Effect of Metal Dispersion in Rh-Based Zeolite and SiO₂ Catalysts on the Hydroformylation of Olefin Mixtures from Fischer–Tropsch Synthesis

Abstract

This study investigates the hydroformylation of C₅₊ olefins derived from Fischer–Tropsch synthesis (FTS) using Rh-based catalysts supported on zeolites (MFI, MEL) and SiO₂. A series of catalysts were synthesized through two different methods: a one-pot hydrothermal crystallization process, which results in highly dispersed Rh species encapsulated within the zeolite framework (Rh@MFI, Rh@MEL), and an impregnation method that produces larger Rh nanoparticles exposed on the support surface (Rh/MFI, Rh/MEL, Rh/SiO₂). Characterization techniques such as BET, TEM, and FTIR were employed to evaluate different catalysts, revealing significant differences in the dispersion and accessibility of Rh species. Owing to its more accessible mesoporous structure, Rh/SiO₂ with a pore size of 5.6 nm exhibited the highest olefin conversion rate (>90%) and 40% selectivity to C₆₊ aldehydes. In contrast, zeolite-encapsulated catalysts exhibited higher selectivity for C₆₊ aldehydes (~50%) due to better confinement and linear aldehyde formation. This study also examined the influence of FTS byproducts, including paraffins and short-chain olefins, on the hydroformylation reaction. Results showed that long-chain paraffins had a negligible effect on olefin conversion, while the presence of short-chain olefins, such as propene, reduced both olefin conversion and aldehyde selectivity due to competitive adsorption. This work highlights the critical role of catalyst design, olefin diffusion, and feedstock composition in optimizing hydroformylation performance, offering insights for improving the efficiency of syngas-to-olefins and aldehydes processes.

1. Introduction

Syngas, a mixture of H₂ and CO, is primarily derived from carbonaceous resources such as coal, natural gas, and biomass [1,2,3]. It serves as a versatile feedstock for the production of high-value chemicals. Among the various conversion pathways, Fischer–Tropsch synthesis (FTS) is widely utilized to transform syngas into diverse products, including alkanes [4,5], olefins [6,7,8], and high-value feedstocks such as C₆₊ alcohols [9,10]. These C₆₊ alcohols have extensive applications, including their use as plasticizers, detergents, and pharmaceutical precursors, with their value increasing proportionally to the carbon chain length [11,12,13]. Extensive research has focused on the conversion of syngas to alcohols, with particular emphasis on C₆₊ alcohols. However, this process involves competitive reaction pathways [14,15,16]. In a typical FTS sequence, CO is first dissociated on the catalyst surface to generate carbon species, followed by carbon–carbon coupling to elongate the chain. Non-dissociative CO insertion into the intermediate species and subsequent hydrogenation yield alcohols [17,18,19,20]. During this process, dissociative and non-dissociative CO activation compete. Stronger CO dissociation favors the production of long-chain olefins, whereas enhanced non-dissociative activation promotes the formation of lower-carbon alcohols [21]. Despite numerous advancements in catalyst design, the selectivity for C₆₊ alcohols remains below 30%, underscoring the challenge posed by this competitive reaction [16,17,22,23].

To address this challenge, a two-step reaction process has been proposed. In the first step, syngas is converted into C₅₊ olefins via FTS, and in the second step, the hydroformylation of the C₅₊ olefins produces C₆₊ aldehydes [24], which can then be readily hydrogenated to form alcohols. For this approach to be viable, the FTS reaction must achieve high selectivity for C₅₊ olefins to ensure a sufficient supply of feedstocks for hydroformylation. Recent studies have reported promising catalyst systems for this purpose [25,26,27,28,29,30]. For instance, the NaRu/SiO₂ catalyst has demonstrated remarkable efficiency in catalyzing syngas conversion, achieving over 70% selectivity for C₅₊ olefins [26]. Recently, Rh@MEL [31] and Rh@MFI [32] catalysts were developed to catalyze the hydroformylation of olefins with high selectivity of linear products, wherein the zeolite is no longer the catalyst but the scaffold to induce regioselectivity. Such catalyst system could induce further improved product selectivity to the final products. However, the Fischer–Tropsch process generates byproducts, such as short-chain olefins, alkanes, and water, which mix with C₅₊ olefins and serve as the feedstock for hydroformylation.

The complex composition of the feedstock raises several critical challenges. Byproducts from the FTS reaction, including short-chain olefins, alkanes, and water, may influence the hydroformylation of C₅₊ olefins, potentially affecting the overall reaction efficiency and selectivity. Consequently, the efficiency of hydroformylation with a mixed feedstock may not be comparable with that achieved with pure olefins. Additionally, the diffusion behavior of C₅₊ olefins is likely influenced by the properties of the hydroformylation catalyst, necessitating an exploration of various heterogeneous catalysts to determine their suitability for this process. Addressing these challenges is essential to enable a seamless tandem reaction using heterogeneous catalysts, representing a significant research gap that requires further investigation.

In this work, we investigated the catalytic hydroformylation performance of C₅₊ olefins within FTS product mixtures containing various components (mainly including long-chain alkanes, short-chain alkenes, and water). A series of Rh catalysts with differing Rh loading methods were synthesized to catalyze the hydroformylation reaction. Characterization techniques, including transmission electron microscopy (TEM) and diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS), were employed to analyze the Rh particle size and spatial distribution within the catalysts. Feedstock mixtures with different compositions were used to systematically examine their effects on the hydroformylation process. The impact of diffusion behaviors of the long-chain olefins was evaluated by comparing catalysts with distinct Rh species positions within the catalyst structure. Additionally, the catalytic performance was studied in the presence of water or propene to assess their impact on the hydroformylation of C₅₊ olefins, as these molecules may compete with C₅₊ olefins for catalytic sites. Insights into the influence of feedstock diffusion, short-chain olefins, and water on the hydroformylation reaction provide valuable guidance for the future development of heterogeneous catalysts aimed at the efficient conversion of FTS products to aldehydes. Such knowledge will be instrumental in the design of catalysts for the conversion of syngas into C₆₊ alcohols.

2. Results and Discussion

2.1. Catalyst Synthesis and Characterization

A series of Rh catalysts with varying Rh loading methods were synthesized to catalyze the hydroformylation reaction. Rh@MFI and Rh@MEL catalysts were prepared via a one-pot hydrothermal crystallization process, while Rh/MFI, Rh/MEL, and Rh/SiO₂ catalysts were synthesized using the impregnation method. The MFI and MEL zeolite supports were synthesized through the same one-pot hydrothermal crystallization process without metal loading, ensuring that the morphology of the zeolites remained consistent. The crystalline structures of the prepared catalysts were confirmed by powder X-ray diffraction (XRD) analysis (Figure 1). No characteristic peaks of SiO₂ were observed, confirming its amorphous nature. However, distinct peaks for RhO₂ were observed at 27° and 36°, indicating the presence of Rh species. The diffraction patterns of the Rh@MFI and Rh@MEL samples aligned well with the characteristic peaks of MEL and MFI, suggesting the successful formation of the zeolite structures during the hydrothermal crystallization process [31,32]. These diffraction patterns also confirmed that Rh loading did not alter the zeolite crystalline structures. A comparison of the XRD patterns of the catalysts synthesized via the impregnation and one-pot hydrothermal methods reveals that neither method affects the zeolite structure, and notably, peaks corresponding to Rh oxides were absent in the zeolite-supported catalysts, indicating better Rh dispersion compared with Rh/SiO₂.

To examine the morphologies of the various catalysts, scanning electron microscopy (SEM) was performed, and the results are shown in Figure 2. The Rh@MEL and Rh/MEL samples exhibited a regular block-like shuttle shape, while the Rh@MFI and Rh/MFI samples displayed a block-like spherical shape [31,32]. These similar morphologies can be attributed to the identical one-pot hydrothermal synthesis conditions, with the exception of metal loading. The catalysts supported on the same type of zeolite exhibited similar morphologies, providing a fair basis for comparing their catalytic performances. In contrast, the Rh/SiO₂ samples displayed irregular shapes due to the amorphous nature of the SiO₂ support. To further investigate the metal loading and porous structure of the catalysts, inductively coupled plasma optical emission spectrometry (ICP-OES) and N₂ physisorption (BET analysis) were employed to determine the elemental compositions and surface areas of the catalysts (

Table 1. Surface area and elemental composition analysis results of catalysts.

Catalysts	Surface Area (m2/g)			Pore Sizes (nm)	ICP Elemental Analysis
	Total	External	Micro		Rh (wt %)
Rh@MFI	291	7	284	0.53	0.21
Rh@MEL	303	7	296	0.48	0.20
Rh/MFI	289	6	283	0.54	0.23
Rh/MEL	299	8	291	0.50	0.20
Rh/SiO2	223	182	40	5.60	0.21

). The ICP-OES results showed similar Rh loadings across the different catalyst types, ensuring a fair comparison of their catalytic activities in subsequent sections. The micropore volume in MFI and MEL zeolites was close to 97.5%, whereas in SiO₂, it was only 18%, with the majority of the pores being mesopores. The nitrogen adsorption−desorption isotherms also confirmed this point (Figure 3). As shown in Figure 3, the Rh/SiO₂ catalyst exhibits a distinct hysteresis loop, indicating that it is a mesoporous material. In contrast, no significant hysteresis loop is observed for the zeolite catalysts, suggesting the presence of a large number of uniformly distributed micropores within these catalysts. Moreover, the loading mode of Rh species does not affect the microporous structure of the zeolites. This is expected, as the channels of MFI and MEL zeolites feature 10-membered rings with a pore size of approximately 0.55 nm, while the pore size of the SiO₂ samples, as determined by BET analysis, is 5.6 nm. These results suggest that catalysts supported on MFI and MEL zeolites provide greater spatial confinement, thereby imposing a higher diffusion barrier for C₅₊ mixed olefins.

The characterization results above confirm that the synthesis of the catalyst supports met expectations, and the Rh loadings are consistent across the different catalysts. Given that the Rh species serve as the active sites for the hydroformylation reaction, it is essential to examine their dispersion within the catalysts. To investigate this, the catalysts were characterized using high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) and transmission electron microscopy (TEM), with the results presented in Figure 4. TEM images of Rh/SiO₂, Rh/MFI, and Rh/MEL catalysts reveal distinct Rh clusters with sizes of 0.8 nm, 1.8 nm, and 1.2 nm, respectively. Compared with SiO₂, Rh species on MFI and MEL zeolites, when prepared via impregnation, exhibit larger particle sizes, suggesting that Rh species are better dispersed on SiO₂ than on MFI and MEL zeolites [33,34,35,36]. In contrast, no obvious Rh species are observed in the TEM images of Rh@MFI and Rh@MEL. However, energy dispersive X-ray spectroscopy (EDS) scanning images reveal a uniform distribution of Rh, indicating that the Rh species exist as sub-nanometer clusters or isolated Rh atoms, which are not readily observable in the STEM images. This finding aligns with the expectation that small Rh species may not be visible in high-resolution TEM images due to their extremely small size and uniform distribution within the zeolite supports [37,38,39,40].

Since Rh species were not detectable in the HAADF-STEM images of Rh@MFI and Rh@MEL catalysts, we employed CO adsorption FTIR to further confirm the dispersion of Rh and investigate its local environment. This technique can distinguish between isolated Rh atoms and Rh nanoparticles. Figure 5 shows the presence of three distinct IR bands. The band centered around 2055 cm⁻¹ is attributed to CO linearly adsorbed on Rh clusters [41,42]. The bands at 2082 cm⁻¹ and 2007 cm⁻¹ correspond to the symmetric and asymmetric vibrations of geminal dicarbonyls on isolated Rh atoms [43,44]. A broad band spanning the 1900–2000 cm⁻¹ region [45,46], which is associated with CO adsorbed in a bridge-like structure on Rh clusters, was almost absent. These findings confirm the presence of both isolated Rh atoms and Rh clusters in the Rh@MFI and Rh@MEL catalysts, with a notable proportion of isolated Rh atoms [47]. This result is consistent with the observations made in the HAADF-STEM analysis.

2.2. Catalytic Performance Evaluation

The NaRu/SiO₂ catalyst was utilized for the conversion of syngas into hydrocarbons, with a particular focus on maximizing the production of C₅₊ olefins. As demonstrated in a previous study [26], Na serves as a promoter by enhancing the dispersion of Ru nanoparticles. The uniform distribution of Na on the catalyst surface facilitates strong electronic interactions between Na and Ru nanoparticles. Na ions donate electrons to Ru atoms, enriching the Ru surface with electrons, which enhances CO adsorption at Ru sites while suppressing hydrogen reactivity. This electron transfer effect reduces the likelihood of secondary olefin hydrogenation, thereby increasing the olefin yield. Consequently, the NaRu/SiO₂ catalyst exhibits high selectivity for C₅₊ olefins during the catalytic conversion of syngas.

To optimize C₅₊ olefin production, the NaRu/SiO₂ catalyst with 5 wt% Ru loading and a Na/Ru molar ratio of 0.5 was employed for Fischer–Tropsch synthesis (FTS). The reaction was carried out under conditions of 260 °C, 1 MPa, and a H₂/CO ratio of 2 (H₂/CO/N₂ = 64/32/4). The results showed a CO conversion of 42.3% and an olefin selectivity of 66.4%, with C₅₊ olefins accounting for 50.8% of the olefin fraction. The selectivities for CH₄ and CO₂ were 3.0% and 2.7%, respectively (Figure 6a). The organic phase of the FTS products was then used as feedstock for the hydroformylation reaction. Specifically, 0.3 g of FTS products were mixed with 5 mL of toluene, and the reaction was carried out at 80 °C, 3 MPa, and a H₂/CO ratio of 1 (H₂/CO/Ar = 45/45/10).

To assess the effect of olefin diffusion on the reaction performance, a series of Rh catalysts with different Rh loading methods were employed. Rh@MFI and Rh@MEL catalysts were prepared via a one-pot hydrothermal crystallization process, resulting in highly dispersed Rh species encapsulated within the zeolite framework. This arrangement allows the zeolite framework to guide the reaction pathway of intermediates confined in the space between the zeolite structure and Rh centers, promoting the exclusive formation of linear aldehyde products [31,32]. In contrast, Rh/MFI, Rh/MEL, and Rh/SiO₂ catalysts were synthesized via impregnation, resulting in larger Rh nanoparticles exposed on the external surface of the support, where they are more accessible to olefin molecules. In the hydroformylation reaction, olefins must diffuse to the Rh sites for conversion into aldehydes. The kinetic diameters of normal olefins ranging from C₅ to C₁₂ are between 0.4 and 0.6 nm. The zeolite structure may impede olefin diffusion, thus affecting their accessibility to catalytic sites [48,49].

The catalytic performances of these catalysts in the hydroformylation of C₅₊ mixed olefins are shown in Figure 7a. The Rh@MEL catalyst exhibited higher C₅₊ olefin conversion and C₆₊ aldehyde selectivity compared with Rh@MFI. Although both MFI and MEL zeolites are microporous materials with similar micropore volumes (Table 1), MEL zeolite features two sets of straight channels, while MFI has one straight and one sinusoidal channel [50,51,52]. The sinusoidal channels in MFI impose greater diffusion resistance for C₅₊ olefins, requiring the olefins to overcome additional barriers to access Rh sites. Consequently, Rh@MEL displayed higher catalytic activity compared with Rh@MFI.

When Rh was loaded onto the support via the impregnation method (Rh/MEL, Rh/MFI, and Rh/SiO₂), the catalytic activity was significantly higher, indicating that olefin diffusion plays a critical role in determining catalytic performance. Olefin conversion was found to be higher over Rh/SiO₂ (>90%) than Rh/MEL and Rh/MFI. A plausible explanation for this is that a significant fraction of Rh atoms is embedded within the zeolite framework in Rh/MEL and Rh/MFI, making them less accessible to the olefin feedstock. In contrast, SiO₂ is mesoporous, and the Rh sites on SiO₂ are more easily reached by olefins, resulting in the higher conversion of C₅₊ olefins. This suggests that the zeolite framework introduces greater diffusion resistance for C₅₊ olefins. Despite the better Rh dispersion in the zeolite structure, the diffusion constraints make it more difficult for C₅₊ olefins to access these active sites, leading to lower activity in the hydroformylation reaction. Although the Rh@MEL catalyst displayed higher selectivity for C₆₊ aldehydes than Rh/SiO₂, the conversion of long-chain olefins was much lower over Rh@MEL. Therefore, enhanced feedstock diffusion is essential for efficient hydroformylation over Rh@MFI and Rh@MEL catalysts, which is crucial for the tandem conversion of syngas to aldehydes and alcohols.

To facilitate a more accurate comparison of the catalytic activities, we controlled the olefin conversion rate to approximately 10% and compared the selectivity of aldehydes with the results presented in Figure 7b. Similar to the findings above, the Rh@MFI and Rh@MEL catalysts exhibited higher selectivity. This is primarily attributed to the confinement effect within the zeolite channels, which restricts the isomerization of olefins within the pores, thereby facilitating their transformation into aldehydes. In contrast, the lack of such confinement on Rh/MEL, Rh/MFI, and Rh/SiO₂ catalysts allows olefins to more readily undergo isomerization, resulting in lower selectivity for aldehydes.

We compare the catalysts in this work with those in published works. For Rh/MEL, Rh/MF,I and Rh/SiO₂ catalysts, the selectivity of aldehyde is basically maintained between 35% and 50%, which is better than most catalysts prepared through the conventional impregnation method [32]. In contrast, the aldehyde selectivity of Rh@MEL and Rh@MFI catalysts is almost the same as that of Rh@S-1 catalysts prepared through solvo-free hydrothermal synthesis methods reported in the literature [53], both of which are about 50%. However, the conversion rate of our catalyst is slightly lower than that of other Rh@MEL and Rh@MFI prepared through the same method [54]. We speculate that this is due to the effect of the concentration of specific olefins in the feedstock on their conversion and aldehyde selectivity. In the studies reported in the literature, the raw materials used were all single olefin instead of C₅₊ olefin mixture, which we believe is the reason why the aldehyde selectivity observed in this work is lower than that reported in the literature.

In addition to C₅₊ olefins, the organic phase of the FTS products also consists of paraffins of varying chain lengths and a significant proportion of short-chain olefins such as ethene and propene. These components can adsorb onto the active sites of the hydroformylation catalyst, reducing the availability of catalytic active sites for C₅₊ olefins and lowering their conversion. Furthermore, FTS generates substantial amounts of water, which can interact with the active sites of the hydroformylation catalyst, potentially altering the reaction pathways for C₅₊ olefins. To explore the effects of short-chain olefins, alkanes, and water on the hydroformylation of C₅₊ olefins, experiments were conducted using the organic phase of the FTS products obtained over the NaRu/SiO₂ catalyst. These studies aimed to evaluate how these substances influence catalytic performance and provide insights for optimizing the tandem reaction process.

To investigate the effect of water on hydroformylation, 1 mL of water was added to the reaction system. The results, shown in Figure 8a, indicate that water had a minimal impact on both the conversion of C₅₊ olefins and the selectivity for C₆₊ aldehydes. These findings suggest that water has a negligible influence under the conditions studied, supporting the feasibility of coupling the FTS and hydroformylation reactions for efficient syngas conversion to C₆₊ aldehydes.

The organic phase of the FTS products contains significant amounts of long-chain paraffins, which may influence the hydroformylation of long-chain olefins. To assess this, we prepared a mixture of C₅–C₁₂ olefins that mirrored the mass distribution found in the FTS products (Figure 8b). Using this mixed olefin feedstock, hydroformylation was conducted at 80 °C, 3 MPa, and a H₂/CO ratio of 1 (H₂/CO/Ar = 45/45/10). The results, shown in Figure 6b, indicate a slight increase in both the conversion of C₅₊ olefins and the selectivity for C₆₊ aldehydes compared with the control experiment using FTS products as feedstock. This modest improvement is likely due to the dilution effect of paraffins in the FTS products, which could have reduced the competition for catalytic sites. However, under the conditions of this study, the overall impact of long-chain alkanes on the hydroformylation of C₅₊ olefins was negligible.

To further explore the influence of short-chain olefins on the hydroformylation of C₅₊ olefins, propene was chosen as a co-fed component. The reaction was conducted with a gas feedstock composition of H₂/CO/propylene/Ar = 45/45/2/8. The results, shown in Figure 8c, demonstrate that the presence of propene led to a decrease in both the conversion of C₅₊ olefins and the selectivity for C₆₊ aldehydes, compared with the control reaction (H₂/CO/Ar = 45/45/10). A plausible explanation for this is that propene, due to its smaller molecular size and faster diffusion rate, is more readily adsorbed onto the catalyst’s active sites. This preferential adsorption reduces the number of available active sites for C₅₊ olefins, thereby lowering their conversion. Furthermore, the reduced availability of catalytic sites may increase the likelihood of alternative reaction pathways for C₅₊ olefins, such as hydrogenation or isomerization, which can lead to a decrease in selectivity for C₆₊ aldehydes. These findings highlight the competitive adsorption effects of short-chain olefins and underscore the importance of catalyst design in tandem reactions. It is crucial to optimize catalyst properties to minimize these competitive effects and enhance the selective conversion of C₅₊ olefins to C₆₊ aldehydes.

3. Experimental Section

3.1. Materials Synthesis

NaRu/SiO₂ Preparation: The catalyst was prepared by incipient wetness impregnation. First, 0.3173 g nitrosyl ruthenium nitrate (Ru ≥ 31.3%, Alfa) was dissolved in 5.3 mL deionized water to prepare a nitrosyl ruthenium nitrate solution according to the required volume for the incipient wetness impregnation of 2 g of aerosol silica (99.9%, Aladdin, Brussels, Belgium). Then 0.0437 g NaNO₃ (AR, Beijing Tongguang Fine Chemical Co., Beijing, China) was added to the nitrosyl ruthenium nitrate solution to form the impregnation precursor solution. In order to better dissolve ruthenium nitrosyl nitrate and NaNO₃, the precursor solution was ultrasonic for 30 min. Then the SiO₂ support was impregnated with the above precursor solution and stirred at room temperature until impregnated completely. The sample was dried at 80 °C for 12 h and calcined in air at 400 °C for 4 h to obtain a NaRu/SiO₂ catalyst.

Rh@MFI Preparation: An amount of 4.06 g of TPAOH (tetrapropyl ammonium hydroxide, 40 wt% in H₂O, Sigma-Aldrich, St. Louis, MO, USA), 0.0725 g of KOH (AR, Beijing Tongguang Fine Chemical Co.), and 9.05 g of deionized water were mixed at room temperature to obtain a TPAOH solution. Then, 4.17 g TEOS (tetraethyl orthosilicate, >99%, Alfa Aesar, Ward Hill, MA, USA) was added, and the mixture was stirred at room temperature for 6 h. A precursor of 0.0051 g RhCl₃·xH₂O (for an Rh loading of 0.21%) dissolved in 0.1 mL deionized water and 0.1 mL EDA (ethylenediamine, ≥99.5%, Sigma-Aldrich) was prepared. The precursor solution was then added to the TEOS–TPAOH–water mixture and stirred for 30 min at room temperature. The resulting mixture was transferred to a 50 mL PTFE-lined autoclave and heated at 100 °C for 24 h under static conditions. Afterward, the product was washed with deionized water and ethanol (95%, Innochem, Beijing, China) until the pH reached 6–7, followed by drying at 60 °C overnight. The dried sample was calcined in a muffle furnace at 560 °C, with a ramping rate of 2 °C/min, for 8 h and reduced with H₂ (10% H₂ in Ar) at 500 °C for 4 h. The final catalyst was labeled Rh@MFI. The molar ratios of the raw materials used for the synthesis of the catalyst are as follows: 799 TPAOH: 2002 SiO₂: 129 KOH: 63811 H₂O: 1 Rh₂O₃: 150 EDA.

Rh@MEL Preparation: An amount of 3.88 g of TBAOH (tetrabutylammonium hydroxide, 40 wt% in H₂O, Sigma-Aldrich), 0.0303 g of KOH, and 2.32 g of deionized water were mixed at room temperature to obtain a TBAOH solution. Then, 4.13 g TEOS was added, and the mixture was stirred at room temperature for 6 h. A precursor of 0.0068 g RhCl₃·xH₂O (for an Rh loading of 0.20%) dissolved in 0.1 mL deionized water and 0.1 mL EDA was prepared. The precursor solution was then added to the TEOS–TBAOH–water mixture and stirred for 30 min at room temperature. The resulting mixture was transferred to a 50 mL PTFE-lined autoclave and heated at 100 °C for 24 h under static conditions. Afterward, the product was washed with deionized water and ethanol until the pH reached 6–7, followed by drying at 60 °C overnight. The dried sample was calcined in a muffle furnace at 560 °C, with a ramping rate of 2 °C/min, for 8 h and reduced with H₂ (10% H₂ in Ar) at 500 °C for 4 h. The final catalyst was labeled Rh@MEL. The molar ratios of the raw materials used for the synthesis of the catalyst are as follows: 399 TBAOH: 1321 SiO₂: 36 KOH: 17215 H₂O: 1 Rh₂O₃: 100 EDA.

Rh/MEL, Rh/MFI, and Rh/SiO₂ Preparation: The catalysts were prepared through wetness impregnation. In the preparation process of Rh/MFI and Rh/MEL described above, no metal precursor solution was added, but the other processes remained the same, resulting in the preparation of MFI and MEL samples. In a typical synthesis process, 4.36 mg of RhCl₃ is dissolved in 1.45 mL of deionized water to prepare a metal precursor solution, and then 1 g MFI or MEL or SiO₂ samples are added to the solution and continuously stirred at room temperature until the impregnation is complete. The sample is then dried at 80 °C for 12 h, calcined in air at 560 °C for 8 h, and finally reduced with H₂ (10% H₂ in Ar) at 500 °C for 4 h. The obtained samples were labeled as Rh/MFI, Rh/MEL, and Rh/SiO₂ catalysts.

3.2. Characterization Methods

Powder X-ray diffraction (XRD) patterns were measured using a Bruker D8 diffractometer (Bruker, Saarbrücken, Germany) using Cu Kα (λ = 0.15406 nm) radiation. The scanning range was 5–50° with a step size of 0.02°.

Inductively coupled plasma emission spectrometry (ICP-OES) analysis was performed using a PerkinElmer Optima 2100DV spectrometer (PerkinElmer, Waltham, MA, USA) equipped with a charge-coupled device detector to analyze the Rh content in the samples.

N₂ physical adsorption desorption analysis of Rh/SiO₂ was performed using a MicroActive ASAP 2420 physical adsorption analyzer (Micromeritics, Norcross, GA, USA), while the analyses of Rh/MFI, Rh/MEL, Rh@MFI, and Rh@MEL were performed using a MicroActive ASAP 2020 physical adsorption analyzer (Micromeritics, Norcross, GA, USA). Approximately 100 mg samples were pretreated under vacuum conditions at 350 °C for 5 h, and the N₂ isotherms were determined at −196 °C. The specific surface areas and pore size of Rh/SiO₂ were determined through the Brunauer–Emmett–Teller (BET) method and t-plot method, respectively. The surface areas of Rh/MFI, Rh/MEL, Rh@MFI, and Rh@MEL were determined through the Horvath–Kawazoe method.

Scanning electron microscopy (SEM) analysis was performed using an FEI Quanta 400 thermal field emission scanning electron microscope for electron microscopic observation at a voltage of 10 kV. Before microscope observation, an appropriate amount of the sample was put into a 10 mL centrifuge bottle with 5 mL of ethanol. The mixture was then subjected to ultrasound for 1 h to disperse the sample. A total of 5 drops of the mixture were added to a silicon support plate with a dropper. The silicon support plate with the sample was fixed to a conductive adhesive for gold spraying prior to the SEM analysis.

Transmission electron microscopy (TEM) and high-angle dark field scanning electron microscopy (HAADF-STEM) were used to observe the distribution of Rh species in the samples. A small number of samples were fully ground and dispersed in 7 mL dichloromethane (CH₂Cl₂) solution. After 30 min of ultrasound, 3–5 drops of suspension were applied to the ultrathin carbon film copper grid with an eyedropper. An FEI-Tecnai-Talos transmission electron microscope was used for observation with an acceleration voltage of 200 kV and a resolution of 1.4 Å.

Infrared spectra were collected using a Bruker VERTEX 70 infrared spectrometer (Bruker, Dresden, Germany). Spectra were obtained at a resolution of 4 cm⁻¹. The spectra of Rh@MEL and Rh@MFI catalysts upon CO adsorption were recorded in an environmental chamber. The sample (~30 mg) was first fully ground and pressed into translucent flakes. Then the sample was put into the chamber, heated to 500 °C under 10% H₂ (H₂/He = 10/90) flow, and pretreated for 1 h. Then it was cooled to room temperature and then purged with He for 10 min. After purging, the background spectrum was collected. CO flow was allowed into the sample chamber at a flow rate of 20 sccm for adsorption. During the CO adsorption process, the sample infrared spectra were collected every 30 s until the CO adsorption equilibrium (about 25 min). Then He flow was injected for purging. During the purging process, the infrared spectra of the sample were collected every 30 s until the spectrum was stable (about 25 min). The infrared spectrum of the sample after complete desorption was recorded.

3.3. Catalytic Performance Evaluation

Fischer–Tropsch Synthesis. The catalyst was evaluated using a micro-fixed bed reactor. The stainless-steel reaction tube had a quartz tube with an inner diameter of 6 mm, and the stainless-steel tube was inserted into the quartz tube to detect the temperature. Typically, 0.1 g catalyst is diluted with 0.05 g silicon carbide, and the mixture is loaded into the constant temperature zone of the reactor. Before the catalytic reaction, the synthesis gas with a H₂/CO ratio of 2/1 (H₂/CO/N₂ = 64/32/4) was injected into the reactor at a flow rate of 5 mL/min (WHSV = 3000 mL/g_cat) after being reduced at 450 °C for 4 h with H₂ (40 mL/min). Unless otherwise specified, the reaction pressure is 1 MPa and the H₂/CO ratio is 2. After passing through a hot trap (393 K) and a cold trap (273 K), the gaseous effluent was analyzed online using an Agilent 6890N apparatus equipped with two detectors. A packed column (PoraPak Q) and two capillary columns (HP-PLOT/Q and HP-MOLESIEVE) were connected to a thermal conductivity detector (TCD) with Ar as the carrier gas for the analysis of H₂, N₂, CO, CH₄, and CO₂. An alumina capillary column (HP-AL/S) was connected to a flame ionization detector (FID) with N₂ as the carrier gas for the analysis of hydrocarbon gaseous products. The specific methods were as follows: The initial column temperature was set at 40 °C and held for 3 min, then ramped to 75 °C at a rate of 5 °C/min. The temperature was then increased to 200 °C at a rate of 10 °C/min, and held at 200 °C for 20 min. The split flow rate at the injection port was 37.9 mL/min with a split ratio of 4:1. Peak areas were obtained by integrating the peaks of each detected substance on the chromatogram. Since CH₄ was detectable on both the TCD and FID detectors, the peak areas of CH₄ from the two detectors were used to compare the results obtained from them. The molar fraction of each substance was determined using a calibration curve. The molar amount of each substance in the gas mixture was calculated per unit time based on the flow rate of the internal standard (N₂) and its chromatographic peak area. Finally, the conversion of feedstock molecules and the selectivity of products were calculated from the molar amount of each component in the gas mixture. The aqueous products, liquid oil products, and solid wax products were collected from a cold trap and a hot trap. The aqueous products and the liquid oil products were analyzed off-line using an Agilent 7890A gas chromatograph (Agilent Technologies, Santa Clara, CA, USA). An AB-InoWax column connected to an FID was employed with N₂ as the carrier gas for the analysis of the aqueous products. Additionally, an HP-PONA capillary column connected to an FID was employed with N₂ as the carrier gas for the analysis of the oil phase products. The solid wax products were analyzed off-line using an Agilent 8890 gas chromatograph. The wax product was dissolved in CS₂ and analyzed using an Ultra ALLOY⁺-1 (Frontier Laboratories, Koriyama, Fukushima, Japan) column with an FID using N₂ as carrier gas.

CO conversion (X_CO) and product selectivity (S_i) were calculated using the following equation:

$${\mathrm{X}}_{\mathrm{C}\mathrm{O}}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{C}\mathrm{O},\mathrm{i}\mathrm{n}}-{\mathrm{F}}_{\mathrm{C}\mathrm{O},\mathrm{o}\mathrm{u}\mathrm{t}}}{{\mathrm{F}}_{\mathrm{C}\mathrm{O},\mathrm{i}\mathrm{n}}}}\times 100\%$$

$${\mathrm{S}}_i={\displaystyle \frac{N_i\times n_i}{\sum (N_i\times n_i)}}\times 100\%$$

where F_CO,in and F_CO,out represent moles of CO at the inlet and the outlet, respectively; S_i denotes the selectivity of product i on a carbon basis; N_i is the molar fraction of product i; and n_i is the carbon number of product i.

Hydroformylation Reaction. The catalyst was evaluated using a 50 mL high-pressure reactor. In a typical reaction, 5 mL of toluene is added to the high-pressure reactor as a solvent, followed by 0.03 g of catalyst and 0.3 g of feedstock (the organic phase product separated from the cold trap product from the FTS reactor). The reactor was sealed, purged 3 times with 1 MPa CO, filled with a 3 MPa H₂/CO ratio of 1 (H₂/CO/Ar = 45/45/10) syngas, heated to 80 °C, and maintained at a 1000 rpm stirring rate for 1 h. After the reaction, 1.5 g of the product was collected in a chromatographic bottle, which was subsequently analyzed via gas chromatography (Agilent 8890, equipped with an HP-5MS capillary column). The tail gas was collected using a 30 mL gas sampling bag and analyzed using an Agilent 8890 GC equipped with both a TCD and an FID. The permanent gases (Ar, CO, and H₂) and hydrocarbons (including propene, propane) in the tail gas were separated by an HP-5MS and a GasPro capillary column, respectively. Ar was used as the internal standard for quantification. The specific methods were as follows: The initial column temperature was set at 50 °C and held for 5 min. Subsequently, the temperature was increased to 200 °C at a rate of 5 °C/min and held for 10 min. Finally, the temperature was raised to 250 °C at a rate of 10 °C/min and maintained at 250 °C for 10 min. The split flow rate at the injection port was 18.9 mL/min with a split ratio of 4:1. The peak area for each substance was obtained by integrating the chromatogram peaks. Subsequently, we employed GC–MS (Agilent 5977B GC/MSD, equipped with an HP-PONA capillary column) to identify individual components in the gaseous and liquid products, followed by the determination of the relative correction factors for each component. The specific methods are as follows: The initial column temperature was set at 30 °C and held for 10 min. Subsequently, the temperature was increased to 300 °C at a rate of 3 °C/min and maintained at 300 °C for 30 min. The split flow rate at the injection port was 15 mL/min with a split ratio of 50:1. The molar fraction of each substance in the gas product and the liquid product was determined using a calibration curve. The molar amounts of each substance in the gas product and liquid product were then calculated based on the internal standard and chromatographic peak areas. Finally, the conversion of raw materials and the selectivity of products were calculated based on the molar amount of each component. The conversion of C₅₊ olefins (X_C5+) and the selectivity of C₆₊ aldehydes (S_C6+) were calculated using the following equations:

$${\mathrm{X}}_{\mathrm{C}5+}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{C}5+,\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}}-{\mathrm{F}}_{\mathrm{C}5+,\mathrm{a}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{r}}}{{\mathrm{F}}_{\mathrm{C}5+,\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}}}}\times 100\%$$

$${\mathrm{S}}_{\mathrm{C}6+}={\displaystyle \frac{{\mathrm{F}}_{\mathrm{C}6+}}{{\mathrm{F}}_{\mathrm{C}5+,\mathrm{b}\mathrm{e}\mathrm{f}\mathrm{o}\mathrm{r}\mathrm{e}}-{\mathrm{F}}_{\mathrm{C}5+,\mathrm{a}\mathrm{f}\mathrm{t}\mathrm{e}\mathrm{r}}}}\times 100\%$$

where F_C5+,before and F_C5+,after represent the moles of C₅₊ olefins before and after the reaction, respectively, while F_C6+ represents the moles of C₆₊ aldehydes after the reaction.

4. Conclusions

In this study, we investigated the catalytic conversion of syngas to C₅₊ aldehydes via hydroformylation, utilizing a series of Rh-based catalysts supported on different zeolite frameworks (MFI and MEL) and SiO₂. Our results demonstrate that the dispersion and accessibility of Rh species significantly influence the catalytic performance. The Rh@MFI and Rh@MEL catalysts, prepared via a one-pot hydrothermal crystallization process, exhibited higher selectivity for C₆₊ aldehydes (~50%), with Rh species being highly dispersed and confined within the zeolite frameworks. In contrast, catalysts prepared via impregnation, such as Rh/MFI, Rh/MEL, and Rh/SiO₂, displayed larger Rh nanoparticles (0.8–1.8 nm) exposed on the support surface, leading to a more accessible but less selective reaction pathway for C₅₊ olefins. The impact of olefin diffusion within the catalyst structure was a critical factor in determining the efficiency of the hydroformylation process. While zeolite-supported catalysts provided greater spatial confinement and enhanced selectivity for linear aldehyde formation, their performance was hindered by diffusion limitations, particularly in the case of MFI zeolite with a pore size of 0.55 nm. On the other hand, Rh/SiO₂ catalysts with a pore size of 5.6 nm, though exhibiting lower selectivity (~40%), showed higher olefin conversion (>90%) due to the more accessible mesoporous structure.

Moreover, this study explored the effect of FTS byproducts, including paraffins and short-chain olefins, on the hydroformylation reaction. Although the presence of long-chain paraffins had a negligible effect, the co-feeding of short-chain olefins, such as propene, reduced the catalytic efficiency of C₅₊ olefin conversion. This was attributed to competitive adsorption, where smaller olefins preferentially occupy active sites, thus hindering the conversion of longer olefins. Overall, this work highlights the importance of both catalyst design and reaction conditions in optimizing the hydroformylation of C₅₊ olefins. Further improvements in feedstock diffusion and catalyst accessibility, especially for zeolite-supported catalysts, could significantly enhance the tandem conversion of syngas to value-added aldehydes and alcohols. These findings provide valuable insights for the development of more efficient catalysts for syngas-based processes, paving the way for more sustainable and selective hydroformylation reactions.