English

• Propylene is a significant raw material for producing various chemicals, such as polypropene, acrylic acid, acrylonitrile, octanol, butanol, and propene oxide [1]. However, the ever-increasing demand for propylene poses a serious challenge in meeting the supply and demand gap. Propane dehydrogenation (PDH), which can effectively convert propane into propylene with high selectivity and stability, has gained much attention due to its great capacity to mitigate the tense demand for propylene, providing a promising route compared to the traditional petroleum-based routes [2].

  However, the high reaction temperature (typically >650 ℃) poses a new challenge in catalyst design [3–5]. The sintering of metal active sites always results in poor catalytic activity and stability. It is essential to develop new chemical processes to reduce energy consumption and increase propylene yield. Solar energy is a kind of green and pollution-free energy, showing a significant advantage in saving fossil fuels [6]. The well-designed catalyst can effectively absorb the light radiation, and the photo-generated carriers would separate into hot electrons and holes, which can be involved in the oxidation reaction and reduction reaction respectively [7,8]. However, exclusive energy input of light cannot effectively activate the C—H bond due to its strong activation constraints (~420 kJ/mol).

  Photo-assisted thermal catalysis can couple the phonon and photon catalyzing the chemical reactions, moreover, it could pump the metal's energy to the specific vibration mode of the adsorbate. However, to maximize the utilization of light energy, optimizing the reaction conditions is a practicable route [9,10]. For instance, Liu et al. [11] found that the maximum overall light enhancement can only be achieved in the optimal light intensity and reaction temperature. The author claims the non-thermal effect has a close relationship with the light intensity and reaction temperature, while the photo-to-thermal effect is more complicated. Linic et al. [12,13] also found that a certain temperature would afford electron momentum, which is conducive to the generation of electron-hole pairs. The above examples indicate coupling the phonons and photos could effectively pump metal's energy to one specific vibration mode of adsorbate, thus achieving a high energy utilization in photo-assisted thermal catalysis.

  Besides optimizing the reaction conditions, another point is to construct a well-designed catalyst. The elaborate control facets of the catalyst have a close relationship with its optical and textural properties [14–19]. For instance, Liu et al. [20] found that CeO₂ morphology strongly affects NbO_x–CeO₂ interactions and structures of NbO_x/CeO₂ catalysts, the 0.6Nb/r-CeO₂–500 is most active in catalyzing low-temperature ODHP reaction. Gong et al. [21] also found that the VO₃/CeO₂(111) catalyst created new empty localized states that are mainly constituted of O 2p orbitals of the ceria-supported VO₃ species, which were determined to be crucial for the cleavage of the first C–H bond of the propane molecule. The light absorbance capacity of CeO₂ also strongly depends on its morphology, such as the sheet CeO₂ has better photochemical properties due to its bigger surface [22], which can provide more active sites to interact with light radiation. However, the above different methods have been proven effective in optimizing the optical properties and regulating C−H bond activation in PDH reactions. The elaborate control of the morphology of CeO₂ crystallites, the V atom's location on certain facets, and the structure-activity relationship, are still not systematically investigated in the photo-assisted thermal propane dehydrogenation process.

  Herein, we constructed four supports with different morphologies and systematically investigated their structure-activity relationship. The monomeric VO_x species were successfully loaded on the R-CeO₂ support, and it achieved the biggest net growth rate. The F-CeO₂ catalyst has the highest growth ratio. The kinetic result shows the non-thermal effect dominates the net growth rate. Exclusively increasing the temperature would lower the propane adsorption and the non-thermal effect strengthens the propane adsorption and activation. The light radiation TPSR experiment indicates that light radiation selectively boosts the first C—H bond activation rather than the second C—H bond. DFT calculations indicate the monomeric VO_x species would first accept electrons from the adsorbed propane to initiate the 1^st C—H bond activation. When it arrives at the transitional state, the metal sites would donate their electrons to the C₃H₇ fragment, finally arriving at the final state. A good linear relationship between the energy barrier and the exchanged electron number was found as a good descriptor for the first time. It shows that the lower the energy barrier, the more electrons transfer. This paper emphasizes the significance of the exchanged electron number in the activation of the C—H bond, and the photo-assisted thermal catalysis method can afford more electron exchange channels to strengthen the intrinsic activity of propane dehydrogenation.

  We first utilized the SEM and TEM images to investigate the morphology and lattice fringe of different CeO₂ supports. It can be seen that all these particles are monodisperse particles with a uniform distribution. The C—CeO₂ sample as shown in Fig. 1a is cube-like with an edge length of 1.2 µm and it comprises small nanoparticles, indicating the C—CeO₂ has a rough surface [23]. The B-CeO₂ sample in Fig. 1b shows poor agglomeration among the particles. Its distribution in size is homogeneous and the shape appears rounded. According to past reports, van der Waal's force is responsible for the formation of CeO₂ agglomerates, which also can be seen in Fig. 1c, the flower-like CeO₂ support comprised of the packed CeO₂ pillar with a diameter of about 10 µm, the agglomeration of the pillar makes the flower-like support has a bigger surface area (29.8 cm²/g) than the C—CeO₂ (7.1 cm²/g), but it is lower than the B-CeO₂ (98.6 cm²/g) and the R-CeO₂ (112.7 cm²/g). The R-CeO₂ catalyst (Fig. 1d) is comprised of packed rods with no left particles, indicating the alkaline concentration and hydrothermal temperature are enough to completely transform the nucleation to the CeO₂ nanorods.

  Figure 1

  

  Figure 1.  The SEM image of the (a) C—CeO₂, (b) B-CeO₂, (c) F-CeO₂ and (d) R-CeO₂. The TEM image of the (e) C—CeO₂, (f) B-CeO₂, (g) F-CeO₂ and (h) R-CeO₂.

  DownLoad: Full-Size Img PowerPoint

  TEM images further confirmed that the four samples consisted of well-defined structures having a rectangular form. The mapping results show the successful loading of V atoms (Fig. S1 in Supporting information). The TEM images clearly show the lattice fringe of these four supports. In the C—CeO₂ catalyst (Fig. 1e), the lattice spacing parallel of 0.27 nm can be ascribed to the (200) plane [23,24], and the other three samples (Figs. 1f-h) have the same distance of 0.31 nm, which can be attributed to the (111) plane of CeO₂, implying the (111) plane is the most popular plane in the three samples.

  The V atoms were loaded on the various supports by the impregnation method. The VO_x polymerization has a close relationship with PDH activity. Thus, the crystal structure and morphology of different CeO₂ supports were investigated. In XRD patterns (Fig. 2a), all the supports exhibit typical CeO₂ nanocrystal peaks such as 2θ = 28.549° (111), 2θ = 33.104° (200), 2θ = 47.496° (220) and 2θ = 56.379° (311). No crystalline impurity was detected from the XRD spectrum [25]. It can be seen that V/C—CeO₂ has the highest crystallinity, while the other three samples show similar low crystallinity. All the samples did not show any V₂O₅ peaks, implying the VO_x has a great dispersion on all the supports.

  Figure 2

  

  Figure 2.  The (a) XRD patterns, (b) N₂ adsorption-desorption isotherms, (c) FTIR spectrum, (d) UV–vis spectrum and (e) derived Tauc plots. (f) The photoluminescence spectrum of the V/C—CeO₂, V/B-CeO₂, V/F-CeO₂ and V/R-CeO₂ samples.

  DownLoad: Full-Size Img PowerPoint

  N₂ physical adsorption isotherms (Fig. 2b) were conducted to obtain the pore size, pore volume, and surface area of the four supports. The results are summarized in Table 1, it can be seen the pore size is 10.1 nm (V/C—CeO₂), 5.8 nm (V/B-CeO₂), 10.3 nm (V/F-CeO₂) and 5.4 nm (V/R-CeO₂), the pore volume is 0.04 cm³/g (V/C—CeO₂), 0.16 cm³/g (V/B-CeO₂), 0.14 cm³/g (V/C—CeO₂) and 0.11 cm³/g (V/R-CeO₂). The V/R-CeO₂ and V/F-CeO₂ both show typical IV-type curves [26], while the V/C—CeO₂ and V/B-CeO₂ both show typical II-type curves. The surface area order is V/R-CeO₂ (112.7 m²/g) > V/B-CeO₂ (98.6 m²/g) > V/F-CeO₂ (29.8 m²/g) > V/C—CeO₂ (7.1 m²/g). The V/R-CeO₂ has the biggest surface area, indicating the VO_x has the greatest dispersion on the R-CeO₂ support.

  Table 1

  

  Table 1.  Physicochemical properties of as-prepared catalysts.

  DownLoad: CSV

  The monomeric VO_x species were confirmed by the FTIR (Fig. 2c). The peak at 3428 cm^-1 is attributed to the surface hydroxyl group [27]. The peaks at 1623 cm^-1 and 1312 cm^-1 are ascribed to the Ce-O bond vibration [28]. The 1044 cm^-1 and 977 cm^-1 are attributed to the V=O and V-O vibration [28,29]. The V/B-CeO₂ has the biggest area of the V=O peak, indicating the VO_x species were polymeric on the B-CeO₂ support, which is consistent with its lowest surface. The 785 cm^-1 at C—CeO₂ is attributed to the CeVO₄ species, indicating the VO_x species has a strong interaction with the C—CeO₂ support. The R-CeO₂ has a bigger area of the 977 cm^-1 (V-O bond), and it has a lower area of the 1044 cm^-1 (V=O bond), indicating the R-CeO₂ has the best dispersion of the VO_x species.

  The optical properties were also investigated using the UV–vis spectrum and photoluminescence spectrum. As seen in Figs. 2d and e, the light absorption capacity follows the order: V/F-CeO₂ (2.7 eV) > V/R-CeO₂ (2.8 eV) > V/B-CeO₂ (2.9 eV) > V/C—CeO₂ (3.2 eV), implying a good dispersion of VO_x species conducing to the light adsorption, which may be associated with its LSPR effect [23]. The photo-generated carriers migration capacity was investigated by the photoluminescence spectrum (Fig. 2f). The V/R-CeO₂ has the best migration capacity, and the V/C—CeO₂ has the worst carrier separation ability. We speculate that a good dispersion of V species both favors light adsorption and carrier migration capacity. Following this, we conducted a series of activity evaluations and kinetic tests to confirm and elucidate the non-thermal effect.

  The catalytic activity tests were conducted and the result can be seen in Fig. S2 (Supporting information). It shows the V/R-CeO₂ has the highest propane conversion under both dark and light conditions. We also obtained the growth ratio brought by the light radiation. It can be seen that V/F-CeO₂ has the highest growth ratio. The ratio order is V/F-CeO₂ (39.2%) > V/R-CeO₂ (19.8%) > V/B-CeO₂ (13.0%) > V/C—CeO₂ (5.6%). The order indicates a close relationship with the catalyst's light absorption and carrier migration capacity.

  To confirm that the non-thermal effect dominates the PTPDH process instead of the photo-to-thermal effect, the TPSR experiment was conducted to investigate the C₃H₈ and H₂ signals under dark and light conditions. As shown in Fig. 3a, it shows the propane onset temperature is 331 ℃ without light radiation, while after the light radiation, the propane disappears quickly (~233 ℃ under light conditions as shown in Fig. 3b). The propylene appearance temperature is ~423 ℃ under dark conditions, while with the assistance of light, it reduces to ~372 ℃. The above results indicate that light radiation helps the activation of propane and the generation of propylene. However, the H₂ onset temperature is ~240 ℃ no matter the conditions. This result strongly demonstrates that the light radiation would not affect the H₂ dissociation or H coupling process, and this reverse reaction would not strongly affect the PDH process. Thus, the rate-determined step on the V/R-CeO₂ is most likely the first C—H bond activation process.

  Figure 3

  

  Figure 3.  The temperature-programmed surface reaction of propane dehydrogenation over V/R-CeO₂ under (a) dark or (b) light conditions. The C₃H₈ reaction orders of the four samples under (c) dark or (d) light conditions.

  DownLoad: Full-Size Img PowerPoint

  Kinetic experiments were further conducted to confirm the non-thermal effect [30]. Propane reaction orders under dark or light conditions were obtained to exclude the photo-to-thermal effect. The derivation process can be found in Section 1 in Supporting information. As shown in Fig. 3c, exclusively increasing the reaction temperature would boost the propane reaction order, according to past reports [31,32], the lower the propane reaction order, the stronger the adsorption and activation of the propane molecule gets. Thus, it can be deduced that increasing the temperature would suppress the propane adsorption and activation. After the light radiation (Fig. 3d), the reaction order decreases slightly, indicating the non-thermal effect can beat the photo-to-thermal effect. Thus, the non-thermal effect would boost the propane adsorption and activation.

  The apparent activation energy was obtained as shown in Fig. S3 (Supporting information). Increasing the propane flow rate would suppress its intrinsic activity. After the light radiation, the intrinsic activity would be boosted, indicating the non-thermal effect can effectively accelerate the electron transfer rate between the adsorbate and the metal centers [33,34]. The stability test (Fig. S4a in Supporting information) indicates the catalysts have good stability under light radiation conditions. Furthermore, thermogravimetry analysis (Fig. S4b in Supporting information) was conducted and it shows the PTPDH has less coke than the PDH with a similar conversion of propane. Both experiments indicate the PTPDH process is more stable.

  DFT calculations were conducted to investigate the detailed electronic effect on the PTPDH process [35,36]. Two reaction pathways were considered and the potential energy surface (PES) can be found in Fig. 4a. The monomeric VO₃ species was considered the active site in the PTPDH process and the detailed models are shown in Fig. 4b. Model I is one VO₃ loaded on the CeO₂ surface without a dangling O atom, and the model II is one VO₃ species loaded on the CeO₂ surface with a dangling O atom. Model I shows a lower energy barrier (0.88 eV) for the 1^st dehydrogenation step but a higher energy barrier (1.34 eV) for the 2^nd step, while model II has a higher energy barrier (1.88 eV) for the 1^st dehydrogenation step and a lower energy barrier (0.71 eV) of the 2^nd step. According to the Sabatier principle [37], a successful catalytic cycle in heterogeneous catalysis always originates from neither too strong nor too weak adsorption of intermediate. Thus, the higher the 1^st step must accompany with the lower the 2^nd step. This unbalanced kinetic condition would significantly suppress the overall PDH production.

  Figure 4

  

  Figure 4.  (a) The potential energy surface of pathway I and pathway II. (b) The detailed reaction configurations in the propane dehydrogenation process.

  DownLoad: Full-Size Img PowerPoint

  The DOS analysis was conducted to illustrate the effect of the light radiation on the PDH as shown in Fig. S5 (Supporting information) [38]. When the propane molecule was adsorbed on the surface of model I, there was no occupation of the C and H atoms (Fig. S5a). When the first H atom was dissociated, the occupation of the C and H atoms appeared at ~−2.5 eV (Fig. S5b). Such a lower energy level rationalizes the higher energy barrier of the 2^nd step. As for model II, the occupation of the C and H atoms appeared at ~−1.3 eV (Fig. S5d) and ~2.6 eV (Fig. S5e) respectively. When the first H atom was dissociated, the occupation of the C and H atoms appeared at ~−1.2 eV The higher energy level (−1.2 eV > −2.5 eV) accompanies a lower energy barrier in the 2^nd step. According to the Frontier molecular orbit theory [39,40], the chemical reactivity is only decided by the relative energy level of the HOMO and LUMO of the hybrid orbits. Thus, an appropriate electronic wave function overlap confers a neither too strong nor too weak interaction between the metal atoms and adsorbate. We speculated the high energy level of C and H atoms is more favorable for the influx of electrons.

  Thus, we tracked the electronic density of the metal active site within the PDH process. In pathway I, the metal center first accepts the electron to arrive at the transitional state, then releases its electrons to arrive at the 1^st state (Fig. S6 in Supporting information). The electronic fluctuation of pathway I (0.6 e^-) is bigger than pathway II (0.1 e^-), indicating more electron influx lowered the energy barrier. As for the 2^nd step, in pathway I, it first releases its electron to arrive at the transitional state and then releases electrons to arrive at the 2^nd state. While in pathway II, the metal would accept the electrons from the adsorbate and release the electrons to arrive at the 2^nd state. Pathway I has a higher energy barrier in the 2^nd step than Pathway II, which may come from the subsequent electron release behavior, leading to the metal center in an electron-deficient state. In Pathway II, the metal would first accept the electrons, the electron-rich state lowers the energy barrier of the 2^nd step. In all, it can be deduced the electron can help lower the reaction barrier and boost the PDH performance.

  The exchanged electron number was correlated to its energy barrier within the 1^st, and the 2^nd C—H bond activation or the H coupling desorption processes (Figs. S5c and f). The negative relationship indicates the more exchanged electron numbers, the lower the active energy barrier of C—H bond activation is.

  In conclusion, we prepared four CeO₂ supports with different morphologies, a series of characterization results indicate that R-CeO₂ has the biggest surface area, leading to the best dispersion capacity for VO_x species and good light absorption capacity. The kinetic result shows that exclusively increasing the temperature would suppress the 1^st step, while light radiation can boost overall propylene production. The light radiation TPSR experiment demonstrates that light radiation selectively boosts the 1^st step and thus balancing the kinetic conditions. DFT calculations indicate that light accelerates the electron transfer behavior within the 1^st dehydrogenation process. More electrons were involved in the 1^st step and thus lowering the reaction energy barrier. The linear relationship between the energy barrier and the exchanged electron number supports this point. This paper emphasizes the significance of the electron exchange in the activation of the C—H bond, and the photo-assisted thermal catalysis method can afford more electron exchange channels to strengthen the intrinsic activity of propane dehydrogenation.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Xiangyang Ji: Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Yishuang Chen: Project administration, Methodology, Data curation, Conceptualization. Peng Zhang: Supervision, Software, Funding acquisition. Shaojia Song: Methodology, Conceptualization. Jian Liu: Visualization, Methodology, Investigation. Weiyu Song: Writing – review & editing, Writing – original draft, Visualization, Conceptualization.

  Acknowledgments

  We gratefully acknowledge the financial support from the National Key R&D Program of China (Nos. 2021YFA1501301, 2021YFC2901100) and the National Natural Science Foundation of China (Nos. 22178381, 22035009).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110719.

  
  
• 1. [1]

     S. Chen, X. Chang, G. Sun, et al., Chem. Soc. Rev. 50 (2021) 3315–3354. doi: 10.1039/D0CS00814A

  2. [2]

     T. Yang, F. Su, D. Shi, et al., Chin. Chem. Lett. 36 (2025) 110444. doi: 10.1016/j.cclet.2024.110444

  3. [3]

     L. Liu, M. Lopez-Haro, C.W. Lopes, et al., Nat. Mater. 18 (2019) 866–873. doi: 10.1038/s41563-019-0412-6

  4. [4]

     L. Liu, M. Lopez-Haro, C.W. Lopes, et al., Nat. Catal. 3 (2020) 628–638. doi: 10.1038/s41929-020-0472-7

  5. [5]

     L. Liu, M. Lopez-Haro, C.W. Lopes, et al., J. Catal. 391 (2020) 11–24. doi: 10.1016/j.jcat.2020.07.035

  6. [6]

     C. Gu, H. Ji, K. Xu, et al., Chin. Chem. Lett. (2024), doi:10.1016/j.cclet.2024. 110565.

  7. [7]

     X. Ji, Y. Ma, X. Sun, S. Song, et al., ACS Appl. Nano Mater. 6 (2023) 6354–6364. doi: 10.1021/acsanm.3c00794

  8. [8]

     R. Zhang, H. Wang, S. Tang, et al., ACS Catal. 8 (2018) 9280–9286. doi: 10.1021/acscatal.8b02441

  9. [9]

     X. Zhang, X. Li, D. Zhang, et al., Nat. Commun. 8 (2017) 14542. doi: 10.1038/ncomms14542

  10. [10]

      X. Li, X. Zhang, H.O. Everitt, J. Liu, Nano Lett. 19 (2019) 1706–1711. doi: 10.1021/acs.nanolett.8b04706

  11. [11]

      Z. Geng, Y. Yu, A.J. Offen, J. Liu, Nat. Catal. 6 (2023) 1241–1247. doi: 10.1038/s41929-023-01045-9

  12. [12]

      P. Christopher, H. Xin, A. Marimuthu, S. Linic, Nat. Mater. 11 (2012) 1044–1050. doi: 10.1038/nmat3454

  13. [13]

      S. Linic, S. Chavez, R. Elias, Nat. Mater. 20 (2021) 916–924. doi: 10.1038/s41563-020-00858-4

  14. [14]

      W.Z. Yu, W.W. Wang, S.Q. Li, et al., J. Am. Chem. Soc. 141 (2019) 17548–17557. doi: 10.1021/jacs.9b05419

  15. [15]

      S. Yang, L. Gao, J. Am. Chem. Soc. 128 (2006) 9330–9331. doi: 10.1021/ja063359h

  16. [16]

      X. Xiang, L. Wu, J. Zhu, et al., Chin. Chem. Lett. 32 (2021) 3215–3220. doi: 10.1016/j.cclet.2021.03.064

  17. [17]

      Y. Shan, X. Deng, X. Lu, et al., Chin. Chem. Lett. 33 (2022) 5158–5161. doi: 10.1016/j.cclet.2022.03.010

  18. [18]

      B. Liu, Q. Lia, X. Du, et al., J. Alloys Compd. 509 (2011) 6720–6724. doi: 10.1016/j.jallcom.2011.03.156

  19. [19]

      Y. Mei, Y. Zhang, J. Li, et al., J. Alloys Compd. 904 (2022) 163879. doi: 10.1016/j.jallcom.2022.163879

  20. [20]

      Y. Liu, L. Luo, Y. Gao, W. Huang, Appl. Catal. B: Environ. 197 (2016) 214–221. doi: 10.1016/j.apcatb.2016.03.011

  21. [21]

      C. Huang, Z.Q. Wang, X.Q. Gong, Chin. J. Catal. 39 (2018) 1520–1526. doi: 10.1016/S1872-2067(18)63072-4

  22. [22]

      Y. Liu, Z. Shen, J. Song, et al., Chin. Chem. Lett. 31 (2020) 2747–2751. doi: 10.1016/j.cclet.2020.06.016

  23. [23]

      B. Feng, D. Li, J. Xiong, et al., Appl. Catal. B: Environ. 325 (2023) 122337. doi: 10.1016/j.apcatb.2022.122337

  24. [24]

      Y. Wang, S. Chen, J. Sun, et al., Sci. China Mater. 66 (2023) 1062–1070. doi: 10.1007/s40843-022-2214-8

  25. [25]

      B.M. Reddy, A. Khan, P. Lakshmanan, et al., J. Phys. Chem. B 109 (2005) 3355–3363. doi: 10.1021/jp045193h

  26. [26]

      H. Ning, K. Xie, B. Jiang, et al., J. Rare Earths (2024), doi:10.1016/j.jre.2024.06. 029.

  27. [27]

      T. Wu, G. Liu, L. Zeng, et al., AIChE J. 63 (2017) 4911–4919. doi: 10.1002/aic.15836

  28. [28]

      G. Qi, R.T. Yang, J. Phys. Chem. B 108 (2004) 15738–15747. doi: 10.1021/jp048431h

  29. [29]

      M.V. Bosco, M.A. Bañares, et al., J. Mol. Catal. A: Chem. 408 (2015) 75–84. doi: 10.1016/j.molcata.2015.07.012

  30. [30]

      X. Li, H.O. Everitt, J. Liu, Nano Res. 12 (2019) 1906–1911. doi: 10.1007/s12274-019-2457-x

  31. [31]

      S. Song, K. Yang, P. Zhang, et al., ACS Catal. 12 (2022) 5997–6006. doi: 10.1021/acscatal.2c00928

  32. [32]

      J. Zhu, M.-L. Yang, Y. Yu, et al., ACS Catal. 5 (2015) 6310–6319. doi: 10.1021/acscatal.5b01423

  33. [33]

      W. Wang, Q. Song, Q. Luo, et al., Nat. Commun. 14 (2023) 2493. doi: 10.1038/s41467-023-38211-3

  34. [34]

      Z. Xiao, P. Li, H. Zhang, et al., Fuel 362 (2024) 130906. doi: 10.1016/j.fuel.2024.130906

  35. [35]

      J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865. doi: 10.1103/PhysRevLett.77.3865

  36. [36]

      G. Henkelman, B.P. Uberuaga, H. Jónsson, J. Chem. Phys. 113 (2000) 9901–9904. doi: 10.1063/1.1329672

  37. [37]

      Z.J. Zhao, S. Liu, S. Zha, et al., Nat. Rev. Mater. 4 (2019) 792–804. doi: 10.1038/s41578-019-0152-x

  38. [38]

      P. Wang, X. Zhang, R. Shi, et al., Nat. Commun. 15 (2024) 789. doi: 10.1038/s41467-024-45031-6

  39. [39]

      G. Klopman, J. Am. Chem. Soc. 90 (1968) 223–234. doi: 10.1021/ja01004a002

  40. [40]

      J. Yu, N.Q. Su, W. Yang, JACS Au 2 (2022) 1383–1394. doi: 10.1021/jacsau.2c00085

• 

  Figure 1  The SEM image of the (a) C—CeO₂, (b) B-CeO₂, (c) F-CeO₂ and (d) R-CeO₂. The TEM image of the (e) C—CeO₂, (f) B-CeO₂, (g) F-CeO₂ and (h) R-CeO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  The (a) XRD patterns, (b) N₂ adsorption-desorption isotherms, (c) FTIR spectrum, (d) UV–vis spectrum and (e) derived Tauc plots. (f) The photoluminescence spectrum of the V/C—CeO₂, V/B-CeO₂, V/F-CeO₂ and V/R-CeO₂ samples.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The temperature-programmed surface reaction of propane dehydrogenation over V/R-CeO₂ under (a) dark or (b) light conditions. The C₃H₈ reaction orders of the four samples under (c) dark or (d) light conditions.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) The potential energy surface of pathway I and pathway II. (b) The detailed reaction configurations in the propane dehydrogenation process.

  下载: 全尺寸图片 幻灯片

  Table 1.  Physicochemical properties of as-prepared catalysts.

  下载: 导出CSV
•