CO₂ Oxidative Dehydrogenation of Propane to Olefin over Cr-M (M = Zr, La, Fe) Based Zeolite Catalyst

Abstract

CO₂ oxidative dehydrogenation of propane (CO₂-ODHP), being not only favorable for olefin production but also beneficial for CO₂ emission control, has recently attracted great attention. Here, a series of single metal (Cr) and bimetal (Zr, La, Fe) modified ZSM-5 zeolites were prepared via an impregnation method. It was found that the bimetal modified ZSM-5 possessed much higher C₃H₈ and CO₂ conversion than that of monometallic modified Cr_3%-ZSM-5 (Cr_3%-Z5), especially for Cr_3%Zr_2%-ZSM-5 (Cr_3%Zr_2%-Z5), which displayed the highest activity (65.4%) and olefin yield (1.65 × 10³ μmol·$g_{cat}^{-1}$ h⁻¹). Various characterizations were performed, including XRD, N₂ adsorption-desorption, H₂-TPR, Raman, XPS, HAAD-STEM, and TEM. It was revealed that Zr not only favored an improvement in the redox ability of Cr, but also contributed to the surface dispersion of loaded Cr species, constituting two major reasons explaining the superior activity of Cr_3%Zr_2%-Z5. To further improve CO₂-ODHP catalytic behavior, a series of Cr_3%-ZSM-5@SBA-15-n composite zeolite catalysts with diverse (ZSM-5/SBA-15) mass ratios were prepared (Cr_3%-ZS-n, n = 0.5, 2, 6, 16), which screened out an optimum mass ratio of six. Based on this, the Cr_3%Zr_2%-ZS-6 compound was further prepared, and it eventually achieved even higher CO₂-ODHP activity (76.9%) and olefin yield (1.72 × 10³ μmol·$g_{cat}^{-1}$ h⁻¹). Finally, the CO₂-ODHP reaction mechanism was further investigated using in situ FTIR, and it was found that the reaction followed the Mars–van Krevelen mechanism, wherein CO₂ participated in the reaction through generation of polydentate carbonates. The Cr⁶⁺ constituted as the active site, which was reduced to Cr³⁺ after the dihydrogen reaction, and was then further oxidized into Cr⁶⁺ by CO₂, forming polydentate carbonates, and thus cycling the reactive species Cr⁶⁺. Additionally, assisted by a Brönsted acid site (favoring breaking of the C-C bond), C₂H₄ and CH₄ were produced.

1. Introduction

Low-carbon olefins, especially ethylene and propylene, can be used as basic raw materials in the synthesis of polymeric materials and fine chemicals, which play an important role in the development of the country’s economy [1,2,3,4]. However, given the scarcity of oil resources and rising oil prices, the technology for producing olefins from oil needs further innovation. In recent years, the global demand for olefins has gradually increased [5]. Compared with the traditional olefin production process of cracking and splitting petroleum products, the use of low-carbon alkanes (e.g., propane) to produce olefins has the advantages of low production costs, high yields, and efficient use of resources. Olefin production from low-carbon alkanes can be divided into two types: direct alkane to olefin production, and alkane oxidation to olefin. The oxidative dehydrogenation of alkanes is an exothermic reaction that is not limited by thermodynamics [6,7]. By utilizing CO₂ as an oxidant, the conversion rate of alkane, as well as the selectivity of olefins, can be improved [4,8,9]. Moreover, it also favors the reduction of associated CO₂ emissions, which is a typical greenhouse gas that contributes to the global warming problem.

The oxidative dehydrogenation of alkanes (e.g., propane and ethane) by CO₂ (CO₂-ODHP) has been widely investigated, including through the use of Pt [10,11], Cr [12], Co [13], Ga [14,15], Mo [2,16], and B-based [17] catalysts. In comparison to other catalysts, Cr-based catalysts have the advantages of low cost and good reactivity [12]. However, problems such as carbon accumulation and easy accumulation of Cr particles occur during the catalytic reaction. The introduction of auxiliary metals into the catalyst and the selection of suitable supports to improve the dispersion of Cr species constitute two approaches to develop highly efficient Cr-based catalysts. For example, taking advantage of the large surface area and ordered pore structure of ZSM-5, with the pore size of 0.56 × 0.51 nm being larger than the molecule dynamics diameter (0.47 nm) of propane molecules, Wang et al. [18] investigated propane dehydrogenation over Cr loaded ZSM-5 (Cr-ZSM-5) zeolite, the Brönsted acid sites of which have also been reported to be favorable for the adsorption and reaction of propane. Igonina et al. [19] impregnated 5 wt.% Cr on MCM-41 using the first impregnation method, which resulted in 20% conversion of C₃H₈ and 76% selectivity of C₃H₆ at 650 °C. Singh et al. [20] used ZrO₂ as a support loaded with 2.5 wt.% Cr and achieved 16% conversion of C₃H₈ and 86.6% selectivity of C₃H₆ at 550 °C. Zhang et al. [21] introduced 5 wt.% Sn into an Al₂O₃ catalyst loaded with 3 wt.% Cr in order to improve the stability of the catalyst. The initial conversion of C₃H₈ in the catalyst was 40%, and after 10 h of reaction, the conversion of C₃H₈ was stable at 32%. Moreover, one type of composite molecular sieve of ZSM-5@SBA-15 was synthesized by adding ZSM-5 into the mother liquor of SBA-15 and was then applied for propane dehydrogenation [22,23], displaying superior catalytic activity in comparison to that of the single ZSM-5 molecular sieve.

Inspired by the above literature, a series of Cr-based ZSM-5 catalysts with different additives (Zr, La, Fe) were prepared by the impregnation method, with the aim of further improving the intrinsic CO₂-ODHP activity of Cr-ZSM-5. Additionally, in order to further improve the reaction performance, a series composite molecular sieve catalyst (Cr_3%-ZSM-5@SBA-15 and Cr_3%Zr_2%-ZSM-5@SAB-15) was further prepared based on the hydrothermal synthesis method, in which the specific mass ratios of ZSM-5 to SBA-15 (Cr_3%-ZS-n, n = 0.5, 2, 6, 16, representing the mass ratio) was detailed and adjusted, eventually achieving a best mass ratio of 6:1 and showing C₃H₈ conversion of 76.9% and an olefin yield of (1.72 × 10³ μmol·$g_{cat}^{-1}$·h⁻¹) from Cr_3%Zr_2%-ZS-6. The structure–activity relationship was further illustrated by a series of characterizations (XRD, N₂ adsorption-desorption, H₂-TPR, Raman, XPS, HAAD-STEM, and TEM). The present work has developed an efficient Cr_3%Zr_2%-Z5 catalyst for the CO₂-driven oxidative dehydrogenation of propane (ODHP). Moreover, a Cr_3%Zr_2%-ZS-6 composite zeolite catalyst with promoted ODHP activity has been further developed. The present work favors the development of other highly efficient CO₂-ODHP catalyst designs.

2. Results and Discussion

2.1. Characterization of Catalysts

X-ray diffraction (XRD) was used to characterize the crystal structure of the synthesized samples. Figure 1a displays the large angle XRD characterization results of the synthesized catalyst samples. As can be seen, both the [Cr_3%-Z5, Cr_3%M_2%-Z5 (M = La, Zr, Fe)] and composite catalyst samples (Cr_3%-ZS-n and Cr_3%Zr_2%-ZS-6) display the characteristic MFI crystal structure at the 2θ of 8.0°, 8.9°, and 23.7° [24,25,26]. The loaded metal species did not significantly affect the associated crystal structure during the synthesis process, and no obvious peaks of metal oxide species, such as CrO₃ or Cr₂O₃, were observed, probably due to their good dispersion over the zeolite samples [27]. Additionally, the diffraction peaks of MFI became sharp as the mass ratio of m_ZSM-5/m_SBA-15 increased for the Cr_3%-ZS-n composite. Figure 1b displays the small-angle XRD of the composite catalysts, where the characteristic diffraction patterns of SBA-15 at the 2θ of 0.8°, 1.5°, and 1.8°, corresponding to the crystalline facets of (100), (110), and (200), respectively, can be observed [27,28,29]. This finding indicates the successful synthesis of SBA-15. As noted, the intensity of the diffraction pattern of SBA-15 gradually decreased as the mass ratio of m_ZSM-5/m_SBA-15 increased. Figure 1c displays the N₂ adsorption-desorption curves of Cr_3%-ZS-6, which displayed a typical IV curve [23], further verifying the successful synthesis of composite zeolite catalysts with mesoporous structures. The specific surface areas of the synthesized samples were characterized using the N₂ adsorption-desorption method, as listed in

Table 1. Specific surface area and Cr⁶⁺ ratio of the prepared Cr-based zeolite catalysts.

Sample	SBET a, m2/g	Cr6+/(Cr6+ + Cr3+)
Cr3%-Z5	367.7	46.9%
Cr3%Zr2%-Z5	324.2	71.1%
Cr3%La2%-Z5	363.2	68.2%
Cr3%Fe2%-Z5	376.2	73.9%
Cr3%-ZS-0.5	456.2	—
Cr3%-ZS-2	469.8	—
Cr3%-ZS-6	505.7	—
Cr3%-ZS-16	403.0	—

^a S_BET: BET specific surface area.

. The addition of the second metallic element of M (M = Zr, La, Fe) did not seriously affect the specific surface area of Cr_3%M_2%-Z5 in comparison to that of Cr_3%-Z5 (367.7 m² g⁻¹). However, much higher surface areas can be achieved for the composite zeolite catalysts (Cr_3%-ZS-n) relative to those of Cr_3%-ZSM-5 and Cr_3%M_2%-Z5, especially for Cr_3%-ZS-6, which showed the highest value of 505.7 m²/g. This result indicates that encapsulation by SBA-15 can efficiently improve specific surface area, which would be favorable for the CO₂-ODHP reaction.

H₂-TPR was conducted to characterize the specific chemical states of the loaded Cr and M (Zr, La, Fe) species over the synthesized catalyst samples (see Figure 1d). For Cr_3%-Z5, the broad peak centered at T of 300–400 °C can be attributed to the reduction of CrO₃ species, and the weak reduction peak at T of 400–500 °C can be attributed to the reduction of Cr⁶⁺ [30,31]. As for the Cr_3%M_2%-Z5 samples, in addition to the abovementioned reduction peaks of CrO₃ and Cr⁶⁺, the additional reduction peaks of the (Zr, La, Fe) species could also be clearly identified. Specifically, for the Cr_3%Fe_2%-ZSM-5 sample, the reduction peak around 400 °C was related to the reduction of Fe₂O₃ to Fe₃O₄; the peak around 500 °C was attributed to the reduction of Fe₃O₄ to FeO; and the peak around 600 °C was ascribed to the reduction of FeO to Fe [32]. As for Cr_3%Zr_2%-Z5 and Cr_3%La_2%-Z5, the additional reduction peaks at T of 300–400 °C and 400–500 °C were related to the reductions of ZrO₂ and La₂O₃, respectively [33,34]. As noted above, after the introduction of Zr, the reduction peaks of Cr (CrO₃ → Cr₂O₃) were shifted to lower temperatures (450 → 290 °C), and also developed a larger reduction peak area. This was closely related to the fact that ZrO₂, which is a P-type semiconductor, formed a strong interaction with the active metal Cr, favoring the reduction of CrO₃, which is conducive to CO₂-ODHP. Similarly, in the Cr_3%La_2%-Z5 and Cr_3%Fe_2%-Z5 scenarios, it is evident that the presence of La and Fe additives also promoted the reduction of CrO₃ to Cr₂O₃, although to a lesser extent compared to Zr. This constitutes one of the major reasons leading to the lower CO₂-ODOH activity of Cr_3%Zr_2%-Z5 than that observed for Cr_3%La_2%-Z5 and Cr_3%Fe_2%-Z5 (as will be stated later). In addition to this, for the composite zeolite catalyst Cr_3%Zr_2%-ZS-6, it was found that, interestingly, the reduction peaks of CrO₃ and Cr⁶⁺ further shifted to lower temperatures (450 → 200 °C) due to encapsulation by SBA-15. This favored the dispersion of the Cr species. This is also one of the main reasons for the further improvement in the CO₂-ODH activity in this sample compared to that of Cr_3%Zr_2%-Z5.

Raman spectroscopy was further used to characterize the loaded Cr species over the synthesized samples, as shown in Figure 1e, where the peak centered at 550 cm⁻¹ can be attributed to the Cr-O vibration of Cr₂O₃ species over Cr_3%-Z5. In addition to the peak at 550 cm⁻¹, the Raman peaks corresponding to Cr₂O₇²⁻ (350–400 cm⁻¹ of Cr_3%Zr_2%-Z5), Cr₃O₁₀²⁻ (850 cm⁻¹ of Cr_3%La_2%-Z5) and Cr₄O₁₃²⁻ (900 cm⁻¹ of Cr_3%Zr_2%-Z5) appeared after the introduction of Zr and La [35]. Only one Raman peak of Cr₂O₇²⁻ (350–400 cm⁻¹) emerged over the composite Cr_3%-ZS-6. The above results indicate that the addition of the second metal, as well as encapsulation by SBA-15 (generating composite catalyst), could significantly affect the valence state of CrO_x. According to a previous report in the literature [36], the formed Cr₂O₇²⁻ and Cr₄O₁₃²⁻ species would be more active for the ODHP reaction relative to that of Cr₂O₃. Thus, the results there (Figure 1) give a further indication of the much higher CO₂-ODHP of Cr_3%Zr_2%-Z5 and Cr_3%-ZS-6 relative to that of Cr_3%-Z5. As noted above, no characteristic Raman peaks of CrO_x species could be found over Cr_3%Fe_2%-Z5, which was probably related to the tendency of FeO_x to react with oxygen in air to form an oxide film, which eventually interferes with Raman spectroscopic analysis.

The valence states of the active metal Cr over the synthesized samples were further analyzed by XPS (Figure 1f), with the derived Cr⁶⁺/(Cr⁶⁺+Cr³⁺) ratios being listed in Table 1. As can be seen, the Cr 2p1/2 for the investigated samples could be split into three peaks, of which the characteristic peaks with electron binding energies at 575.0 eV were attributed to Cr³⁺, and the two other peaks with electron binding energies of 577.2 and 579.0 eV were attributed to Cr⁶⁺ [37]. As listed in Table 1, the Cr⁶⁺/(Cr⁶⁺+Cr³⁺) ratios of bimetallic modified ZSM-5 samples (68.2–73.9%) were much higher than that of monometallic Cr_3%/ZSM-5 (46.9%), indicating that the second metal additives favored the formation of Cr⁶⁺ species, which have much higher activity than that of Cr³⁺. In this regard, the addition of the second metal promoted the Cr⁶⁺ ratio relative to the single-metal-promoted Cr_3%-Z5, which is another important reason leading to the much higher CO₂-ODHP activity. As noted, no XPS signal of Cr species was observed over Cr_3%-ZS-6, which could be related to the encapsulation of Cr_3%-ZSM-5 by SBA-15, leading to the Cr being undetectable.

Figure 2 shows the HAADF-STEM, HRTEM, and EDS-mapping results of Cr_3%-Z5 (Figure 2a–c), Cr_3%Zr_2%-Z5 (Figure 2d–f) and Cr_3%Zr_2%-ZS-6 (Figure 2g–i), which provide information on the crystalline structure and surface distribution of the active metal Cr. As compared in Figure 2a,d,g, the Cr agglomerates can be clearly observed over Cr_3%-Z5; however, they cannot be seen over the samples of Cr_3%Zr_2%-Z5 and Cr_3%Zr_2%-ZS-6. This finding is well supported by the EDS-mapping results shown in Figure 2c,f,i, displaying the aggregated Cr nanoparticles over Cr_3%-Z5 and the monodispersed Cr over Cr_3%Zr_2%-Z5 and Cr_3%Zr_2%-ZS-6. This finding gives us a clue that the addition of Zr (Figure S1), as well as encapsulation by SBA-15, could favor the dispersion of loaded Cr species. In addition, the obvious lattice stripes at 0.375 nm belonging to the (151) plane of ZSM-5 [38] can be clearly observed for the investigated samples (Figure 2b,e,h), and the regular pore structure of SBA-15 [29], with a clear interface between SBA-15 and ZSM-5, can also be observed in Figure 2h, verifying the successful synthesis of the Cr_3%Zr_2%-ZS-6 composite.

2.2. Reaction Mechanism Investigation

In situ FTIR was employed to investigate the CO₂-ODHP reaction mechanism over the best-performing sample of Cr_3%Zr_2%-ZS-6 at T of 250–500 °C, as shown in Figure 3. Several intermediate species with IR vibration peaks centered at 1700–1350 cm⁻¹ were identified, including the methyl radical [CH₃*] in the range of 1400–1370 cm⁻¹ [34], polydentate carbonate species around 1468 cm⁻¹ [39], H₂O around 1640 cm⁻¹, and olefins with the telescopic vibrational peak of C=C around 1650 cm⁻¹ [40]. As noted, the other peaks in the range of 3600–3750 cm⁻¹ belonged to hydroxyl species [OH*], and the 1501 cm⁻¹ peak was attributed to be the telescopic vibrational peak of the Brönsted acid site [41]. Based on the above derived intermediate information, as well as reports from the literature [36], it can be proposed that CO₂-ODHP would follow the Mars–van Krevelen mechanism [42]. First, CO₂ is adsorbed on the support at a low temperature (T of 250 °C), forming polydentate carbonates (1468 cm⁻¹). The reaction from propane to olefin then begins with an increase in temperature (T > 300 °C), where the polydentate carbonates participate in the reaction with a gradual decrease in the IR signal (1470 cm⁻¹) and a simultaneous gradual increase in the IR signal of olefin (v_C=C of 1649 cm⁻¹) and H₂O (v_H2O of 1625 cm⁻¹). As reported [43], Cr⁶⁺ is the activation site for the propane dehydrogenation reaction, and the C-H bond broken to produce radicals of C₃H₇* (1370–1400 cm⁻¹, Figure 3) and H* is reported to be the rate-determining step (RDS). In addition to C₃H₆, C₂H₄ and CH₄ were also observed during activity measurement. This can be attributed to the strong C-C bond breaking effect of the Brönsted acid site (around 1540 cm⁻¹) of the ZSM-5 support. The interaction of C₃H₈ with a Brönsted acid site leads to the breaking of the C-C bond, generating radicals of ethyl (C₂H₅*) and (CH₃*) methyl (1400–1370 cm⁻¹, Figure 3). Subsequently, the broken C-H bond of C₂H₅* leads to the formation of C₂H₄ and the dissociated proton H*, which further interacts with the radical of CH₃* to produce CH₄. Based on the above discussion, we can conclude that the Cr⁶⁺ is the activation site for propane dehydrogenation, where CO₂ participates in the reaction as polydentate carbonates. Assisted by the Brönsted acid site (favoring C-C bond breaking), C₂H₄ and CH₄ are produced. As noted, after the dehydrogenation reaction, Cr⁶⁺ is reduced to Cr³⁺, which is then further oxidized to Cr⁶⁺ by CO₂, forming polydentate carbonates and thus cycling the reactive species Cr⁶⁺. The CO₂/C₃H₈ adsorption over ZrO2 [111] and CrO3 [110] was further simulated by DFT to further illustrate the role played by Zr during CO₂-ODHP. The results are shown in Tables S1 and S2. As can be seen, the adsorption energies of CO₂ and C₃H₈ on the ZrO₂ were −0.14 eV and −0.03 eV, and those of CrO₃ were −0.09 and −0.06 eV, respectively. Obviously, the ZrO₂ exhibited a much lower CO₂ adsorption energy (−0.14 eV) than that of CrO₃ (−0.09 eV). This finding indicates that the introduced Zr also functioned as the active site during CO₂-ODHP, favoring CO₂ adsorption over the Cr-based ZSM-5 zeolite.

2.3. Catalytic Activity of Cr-Based Catalysts

Figure 4a,b display the propane conversion, CO₂ conversion, olefin selectivity, and olefin yield of the synthesized samples of Cr_3%M_2%-Z5 (M = Zr, La, Fe), where the Cr loading was 3 wt.% and the additive metal loading was 2 wt.%. The proportion of each product (C₃H₆, C₂H₄, C₂H₆, CH₄) is depicted in Figure S2a. As can be seen, adding additional metals, such as Zr, La, and Fe, increased C₃H₈ conversion and olefin yield; this was accompanied by a slight decrease in olefin selectivity (62.9 → 58.6%). Among the samples, Cr_3%Zr_2%-Z5 displayed the best CO₂-ODHP catalytic behavior, achieving C₃H₈ conversion of 65.4% and an olefin yield of 1.65 × 10³ μmol·$g_{cat}^{-1}$·h⁻¹. Figure 4c,d display the activity measurement results for the composite catalysts of Cr_3%-ZS-n (n = 0.5, 2, 6, 16) and Cr_3%Zr_2%-ZS-6, wherein the ZSM-5: SBA-15 ratio of 6:1 led to much higher C₃H₈ conversion (73.7%) and olefin yield (1.69 × 10³ μmol·$g_{cat}^{-1}$·h⁻¹) than those of the other catalysts. The distribution of each element (C₃H₆, C₂H₄, C₂H₆, CH₄) in the product is shown in Figure S2b, showing a slight decrease in olefin selectivity up to 52.0%. Moreover, the C₃H₈ conversion and olefin yield was further improved to 76.9% and 1.72 × 10³ μmol·$g_{cat}^{-1}$·h⁻¹, respectively, over the bimetallic composite molecular sieve of Cr_3%Zr_2%-ZS-6 (mass ratio of 6.0) due to the superior intrinsic CO₂-ODHP activity of Cr_3%Zr_2%-Z5 compared to that of Cr_3%-Z5 (Figure 4a,b). It can be seen from Figure S3 that the activity of the catalysts (Cr_3%-Z5, Cr_3%Zr_2%-Z5, and Cr_3%Zr_2%-ZS-6) decreased slightly after 180 min of reaction time.

3. Material and Methods

3.1. Preparation of Cr-Based ZSM-5 Zeolite Catalyst

Approximately 0.476 g of Cr(NO₃)₃·9H₂O and a certain amount of nitrate compounds of the additive metals were weighed into a 250 mL tomato-shaped bottle. Then, 50 mL of deionized water and 2 g of ZSM-5 carrier with an Si:Al ratio of 50 were added and stirred in a water bath at 90 °C for 4 h. The tomato-shaped bottle containing the sample was mounted on a spinner to remove water, and was thereafter placed in an oven at 60 °C and dried overnight. Finally, the sample was calcined in a muffle furnace at 550 °C under air atmosphere for 6 h to obtain the Cr_3%-Z5 and Cr_3%M_2%-Z5 (M = Zr, La, Fe) samples.

3.2. Preparation of Cr-Based ZSM-5@SBA-15 Composite Zeolite Catalyst

SBA-15 was prepared by using triblock copolymer P123 as the structure-directing agent and tetraethyl silicate TEOs as the silicon source. First, 0.326 g of P123 was dissolved in 50 mL of 1.6 mol/L hydrochloric acid and stirred for 15 min in a water bath at 40 °C. Then, 0.75 mL of tetraethyl silicate was added and stirred continuously for 40 min to form the mother solution of SBA-15. A certain amount of Cr(NO₃)₃·9H₂O and ZSM-5 molecular sieves (Si:Al = 50) were added to the mother solution of SBA-15 according to the mass ratios of ZSM-5 to SBA-15 of 0.5, 2, 6, and 16, respectively, and stirred for 24 h. After that, the samples were placed in a crystallization tank at 100 °C for 24 h. Then, after washing three times with water and alcohol, centrifugation, drying (60 °C for 12 h), and calcination (550 °C for 6 h), the Cr_3%-ZS-n (n = 0.5, 2, 6, 16) was obtained. A similar procedure was also applied during Cr_3%Zr_2%-ZS-6 preparation, utilizing Zr(NO₃)₃·5H₂O as the Zr precursor.

3.3. Characterizations

X-ray diffraction (XRD) was performed on a D8 Advance instrument (Karlsruhe, Germany) to analyze the crystalline phase structure of the catalyst samples using a Cu Kα radiation source in the range of 5–60°. Specific surface area analysis (BET) was carried out on an Autosorb iQ (Boynton Beach, FL, USA) adsorption apparatus to analyze the specific surface area of the catalyst samples, and N₂ adsorption and desorption curves were obtained using an Autosorb iQ adsorption apparatus. The samples were degassed at 300 °C for 6 h. H₂ programmed temperature rising reduction (H₂-TPR) was carried out on a fully automated multicurrent apparatus, the tp-5080-B (Tianjin, Beijing), and was used to analyze the composition of the metal oxide species of the samples. A 0.1 g sample was first pretreated at 100 °C for 10 min, then cooled to room temperature to level the baseline, and finally heated from 100 to 900 °C at a rate of 10 °C/min. Raman spectroscopy (Raman) was performed on a LabRam HR Evolution Raman spectrometer (Paris, France) in the range of 100 to 1200 cm⁻¹ with a laser wavelength of 532 nm. X-ray photoelectron spectroscopy (XPS) was performed on a Thermo Scientific K-Alpha X-ray photoelectron spectrometer, Waltham, MA, USA, to analyze the material composition of the samples. Transmission electron microscopy (TEM) was performed using a JEOL JEM-F200 transmission electron microscope (Tokyo, Japan) to observe the structural compositions of the catalyst samples, which were first sonicated in ethanol for 60 min and then added drop-wise to a copper grid to prepare the samples for measurement. In situ Fourier infrared spectroscopy (FTIR) was carried out using a BRUKER (Billerica, MA, USA) TENSOR II infrared spectrometer and was used to measure the functional groups and chemical compositions etc. produced during the reaction. Infrared variable temperature experiments were conducted, in which the samples to be measured were pressed and first pretreated with H₂ at 500 °C for 60 min, before being cooled to room temperature. Then, a mixed gas comprising 5% C₃H₈ and 5% CO₂, balanced by He (total flow rate of 20 mL/min), was introduced into the sample under temperatures of 250, 300, 350, 400, 450, and 500 °C for 30 min, with the IR signal being collected each minute.

3.4. Catalytic Test

The CO₂-ODHP activity measurement was carried out in a continuous flow fixed bed reactor under atmospheric pressure and T of 600 °C. The volume ratio of the feed gas was set to be C₃H₈:CO₂:He = 2:2:16. The gas hourly velocity (GHSV) was 12,000 h⁻¹. As noted, before the reaction, 0.1 g of catalyst (40–60 mesh) was initially pretreated using 5% H₂ in He for 60 min at 600 °C. The composition of the reacted gas, as well as the product, were analyzed on-line using an SP-7890 gas chromatograph with a TCD detector (Shimadzu, Kyoto, Japan). The propane conversion, CO₂ conversion, olefin selectivity, and olefin yield were calculated by Equations (1)–(4), as shown below:

$${\mathrm{X}}_{{\mathrm{C}}_3{\mathrm{H}}_8}=\frac{{\mathrm{C}}_3{\mathrm{H}}_8\left(\mathrm{i}\mathrm{n}\right)-{\mathrm{C}}_3{\mathrm{H}}_8\left(\mathrm{o}\mathrm{u}\mathrm{t}\right)}{{\mathrm{C}}_3{\mathrm{H}}_8\left(\mathrm{i}\mathrm{n}\right)}\times 100\%$$

(1)

$${\mathrm{X}}_{\mathrm{C}{\mathrm{O}}_2}=\frac{\mathrm{C}{\mathrm{O}}_2\left(\mathrm{i}\mathrm{n}\right)-\mathrm{C}{\mathrm{O}}_2\left(\mathrm{o}\mathrm{u}\mathrm{t}\right)}{\mathrm{C}{\mathrm{O}}_2\left(\mathrm{i}\mathrm{n}\right)}\times 100\%$$

(2)

$${\mathrm{S}}_{\mathrm{o}\mathrm{l}\mathrm{e}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{s}}=\frac{3{\mathrm{C}}_3{\mathrm{H}}_6+2{\mathrm{C}}_2{\mathrm{H}}_4}{3{\mathrm{C}}_3{\mathrm{H}}_6+2{\mathrm{C}}_2{\mathrm{H}}_6+2{\mathrm{C}}_2{\mathrm{H}}_4+\mathrm{C}{\mathrm{H}}_4}\times 100\%$$

(3)

$$\mathrm{Y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d} \mathrm{o}\mathrm{f} \mathrm{o}\mathrm{l}\mathrm{e}\mathrm{f}\mathrm{i}\mathrm{n}\mathrm{s}=\frac{∆PPm\times Fin\times 60\times 273\times {10}^4}{T\times M_{cat}\times 22.4\times {10}^3}$$

(4)

where F_in represents the flow rate of gases introduced (C₃H₈ and CO₂); ΔPPm represents the increase in propylene and ethylene; T represents reaction temperature; and M_cat represents the mass of catalyst.

4. Conclusions

In conclusion, a series of Cr-based zeolite catalysts were prepared for CO₂-ODHP with the aim of providing a type of candidate catalyst that not only favors propane conversion to olefin, but also contributes to CO₂ emission control. It was found that the addition of a second metal (Zr, La, or Fe) enhanced CO₂-ODHP activity relative to that of single metal Cr promoted ZSM-5 (Cr_3%-Z5), with Zr exhibiting the most pronounced effect. Characterizations using H₂-TPR, XPS, Raman spectroscopy, and HAAD-STEM suggested that the addition of Zr improved the redox ability of Cr, increased the Cr⁶⁺ species ratio (which acted as active sites for the reaction), and favored the dispersion of Cr species over the ZSM-5 support, which together constituted three major factors leading to the superior CO₂-ODHP activity of Cr_3%Zr_2%-Z5. In addition, one zeolite composite, in which Cr_3%Zr_2%-Z5 was encapsulated by SBA-15 (m_Cr3%-ZSM-5/m_SBA-15 mass ratio of 6), was prepared using the hydrothermal synthesis method, which further promoted CO₂-ODHP activity due to the large surface area of SBA-15 favoring the dispersion of active Cr species. Finally, the CO₂-ODHP reaction mechanism was investigated using in situ FTIR. It was found that the reaction followed the Mars–van Krevelen mechanism, wherein the CO₂ was initially adsorbed over the Cr_3%Zr_2%-ZS-6, forming polydentate carbonates, nand then participated in the reaction through reacting with dissociated H* from C₃H₈ to produce H₂O and C₃H₆. Additionally, assisted by the Brönsted acid site (favoring C-C bond breaking), C₂H₄ and CH₄ were generated. In general, the present work investigated CO₂-ODHP over a series of Cr-based zeolite catalysts, based on which one Cr_3%Zr_2%-ZS-6 candidate composite catalyst was screened out. The present work favors the development of CO₂-ODHP techniques.