English

• Natural gas has been widely used as a clean energy source in industrial manufactures and daily routine [1,2]. Methane (CH₄) is the main ingredient of natural gas and has harmful effects on the environment due to its strong greenhouse effect [3–5]. Therefore, there is an urgent need for conversion of CH₄ to high-value products or liquid fuels, which is crucial for storage and transportation of CH₄. At present, the catalytic CH₄ conversion is mainly based on thermocatalysis [6,7], photocatalysis [8,9], electrocatalysis and enzymecatalysis [10–15]. Among them, the thermocatalysis approach has been widely used in industry because of its relatively low cost and high efficiency [6,7]. The dissociation of CH₄ molecule is a key step for CH₄ conversion, and one of the most suitable catalysts are group-VIIIB transition metals, especially Pt-based nanoparticles, nanoshells, and single atoms [16,17].

  The previous studies have reported that platinum has decent catalytic performances for CH₄ conversion to produce hydrogen [18,19]. Rategarpanah et al. demonstrated that the addition of a small amount of Pt to the Ni-Cu nanoalloys supported on MgO·Al₂O₃ powders can considerably promote the CH₄ activation, in which the CH₄ conversion rate increases from 70% to 80%, and the H₂ formation rate increases from ~1.35 mol g⁻¹ min⁻¹ to ~1.55 mol g⁻¹ min⁻¹ [19]. Moreover, Hosseini et al. [20] reported that the Fe₂O₃@CeO₂ core-shell structure loaded with Pt nanoparticles achieves about 80% conversion rate in the dry reforming of methane (DRM), leading to the production of CO_x-free (< 2 ppm) H₂ in the oxidation step by steam with a cyclically stable yield (10.8 mmol/g catalyst-oxygen carrier). Comparing with nanoparticles, dispersed Pt single atom or few atoms exhibits outstanding capabilities for CH₄ conversion [21–26]. For instance, Yan et al. [22] reported that Pt single atoms supported on Mn₂O₃ owns excellent catalytic activity for CH₄ oxidation with 90% conversion rate at 603 ℃. Gao et al. [26] showed that Pt single atoms supported on La₂O₃ precursor with carbon species (Pt₁/LP) have improved activity for CH₄ conversion with regard to Pt nanoparticles. When Pt₁/LP was supported on CeO₂ surface obtained by using precipitation method, the CH₄ conversion rate can be further enhanced to 82%.

  It is known that single-atom catalysis faces the problem of low density of active sites (< 1 wt%) on the substrate, which limits the catalysis of complicated or multi-electron reactions [27]. Atomically thin layers of transition metals combine the advantages of nanoparticle and single-atom catalysts, providing a large number of active sites as well as maximizing the atomic utilization [28]. Furthermore, the synergistic effect between metal overlayers and the underlying substrates can boost the surficial activities [29–32]. So far, Pt overlayers supported on various transition metal oxides (TiO₂, ZrO₂, CeO₂ and so on) have been synthesized in laboratory [33–41]. Yang et al. predicted that Pt monolayer on V₂C MXene exhibits high activity for oxygen reduction reaction owing to the strong metal-support interaction [30]. Later, Li et al. successfully synthesized Pt nanolayers with a single or double atomic layer thickness on Mo₂TiC₂T_x MXene, which can catalyze non-oxidative coupling of CH₄ to ethane/ethylene [42]. These results shed light in the extraordinary catalytic properties of atomically thin Pt overlayers. However, the principles for tuning the activity of Pt overlayers and the role of different substrates await being understood.

  Herein we explore the catalytic behavior of Pt monolayer on transition metal oxides TiO₂, ZrO₂ and CeO₂ for CH₄ conversion by first principles calculations. Among them, Pt monolayer supported on metal-terminated CeO₂ enables direct conversion of CH₄ to methanol (CH₃OH) and suppresses coke formation. The surficial activities of Pt monolayer sensitively depend on the type of substrate, which regulates the d-band center of Pt monolayer. The kinetic barrier of reaction is linearly correlated with the energy difference between d-band center of Pt monolayer and bonding states of adsorbed molecules, which serves as a key descriptor for precise design of novel catalysts for liquid fuel storage.

  Density functional theory (DFT) calculations were performed by using the Vienna ab initio simulation package (VASP) [43], with the projector augmented wave potentials [44], and the generalized gradient approximation parameterized by Perdew, Burke and Ernzerhof (GGA-PBE) for the exchange and correlation functional [45]. The plane wave basis set with an energy cutoff 500 eV was used. The Brillouin zone was sampled by a 3 × 3 × 1 Monkhorst-Pack k-point mesh. All structures were optimized using a convergence criterion of 10⁻⁴ eV and 0.02 eV/Å for total energy and force thresholds, respectively. The DFT-D3 dispersion correction scheme of Grimme was adopted to describe the van der Waals interactions in these layered systems [46]. Kinetic barriers and transition states for CH₄ conversion were simulated using the climbing-image nudged elastic band (CI-NEB) method implemented in VASP [47]. An effective Hubbard parameter U_eff = U − J = 5 eV was used for Ce atoms to consider the strong on-site Coulombic interaction of the Ce 4f orbital [48]. The number of charge transfer between Pt monolayer and underlying substrates was evaluated using the Bader charge analysis [49]. The electronic structures were calculated by DS-PAW program [50].

  For the growth of Pt nanofilms in the experiment, transition metal oxides have been widely utilized as substrates [34–42]. Here we considered Pt monolayer on the O-terminated rutile TiO₂(110) and metal-terminated cubic ZrO₂(100) surfaces, which have been synthesized in laboratory [35–40]. For the model of Pt on TiO₂ (hereafter noted as Pt/TiO₂), the supercell consists of 4 × 2 unit cells of rutile TiO₂(110), with lattice vectors of u = 5a₁ − a₂ and v = 2a₁ − 5a₂ for Pt(111) monolayer. For Pt/ZrO₂, we adopted the supercell comprising Pt(111) monolayer (u = 2a₁ + a₂ and v = a₂) on 3 × 4 unit cells of cubic ZrO₂(110). We also considered cubic CeO₂ as a possible substrate, which has been widely used to support nanofilms, nanoparticles and single atoms for catalysis [51,52]. According to the previous experimental characterization, the surface of CeO₂ sensitively depends on the growth condition [53,54]. In the oxidative atmosphere, CeO₂ tends to form the O-terminated surface, while Ce-termination or a mixture of O- and Ce-termination may occur for CeO₂ prepared in the reductive atmosphere [54,55]. Moreover, the CeO₂(111) surface is non-polar perpendicular to the surface, exhibiting outstanding stability and well-defined geometrical structures compared with the other low-index surfaces [55,56]. Therefore, we constructed Pt(111) monolayer on the O-terminated and Ce-terminated CeO₂(111) to explore the substrate effect. The supercell of Pt/CeO₂ includes 3 × 3 unit cells of Pt(111) monolayer and 4 × 4 unit cells of CeO₂(111) substrate. The lattice of Pt overlayer is stretched or compressed to match that of the substrate, resulting in lattice mismatch of 0.54%−3.21%. To elucidate the strain effect on the electronic structure and catalytic property of Pt overlayer, we perform test calculations on Pt(111) monolayer by applying a biaxial stretch strain ranging from 0% to 3.5%. The d-band center is lifted by 0.24 eV toward Fermi level and the adsorption energy of CH₃^* species is weakened by 0.08 eV. Therefore, the strain on Pt overlayer induced by substrates may have a minor effect on the catalytic properties.

  The structures of Pt overlayer on cubic ZrO₂(100), cubic CeO₂(111) and rutile TiO₂(110) are shown in Fig. 1. The detailed structure parameters are presented in Table 1. The interlayer distance ranges from 1.73 Å to 2.35 Å, which shows the typical covalent interaction between Pt(111) monolayer and underlying TMOs substrates. These interfacial covalent interactions can also be characterized by the differential charge densities, showing the electron accumulation at the interfacial region (Figs. 1e-h). It is worth noting that Pt monolayer grown on TiO₂ exhibits a relatively large deformation with a vertical buckling of 1.51 Å for the Pt monolayer. On the other three substrates, Pt monolayer undergoes slight deformation with small vertical buckling of 0.27~0.42 Å. The optimized Pt-Pt bond lengths of Pt/TiO₂, Pt/ZrO₂, Pt/CeO₂ (Ce) and Pt/CeO₂ (O) are in the range of 2.47–2.78, 2.64–2.78, 2.78–2.98 and 2.71–2.90 Å, respectively, comparable with that of Pt(111) surface (2.78 Å).

  Figure 1

  

  Figure 1.  Geometrical structures of Pt/TMOs hybrid systems. Atomic structures of Pt/TMOs (a-d) and corresponding differential charge density distributions (e-h). White lines labeled the lattice. The Pt, O, Ti, Zr, and Ce atoms are shown by dark blue, pink, green, purple and blue balls, respectively.

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  The key parameters of Pt/TMOs hybrid systems including lattice mismatch (δ), interlayer distance (d), vertical buckling of Pt monolayer (Δd_Pt), Pt-Pt bond length (h_Pt-Pt), formation energy per Pt atom in Pt monolayer (E_form), charge transfer between Pt monolayer and the substrate (CT), the d-band center of Pt monolayer (ε_d).

  DownLoad: CSV

  To characterize the thermostabilities of Pt/TMOs hybrid systems, we calculated their formation energy per Pt atom (E_form), as follows (Eq. 1):

  (1)

  where E_total, E_PtML, and E_sub denote the total energy of the Pt/TMOs system, Pt monolayer and TMOs substrate, respectively; N_Pt is the number of Pt atoms. As a result, the formation energy of these Pt/TMOs hydride systems range from −0.66 eV/atom to −2.76 eV/atom, signifying the strong bonding interaction between Pt monolayer and underlying TMOs substrates. It is worth noting that the E_form of Pt/CeO₂(Ce) is −2.04 eV, which was competitive to or much lower than that of experiment-prepared Pt/TiO₂ (E_form = −1.24 eV) and Pt/ZrO₂ (E_form = −2.76 eV). According the Bader charge analysis [49], charge transfer is from Pt monolayer to TiO₂ substrate (0.17 e per Pt atom), whereas Pt monolayer gain electrons from the other substrates (0.02–0.34 e per Pt atom) (Table 1). Figs. 1e-h depicts the two-dimensional charge distribution between Pt monolayer and substrates. The interfacial charge transfer is correlated to the electronegativity of atoms. The electronegativities of Ti, Zr, Ce, O and Pt are 1.54, 1.33, 1.12, 3.34 and 2.28, respectively. For Pt/ZrO₂ and Pt/CeO₂ (Ce) systems, due to the high electronegativity of Pt compared with Zr and Ce, Pt atoms of Pt/ZrO₂ and Pt/CeO₂ (Ce) gains some electrons (0.34 and 0.33 e per Pt atom) from the underlying Zr and Ce, respectively. For Pt/TiO₂, the electronegativity of O exceeds electronegativity of Pt, and hence each Pt atom loses electrons of 0.17 e. In the Pt/CeO₂ (O) system, due to the electronegativity of Ce is less than electronegativity of Ti, Ce atoms affords more electrons to O atoms, which leads to the charge transfer from O to Pt atoms.

  A lot of experimental and theoretical studies have indicated that the Pt-based nanomaterials can achieve a decent chemical activity for CH₄ activation and unique selectivity for yielding CH₃OH product [57,58]. We investigate the catalytic process of CH₄ activation and conversion with H₂O on the Pt/TMOs hybrid systems as well as the bulk-phase Pt(111) surface for reference. According to the previous studies [59], the pathway of CH₄ reaction with H₂O to yield CH₃OH mainly includes three relevant processes (Eqs. 2 and 3):

  (2)

  (3)

  (4)

  where ^* represents an adsorption site on the catalyst surface. Alternatively, the CH₃^* intermediate can be further dehydrogenated via Eq. 5:

  (5)

  which provides the opportunity to form olefin.

  Fig. 2 shows the reaction energy (ΔH) for each reaction step, where positive (negative) numbers represent endothermic (exothermic) reaction. For Eq. 2, Pt/TiO₂, Pt/CeO₂ (Ce) and Pt/CeO₂ (O) have reaction energies of −1.26 eV, −0.61 eV and −1.54 eV, respectively. In contrast, CH₄ dehydrogenation is unfavorable on Pt/ZrO₂ with a reaction energy of 1.11 eV. The reaction energies for Eqs. 4 and 5 determine the tendency of CH₃^* to combine with OH^* to form CH₃OH or to be further dehydrogenated. The results show that CH₃^* prefers to react with OH^* to yield CH₃OH product on Pt/TiO₂, Pt/ZrO₂ as well as Pt(111) surface, while Pt/CeO₂ with strong chemical activity can dehydrogenate CH₃^* to CH₂^* intermediate. In contrast to other systems, the reaction energies of two Eqs. 4 and 5 on the Pt/CeO₂ (Ce) system are both close to 0. The kinetic barrier dominates selectivity, as will be discussed later. We examined the reaction Eq. 3 to check whether OH^* can be produced by dissociation of H₂O on these systems.

  Figure 2

  

  Figure 2.  Kinetic process of CH₄ conversion and H₂O dissociation. Energy diagrams of (a) CH₄ activation and (b) H₂O dissociation on Pt/TMOs. The reaction heat (ΔH), kinetic barriers (E^a), and the corresponding structure configurations are given in insets. Histogram of reaction energies and kinetic barriers for (c) CH₄ → CH₃^* + H^*, (d) H₂O → OH^* + H^*, (e) CH₃^* + OH^* → CH₃OH and (f) CH₃^*→ CH₂^* + H^* processes on various Pt/TMOs systems and bulk-phase Pt(111).

  DownLoad: Full-Size Img PowerPoint

  To further evaluate the catalytic performance of each system, we calculated the kinetic barriers for Eqs. 2–5. Figs. 2a and b show the schematic diagrams of CH₄ dehydrogenation and H₂O dissociation on Pt/CeO₂ (Ce) and Pt/TiO₂, respectively. The values of kinetic barriers are represented by bars in Figs. 2c-f. Fig. 3a shows the complete reaction pathways of Pt/CeO₂ (Ce) and the corresponding adsorption conformation. The detailed information of other systems is shown in Fig. 3b. The transition states of Eqs. 2–5 were represented as TS1, TS2, TS3 and TS4, respectively. The kinetic barriers of Pt/TiO₂ are 1.65, 0.77, 1.63 and 1.25 eV for TS1, TS2, TS3 and TS4, respectively. TS2 has a much smaller kinetic barrier than those of other reactive steps. Too easy decomposition of H₂O may occupy the active sites and leads to inactivation of catalyst. The kinetic barriers of Pt/ZrO₂ are 1.49, 1.96, 1.21 and 1.58 eV for TS1, TS2, TS3 and TS4, respectively. The high kinetic barrier of H₂O dissociation results in low activity for Pt/ZrO₂, due to the deficiency of OH^* for CH₃OH production. The kinetic barriers of the Pt/CeO₂ (Ce) system are 1.05, 0.91, 0.68 and 1.37 eV for TS1, TS2, TS3 and TS4, respectively. The rate-determining step (RDS) of Pt/CeO₂ (Ce) system is TS1 with kinetic barrier of 1.05 eV, while the RDS of bulk-phase Pt(111) is TS3 with larger kinetic barrier of 1.53 eV. Therefore, the CH₄ to CH₃OH conversion is easier to achieve on Pt/CeO₂ (Ce) than that on bulk Pt. On the other hand, the kinetic TS2, TS3 and TS4, respectively, which indicates that CH₃ dehydrogenation to CH₂^* is more favorable than the formation of CH₃OH. To gain a clear understanding of the product selectivity of Pt/CeO₂ (O), we further consider the following dehydrogenation processes (Eqs. 6 and 7):

  (6)

  (7)

  (8)

  Figure 3

  

  Figure 3.  The reaction pathway of CH₄ conversion. (a) The reaction pathway of CH₄ conversion on Pt/CeO₂ (Ce), and the corresponding structure configurations are given in insets. (b) The reaction pathway of CH₄ conversion on other Pt/TMOs systems and bulk-phase Pt(111) surface. The color number is the kinetic barrier for each reaction step.

  DownLoad: Full-Size Img PowerPoint

  The transition states of Eqs. 6–8 were represented as TS5, TS6 and TS7. As shown in Fig. 4, the kinetic barriers of the above processes on Pt/CeO₂ (O) are 0.03, 0.90 and 1.12 eV for TS5, TS6 and TS7, respectively. The CH₂^* intermediate prefers further dehydrogenation to generate CH^*, rather than the coupling with another CH₂^* to form C₂H₄^*. The CH^* species may be further dehydrogenated to produce C^*, which would occupy the active site and poison the catalysts [60]. Therefore, Pt monolayers on various substrates exhibit different catalytic behavior. Pt/CeO₂ (Ce) has higher activity for CH₄ conversion to CH₃OH, while Pt/CeO₂ (O) may be over-reactive and suffer from coke and deactivation.

  Figure 4

  

  Figure 4.  The further dehydrogenation processes on Pt/CeO₂ (O). The colored number is the kinetic barrier for each reaction step.

  DownLoad: Full-Size Img PowerPoint

  To further elucidate the substrate effects, we examined the electronic structures of Pt monolayer on various substrates. We calculated the density of states (DOS) for each system after CH₄ adsorption and compared it with the DOS of free gas phase CH₄. Figs. 5a-d show the DOS of Pt/TMOs. The d-band center of the Pt monolayer is calculated using the following equation (Eq. 9) [61]:

  (9)

  Figure 5

  

  Figure 5.  The orbital interaction of CH₄ adsorbed on Pt/TMOs. (a-d) Local density of states (DOS) of (from left to right) gas-phase and adsorbed CH₄ molecule, and Pt/TMOs. (e) Differential charge density between Pt overlayer and CH₄. (f) The relationship between the kinetic barriers (E^a) and Δε.

  DownLoad: Full-Size Img PowerPoint

  where D(E) is the DOS of the d-band of the Pt monolayer at a given energy E, the integral is taken up to the Fermi level (shifted to zero). The d-band center of the Pt monolayer is indicated by the blue dashed line in Figs. 5a-d. Due to the interfacial charge transfer, the Pt monolayers grown on different substrates have different d-band centers. It is worth noting that the variation of d-band center of Pt monolayer is not only related to the amount of charge transfer, but also depends on the DOS distribution of the substrate. Specifically, the DOS of ZrO₂ is mainly distributed at −2~−7 eV, which leads to the d-band center of Pt monolayer far away from the Fermi level (Fig. 5b). The DOS of CeO₂ (O) is mainly distributed at −1.5~−5 eV, which leads to the d-band center of Pt monolayer closer to the Fermi level (Fig. 5d).

  The activity of Pt overlayer is closely related to its d-band center modulated by the presence of substrate. Figs. 5a-d show the energy level of CH₄ upon adsorbing on the Pt overlayer. For Pt/CeO₂ (O), the anti-bonding (σ^*) orbital level of CH₄ exhibits a notable shift to the Fermi level owing to the charge transfer from Pt overlayer to CH₄. The Bader charge analysis shows that CH₄ gains 0.23 e from d orbital of Pt/CeO₂ (O), while only about 0.02 e charge transfer occurred for the other three systems. The injection of electrons increases the polarization of CH₄ molecule, elongating the C-H bond length to 1.09~1.18 Å and weakening the C-H bond. Fig. 5e depicts the two-dimensional charge distribution between Pt monolayer and CH₄. Fig. 5f displays a linear relationship between the activation energy of C-H bond and the d-band center of Pt overlayer relative to σ^* of CH₄^* (Δε). Roughly speaking, a lower Δε value indicates more charge transfer from d-band center of Pt overlayer to the anti-bonding orbital of adsorbate, thus resulting in a lower activation energy for C-H bond cleavage in both CH₄ and CH₃^* species. Therefore, the energy difference between d-band center of Pt overlayer and the anti-bonding orbital of adsorbate is a possible descriptor to predict the activation energy of C-H bond, which guides the rational selection of proper substrates to tune the catalytic performance of metal overlayers for certain reactions.

  In summary, we investigate the geometrical and electronic structures of Pt monolayer supported on various metal oxide surfaces and exploit them for CH₄ conversion based on first principles calculations. Monolayer Pt on Ce-terminated CeO₂(111) surface is a potential catalyst for CH₄ conversion to CH₃OH with high activity and selectivity. The relationship between chemical activity and electronic structure was established for supported Pt monolayer hybrid systems. The reaction occurs more readily with smaller Δε. Moreover, the substrate with more density of states distributed near the Fermi level induces larger changes of the d-band center of Pt monolayer. These computational results provide useful insights for the design of high-efficiency metal overlayers for direct and selective CH₄ conversion.

  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.

  Acknowledgments

  This work was supported by the National Natural Science Foundation of China (Nos. 11974068, 91961204, and 12004065), the Fundamental Research Funds for the Central Universities of China (No. DUT20LAB110), the Liaoning Provincial Natural Science Foundation of China (No. 2019JH3/30100002), and Key Research and Development Project of Liaoning Province (No. 2020JH2/10500003). The authors gratefully acknowledge the computer resources provided by the Supercomputing Center of Dalian University of Technology and Shanghai Supercomputer Center. We gratefully acknowledge HZWTECH for providing computation facilities.

  
  
• 1. [1]

     Z.R. Chong, S.H.B. Yang, P. Babu, P. Linga, X.S. Li, Appl. Energy 162 (2016) 1633–1652. doi: 10.1016/j.apenergy.2014.12.061

  2. [2]

     S. Faramawy, T. Zaki, A.A.E. Sakr, J. Nat. Gas Sci. Eng. 34 (2016) 34–54. doi: 10.1016/j.jngse.2016.06.030

  3. [3]

     H. Rodhe, Science 248 (1990) 1217–1219. doi: 10.1126/science.248.4960.1217

  4. [4]

     D.J. Wuebbles, K. Hayhoe, Earth Sci. Rev. 57 (2002) 177–210. doi: 10.1016/S0012-8252(01)00062-9

  5. [5]

     R.W. Howarth, R. Santoro, A. Ingraffea, Clim. Change 106 (2011) 679–690. doi: 10.1007/s10584-011-0061-5

  6. [6]

     A. Konieczny, K. Mondal, T. Wiltowski, P. Dydo, Int. J. Hydrog. Energy 33 (2008) 264–272.

  7. [7]

     U.P.M. Ashik, W.M.A. Wan Daud, H.F. Abbas, Renew. Sustain. Energ. Rev. 44 (2015) 221–256. doi: 10.1016/j.rser.2014.12.025

  8. [8]

     L. Yuliati, H. Yoshida, Chem. Soc. Rev. 37 (2008) 1592–1602. doi: 10.1039/b710575b

  9. [9]

     X. Li, C. Wang, J. Tang, Nat. Rev. Mater. 937 (2022) 1–16.

  10. [10]

      N. Spinner, W.E. Mustain, J. Electrochem. Soc. 160 (2013) F1275–F1281. doi: 10.1149/2.071311jes

  11. [11]

      R.S. Kim, Y. Surendranath, ACS Cent. Sci. 5 (2019) 1179–1186. doi: 10.1021/acscentsci.9b00273

  12. [12]

      J. Jang, K. Shen, C.G. Morales-Guio, Joule 3 (2019) 2589–2593. doi: 10.1016/j.joule.2019.10.004

  13. [13]

      M. Merkx, D.A. Kopp, M.H. Sazinsky, et al., Angew. Chem. Int. Ed. 40 (2001) 2782–2807. doi: 10.1002/1521-3773(20010803)40:15<2782::AID-ANIE2782>3.0.CO;2-P

  14. [14]

      R. Balasubramanian, S.M. Smith, S. Rawat, et al., Nature 465 (2010) 115–119. doi: 10.1038/nature08992

  15. [15]

      F.E. Zilly, J.P. Acevedo, W. Augustyniak, Angew. Chem. Int. Ed. 50 (2011) 2720–2724. doi: 10.1002/anie.201006587

  16. [16]

      R. Zhang, L. Song, Y. Wang, Appl. Surf. Sci. 258 (2012) 7154–7160. doi: 10.1016/j.apsusc.2012.04.020

  17. [17]

      H.B. Ma, T. Sheng, W.S. Yu, et al., ACS Catal. 9 (2019) 10159–10165. doi: 10.1021/acscatal.9b02738

  18. [18]

      H. Yoshida, K. Hirao, J. Nishimoto, et al., J. Phys. Chem. C 112 (2008) 5542–5551.

  19. [19]

      A. Rategarpanah, F. Meshkani, Y. Wang, et al., Energy Convers. Manag. 166 (2018) 268–280. doi: 10.1016/j.enconman.2018.04.033

  20. [20]

      D. Hosseini, P.M. Abdala, F. Donat, et al., Appl. Catal. B: Environ. 258 (2019) 117946. doi: 10.1016/j.apcatb.2019.117946

  21. [21]

      P. Xie, T. Pu, A. Nie, et al., ACS Catal. 8 (2018) 4044–4048. doi: 10.1021/acscatal.8b00004

  22. [22]

      D. Yan, J. Chen, H. Jia, Angew. Chem. Int. Ed. 59 (2020) 13562–13567. doi: 10.1002/anie.202004929

  23. [23]

      S. Ding, H.A. Chen, O. Mekasuwandumrong, et al., Appl. Catal. B: Environ. 281 (2021) 119471. doi: 10.1016/j.apcatb.2020.119471

  24. [24]

      X. Zhang, Z. Li, W. Pei, et al., ACS Catal. 12 (2022) 3634–3643. doi: 10.1021/acscatal.1c05695

  25. [25]

      L. Liu, S. Das, Z. Zhang, et al., ACS Appl. Mater. Interfaces 14 (2022) 5363–5375. doi: 10.1021/acsami.1c21550

  26. [26]

      Y. Gao, D. Yan, C. Wang, et al., ACS Appl. Nano Mater. 5 (2022) 2688–2698. doi: 10.1021/acsanm.1c04363

  27. [27]

      H. Jeong, S. Shin, H. Lee, ACS Nano 14 (2020) 14355–14374. doi: 10.1021/acsnano.0c06610

  28. [28]

      P. Liu, Y. Zhao, R. Qin, et al., Science 352 (2016) 797–800. doi: 10.1126/science.aaf5251

  29. [29]

      S.T. Hunt, M. Milina, A.C. Alba-Rubio, et al., Science 352 (2016) 974–978. doi: 10.1126/science.aad8471

  30. [30]

      X. Yang, Y. Zhang, Z. Fu, et al., ACS Appl. Mater. Interfaces 12 (2020) 28206–28216. doi: 10.1021/acsami.0c06174

  31. [31]

      K. Sasaki, N. Marinkovic, H.S. Isaacs, R.R. Adzic, ACS Catal. 6 (2015) 69–76.

  32. [32]

      D.V. Esposito, S.T. Hunt, Y.C. Kimmel, J.G. Chen, J. Am. Chem. Soc. 134 (2012) 3025–3033. doi: 10.1021/ja208656v

  33. [33]

      Y. Zhou, J.M. Perket, J. Zhou, J. Phys. Chem. C 114 (2010) 11853–11860.

  34. [34]

      H.P. Steinruck, F. Pesty, L. Zhang, T.E. Madey, Phys. Rev. B 51 (1995) 2427–2439. doi: 10.1103/PhysRevB.51.2427

  35. [35]

      S. Bonanni, K. Aït-Mansour, H. Brune, W. Harbich, ACS Catal. 1 (2011) 385–389. doi: 10.1021/cs200001y

  36. [36]

      H.B.R. Lee, S.F. Bent, Chem. Mater. 24 (2012) 279–286. doi: 10.1021/cm202764b

  37. [37]

      T.Y. Chang, Y. Tanaka, R. Ishikawa, et al., Nano Lett. 14 (2014) 134–138. doi: 10.1021/nl403520c

  38. [38]

      S. Roberts, R. Gorte, J. Phys. Chem. 95 (1991) 5600–5604. doi: 10.1021/j100167a043

  39. [39]

      V. Pramhaas, M. Roiaz, N. Bosio, et al., ACS Catal. 11 (2021) 208–214. doi: 10.1021/acscatal.0c03974

  40. [40]

      S. Roberts, R. Gorte, J. Chem. Phys. 93 (1990) 5337–5344. doi: 10.1063/1.459653

  41. [41]

      E. Altman, R. Gorte, J. Catal. 110 (1988) 191–196. doi: 10.1016/0021-9517(88)90309-0

  42. [42]

      Z. Li, Y. Xiao, P.R. Chowdhury, et al., Nat. Catal. 4 (2021) 882–891. doi: 10.1038/s41929-021-00686-y

  43. [43]

      G. Kresse, J. Furthmüller, Phys. Rev. B 54 (1996) 11169. doi: 10.1103/PhysRevB.54.11169

  44. [44]

      G. Kresse, D. Joubert, Phys. Rev. B 59 (1999) 1758.

  45. [45]

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

  46. [46]

      S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 132 (2010) 154104. doi: 10.1063/1.3382344

  47. [47]

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

  48. [48]

      M. Nolan, S.C. Parker, G.W. Watson, Surf. Sci. 595 (2005) 223–232. doi: 10.1016/j.susc.2005.08.015

  49. [49]

      G. Henkelman, A. Arnaldsson, H. Jónsson, Comput. Mater. Sci. 36 (2006) 354–360. doi: 10.1016/j.commatsci.2005.04.010

  50. [50]

      P.E. Blöchl, Phys. Rev. B 50 (1994) 17953. doi: 10.1103/PhysRevB.50.17953

  51. [51]

      L. Tao, Y.L. Shi, Y.C. Huang, et al., Nano Energy 53 (2018) 604–612. doi: 10.1016/j.nanoen.2018.09.013

  52. [52]

      J. Li, Y. Tang, Y.Y. Ma, et al., ACS Appl. Mater. Interfaces 10 (2018) 38134–38140. doi: 10.1021/acsami.8b15585

  53. [53]

      J.H. Li, Z.Q. Liu, D.A. Cullen, et al., ACS Catal. 9 (2019) 11088–11103. doi: 10.1021/acscatal.9b03113

  54. [54]

      Q.Y. Wang, Y.X. Li, A. Serrano-Lotina, et al., J. Am. Chem. Soc. 143 (2021) 196–205. doi: 10.1021/jacs.0c08640

  55. [55]

      J. Vecchietti, P. Pérez-Bailac, P.G. Lustemberg, et al., ACS Catal. 12 (2022) 10482–10498. doi: 10.1021/acscatal.2c02117

  56. [56]

      Y.Y. Lin, Z.L. Wu, J.G. Wen, K.R. Peoppelmeier, L.D. Marks, Nano Lett. 14 (2014) 191–196. doi: 10.1021/nl403713b

  57. [57]

      B. Loenders, Y. Engelmann, A. Bogaerts, et al., J. Phys. Chem. C 125 (2021) 2966–2983. doi: 10.1021/acs.jpcc.0c09849

  58. [58]

      Y. Cao, Y. Yang, W. Yu, et al., ACS Appl. Mater. Interfaces 14 (2022) 13344–13351. doi: 10.1021/acsami.1c25203

  59. [59]

      Z. Liu, E. Huang, I. Orozco, et al., Science 368 (2020) 513–517. doi: 10.1126/science.aba5005

  60. [60]

      M. Belgued, P. Pareja, A. Amariglio, H. Amariglio, Nature 352 (1991) 789–790. doi: 10.1038/352789a0

  61. [61]

      W. Pei, S. Zhou, J. Zhao, et al., Nano Energy 76 (2020) 105049. doi: 10.1016/j.nanoen.2020.105049

• 

  Figure 1  Geometrical structures of Pt/TMOs hybrid systems. Atomic structures of Pt/TMOs (a-d) and corresponding differential charge density distributions (e-h). White lines labeled the lattice. The Pt, O, Ti, Zr, and Ce atoms are shown by dark blue, pink, green, purple and blue balls, respectively.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Kinetic process of CH₄ conversion and H₂O dissociation. Energy diagrams of (a) CH₄ activation and (b) H₂O dissociation on Pt/TMOs. The reaction heat (ΔH), kinetic barriers (E^a), and the corresponding structure configurations are given in insets. Histogram of reaction energies and kinetic barriers for (c) CH₄ → CH₃^* + H^*, (d) H₂O → OH^* + H^*, (e) CH₃^* + OH^* → CH₃OH and (f) CH₃^*→ CH₂^* + H^* processes on various Pt/TMOs systems and bulk-phase Pt(111).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  The reaction pathway of CH₄ conversion. (a) The reaction pathway of CH₄ conversion on Pt/CeO₂ (Ce), and the corresponding structure configurations are given in insets. (b) The reaction pathway of CH₄ conversion on other Pt/TMOs systems and bulk-phase Pt(111) surface. The color number is the kinetic barrier for each reaction step.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  The further dehydrogenation processes on Pt/CeO₂ (O). The colored number is the kinetic barrier for each reaction step.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  The orbital interaction of CH₄ adsorbed on Pt/TMOs. (a-d) Local density of states (DOS) of (from left to right) gas-phase and adsorbed CH₄ molecule, and Pt/TMOs. (e) Differential charge density between Pt overlayer and CH₄. (f) The relationship between the kinetic barriers (E^a) and Δε.

  下载: 全尺寸图片 幻灯片

  Table 1.  The key parameters of Pt/TMOs hybrid systems including lattice mismatch (δ), interlayer distance (d), vertical buckling of Pt monolayer (Δd_Pt), Pt-Pt bond length (h_Pt-Pt), formation energy per Pt atom in Pt monolayer (E_form), charge transfer between Pt monolayer and the substrate (CT), the d-band center of Pt monolayer (ε_d).

  下载: 导出CSV
•