Promoting Effect of Cerium Oxide on the Catalytic Performance of Yttrium Oxide for Oxidative Coupling of Methane

Introduction

Methane, which is the main constituent of natural gas, is an abundant hydrocarbon resource for energy and chemicals. Since methane is the most stable hydrocarbon, the selective oxidation to the chemicals seems to be difficult compared with other hydrocarbons. Therefore, in many cases, methane is first partially oxidized to synthesis gas (CO + H₂), and then converted to the chemicals such as methanol, acetic acid, light hydrocarbons and so on. This is well-known process as C1 chemistry.

On the other hand, as for the direct conversion of methane to the chemicals, oxidative coupling of methane (OCM) is regarded as energy efficient and simple chemical process for the production of higher hydrocarbons, especially, ethane and ethylene (Lunsford, 1995). The development of highly active OCM catalyst has been continuously performed over the past few decades (Hutchings et al., 1989; Arndt et al., 2011; Takanabe, 2012; Farrell et al., 2016) since the pioneering works of Keller and Bhasin (1982). The OCM reaction proceeds via the formation of methyl radicals formed by the reaction of methane with active oxygen species and subsequent coupling the methyl radicals (Dubois and Cameron, 1990). Therefore, designing the catalyst surface to create the active oxygen species such as O⁻, O${}_2^{2-}$, and O²⁻ for the OCM reaction must be the important strategy to develop highly active catalysts (Kumar et al., 2016).

Cerium oxide (CeO₂) is well-known material showing reversible oxygen release/absorb properties, so-called OSC (oxygen storage capacity) (Trovarelli, 1996), leading us to the expectation that CeO₂ can promote the activation of methane and then improve the catalytic performance for the OCM reaction. The promoting effect of CeO₂ on the OCM reaction has already been reported by many researchers. For example, Tang et al. (2011) reported that the addition of CeO₂ into Li/MgO caused an increase in the CH₄ conversion at lower reaction temperature with little change in C₂ selectivity. They explained the promoting effect of CeO₂ by oxygen activation caused via the electron transfer between Ce⁴⁺/Ce³⁺ and Li/MgO. A similar promoting effect of CeO₂ was reported for Na/CaO (Pacheco Filho et al., 2000) and Li/MgO (Elkins et al., 2016). In case of alkali-doped catalysts, Li⁺O⁻ and Na⁺O⁻ sites are proposed to participate in oxygen activation (Ito et al., 1985; Lin et al., 1986). Therefore, the role of CeO₂ is suspected to improve the catalytic ability of Li⁺O⁻ and Na⁺O⁻ sites through the electronic interaction. On the other hand, Rane et al. (2006) investigated the additive effect of rare earth elements (La, Sm, Ce, Nd, and Yb) on the OCM activity of CaO, and found that CeO₂ is not good additive for the formation of C₂ products. This is probably due to the high ability of CeO₂ for the complete oxidation of CH₄ to CO₂ (Ferreira et al., 2012; Xu et al., 2018).

Recently, we have investigated that similarity of catalytically active sites for oxidative coupling of methane (OCM) and NO decomposition, and found that the catalysts showing the activity for the latter reaction are also active for the former reaction (Haneda et al., 2018). We also reported that the addition of small amount of CeO₂ into Ba/Y₂O₃ causes a significant increase in the NO decomposition activity (Doi et al., 2015), suggesting the possibility that CeO₂ behaves as effective promoter for the OCM reaction. In this study, we prepared CeO₂-promoted Y₂O₃ as the OCM catalyst by various methods, and found that CeO₂/Y₂O₃ prepared by coprecipitation method showed the highest OCM activity. The active state and the role of CeO₂ in the OCM reaction are discussed on the basis of its catalyst characterizations.

Experimental

Catalyst Preparation

A plural CeO₂-promoted Y₂O₃ catalyst was prepared by coprecipitation, homogeneous precipitation, impregnation and hydrothermal methods. The loading of CeO₂ was fixed at 3 mol%, except for the case of coprecipitation method. The details of CeO₂/Y₂O₃ catalysts as well as Y₂O₃ itself prepared in the present study are summarized in Table 1.

Table 1. Physico-chemical properties of CeO₂/Y₂O₃.

Yttrium Oxide (Y₂O₃)

Y₂O₃ was prepared by precipitation method using yttrium(III) nitrate and ammonia aqueous solution, followed by drying and calcination at 800°C for 5 h in air.

Coprecipitation Method [CeO₂/Y₂O₃(CP)]

An aqueous solution of ammonia (10% NH₄OH, FUJIFILM Wako Pure Chemical Corporation) as a precipitation agent was added to an aqueous solution of yttrium(III) nitrate [Y(NO₃)₃·6H₂O, FUJIFILM Wako Pure Chemical Corporation], and ammonium cerium(IV) nitrate [(NH₄)₂Ce(NO₃)₆, FUJIFILM Wako Pure Chemical Corporation] at room temperature. The precipitate thus obtained was washed with distilled water, followed by drying and calcination at 800°C for 5 h in air. The loading of CeO₂ was changed from 1 to 10 mol%. The samples are abbreviated as CeO₂(x)/Y₂O₃(CP), where x is the loading of CeO₂.

Homogeneous Precipitation Method [CeO₂/Y₂O₃(HP)]

The precipitation of yttrium and cerium hydroxides was obtained by adding urea [CO(NH₂)₂, FUJIFILM Wako Pure Chemical Corporation] to an aqueous solution of yttrium(III) nitrate and cerium(III) nitrate [Ce(NO₃)₃·6H₂O, FUJIFILM Wako Pure Chemical Corporation] and stirring at 90°C for 24 h, and then washed with distilled water, followed by drying and calcination at 800°C for 5 h in air.

Impregnation Method [CeO₂/Y₂O₃(I)]

The deposition of Ce ions onto Y₂O₃ was carried out by impregnating Y₂O₃ powder, which was prepared by precipitation method as mentioned above, with an aqueous solution of cerium(III) nitrate, followed by drying and calcination at 800°C for 5 h in air.

Hydrothermal Method [CeO₂/Y₂O₃(HT)]

The oleate solution prepared by dissolving potassium oleate (C₁₇H₃₃COOK, 19% solution, FUJIFILM Wako Pure Chemical Corporation) with distilled water was added to an aqueous solution of yttrium(III) nitrate and ammonium cerium(IV) nitrate at room temperature under vigorously-stirred condition, followed by addition of ammonia aqueous solution. The mixture solution thus obtained was transferred to a Teflon vessel and then treated at 200°C for 6 h in an autoclave. The product thus obtained was washed with distilled water, followed by drying and calcination at 800°C for 5 h in air.

Catalyst Characterizations

X-ray diffraction (XRD) patterns were recorded using a Rigaku MiniFlex diffractometer with Cu Kα radiation at 30 kV and 15 mA. The scanning was done from 2θ = 15–65° at a speed of 1 deg min⁻¹. The BET surface area of the catalysts was determined by N₂ physisorption at liquid nitrogen temperature using a BELSORP mini-II, after evacuating the samples at 300°C for 1 h. Raman spectra were measured with a Micro-RAM300/NK (Lambda Vision) equipped with a TE-cooled charge coupled device (CCD) detector and a green laser (λ = 532 nm) under the ambient atmosphere. Direct observation of CeO₂/Y₂O₃(CP) samples by TEM was performed with a JEM-2100 (JEOL) operating at an acceleration voltage of 200 kV. H₂-TPR measurement was conducted to estimate the reducibility of the catalysts (Haneda et al., 2015). The TPR profiles were obtained from room temperature to 800°C in a 30 cm³min⁻¹ flow of 5% H₂/Ar at a heating rate of 10 °C min⁻¹. The consumption of H₂ was monitored using a thermal conductivity detector (TCD).

The ¹⁶O/¹⁸O isotopic exchange reaction was carried out in a flow reactor system. An adequate amount of the sample (30 mg) was installed in a quartz reactor, and then pretreated in a flow of 5% O₂/He at 600°C for 1 h. After cooling to 100°C in flowing 5% O₂/He, the sample was purged with He at 100°C for 30 min. The reaction gas composed of 1% ¹⁸O₂/He was fed to the catalyst at a rate of 20 cm³ min⁻¹, and then the temperature was increased up to 600 °C at a heating rate of 10°C min⁻¹. The masses of 32 (¹⁶O₂), 34 (¹⁸O¹⁶O), and 36 (¹⁸O₂) were continuously monitored by a quadrupole mass spectrometer (PFEFFER OminiStar).

Activity Measurement

The catalytic activity for oxidative coupling of methane was evaluated using a fixed-bed continuous flow reactor. The reaction gas composed of 44.4% CH₄ and 11.1% O₂ diluted with He as the balance gas was fed to a 0.1 g catalyst that had been pretreated in situ in a flow of He at 650°C for 2 h at a rate of 45 cm³ min⁻¹. The effect of variation of CH₄/O₂ ratio was also investigated, where O₂ concentration was varied over the range of 5.5–22.2% and CH₄ concentration was fixed at 44.4%. The reaction temperature was increased from 650 to 800°C in steps of 50°C, and the steady-state catalytic activity was measured at each temperature. The effluent gas was analyzed with the use of on-line gas chromatograph equipped with TCD and FID as a detector (Shimadzu GC-2014ATTF) using a Molecular Sieve 13X column (for analysis of O₂, CH₄, and CO), a Porapak QS column (for analysis of CO₂), and a DC-200 (20%)/Shimalite column (for analysis of C₂H₄ and C₂H₆).

Results and Discussion

Activity of CeO₂/Y₂O₃ Prepared by Different Method

Figure 1A shows the XRD patterns of 3 mol% CeO₂/Y₂O₃ prepared by different method. Distinct XRD peaks indexed to the cubic phase of Y₂O₃ were observed for all the samples. No peaks due to CeO₂ were detected because of the low loading (3 mol%), suggesting the presence of CeO₂ nanoparticles. It is of interest that a shift of XRD peaks due to Y₂O₃ toward lower angle was observed for CeO₂/Y₂O₃(HP) (Figure 1B–b), suggesting the formation of CeO₂-Y₂O₃ solid solution. This can be explained by the fact that the radius of Ce³⁺ ion (0.114 nm)/Ce⁴⁺ ion (0.097 nm) is larger than that of Y³⁺ ion (0.09 nm) (Shannon, 1976). Homogeneous precipitation including Ce³⁺ and Y³⁺ ions may be produced under the conditions, where the pH of solution is gradually increased via the hydrolysis of urea.

Figure 1. XRD patterns of (a) CeO₂/Y₂O₃(CP), (b) CeO₂/Y₂O₃(HP), (c) CeO₂/Y₂O₃(I), (d) CeO₂/Y₂O₃(HT) given in the wide range of 2θ = 16 − 64° (A) and in the expanded range of 2θ = 28.0 − 30.5° (B).

Table 1 summarizes the crystallite size of Y₂O₃, which was calculated from the XRD peak due to the (222) plane (2θ = 29.1°) using Scherrer's equation without the subtraction of the instrument broadening, where shape factor of “K” was used as 0.9, and the BET surface area of CeO₂/Y₂O₃. No significant difference in the crystallite size of Y₂O₃ was observed for CeO₂/Y₂O₃(CP), CeO₂/Y₂O₃(HP) and CeO₂/Y₂O₃(I). In accordance with this, these samples gave similar BET surface area in the range of 23–28 m²g⁻¹. On the other hand, the BET surface area of CeO₂/Y₂O₃(HT) was lower than those of other samples, although the crystallite size of Y₂O₃ was evaluated to be small. This conflict might be ascribed to the formation of aggregated Y₂O₃ crystallites with less-porous structure via hydrothermal process (Haneda et al., 2018).

Figure 2 shows the catalytic activity of 3 mol% CeO₂/Y₂O₃ for the OCM reaction. It appears that the addition of CeO₂ into Y₂O₃ by coprecipitation (CP), impregnation (I), and hydrothermal (HT) processes caused an increase in not only CH₄ conversion (Figure 2A) but also C₂ selectivity (Figure 2B) in the entire temperature range, suggesting that CeO₂ is effective additive for the OCM reaction. Since the participation of oxygen species such as O⁻, O${}_2^{-}$, O²⁻, and O${}_2^{2-}$ formed on the catalyst surface in the activation of CH₄ molecule for the OCM reaction has been established by many researchers (Keller and Bhasin, 1982; Borchert and Baerns, 1997; Sekine et al., 2009; Takanabe, 2012; Liang et al., 2014), active oxygen species would be efficiently supplied from the lattice of CeO₂. On the other hand, as seen in Figure 2, CeO₂/Y₂O₃(HP) showed lower CH₄ conversion and C₂ selectivity than Y₂O₃. This is probably ascribed to the formation of CeO₂-Y₂O₃ solid solution revealed from XRD measurements (Figure 1Bb), resulting in ineffective supply of active oxygen species. Taking into account the fact that CeO₂/Y₂O₃(HP) possesses similar physico-chemical properties, such as BET surface area and crystallite size (Table 1), with CeO₂/Y₂O₃(CP) and CeO₂/Y₂O₃(I), it is suspected the presence of the important factors affecting the catalytic activity of CeO₂/Y₂O₃ for the OCM reaction.

Figure 2. Catalytic activity of 3 mol% CeO₂/Y₂O₃ for the OCM reaction. (A) CH₄ conversion and (B) C₂ selectivity.

Effect of CeO₂ Loading on the Activity of CeO₂/Y₂O₃(CP)

Structural Characterization of CeO₂/Y₂O₃(CP)

In order to gain information on the activity controlling factor, the catalytic performance of CeO₂(x)/Y₂O₃(CP) with different CeO₂ loading was examined. Structural characterizations were carried out. In Table 2 are summarized the BET surface area of CeO₂(x)/Y₂O₃(CP). It appears that no significant difference in the BET surface area was obtained in the range of 19–27 m²g⁻¹. Figure 3A shows the XRD patterns of CeO₂(x)/Y₂O₃(CP). No other peaks, except for those assignable to Y₂O₃, were detected irrespective of CeO₂ loading. It is also noteworthy that a shift of XRD peaks by the addition of CeO₂ was not recognized. Figure 3B shows the Raman spectra of CeO₂(x)/Y₂O₃(CP). Raman bands ascribed to F_g + A_g mode of cubic Y₂O₃ with C-type structure (Repelin et al., 1995; Yashima et al., 1997; Ubaldini and Carnasciali, 2008) were observed for all the samples, and their intensities were gradually decreased with an increase in CeO₂ loading. From XRD and Raman measurements, CeO₂ particles were revealed to present as highly dispersion state on the surface and/or in the bulk of Y₂O₃.

Table 2. BET surface area, crystallite size of Y₂O₃ and the amount of H₂ consumption in H₂-TPR of CeO₂(x)/Y₂O₃(CP).

Figure 3. Structural characterization of CeO₂(x)/Y₂O₃(CP) by (A) XRD and (B) Raman. (a) Y₂O₃, (b) CeO₂(1)/Y₂O₃(CP), (c) CeO₂(3)/Y₂O₃(CP), (d) CeO₂(4)/Y₂O₃(CP), (e) CeO₂(5)/Y₂O₃(CP), (f) CeO₂(10)/Y₂O₃(CP).

In order to gain information on the dispersion state of CeO₂ particles, the morphology of CeO₂(x)/Y₂O₃(CP) with the CeO₂ loadings (x) of 0, 3, and 10 mol% was observed by TEM analysis. As can be seen in Figures 4a–c, the addition of CeO₂ into Y₂O₃ did not cause a significant change in the particle morphology, whereas the particle size was gradually decreased. This is in accordance with the crystallite size of Y₂O₃ (Table 2). These results suggest that the aggregated particles observed in TEM images mainly consist of Y₂O₃. Figures 4d–f show the TEM images of CeO₂(x)/Y₂O₃(CP) observed at high magnification. Unfortunately, CeO₂ and Y₂O₃ particles were not clearly distinguished each other because of low CeO₂ loading (< 10 mol%) and the same crystal system (cubic). However, it appears from the comparison of Figures 4d,e that the surface morphology of Y₂O₃ was changed by addition of 3 mol% CeO₂. Namely, Y₂O₃ consists of the particles with smooth surface (Figure 4d), whereas the surface of Y₂O₃ particles in CeO₂(3)/Y₂O₃(CP) seems to be rough (Figure 4e), suggesting the surface interaction between CeO₂ and Y₂O₃. In other words, CeO₂ nanoparticles of which the size is smaller than the detection limit by TEM and XRD might be dispersed on the surface of Y₂O₃ particles. On the other hand, as seen in Figure 4f, CeO₂(10)/Y₂O₃(CP) seems to be composed of a plurality of particles with different surface morphology. CeO₂ nanoparticles might be present not only independently but also weakly interacting with Y₂O₃.

Figure 4. TEM images of (a,d) Y₂O₃, (b,e) CeO₂(3)/Y₂O₃(CP), (c,f) CeO₂(10)/Y₂O₃(CP).

Catalytic Activity of CeO₂/Y₂O₃(CP) for the OCM Reaction

Figure 5 shows the effect of CeO₂ loading on the catalytic activity of CeO₂(x)/Y₂O₃(CP) for the OCM reaction. As mentioned before, the addition of CeO₂ into Y₂O₃ caused an increase in not only CH₄ conversion but also C₂ selectivity in the entire temperature range. As can be seen in Figure 5A, CH₄ conversion was drastically increased by addition of 3 mol% CeO₂. Further increase in CeO₂ loading up to 10 mol% did not cause any change in the CH₄ conversion. In accordance with previous reports (Primet and Garbowski, 2002), CeO₂ seems to be good additive for the activation of CH₄ molecules. On the other hand, as seen in Figure 5B, the effect of CeO₂ loading on the C₂ selectivity was slightly different from that on the CH₄ conversion. The C₂ selectivity was gradually increased with an increase in the CeO₂ loading up. The maximum C₂ selectivity was achieved at 3 mol% CeO₂ loading. However, further increase in CeO₂ loading caused a gradual decrease in the C₂ selectivity. The optimum CeO₂ loading for the formation of C₂H₆/C₂H₄ as a product was found to be 3 mol%. As seen in Table 2, the highest BET surface area was obtained for CeO₂(3)/Y₂O₃(CP). This is in agreement with the catalytic activity. However, although CeO₂(5)/Y₂O₃(CP) was found to possess similar BET surface area with CeO₂(3)/Y₂O₃(CP), the former catalytic activity was clearly lower than that of the latter catalyst. This suggests that the BET surface area is not only the important factor to determine the catalytic activity.

Figure 5. Effect of CeO₂ loading on the activity of CeO₂(x)/Y₂O₃(CP) for the OCM reaction. (A) CH₄ conversion and (B) C₂ selectivity.

Since the CH₄/O₂ ratio is an important parameter to optimize the reaction conditions, the effect of CH₄/O₂ ratio on the catalytic activity of CeO₂(3)/Y₂O₃(CP) was examined. As can be seen in Figure 6, an increase in the CH₄/O₂ ratio caused a decrease in the CH₄ conversion as well as an increase in the C₂ selectivity. This indicates that the reaction gas containing higher concentration of O₂ is favored for the activation of CH₄, resulting in high CH₄ conversion, while is also favored for the formation of undesirable CO₂, leading to low C₂ selectivity. The trade-off between CH₄ conversion and C₂ selectivity is well-known phenomena in the OCM reaction (Ghose et al., 2014; Godini et al., 2014; Ivanov et al., 2014). From the kinetic point of view, the activated CH₄ molecule (CH₃· radical) may preferentially react with O₂ species in the presence of excess O₂. It would be important to explore the optimum reaction conditions to achieve high C₂ productivity.

Figure 6. Effect of CH₄/O₂ ratio on (A) CH₄ conversion and (B) C₂ selectivity in the OCM reaction over CeO₂(3)/Y₂O₃(CP). (Red circles) CH₄/O₂ = 2, (blue triangles) CH₄/O₂ = 4, (green squares) CH₄/O₂ = 8.

Active Sites for the OCM Reaction Over CeO₂/Y₂O₃(CP)

XRD and Raman measurements revealed that CeO₂ particles are present as highly dispersion state on the surface and/or in the bulk of Y₂O₃ (Figure 3). In addition, TEM observation suggested that the dispersion state of CeO₂ particles is different depending on CeO₂ loading (Figure 4). In case of CeO₂(3)/Y₂O₃(CP), CeO₂ nanoparticles might be dispersed on the surface of Y₂O₃ particles, while CeO₂ nanoparticles might be present not only independently but also weakly interacting with Y₂O₃ in CeO₂(10)/Y₂O₃(CP). Therefore, the dispersion state of CeO₂ would be related to the catalytic activity. It is well known that the reduction of CeO₂ particles gradually proceeds from surface to bulk in the wide range from low to high temperature region, respectively (Yao and Yao, 1984; Trovarelli, 1996; Imagawa et al., 2011). In order to obtain an information on the dispersion state of CeO₂, H₂-TPR measurements were carried out. As given in Figure 7a, very broad H₂ consumption peaks in the temperature range of 100–700°C were observed in the H₂-TPR profile of Y₂O₃. This is probably due to the reduction of oxygen species interacting with the defects sites in the lattice of Y₂O₃ with C-type structure. No significant change in the H₂-TPR profile was observed for CeO₂(1)/Y₂O₃(CP) (Figure 7b). This is in agreement with the tendency of the catalytic activity (Figure 5). It is noteworthy that the addition of 3 mol% CeO₂ caused an appearance of distinct H₂ consumption peaks at around 200 and 600°C (Figure 7c). The area of H₂ consumption peak at around 600°C, which can be ascribed to the reduction of oxygen species in the bulk of CeO₂ (Yao and Yao, 1984; Trovarelli, 1996; Imagawa et al., 2011), was gradually increased with an increase in CeO₂ loading up to 10 mol%. In contrast, the low-temperature peak at around 200 °C due to the reduction of surface/subsurface oxygen species on CeO₂ particles (Yao and Yao, 1984; Trovarelli, 1996; Imagawa et al., 2011) was clearly decreased for CeO₂(10)/Y₂O₃(CP). These results suggest that the size of CeO₂ particles dispersed on the surface and/or in the bulk of Y₂O₃ was significantly increased with an increase in CeO₂ loading.

Figure 7. H₂-TPR profiles of CeO₂(x)/Y₂O₃(CP) with different CeO₂ loading. (a) Y₂O₃, (b) CeO₂(1)/Y₂O₃(CP), (c) CeO₂(3)/Y₂O₃(CP), (d) CeO₂(4)/Y₂O₃(CP), (e) CeO₂(5)/Y₂O₃(CP), (f) CeO₂(10)/Y₂O₃(CP).

Table 2 summarizes the amount of H₂ consumption for the reduction of surface/subsurface oxygen species in the temperature range of 100–400°C. It is of interest that the maximum amount of H₂ consumption was obtained for CeO₂(3)/Y₂O₃(CP), and then decreased with CeO₂ loading. Figure 8 shows the relationship between the amount of H₂ consumption for the reduction of surface/subsurface oxygen species and the catalytic activity for the OCM reaction at 750 °C. It appears that the CH₄ conversion and C₂ selectivity were linearly increased with an increase in the amount of H₂ consumption. This clearly indicates that oxygen species on the surface/subsurface of highly dispersed CeO₂ particles can effectively activate CH₄ molecule, resulting in the selective formation of C₂H₆/C₂H₄. On the basis of XRD, Raman, TEM, and H₂-TPR measurements, we can conclude that the creation of highly dispersed CeO₂ particles is one of the strategies for the catalyst design leading to the development of highly active OCM catalysts.

Figure 8. Relationship between the amount of H₂ consumption for the reduction of surface/subsurface oxygen species and the catalytic activity for the OCM reaction at 750°C.

Catalytic Role of CeO₂ in the OCM Reaction

Since CeO₂ is well-known material showing reversible oxygen release/absorb properties (Trovarelli, 1996), the ¹⁶O/¹⁸O isotopic exchange reaction was carried out to clarify the possibility of O₂ activation as catalytically role of CeO₂ in the OCM reaction. Martin and Duprez (1996) and Duprez (2006) investigated the oxygen mobility of various kinds of oxide by using the ¹⁶O/¹⁸O isotopic exchange reaction, and proposed two mechanisms depending on the type of oxide. One group is non-reducible oxide such as Al₂O₃, SiO₂, ZrO₂ and MgO, and the other is reducible oxide such as CeO₂ and CeO₂-ZrO₂. In the former case, the simple hetero-exchange that occurs with the participation of only one oxygen of the oxide at each step of the following equations (1) and (2):

Here, the consecutive evolution of ¹⁸O¹⁶O and ¹⁶O₂ can be observed. On the other hand, the isotopic exchange reaction over reducible oxides takes place via the multiple-hetero exchange mechanism, where the reaction between a molecule dioxygen (¹⁸O₂) in gas phase and two atomic oxygens (¹⁶O) of the solid occurs the following equation (3) or (4):

In this case, ¹⁶O₂ and ¹⁶O¹⁸O seem to be simultaneously evolved.

Figure 9 shows the profiles of ¹⁶O₂ and ¹⁶O¹⁸O evolution and ¹⁸O₂ consumption in the ¹⁶O/¹⁸O isotopic exchange reaction over Y₂O₃ and CeO₂(3)/Y₂O₃(CP). It appears that both catalysts gave similar profiles for each O₂ species. The evolution of ¹⁶O¹⁸O and the consumption of ¹⁸O₂ were simultaneously observed, followed by the ¹⁶O₂ evolution at higher temperature. This suggests that the simple hetero-exchange mechanism is favored on Y₂O₃ and CeO₂/Y₂O₃ with lower CeO₂ loading, although the O₂ activation over CeO₂-containing sample was suspected to be governed by the multiple-hetero exchange mechanism. This is probably due to the high dispersion state of CeO₂ nanoparticles on the surface of Y₂O₃, as revealed by H₂-TPR measurement (Figure 7).

Figure 9. Change in the molecular ratio of ¹⁶O₂ (green triangles), ¹⁶O¹⁸O (red circles) and ¹⁸O₂ (blue squares) during the isotopic exchange reaction of O₂ over (A) Y₂O₃ and (B) CeO₂(3)/Y₂O₃(CP) as a function of temperature.

Dong et al. (2004a,b) investigated the oxygen mobility of CeO₂-ZrO₂ based catalyst with different structural homogeneity, and reported that oxygen species activated at lower temperature via the multiple-hetero exchange mechanism is highly active for the complete oxidation reaction. It can be expected that oxygen species with moderate activity, which was evolved via the simple hetero-exchange mechanism, plays a beneficial role in CH₄ activation (Kumar et al., 2016), leading to the selective oxidation to C₂H₆/C₂H₄. Taking into account the fact that the temperature for the evolution of ¹⁶O¹⁸O and ¹⁶O₂ was significantly lowered by addition of CeO₂ into Y₂O₃ (Figure 9), we can conclude that the catalytically role of CeO₂ in the OCM reaction is to promote the formation of active oxygen species via the simple hetero-exchange mechanism.

Conclusion

The catalytic activity of Y₂O₃ for the OCM reaction was effectively improved by addition of CeO₂. Among the catalysts tested here, CeO₂/Y₂O₃ prepared by coprecipitation method showed the highest activity. The optimum CeO₂ loading was 3 mol%. From the structural characterizations by XRD, Raman and TEM, CeO₂ particles were found to be dispersed on the surface of Y₂O₃ without the formation of CeO₂-Y₂O₃ solid solution. H₂-TPR measurements revealed that the amount of H₂ consumption for the reduction of CeO₂ surface was increased with an increase in CeO₂ loading and then reached the maximum at 3 mol%. Highly dispersed CeO₂ particles was suspected to act as catalytically active sites in the OCM reaction. From the ¹⁶O/¹⁸O isotopic exchange reaction, the beneficial role of CeO₂ in the OCM reaction was concluded to promote the formation of active oxygen species via the simple hetero-exchange mechanism.

Author Contributions

MH: worked on the experimental setup; YK and YN: conducted the experiments; AT: carried out the TEM analysis.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Acknowledgments

This study was supported by a Grant–in–Aid for Scientific Research (No. 16K14046, 18K05294) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References