Effect of Ce Content on the Chemical Looping Oxidative Dehydrogenation of Propane to Propylene over a VO_x-CeO₂/γ-Al₂O₃ Oxygen Carrier

Abstract

The chemical looping oxidative dehydrogenation of propane to propylene (CL-ODHP) replaces molecular oxygen with lattice oxygen (O_latt) in oxygen carriers. This method boosts propylene selectivity by avoiding the deep oxidation of propane. Herein, a series of 10V-XCe/Al oxygen carriers with different Ce contents were prepared to realize different VO_x-CeO_y interactions. The effect of the Ce content in 10V-XCe/Al oxygen carriers on the CL-ODHP reaction was studied and the optimal Ce content was determined. CeO₂ prevents the outward diffusion and evolution of O_latt in VO_x carriers to the adsorbed electrophilic oxygen species (O_elec), effectively inhibiting the loss of O_latt, improving the selectivity of propylene, and extending the lifetime and activity of the oxygen carriers. After characterizing and analyzing the oxygen carriers, it was found that 10V-3Ce/Al has the highest specific surface area, highest oxygen capacity, and lowest reducibility. The 10V-3Ce/Al also delivers the highest oxidative dehydrogenation performance. At 550 °C, the average propylene and CO_x selectivity values of 10V-3Ce/Al were 81.87% and 7.28%, respectively (vs. 62.79% and 25.64% respectively, for 10V/Al). It is demonstrated that 10V-3Ce/Al exhibits good cycle stability with no significant decrease in catalytic performance after 15 cycles. In situ diffuse-reflectance infrared Fourier-transform spectroscopy indicates that CL-ODHP on 10V-3Ce/Al undergoes the Mars-van Krevelen mechanism. The migration and evolution of O_latt in oxygen carriers is controlled by reasonably modifying the metal oxide interactions to improve propylene yield. This work will thus guide the subsequent development of novel and efficient CL-ODHP oxygen carriers.

1. Introduction

Propylene is an important raw material for petrochemicals, and is mainly used to produce chemical materials such as polypropylene, propylene oxide, and acrylonitrile. The demand for propylene has increased with development in the downstream market. The current development in the shale gas industry has increased the availability of low-cost propane, making propylene production from propane appealing [1,2].

The industrial process, direct dehydrogenation of propane to propylene (DHP), is a strong heat-absorption process with restricted thermodynamic equilibrium. Effective propane conversion requires a high temperature, which increases the risk of problems such as catalyst carbon build-up. These inherent drawbacks limit the performance of catalytic dehydrogenation of propane [3,4]. In contrast, oxidative dehydrogenation of propane to propylene (ODHP) is an exothermic reaction not limited by thermodynamics. As ODHP can theoretically achieve high conversion rates at lower temperatures than DHP without carbon accumulation, it has attracted extensive research interest. However, in this process, O₂ deeply oxidizes propane to CO_x, which reduces propylene production. The widespread adoption of ODHP is further limited by the requirement for expensive air-separation equipment. The safety of co-feeding propane and O₂ raises additional concerns [5,6].

The shortcomings of propylene production technologies can be overcome by a new propylene production route called chemical looping oxidative dehydrogenation of propane to propylene (CL-ODHP), which includes oxidation and reduction reactions using oxygen carriers in two reactors. Propane is oxidized to propylene and water by lattice oxygens (O_latt) in the oxygen carrier. The water is separated from the reaction system via condensation, driving the reaction equilibrium in the positive direction and thus increasing the propylene yield. The reduced oxygen carrier enters the air reactor, where it is oxidized and regenerated before entering the dehydrogenation reactor to complete the oxidation and dehydrogenation cycles [7,8]. The CL-ODHP process confers the following advantages over other propylene production processes: (a) the oxygen carrier avoids direct contact between propane and O₂, thereby eliminating explosion risks; (b) the process does not require an air separation device, reducing the investment and operating costs; (c) the reaction occurs without O₂, improving the operability and temperature control; and (d) hydrogen combustion is indirect and flameless, thus reducing CO_x emissions.

However, the CL-ODHP process requires suitable oxygen carriers. Metal oxides have been reported as suitable oxygen carriers for chemical looping processes. Liu et al. [9] found that polymetallic oxide oxygen carriers have high CL-ODHP activity, and by regulating the ratio between metal oxides, the polymetallic oxide oxygen carriers can be made to release O_latt slowly, thus improving the selectivity of low-carbon olefins. After studying the CL-ODHP properties of different metal oxides (MoO_x/γ-Al₂O₃, GaO_x/γ-Al₂O₃, and VO_x/γ-Al₂O₃) as oxygen carriers, Wu et al. [10] found that VO_x/γ-Al₂O₃ oxygen carriers can conduct CL-ODHP reactions at a lower temperature and have higher CL-ODHP activity. Although VO_x/γ-Al₂O₃ has subsequently attracted wide research interest, Jiang et al. [11] found that O_latt in VO_x/γ-Al₂O₃ oxygen carriers are rapidly released and propane is deeply oxidized to CO_x, causing propylene selectivity to deteriorate. This selectivity can be improved by a metal-oxide promoter that regulates the O_latt release rate of the VO_x/γ-Al₂O₃ oxygen carrier. According to previous research, CeO₂ has excellent oxygen storage and release characteristics as well as redox properties, which can change the O_latt-transfer ability of adjacent metal oxides [12,13]. Hedun et al. [14] prepared FeCeO_x as an oxygen carrier via the impregnation method. They reported that CeO₂ inhibits the conversion rate of O_latt to O_elec in Fe-based oxygen carriers, thereby improving propylene selectivity. To similarly regulate the slow release of O_latt and improve propylene selectivity, we add CeO₂ to VO_x/γ-Al₂O₃ oxygen carriers and investigate the synergistic effect of VO_x and CeO₂ oxygen carriers on the CL-ODHP reaction. We also add bimetallic oxides to improve propylene selectivity. CeO₂ increases the oxygen capacity of the carrier and controls the O_latt-release rate in the VO_x/γ-Al₂O₃ oxygen carrier. To determine the optimum Ce content for increasing the propylene yield, we prepared a series of VO_x (10 wt.%)-CeO₂ (X wt.%)/γ-Al₂O₃ oxygen carriers (X = 0, 1, 3, 5, 7) for the CL-ODHP reaction and investigated the effect of Ce content on the physicochemical properties of the oxygen carriers. These properties were then related to the reactivity performance of CL-ODHP.

2. Oxygen Carrier Characterization

2.1. XRD Characterization Results

Figure 1 shows XRD profiles of the fresh 10V/Al, 10V-1Ce/Al, 10V-3Ce/Al, 10V-5Ce/Al, and 10V-7Ce/Al oxygen carriers. The diffraction peaks at 2θ = 45.56° and 67.03° are assigned to mesoporous γ-Al₂O₃ (JCPDS 10-0425) [15]. With increasing Ce content in the oxygen carrier, the peak intensity corresponding to the CeO₂ phase becomes sharper and more intense. The diffraction peaks at 2θ = 28.89°, 32.38°, and 47.85° characterize the cubic crystal structure of the CeO₂ phase, indicating that CeO₂ exists in the cubic fluorite phase. The diffraction peaks of VO_x species are absent, possibly because these species are highly dispersed on the surface of the γ-Al₂O₃ carrier, rendering them undetectable by the XRD instruments [16].

2.2. Analysis of Oxygen Carrier Morphology

Figure 2a–d shows the morphological characteristics of the oxygen carriers detected via SEM. All oxygen carriers appear as spherical particles with numerous surface folds. The surface smoothness of the 10V-XCe/Al oxygen carriers increases with increasing Ce content from 0 to 3 wt.%. When the Ce content reaches 7 wt.% (10V-7Ce/Al), the surface is rendered uneven by particle agglomeration.

The distributions of the Al, V, O, and Ce elements were investigated through mapping energy spectral analysis. Figure 2e,f shows the mapping images of the 10V/Al and 10V-3Ce/Al oxygen carriers, respectively. Each metal can be distinguished in the maps. Comparison of Figure 2e-III and Figure 2f-III shows that Ce introduction improves the uniformity of the V species through the sample. Moreover, Figure 2(e-II,f-II) shows that the introduction of Ce resulted in a denser distribution of O species through the samples, indicating that Ce increases the O content. As shown in Figure 2(f-IV), the Ce species is uniformly distributed on the surface of 10V-3Ce/Al.

Figure 3 shows the TEM of a fresh 10V-3Ce/Al oxygen carrier, where the polycrystalline material consists of small randomly oriented crystal structure domains. The lattice stripe spacing d = 0.31 nm and 0.26 nm correspond to the (111) crystal plane and the (200) crystal plane of CeO₂, respectively [17]. The lattice stripe spacing d = 0.28 nm corresponds to the (301) crystal planes of V₂O₅ [18]. By Figure 2 and Figure 3, it is shown that CeO₂ and V₂O₅ are uniformly dispersed on the surface of the oxygen carrier.

2.3. Pore-Structure Analysis of the Oxygen Carriers

Figure 4 shows the isothermal sorption and desorption curves of the oxygen carriers. The sorption and desorption curves of γ-Al₂O₃ and all oxygen carriers are type IV and exhibit H4-type hysteresis loops with low-pressure adsorption (P/P₀ = 0–0.10, where P and P₀ denote that P is the true pressure of the gas and P₀ is the saturated vapor pressure of the gas at the measured temperature, respectively), indicating that the oxygen carriers contain few micropores. When P/P₀ = 0.40, capillary coalescence occurs, resulting in a rapid increase in the adsorption amount [19].

Table 1. Structural properties of γ-Al₂O₃ and the 10V-XCe/Al oxygen carriers.

Samples	SBET (m2/g)	Vp (cm3/g)	Dp (nm)
γ-Al2O3	206.87	0.25	3.25
10V-Al	119.95	0.18	4.64
10V-1Ce/Al	124.91	0.19	4.95
10V-3Ce/Al	131.10	0.21	5.38
10V-5Ce/Al	104.29	0.16	4.30
10V-7Ce/Al	90.13	0.11	3.75

summarizes the pore-structure parameters of the oxygen carriers. γ-Al₂O₃ exhibits the highest specific surface area (206.87 m²/g), implying that it enhances the oxygen carrier’s performance most favorably. The specific surface area decreases to 119.95 m²/g after impregnating the VO_x species. In the 10V-XCe/Al oxygen carriers, the pore structure parameters first increase and then decrease with increasing Ce content. The specific surface area, pore volume, and pore diameter are maximized in 10V-3Ce/Al (131.10 m²/g, 0.21 cm³/g, and 5.38 nm, respectively). These results indicate that adding a small amount of Ce to the 10V/Al oxygen carrier increases the specific surface area of the oxygen carrier, allowing more uniform dispersion of the active component on the carrier surface. Further, the enlarged pore size benefits propylene desorption [20].

2.4. Thermal Stabilities of the Oxygen Carriers

Figure 5 plots the TGA and differential thermogravimetric (DTG) curves of the 10V-XCe/Al samples (X = 0, 1, 3, 5, 7) as the temperature was increased from 50 °C to 800 °C in inert nitrogen gas. The DTG curves (Figure 5b) present a weight-loss peak at 90–120 °C, which probably represents the evaporation of physical water molecules from the sample surfaces [21]. The oxygen carrier with the highest cerium content (7 Ce) shows the least weight loss (Figure 5). The thermal stability of the oxygen carriers increases with increasing Ce loading, probably because the CeO₂ phase inhibits γ-Al₂O₃ sintering and phase transition at high temperatures [22].

2.5. Reduction Properties of the Oxygen Carriers

The reducibility and oxygen-carrying capacities of the oxygen carriers were characterized using H₂-TPR. The H₂-TPR curves of all oxygen carriers are shown in Figure 6a. The 3Ce/Al oxygen carrier shows a reduction peak near 450 °C, which is attributed to the reduction of Ce⁴⁺, and the reduction peak of CeO₂ phase overlaps with the reduction peak of VO_x species [23]. The 10V/Al oxygen carrier shows an initial large reduction peak near 450 °C, attributed to the reduction of amorphous monomer or polymeric VO_x species phases with oxidation states between V⁵⁺ and V⁴⁺ [24]. Therefore, in the temperature range of 400-550 °C, the H₂ absorption is equal to the sum of H₂ consumed during the reduction of the VO_x phase and the CeO₂ phase. Increasing the Ce loading increasingly shifts the position of the reduction peak from that of 10V/Al. Therefore, CeO₂ changes the reducibility of the oxygen carrier. Moreover, the reducibility decreases gradually with temperature from left to right and its strength decreases in the order of 10V/Al > 10V-7Ce/Al > 10V-5Ce/Al > 10V-1Ce/Al > 10V-3Ce/Al. Gu [25] and Leon [26], who investigated the relationship between oxygen carrier reducibility and propylene selectivity, reported that lower reducibility usually corresponds to higher propylene selectivity.

The oxygen loadings in the oxygen carriers were calculated from the hydrogen consumptions of the carriers. Although the oxygen carrier conducts the H₂-reduction process on its surface, the consumed O_latt is supplied from its interior [24]. As shown in Figure 6b, increasing the Ce content does not monotonically increase the amount of O_latt stored in the carrier; the oxygen loading is maximized in the carrier containing 3 wt.% Ce.

2.6. XPS Elemental Valence Analysis of the Oxygen Carriers

Figure 7a shows the Ce 3d XPS spectra of the oxygen carriers. Due to the splitting of the Ce 3d_3/2 and Ce 3d_5/2 main peaks and the different final states after relaxation, the XPS spectra of Ce species can be deconvoluted into six peaks. Peaks at 904.03 and 885.22 eV were attributed to Ce³⁺, and peaks at 907.04, 900.72, 898.17, and 881.74 eV were attributed to Ce⁴⁺ [27]. The Ce⁴⁺ species improve the electron-transfer channel and optimize the activity of the oxygen-carrier O_latt, thereby facilitating the dehydrogenation reaction [17]. Increasing the Ce loading in the oxygen carrier increases the total area of the Ce 3d peak. The Ce⁴⁺ to (Ce⁴⁺ + Ce³⁺) ratio is largest (57.32%) in 10V-3Ce/Al whereas the percentage of Ce⁴⁺ is smallest (37.64%) in 10V-7Ce/Al. The change in Ce⁴⁺/(Ce⁴⁺ + Ce³⁺) ratio follows the oxygen-loading trend of the oxygen carriers (Figure 6c). However, O_latt is present only in CeO₂ and is absent in Ce³⁺ species [28]. When the mass fraction of Ce content in the oxygen carrier exceeds 3 wt.%, the Ce³⁺ content increases while the Ce⁴⁺ content decreases.

Figure 7b shows the XPS pattern of O 1s of fresh oxygen carriers, the peak at 530.02 eV corresponds to the adsorbed hydroxyl oxygen species (O_OH), the peak at 531.17 eV corresponds to the lattice oxygen species (O_latt), and the peak at 532.39 eV corresponds to the electrophilic oxygen species (O_elec) [29]. As shown in

Table 2. Relative content of oxygen species on the surface of fresh oxygen carriers.

Oxygen Carrier	Olatt/(OOH + Olatt + Oelec)	Oelec/(OOH + Olatt + Oelec)
10V/Al	31.45%	27.78%
10V-1Ce/Al	37.63%	22.01%
10V-3Ce/Al	45.38%	19.03%
10V-5Ce/Al	35.61%	23.51%
10V-7Ce/Al	33.77%	26.08%

, the content of the oxygen carrier O_latt increases significantly with increasing Ce loading, reaching a maximum of 10V-3Ce/Al. More O_latt promotes the selective oxidation of propane to propylene, thus improving propylene selectivity. The content of O_elec decreases with increasing Ce loading, and the highest concentration of Ce⁴⁺ is found in 10V-3Ce/Al oxygen carriers, where Ce⁴⁺ inhibits the conversion of O_latt to O_elec on the surface of the oxygen carriers, thus storing more O_latt for the carriers [27]. Figure 7e shows the O 1s XPS spectra before and after the reaction of 10V-3Ce/Al oxygenates, and the O_latt content of the oxygenate decreased significantly after the reaction. However, the rate of conversion of O_latt to O_elec was inhibited by the presence of Ce⁴⁺, and the oxidative dehydrogenation of propane also consumed O_elec, so the content of O_elec decreased after the reaction.

Figure 7c compares the XPS spectra of V 2p before and after the 10V-3Ce/Al reaction. The peaks at 515.35, 516.65, and 517.64 eV belong to V³⁺, V⁴⁺, and V⁵⁺ species, respectively [30]. In the spectrum of fresh 10V-3Ce/Al, the V³⁺ peak is absent because this species is oxidized during calcination in air. As shown in

Table 3. Valence changes in the V element before and after the 10V-3Ce/Al reaction.

The Valence State of Element V	V5+	V4+	V3+
Before reaction	41.92%	58.08%	0
After Reaction	21.04%	37.76%	41.20

, the reaction significantly changes the content of V⁵⁺ species (the active species of ODHP [31]) in the oxygen carriers. After 15 min of reaction, 20.88% of the V⁵⁺ was completely converted to V³⁺. Figure 7d compares the Ce 3d XPS spectra before and after the 10V-3Ce/Al reaction. Note that O_latt exists in CeO₂ [28]. As shown in

Table 4. Valence changes in the Ce element before and after the 10V-3Ce/Al reaction.

The Valence State of Element Ce	Ce4+	Ce3+
Before reaction	63.27%	36.73%
After reaction	54.96%	45.04%

, the Ce⁴⁺ content decreased from 63.27% before the reaction to 54.96% after the reaction; concomitantly, the Ce³⁺ content increased from 36.73% to 45.04% after the reaction. As the O_latt are slowly released, Ce⁴⁺ is not completely transformed into Ce³⁺. The slow release of O_latt accompanies the valence change in the Ce and V elements. The whole reaction is an oxidation-reduction process, in which the O_latt released by the oxygen carrier catalyzes the oxidation of propane to propylene.

3. CL-ODHP Performance of the Oxygen Carrier

3.1. In Situ DRIFT Investigation of the 10V-3Ce/Al Oxygen Carrier

To further elucidate the catalytic oxidation mechanism, the 10V-3Ce/Al oxygen carrier was subjected to in situ DRIFT experiments in an N₂:C₃H₈ (v/v = 45:5) atmosphere. The vibration peaks at 1355, 1412, 1662, 2963, and 3189 cm⁻¹ in the DRIFT spectra (Figure 8) are assigned to C-H bond stretching, C=O bond stretching, C=C bond stretching, variant Ce-O-V bonds, and V-OH bond stretching, respectively [7]. The peak at 1662 cm⁻¹ is the absorption peak of CO₂ [7]. Note that the area under the C=C bond peak is smaller at 600 °C than at 550 °C, indicating that more olefins are generated at 550 °C than at 600 °C. V⁵⁺ is the main activation site of the C-H bond, and in the presence of V⁵⁺, the propylidene C-H bond is broken and the propyl radical C₃H₇* and the hydrogen radical H* are adsorbed on the surface of the oxygen carrier [26]. When the temperature reaches 600 °C, the CO₂ peak is significantly enhanced. At high temperatures (600 °C), the inhibition of Ce⁴⁺ is weakened, the rate of O_latt evolution to O_elec is enhanced, and the propyl radical C₃H₇* is deeply oxidized to CO_x, while the high temperature (600 °C) weakens the metal-to-metal interaction forces, leading to the disruption of the V-O-Ce bond [20,32]. For this reason, the activity of the oxygen carrier was tested at 550 °C in this study.

3.2. Effect of the Oxygen Carrier on Propane Conversion and Propylene Selectivity

Figure 9a compares the propane conversions of the 10V-XCe/Al oxygen carriers (X = 0, 1, 3, 5, 7). As CL-ODHP consumes O_latt from the oxygen carriers, the propane conversion decreases with increasing reaction time. The average propane conversion rates of 10V/Al, 10V-1Ce/Al, 10V-3Ce/Al, 10V-5Ce/Al, and 10V-7Ce/Al are 16.32%, 18.64%, 24.72%, 20.31%, and 23.16%, respectively, in the time range of 3-15 min. The results show that introducing CeO₂ at 550 °C significantly improves propane conversion. Propane conversion is maximized at a Ce loading of 3 wt.%. This improvement can be explained by oxygen loading in the oxygen carrier combined with a structural pore change. Ce increases the oxygen carrying capacity of the oxygen carrier; in particular, Ce⁴⁺ inhibits the conversion and loss of O_latt and facilitates the oxidation of propane to propylene [17]. In addition, a small amount of Ce increases the specific surface area of the oxygen carrier, exposing more active sites on the surface of the oxygen carrier. The propane then reacts more fully.

Figure 9b compares the propylene selectivity of the 10V-XCe/Al oxygen carriers (X = 0, 1, 3, 5, 7). The propylene selectivity slowly decreases with increasing reaction time and is maximized at 83.42% at a Ce loading of 3 wt.%. The surface of the oxygen carrier hosts two main types of reactive oxygen species: O_elec, which oxidizes propane or propylene to CO_x by depth, and O_latt, which mainly reacts to produce propylene [33]. Previous XPS characterization showed that 10V-3Ce/Al contains more Ce⁴⁺ than the other samples. As Ce⁴⁺ inhibits the conversion of O_latt to O_elec, the 10V-3Ce/Al oxygen carrier retains a large number of O_latt and thus produces more propylene per unit time than the other oxygen carriers.

Figure 9c compares the CO_x selectivity of the 10V-XCe/Al oxygen carriers (X = 0, 1, 3, 5, 7). The 10V/Al oxygen carrier exhibits the highest CO_x selectivity (23.01%): the average CO_x selectivity values of 10V-1Ce/Al, 10V-3Ce/Al, 10V-5Ce/Al, 10V-7Ce/Al, and 3Ce/Al are 9.71%, 7.06%, 15.22%, 19.56%, and 52.42%, respectively. When the 3Ce/Al catalyst carrier (which lacks the VO_x component) reacts with propane, the O_latt species rapidly evolves into O_elec and propane is directly and deeply oxidized to CO_x, basically bypassing the ODHP effect [34].

Figure 9d compares the propylene yield of the 10V-XCe/Al oxygen carriers (X = 0, 1, 3, 5, 7). The average yields of propylene for 10V/Al, 10V-1Ce/Al, 10V-3Ce/Al, 10V-5Ce/Al, and 10V-7Ce/Al were 10.53%, 13.89%, 19.36%, 14.78%, and 15.23%. The propylene yield of 10V-3Ce/Al was improved by 83.86% compared to 10V/Al oxygenate carrier. From BET, it is known that 10V-3Ce/Al has the largest specific surface area, which provides more active for sites for CL-ODHP and facilitates the adequate reaction of propane. From H₂-TPR, 10V-3Ce/Al has the lowest reducibility. According to XPS, 10V-3Ce/Al has the highest O_latt capacity and Ce⁴⁺ content. Ce⁴⁺ inhibits the evolution of O_latt on the surface of the oxygen carrier towards O_elec, weakening the rate of propane being deeply oxidized to CO_x. Therefore, the average yield of propylene from 10V-3Ce/Al oxygen carriers is the highest.

3.3. CL-ODHP Cycling of the 10V/Al and 10V-3Ce/Al Oxygen Carriers

In this experiment, two oxygen carriers, 10V/Al and 10V-3Ce/Al, were cycled through 15 reaction regenerations at 550 °C. After 15 cycles, the propane conversion, propylene selectivity, and CO_x selectivity of the 10V-3Ce/Al oxygen carrier were 22.41%, 81.74%, and 6.88%, respectively (Figure 10), almost unchanged from those of the first dehydrogenation. The cyclic oxidative-dehydrogenation results confirmed the high reactivity and stability of the 10V-3Ce/Al oxygen carrier.

Figure 11a XRD patterns of fresh oxygen carriers, Figure 11b XRD patterns of oxygen carriers after 15 cycles of CL-ODHP. No diffraction peaks of VO_x species were observed before and after the cycle, and the absence of VO_x species peaks could be due to the formation of highly dispersed amorphous VO_x phases or the formation of VO_x microcrystals undetectable by XRD [16]. The peak intensity of the CeO₂ phase of the 10V-3Ce/Al oxygen carrier did not decrease significantly after cycling, indicating that the CeO₂ phase was still stably dispersed on the surface of the oxygen carrier after 15 cycles of CL-ODHP.

Figure 12 shows the TEM of the 10V10V-3Ce/Al 3Ce/Al oxygenate before and after cycling. After 15 cycles of CL-ODHP, there is no sintering on the surface of the 10V-3Ce/Al oxygenate, indicating that the 10V10V-3Ce/Al 3Ce/Al oxygenates have good cycling stability. The lattice stripe spacing d = 0.31 nm corresponds to the (111) crystal plane of CeO₂. The lattice stripe spacing d = 0.28 nm corresponds to the (301) crystal planes of V₂O₅. V₂O₅ phase and CeO₂ phase are stable distributions on the surface of the oxygen carrier, indicating that the 10V-3Ce/Al oxygen carrier has good cycle stability.

4. Mechanism of CL-ODHP

The CL-ODHP follows the Mars-van Krevelen mechanism. The CL-ODHP mechanism was proposed by combining in situ DRIFT and XPS characterization. As shown in Figure 13, the first step of CL-ODHP is the adsorption of propane on the surface of the 10V-3Ce/Al oxygen carrier and the breaking of the propane methylene C-H bond. V⁵⁺ is the dehydrogenation activation site [16]. The rate-limiting step is breakage of the C-H bond of propane to form the propyl radical C₃H₇* and the hydrogen radical H*. Based on changes in XPS O_latt content before and after oxygen carrier reaction combined with previous in situ DRIFTS characterization, VO_x then releases O_latt to selectively oxidize C₃H₇* and H* to C₃H₆ and H₂O, respectively. CeO₂ uniformly dispersed on the surface of the oxygen carrier and controls the release and evolution rate of O_latt, thus avoiding the deep oxidation of C₃H₇*. However, at high temperatures (600 °C), CeO₂ loses its inhibition ability and O_latt rapidly evolves into O_elec, enabling the deep oxidation of C₃H₇* to CO₂ [20]. Similar results were obtained by Gong et al. and Zuo et al. Gong et al. [7] investigated the application of VO_x-TiO₂ oxygen carriers in CL-ODHP. V⁵⁺ is the main C-H bond activation site, and the propyl radical C₃H₇* and the hydrogen radical H* are generated by the breakage of the propane methylene bond under the activation of V⁵⁺. Zuo et al. [20] investigated the application of CeO₂-modified 17.5 Cr/Al oxygen carriers in CL-ODHP. CeO₂ increased the amount of O_latt and enhanced the adsorption of propyl radicals, while Ce⁴⁺ modulated the rate of O_latt evolution to O_elec and propyl radicals were further oxidized to propylene.

5. Experiment

5.1. Preparation of the Oxygen Carriers

Ce (NO₃)·6H₂O, NH₄VO₃, and H₂C₂O₄·2H₂O were purchased from Sinopharm Group Chemical Reagent Co., Ltd., (Shanghai, China). Ce (NO₃)·6H₂O, NH₄VO₃, and H₂C₂O₄·2H₂O were weighed and simultaneously dissolved in deionized water. During this process, the Ce content was adjusted to 0, 1, 3, 5, or 7 wt.%, and the mass ratio of oxalic acid to ammonium metavanadate was fixed at 2:1. The solution was then ultrasonicated for 30 min to ensure complete mixing of the constituents. After immersing the γ-Al₂O₃ carrier, the solution was allowed to stand at room temperature for 12 h and then dried in an oven at 90 °C for 12 h. The dried oxygen carrier was calcined in a 600 °C muffle furnace for 4 h (heating rate: 3 °C/min). Finally, the oxygen carrier was ground and sieved. The prepared samples were expressed as 10V-XCe/Al (VO_x content = 10 wt.%; CeO_y content, X = 0, 1, 3, 5, and 7 wt.%).

5.2. Oxygen-Carrier Performance Tests

Figure 14 shows the CL-ODHP reactor for investigating the activities of the oxygen carriers. The mass of the oxygen carriers was 0.50 g. The temperature was increased to 550 °C at 10 °C/min in an N₂ atmosphere. The N₂ was then purged for 20 min before passing propane through the reactor. The flow rates of the N₂ and C₃H₈ feed gases were 45 and 5 mL/min, respectively. After 3 min of reaction, the reaction gas was collected from the bottom of the quartz tube into a gas-sampling bag and then pumped into a gas chromatograph (A91 Plus Panna, Guangzhou Baiwei Instrument Technology Co., Ltd., Guangzhou, China) for analysis.

The propane conversion, propylene selectivity, CO_x selectivity, and propylene yield were respectively calculated as follows:

Propane conversion rate

$${\mathrm{X}}_{{\mathrm{C}}_3{\mathrm{H}}_8}=\frac{{\mathrm{N}}_{{\mathrm{C}}_3{\mathrm{H}}_6 }+\frac{2}{3}{\mathrm{N}}_{{\mathrm{C}}_2}+\frac{1}{3}{\mathrm{N}}_{{\mathrm{C}}_1}}{{\mathrm{N}}_{{\mathrm{C}}_3}+\frac{2}{3}{\mathrm{N}}_{{\mathrm{C}}_2}+\frac{1}{3}{\mathrm{N}}_{{\mathrm{C}}_1}}\times 100\%$$

(1)

Propylene selectivity

$${\mathrm{S}}_{{\mathrm{C}}_3{\mathrm{H}}_6=} \frac{{\mathrm{N}}_{{\mathrm{C}}_3{\mathrm{H}}_6}}{{\mathrm{N}}_{{\mathrm{C}}_3{\mathrm{H}}_6}+\frac{2}{3}{\mathrm{N}}_{{\mathrm{C}}_2}+\frac{1}{3}{\mathrm{N}}_{{\mathrm{C}}_1}}\times 100\%$$

(2)

CO_x selectivity

$${\mathrm{S}}_{{\mathrm{CO}}_{\mathrm{x}}=}\frac{{\mathrm{N}}_{{\mathrm{CO}}_{\mathrm{x}}}}{{\mathrm{N}}_{{\mathrm{C}}_3{\mathrm{H}}_6}+\frac{2}{3}{\mathrm{N}}_{{\mathrm{C}}_2}+\frac{1}{3}{\mathrm{N}}_{{\mathrm{C}}_1}}\times 100\%$$

(3)

Propylene yield

$${\mathrm{Y}}_{{\mathrm{C}}_3{\mathrm{H}}_6}{=\mathrm{X}}_{{\mathrm{C}}_3{\mathrm{H}}_8}{\times \mathrm{S}}_{{\mathrm{C}}_3{\mathrm{H}}_6}$$

(4)

here, N represents the molar concentration of each substance at the outlet, C₁ denotes CH₄, CO, and CO₂, C₂ denotes C₂H₄ and C₂H₆, and C₃ denotes C₃H₆ and C₃H₈.

5.3. Oxygen Carrier Characterization

The structural and physical phases in the oxygen carrier were analyzed via X-ray diffraction (XRD) using a DX-2007 Advance X-ray diffractometer (Haoyuan Corporation, Dandong, China) with Cu Kα as the X-ray emission source, a tube voltage of 40 KV, and a wide angle-scanning range (5–90° in 5° increments). The structural parameters of the oxygen carriers were characterized using a Micromeritics TriStar 3000 physisorption instrument (Norcross, GA, USA) and the specific surface areas and pore-size distributions were calculated via Brunauer-Emmett-Teller and Barrett-Joyner-Halenda computational modeling, respectively. The sample to be tested was first degassed at 200 °C under vacuum for 3 h. The surface morphological characteristics of the oxygen carriers were observed under a scanning electron microscope (SEM) (JSM-IT100; JOEL Corporation, Tokyo, Japan) operated at 20 kV.

H₂ temperature-programmed reduction (H₂-TPR) was tested with a Micromeritics Autochen 2920 chemisorption instrument (Mack Corporation, USA). For this test, 50 mg of the sample was weighed and treated in a He atmosphere at 200 °C for 1 h. The treated sample was cooled to room temperature and the atmosphere was switched to 8% H₂/Ar, then purged until stabilization was achieved. After stabilization, the sample was heated to 900 °C and the H₂-TPR curve was obtained by recording the hydrogen consumption signal with a thermal conductivity detector (TCD). The elemental valence states of the oxygen carriers were characterized using X-ray electron spectroscopy (XPS) (Kratos Analytical, Manchester, UK) with monochromatic AlKα (hv = 1486.65 eV) as the excitation source. The sample to be tested was pressed and pretreated under the reaction conditions before loading it into the test bag. The thermal stabilities of the oxygen carriers were analyzed via thermogravimetric analysis (TGA) using an STA-409PC instrument (Netz, Selb, Germany) under an N₂ atmosphere. The temperature was increased from the programmed 30 °C to 800 °C and the tail-gas concentration was detected using TCD.

In situ diffuse-reflectance infrared Fourier-transform spectroscopy (DRIFTS) data were collected using a DRIFTS instrument equipped with a mercury-cadmium-telluride detector and an in situ reaction cell (Thermo Fisher Nicolet, Waltham, MA, USA). The resolution of the data collection was 4 cm⁻¹. Prior to analysis, the samples were dried at 600 °C for 30 min under an N₂ (45 mL/min) atmosphere. The background spectra were then collected at the required temperature while the temperature of the in situ reaction tank naturally declined to 30 °C. Subsequently, the oxygen carrier absorbed N₂: C₃H₈ (v/v = 45/5) gas mixture for 20 min. During measurements, the temperature was ramped at 3 °C/min. Each temperature point was stabilized for 10 min.

6. Conclusions

In this study, a series of 10V-XCe/Al oxygen carriers (X = 0, 1, 3, 5, 7) for CL-ODHP was prepared through an impregnation technique. The effects of Ce content on the catalytic activity and physicochemical properties of 10V-XCe/Al were investigated. The main findings are summarized below.

• Introducing CeO₂ increases the O_latt capacity of VO_x/γ-Al₂O₃ oxygen carriers and prolongs their catalytic oxidation activity. A certain amount of CeO₂ is uniformly dispersed on the surface of VO_x/γ-Al₂O₃ oxygen carriers, inhibiting the migration of O_latt species to electrophilic oxygen species, thus improving propylene selectivity.
• Among the prepared samples, 10V-3Ce/Al exhibits the highest specific surface area, the highest oxygen capacity, and the lowest reducibility. The 10V-3Ce/Al oxygen carrier also delivers the highest oxidative dehydrogenation performance.
• In the VO_x/γ-Al₂O₃ oxygen carrier with the CeO₂ additive (10V-3Ce/Al), the propylene selectivity improves to 83.42% (from 62.59% in 10V/Al) and the CO_x selectivity reduces to 6.88% (from 22.76% in 10V/Al).