particularly between the physically mixed and co-impregnated samples, the catalysts were characterized using XPS and CO-FTIR (see Section 2.2). The XPS analysis results of the physically mixed granules showed that the Pt-Sn intermetallic NPs were not disturbed (Figure 3a,b), which was further confirmed using CO-FTIR (Figure 3c), where a peak related to CO chemisorbed on Pt₁Sn₁ was observed. It is well-known that Pt-Sn intermetallic NPs usually result in C–H bond breaking in alkane dehydrogenation with the concomitant generation of H₂. The addition of CeO₂@SiO₂ can catalyze the reverse water–gas shift reactions, where CO₂ reacts with the generated H₂ to produce syngas. The removal of H₂ might facilitate the C₅-paraffin dehydrogenation reaction, leading to a dramatic increase in C₅ conversion. In the case of the co-impregnated 1Pt-0.6Sn/75CeO₂@SiO₂ catalyst, the comparatively strong interaction between Ce and Pt, which formed a Pt–O–Ce bond instead [14], restricted the Pt and Sn interaction to form intermetallic NPs. This resulted in SnO₂ formation, as confirmed in the XPS analysis (Figure 3b). The CO-FTIR results also showed the formation of large Pt ensembles, which prefer C–C bond cleavage over C–H bond cleavage (Figure 3c) [2]. Therefore, the co-impregnated samples exhibited a lower olefin selectivity as compared to 1Pt0.6Sn@SiO₂.

We further controlled the nanoscale intimacy between PtSn and CeO₂ by shortening the distance between the two components from milliscale to nanoscale. To this end, we followed different treatments including a dual-bed combination, physical mixing, and ball milling (Scheme 1). Considering other possible distributions of the bi-functional catalyst, the dual-bed combination and ball milling were also investigated for the CO₂-ODH of pentane, which returned a C₅ conversion of 25.9% and 24.7% with a C₅ total olefin selectivity of 62.3% and 62%, respectively (Figure 5). These activities were lower compared to those of the physically mixed catalyst. The dual-bed combination had slightly higher activity compared with 1Pt-0.6Sn@SiO₂, which could be due to the limited interaction between the two bed surfaces inside the reactor. On the other hand, the ball milling process damaged the pore structure of the 1Pt-0.6Sn@SiO₂ and 75CeO₂@SiO₂ samples, leading to a decrease in the specific surface area from ~486 m² g⁻¹ to 147 m² g⁻¹ (see below). Due to the collapse of the structure, the ball milling catalysts might exhibit lower activity than the physically mixed catalysts.

It is well-known that Cu supported on alumina (Cu/Al) promotes reverse water–gas shift reactions [25,26]. In this study, we also examined the effect of Cu/Al in the C₅ ODH reaction. Zinc (Zn) was added to enhance the Cu dispersion. For comparison, Ce-Zn/Al was also prepared. These catalysts were individually mixed with 1Pt-0.6Sn@SiO₂ and were analyzed for C₅ ODH (Figure 6). The catalytic results show that with the addition of Ce-Zn/Al, the initial activity (1 h time on stream) was almost doubled, with a C₅ conversion of 44%, but continuously decreased with the time on stream. On the other hand, the bi-functional catalyst composed of 1Pt-0.6Sn@SiO₂ and Cu-Zn/Al returned an initial conversion of 22% (1 h time on stream), where slow deactivation was observed after 2 h of time on stream. These results indicate that the addition of a metal, which has the ability to actively participate in the reverse water–gas shift (RWGS) reaction, drives the equilibrium concentration towards the products, leading to an enhanced yield of C₅ olefins.

3. Materials and Methods

3.1. Materials

The following materials were used: Chloroplatinic acid hexahydrate (H₂PtCl₆·6H₂O, ACS reagent, ≥37.50% Pt basis, Aldrich, St. Louis, MO, USA), Tin(II) chloride dihydrate (SnCl₂·2H₂O, ACS reagent, 98% trace metal basis Thermoscientific, Shanghai, China), Ammonium Cerium (IV) nitrate (Ce(NH₄)₂(NO₃)₆, ACS reagent, 99%), LUDOX-AS-40 (Aldrich, St. Louis, MO, USA), CARiACT Grade G-6 (denoted as SiO₂) (size: 75–250 µm, Fuji Silysia Chemical LTD, Japan), and Cerium oxide (99.99%) 10 nm, CAS NO: 1306-38-3 (Avention, Siheung, Republic of Korea).

3.2. Catalyst Preparation

3.2.1. Preparation of Pt-Sn Supported on SiO₂ (1Pt-0.6Sn@SiO₂)

The silica-supported Pt-Sn catalyst was prepared using the incipient wetness impregnation method reported by Motagamwala et al. [12]. Firstly, H₂PtCl₆·6H₂O and SnCl₂·2H₂O sources were stirred in 2 mL of 0.1 M HCl solution for 3 h, followed by the addition of 1 mL deionized water. The solution obtained was loaded on the SiO₂ support. This sample was collected vacuum-dried for 12 h at room temperature, and calcined in air for 2 h at 350 °C. The calcined catalyst was labeled as 1Pt-0.6Sn@SiO₂ and stored in a desiccator. The Pt content was 1 wt%, and that of Sn was 0.6 wt%; this ratio was maintained constant throughout the study.

3.2.2. Preparation of CeO₂@SiO₂ Catalyst

CeO₂@SiO₂ catalysts were prepared following a simple two-step procedure. Generally, an appropriate amount of the cerium source (Ce(NH₄)₂(NO₃)₆) was stirred in 5 g of H₂O. After 20 min, 1 g of LUDOX-AS-40 was added in a dropwise manner. After 30 min of stirring, diluted ammonia solution (35%) was added dropwise to the reaction mixture to attain a pH of ~8–10. After filtration and washing with distilled water, the sample was dried at 100 °C for 12 h, followed by calcination at 500 °C for 4 h. Samples with different cerium contents were prepared and labeled as xCeO₂@SiO₂ (x = 5, 25, 50, 75, 100), characterized, and stored for further use.

3.2.3. Preparation of Pt-Sn/CeO₂@SiO₂ Catalyst (Co-Impregnated)

For the co-impregnated sample, Pt-Sn solution was prepared following the procedure discussed in Section 3.2.1 and subsequently applied to 75CeO₂@SiO₂ via incipient wetness impregnation. The remaining protocol was the same as that discussed in Section 3.2.1. The catalyst prepared was labeled as a co-impregnated catalyst.

3.3. Characterization

N₂ adsorption–desorption was measured using a Micrometrics Tristar II 3020 instrument (Norcross, GA, USA) at −196 °C. All samples were degassed at 250 °C for 3 h before analysis. The Brunauer–Emmett–Teller (BET) equation was used for the surface area calculation using an adsorption branch obtained within a relative pressure (P/P₀) range of 0.05 to 0.3. The powder X-ray diffraction (XRD) patterns were investigated using a Rigaku Miniflex instrument (Rigaku corporations, Tokyo, Japan) at 30 kV and 15 mA with a Cu Kα source (λ = 0.154 nm). STEM images were obtained using a FEI Titan environmental transmission electron microscope (ETEM) G2 (Hillsboro, OR, USA) at an acceleration voltage of 300 kV. X-ray photoelectron spectroscopy (XPS) was conducted using a Thermo Scientific (NEXSA) instrument (Waltham, MA, USA)with an Al–Kα source (1486.6 eV). The in situ CO-FTIR was analyzed using a Shimadzu IR Tracer-100 spectrometer (Kyoto, Japan), with an in situ IR cell. H₂-TPR analysis was performed on catalysts using a BELCAT-M instrument (MicrotracBEL, Osaka, Japan). The catalyst was pretreated in an Ar flow at 300 °C for 1 h and cooled, and the analysis was carried out in a 10%H₂/Ar flow at 50mL/min. The temperature was increased to 800 °C with a ramping rate of 5 °C/min.

3.4. Catalytic Test

The C₅ ODH was performed at 500 °C using a fixed-bed downflow quartz reactor (L = 45 cm, i.d. = 1 cm). Before the activity measurement, 0.2 g of catalyst (0.1 g 1Pt-0.6Sn@SiO₂ and 0.1 g 75CeO₂@SiO₂) was exposed to 10 vol.%H₂/Ar at 550 °C. After holding for 2 h, the 10 vol.%H₂/Ar flow was switched with Ar, and the reactor was cooled to 500 °C. A laboratory-designed Pyrex bubbler was used for C₅-alkane. The C₅ saturator was set to −16 °C to obtain a weight hour space velocity (WHSV) = 3.88 h⁻¹. A feed of Ar and CO₂ with a total of 16 mL/min (8 mL each) was passed through the bubbler followed by the catalyst bed, and fixed in the center of the reactor with the feed composition of Ar:CO₂:C₅ = 4:4:1. The reactor outlet line was directly connected to an online GC (gas chromatography) system (Agilent: 7890B, Santa Clara, CA, USA) equipped with flame ionization and thermal conductivity detectors. For the hydrocarbon separation, an HP-Al/S column was used, and for the CO₂ and CO separation, an Rt-U Bond capillary column from Restek (Bellefonte, PA, USA) was used.

The C₅-alkane and CO₂ conversion and the product selectivity were obtained using the equations below:

C₅ conversion:

$$X_{C_5{H}_{12}}=\frac{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{12}{}_{\left(\mathrm{in}\right)}-{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{12}{}_{\left(\mathrm{out}\right)}}{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{12}{}_{\left(\mathrm{in}\right)}}\times 100\%$$

(1)

C₅ olefin selectivity:

$$S_{C_5{H}_{10-x}}=\frac{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{10-x}{}_{\left(\mathrm{out}\right)}}{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{12}{}_{\left(\mathrm{in}\right)}-{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{C}}_5{\mathrm{H}}_{12}{}_{\left(\mathrm{out}\right)}}\times 100\%$$

(2)

where $C_5{H}_{10-x}$ represents different products formed in pentane oxidative dehydrogenation with CO₂ (10 − x = number of hydrogens, and x = 0, 2).

The CO₂ conversion was calculated using the equation below:

$$X_{{\mathrm{CO}}_2}=\frac{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{CO}}_2{}_{\left(\mathrm{in}\right)}-{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{CO}}_2{}_{\left(\mathrm{out}\right)}}{{\mathrm{Peak}\mathrm{area}\mathrm{of}\mathrm{CO}}_2{}_{\left(\mathrm{in}\right)}}\times 100\%$$

(3)

4. Conclusions

Mesoporous silica (SiO₂)-supported Pt-Sn and CeO₂ catalysts (1Pt-0.6Sn@SiO₂ and 75CeO₂@SiO₂) were prepared and utilized for C₅ ODH using different integration patterns to obtain better activity with high C₅ olefin selectivity. Among all the catalysts, the physically mixed bi-functional catalyst returned a higher C₅ conversion of 32.8%, with a C₅ total olefin selectivity of 68.7%, while for the co-impregnated catalyst, the activity was even lower than that of 1Pt-0.6Sn@SiO₂. The XPS and CO-FTIR results confirmed that for the physically mixed catalyst, the Pt-Sn intermetallic NPs were not disturbed. Additionally, the existence of CeO₂@SiO₂ in the vicinity effectively enhanced the activity, which was also attributed to the rise in CO₂ conversion, which enhanced the C₅ ODH. On the other hand, in the case of the co-impregnated catalyst, the strong Pt–Ce interaction hindered the formation of Pt-Sn intermetallic NPs. Moreover, the addition of Cu-Zn/Al supplemented the RWGS reaction, which elevated the catalyst stability by pushing the equilibrium towards the products.