La₂O₃ catalysts with diverse spatial dimensionality for oxidative coupling of methane to produce ethylene and ethane

Abstract

We investigated the catalytic behaviour of La₂O₃ catalysts with different spatial dimensionality, such as zero-, one-, two- and three-dimensional structure, for the oxidative coupling of methane. The La₂O₃ catalyst with two-dimensional structure exhibited the highest conversion of methane with the most ethylene and ethane produced among these catalysts. The collective contribution from the exposed specific facets, more electron deficient oxygen species and more moderate surface basic sites on the two-dimensional La₂O₃ catalyst account for its highly efficient catalysis in the oxidative coupling of methane to produce ethylene and ethane.

---

With the widespread discovery of natural gas resources and the depletion of oil reserves, methane, which has diverse chemical applications, has drawn increasing attention worldwide for several decades. Oxidative coupling of methane (OCM) for valuable hydrocarbons has been considered as a promising process and a directly effective utilization pattern of natural gas.^1–3 Even if the prospect is attractive and routine, and increasing study is competently carried out, difficulties, such as the inefficient activity of the C–H bond, the low yield of hydrocarbons and the high reaction temperature, hinder the flourishing development of the OCM process. Since first reported by Keller and Bhasin, thousands of catalysts have been studied and elaborated for oxidative coupling of methane⁴ such as alkali metal oxides,⁵ rare earth oxides,^6–10 perovskite-type complex oxides,¹¹ and Mn–Na₂WO₄/SiO₂. Among these catalysts, La₂O₃ has been widely studied due to its relatively excellent catalytic properties such as low ignition temperature of methane and long-term stability. Well-faceted La₂O₃ nanorods prepared in our laboratory obtained higher activity compared to nanoparticles and further showed no decrease of C₂ hydrocarbon (ethane and ethylene) yield during the 200 hour stability experiment.¹² However, no further significant advance in the precisely-defined structure of La₂O₃ catalysts for OCM reaction has been reported in the related literature. Especially, La₂O₃ with controlled dimensionality, from zero-dimensional to three-dimensional, has rarely been considered in this catalytic reaction, in that OCM is always carried out at much high temperature leading to the sintering of most catalysts. The catalytic properties can often be significantly improved by the reasonable control of nanoscale shape and size.¹³ Therefore, it is worthwhile synthesizing La₂O₃ materials with different morphologies and structures to be used as catalysts for the OCM reaction. Our efforts aim to make La₂O₃ catalysts of different dimensional architectures (zero-dimensional (0D), one-dimensional (1D), two-dimensional (2D) and three-dimensional (3D)) and study the effects of spatial dimensionality of the catalyst on the conversion of methane and the selectivity of hydrocarbons.

Herein, we synthesized La₂O₃ catalysts with diverse spatial structures (0D to 3D), i.e., nanoparticles, nanorods, nanosheets, and nanoflowers via feasible solvothermal routes (see the ESI†). These La₂O₃ catalysts indeed displayed the distinct performance for the conversion of methane and the selectivity of hydrocarbons at mild reaction temperatures. It is noted that La₂O₃ hexagonal sheets with 2D structure obtained the higher conversion of methane with more ethylene when compared with 0D nanoparticles, 1D nanorods and 3D nanoflowers catalysts. Interestingly, the catalytic yield for C₂ hydrocarbons (ethylene and ethane) obtained from the OCM process over these La₂O₃ catalysts follows the order 2D > 3D = 1D > 0D. The more effective catalysis of the 2D La₂O₃ catalyst should be due to the collective contribution of the (120) oriented surface, more electron deficient oxygen species (peroxide ions and superoxide ions) on the surface and more moderate surface basic sites, which could benefit the activation of methane and the formation of ethane and ethylene.

Direct information about typical shapes of La₂O₃ products gained under different experimental conditions is indicated by scanning electron microscopy (SEM) (Fig. 1). As shown in Fig. 1a, the La₂O₃ catalyst consists of nanoparticles with average diameter between 200 and 300 nm. The X-ray diffraction pattern from Fig. S1† shows the synthesized La₂O₃ nanoparticles typically match the hexagonal phase of La₂O₃ (JCPDS 05-0602) with crystal unit cell parameters a = 0.3930 nm and c = 0.6109 nm. From the high resolution transmission electron microscopy (HRTEM) image of the nanoparticles, the lattice structure is discerned with spacings of 0.30, 0.31 and 0.34 nm, close to the lattice spacings measured for (101), (002) and (100) planes (Fig. 2a). For the 1D structure, the average length of these rods is about 400 nm and the diameter is approximately 20 nm, as shown in Fig. 1b. The crystal lattice fringe of a single nanorod is 0.34 nm apart in Fig. 2b, which agrees with the d value of the (100) lattice planes of hexagonal La₂O₃. It is obvious that the inset of Fig. 2b was detected along the [001] direction, which means the La₂O₃ nanorods are mainly enclosed by (001) facets and the terminal ends of the nanorods are bounded by (120) facets.

Fig. 1 SEM images of La₂O₃ (a) nanoparticles, (b) nanorods, (c) nanosheets, and (d) nanoflowers.

Fig. 2 HRTEM images of La₂O₃ (a) nanoparticles, (b) nanorods, (c) nanosheets, (d) nanoflowers. The insets are the corresponding crystal morphology diagrams. (e and f) Surface atoms arrangement of La₂O₃ nanocrystals with (120) and (001) planes, where the red and blue spheres are oxygen and lanthanum atoms, respectively.

With regard to the 2D structure, they look like elongated hexagons with microscale edge lengths (Fig. 1c). These hexagonal sheets are self-assembled with short rods. The interplanar spacings of 0.31 and 0.30 nm on the HRTEM image of an isolated nanosheet have lattice fringe directions attributed to (002) and (101), as observed from Fig. 2c. The exposed facet is perpendicular to the two facets and can be reconstructed in a stimulation diagram in the inset of Fig. 2c, i.e., the exposed facet in the 2D nanosheet is the (120) plane. In terms of the 3D structure, as displayed in Fig. 1d, these are constructed with leaf-like plates and appear as flowers. The HRTEM image of the leaf shows that the interplanar spacing of 0.34 nm in Fig. 2d has a lattice fringe direction attributed to (100) of hexagonal La₂O₃. From the inset of Fig. 2d it is measured that the dominant exposed planes of La₂O₃ nanoflowers are of (001) orientation. From Fig. S1,† it can be found that the four samples match the hexagonal crystal phase of La₂O₃ well.

To gain insight into the spatial structure effect of La₂O₃ catalysts (from 0D to 3D) on oxidative coupling of methane, we assessed their catalytic performance and summarized this in Fig. 3. It is clearly noted that the catalytic activity for methane conversion of these La₂O₃ catalysts follows the order 2D nanosheets > 1D nanorods = 3D nanoflowers > 0D nanoparticles (Fig. 3a). The selectivity toward ethylene is also shown in the descending sequence: 2D nanosheets > 1D nanorods = 3D nanoflowers > 0D nanoparticles, as expressed in the upper panel of Fig. 3b. The selectivity of ethane over 1D, 2D and 3D catalysts is higher than that on the 0D catalyst at mild reaction temperatures, as displayed in the lower panel of Fig. 3b. A higher ratio of ethylene to ethane was achieved on 2D nanosheets than on the other three catalysts. The ethylene/ethane variation tendency with temperature increased, i.e., the proportion of ethylene increased gradually, whereas the ethane proportion decreased. This suggests that the high temperature favors the formation of ethylene mainly from the cracking of ethane on the surface of these catalysts. For the selectivity of CO and CO₂ from the further oxidation of methane and C₂ hydrocarbons, more CO and CO₂ products could be created over La₂O₃ nanoparticles in Fig. 3c. The selectivity of CO₂ is higher than that of CO over the four catalysts, indicating that the complete oxidation of methane dominated among the coupling reaction of methane. The valuable products, such as ethylene and ethane, in which the former is more useful than the latter in chemical industry, are desired in the OCM reaction. From Fig. 3d, in particular, 2D La₂O₃ hexagonal sheets obtain a 15% yield of ethylene and ethane at 550 °C under a flow rate of 240 mL min⁻¹ of 75% methane and 25% oxygen, while 1D nanorods and 3D nanoflowers catalysts gave a similar yield of C₂ hydrocarbons (12%), and 0D catalyst exhibited a negligible C₂ yield (0.6%) under other identical catalytic conditions. Interestingly, only silica sand exhibited a different catalytic behaviour from La₂O₃ catalysts, as shown in Fig. S3.† C₂ selectivity over silica decreased with the increase of reaction temperature and the C₂ yield is much lower when the reaction temperature is below 800 °C. The next studies will lay great stress on suppressing the deep oxidation of methane and hydrocarbons occurring in the gas phase at the catalysts' surface.

Fig. 3 Catalytic performance of La₂O₃ catalysts: (a) CH₄ conversion, (b) C₂H₄ and C₂H₆ selectivity, (c) CO selectivity and CO₂ selectivity (d) the ratio of C₂H₄/C₂H₆ (conv.: conversion; sel.: selectivity).

To elucidate the shape-specific effects of the 2D La₂O₃, above all, the structural factor is taken into account. HRTEM studies have revealed the distinct exposed facets in the four catalysts. La₂O₃ nanosheets are mainly exposed (120) planes, while the major exposed crystal planes of both La₂O₃ nanorods and nanoflowers are of the (001) orientation. The representative surface atoms arrangement of La₂O₃ nanocrystals with (120) and (001) crystal planes are shown schematically in Fig. 2e and f, respectively. The number of missing neighbors of a unit cell in {120} and {001} planes is 5 and 4, respectively, which implies that in the (120) surface of the La₂O₃ nanosheets, the surface atoms‘ low coordination character might provide a favorable environment for the activation of methane and oxygen molecules.

Furthermore, investigations of surface element compositions, surface basic sites/strength, and surface area were performed to give a deep understanding of the different activity and selectivity observed experimentally over La₂O₃ catalysts with different spatial dimensionalities (from 0D to 3D). N₂ adsorption–desorption isotherms and pore size distributions of the prepared La₂O₃ catalysts were taken, as shown in Fig. 4a, and their specific surface areas and pore properties are summarized in Table 1. Obviously, the specific surface areas of different dimensionality La₂O₃ catalysts are 7.2, 18.0, 12.0, and 9.5 m² g⁻¹ for 0D, 1D, 2D and 3D, respectively, which is not in agreement with their catalytic performance in the OCM, i.e., the higher surface area usually shows higher catalysis. Thus, we speculate that catalytic performance of these catalysts seems related not to the specific surface area but to the density of active sites, and a higher specific area means a higher residence time of intermediate species, resulting in an increased possibility of complete oxidation and a decreased C₂ hydrocarbon selectivity. As the surface area cannot account for the better catalytic performance of the 2D La₂O₃ catalyst, we therefore investigate the reason based on the surface element compositions and surface basic sites and strengths.

Fig. 4 (a) N₂ adsorption–desorption isotherms of fresh La₂O₃ catalysts. Insets are corresponding pore size distributions (note: the vertical ordinate appoints to dV/dD (cm³ g⁻¹)). (b) CO₂-TPD profiles of La₂O₃ catalysts with different spatial dimensionalities.

Table 1 Surface element compositions, porous texture, catalytic performance for OCM to produce ethylene and ethane^a

Catalysts	CH4 conv. (%)	C2 sel. (%)	C2 yield (%)	SBET (m^2 g^−1)	Pore size (nm)	(O^− + O2)/O^2−	Exposed facets
a Note: reaction temperature: 550 °C; gas hourly space velocity: 72 000 h^−1; ratio CH4/O2 = 3.
0D	9.9	6.4	6.3	7.2	20.8	1.24	—
1D	29	40.5	11.7	18.0	77.9	1.51	(001)
2D	32.3	45.9	14.8	12.0	75.2	2.12	(120)
3D	28.9	41.6	12.0	9.5	52.5	1.60	(001)

---

It has been reported that the efficiency of C₂ formation is directly correlated to the available basic sites.^14–17 CO₂-TPD (temperature programmed desorption) profiles are often used to characterize surface basic properties and results for the four catalysts are shown in Fig. 4b. In the case of the catalysts, the main CO₂ desorption peak occurred in the range 200–600 °C, which means that they are all moderately basic sites. Under the applied conditions, 2D La₂O₃ nanosheets exhibited maximum desorption of the CO₂ molecule and the second maximum CO₂ desorption was detected on the surface of La₂O₃ nanorods, while low CO₂ desorption occurred on nanoflowers and nearly negligible CO₂ desorption on nanoparticles. 2D La₂O₃ nanosheets contained more surface basic sites and thus expressed better behaviour in the selectivity of ethylene and ethane compared to catalysts with 0D, 1D and 3D structures. This is due to the fact that moderate surface basic sites are beneficial to the production of C₂ hydrocarbons.¹⁰

The electronic properties of the metal oxide and the adsorbed species on the catalysts also play a non-negligible role in the C–H bond breaking and the coupling of C–C bonds from methane.^18–22 In the O 1s X-ray photoelectron spectroscopy (XPS) profiles in Fig. 5a, two peaks were recorded for the four catalysts, which can be decomposed into four components, i.e., superoxide ions O₂⁻, peroxide ions O⁻, lattice oxygen O²⁻ and carbonate CO₃²⁻. Surface-adsorbed oxygen species (O⁻ and O₂⁻) are widely considered as efficient species to enhance the C₂ hydrocarbon selectivity, whereas lattice oxygen (O²⁻) is the cause of CO and CO₂ produced by complete oxidation. The ratio of the peak intensities of the surface-adsorbed oxygen species (O⁻ and O₂⁻) to lattice oxygen (O²⁻) is calculated as 1.24 for 0D nanoparticles, 1.51 for 1D nanorods, 2.12 for 2D nanosheets and 1.60 for 3D nanoflowers, respectively, as shown in Table 1. Therefore, taking into considering these ratio values, high selectivity toward ethylene and ethane can be achieved over La₂O₃ nanosheets, which is in good agreement with the experimental results, whereas high selectivity toward CO and CO₂ was obtained on nanoparticles. To understand the electronic properties of La₂O₃ with different structures on the coupling reaction of methane we set out to characterize La 3d by XPS. As shown in Fig. 5b, we can easily find two peaks located at approximately 835 and 838 eV, which are assigned to La 3d_5/2 XPS profiles. When compared to the La 3d_5/2 binding energy of 0D, 1D and 3D La₂O₃, the highest magnitude of the shift towards low binding energy of 2D La₂O₃ shows that unique electron properties occur on 2D La₂O₃ nanosheets, which might enhance the catalytic performance. One question remains unanswered, however, namely, which of the activity of methane or the selectivity of C₂ hydrocarbons can be improved by La₂O₃ with the low binding energy of La 3d. The result is interesting and further study is in progress.

Fig. 5 (a) O 1s binding energy spectra and (b) La 3d_5/2 binding energy spectra of La₂O₃ catalysts.

As is well known, the oxidative coupling of methane is exothermic and the inner temperature of the reactor is much higher, as obtained from the association curve of feed and bed temperatures (Fig. S2a†). Though the feed temperature is under 800 °C, the temperature on the surface of the catalysts is always above 900 °C (Fig. S2b†). At so high a temperature, it is crucial that the structure of the catalyst is stable during the OCM reaction. The SEM studies of spent catalysts indicated that 2D La₂O₃ nanosheets seem to be robust compared to the other three catalysts which are slightly sintered after catalytic reaction to a certain extent (Fig. 6a–d), which can result from the decrease of the surface areas of La₂O₃ catalysts after the catalytic reaction (Fig. S4†). Moreover, the stability test of catalysts with reaction time also demonstrates the robust nature of the 2D La₂O₃ catalyst, as presented in Fig. S5.† The robustness of the 2D La₂O₃ catalyst is of particular importance to put it to real-world use.

Fig. 6 (a–d) SEM images of La₂O₃ catalysts after OCM reaction.

Conclusions

In this study, we report the catalytic performance of La₂O₃ catalysts with different dimensional architectures, such as 0D nanoparticles, 1D nanorods, 2D hexagonal sheets and 3D nanoflowers, on the oxidative coupling of methane. It is found that the La₂O₃ catalyst with 2D structure is more effective than corresponding 0D, 1D and 3D catalysts for the conversion of methane and selectivity of ethylene and ethane, wherein the yield of ethylene and ethane could reach 15% at 550 °C. The investigation implies that not only do the exposed specific planes of La₂O₃ allow improvement of the catalytic activity, but also the surface-adsorbed oxygen and surface basic sites and strengths make a contribution in oxidative coupling of methane. Activity and selectivity control on the catalytic reaction by tuning the dimensionality of the catalyst might aid in improving the studies of shape-specific effects of catalysts in useful real-world reactions.

Acknowledgements

We are grateful for financial support by the National Natural Science Foundation of China (21273151), Shell Global Solutions International B. V. and Hundred Talent Program of CAS. We thank Dr Carl Mesters, Dr Alexander van der Made, Dr Tim Nisbet and Dr Sander van Bavel from Shell for helpful discussions.

References

1. K. Otsuka, K. Jinno and A. Morikawa, J. Catal., 1986, 108, 353–359.
2. J. S. Lee and S. T. Oyama, Catal. Rev., 1988, 30, 249–280.
3. Y. Amenomiya, V. I. Birss, M. Goledzinowski, J. Galuszka and A. R. Sanger, Catal. Rev., 1990, 32, 163–227.
4. G. E. Keller and M. M. Bhasin, J. Catal., 1982, 73, 9–19.
5. D. J. Driscoll, W. Martir, J. Wang and J. H. Lunsford, J. Am. Chem. Soc., 1985, 107, 58–63.
6. J. M. DeBoy and R. F. Hicks, Ind. Eng. Chem. Res., 1988, 27, 1577–1582.
7. T. W. Elkins, B. Neumann, M. Bäumer and H. E. Hagelin-Weaver, ACS Catal., 2014, 4, 1972–1990.
8. B. Neumann, T. W. Elkins, A. E. Gash, H. Hagelin-Weaver and M. Bäumer, Catal. Lett., 2015, 145, 1251–1261.
9. P. Schwach, W. Frandsen, M.-G. Willinger, R. Schlögl and A. Trunschke, J. Catal., 2015, 329, 560–573.
10. Y. Hou, W. Han, W. Xia and H. Wan, ACS Catal., 2015, 5, 1663–1674.
11. J. E. Elshof, H. J. M. Bouwmeester and H. Verweij, Appl. Catal., A, 1995, 150, 195–212.
12. J. Song, Y. Sun, R. Ba, S. Huang, Y. Zhao, J. Zhang, Y. Sun and Y. Zhu, Nanoscale, 2015, 7, 2260–2264.
13. P. Huang, Y. Zhao, J. Zhang, Y. Zhu and Y. Sun, Nanoscale, 2013, 5, 10844–10848.
14. A. G. Dedov, G. D. Nipan, A. S. Loktev, A. A. Tyunyaev, V. A. Ketsko, K. V. Parkhomenkc and I. I. Moiseev, Appl. Catal., A, 2011, 406, 1–12.
15. V. H. Rane, S. T. Chaudharia and V. R. Choudhary, J. Chem. Technol. Biotechnol., 2006, 81, 208–215.
16. F. Papa, P. Luminita, P. Osiceanu, R. Birjega, M. Akanec and I. Balinta, J. Mol. Catal. A: Chem., 2011, 346, 46–54.
17. Y. Hou, Y. Lin, Q. Li, W. Weng, W. Xia and H. Wan, ChemCatChem, 2013, 5, 3725–3735.
18. A. J. Nagy, G. Mestl and R. Schlogl, J. Catal., 1999, 188, 58–68.
19. Z. Kalenik and E. E. Wolf, Catal. Today, 1992, 13, 255–264.
20. J. X. Wang and J. H. Lunsford, J. Phys. Chem., 1986, 90, 3890–3891.
21. W. Ding, Y. Chen and X. Fu, Catal. Lett., 1994, 23, 69–78.
22. M. S. Palmer, M. Neurock and M. M. Olken, J. Am. Chem. Soc., 2002, 124, 8452–8461.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra01805j

---