Elsevier

Surface Science

Volume 614, August 2013, Pages 53-63
Surface Science

A first-principles study of CO hydrogenation into methane on molybdenum carbides catalysts

https://doi.org/10.1016/j.susc.2013.04.001Get rights and content

Highlights

  • DFT calculations were performed to study CO hydrogenation to methane on Mo2C.

  • Reaction mechanism is CO  HCO  H2CO  H2COH  CH2  CH3  CH4 on Mo2C.

  • CH3 + H  CH4 is the rate-determining step.

  • Barrier of CH3 + H  CH4 on fcc-Mo2C (100) is lower than on hcp-Mo2C (101).

Abstract

The reaction mechanisms for the CO hydrogenation to produce CH4 on both fcc-Mo2C (100) and hcp-Mo2C (101) surfaces are investigated using density functional theory calculations with the periodic slab model. Through systematic calculations for the mechanisms of the CO hydrogenation on the two surfaces, we found that the reaction mechanisms are the same on both fcc and hcp Mo2C catalysts, that is, CO  HCO  H2CO  H2COH  CH2  CH3  CH4. The activation energy of the rate-determining step (CH3 + H  CH4) on fcc-Mo2C (100) (0.84 eV) is lower than that on hcp-Mo2C (101) (1.20 eV), and that is why catalytic activity of fcc-Mo2C is higher than hcp-Mo2C for CO hydrogenation. Our calculated results are consistent with the experimental observations. The activity difference of these two surfaces mainly comes from the co-adsorption energy difference between initial state (IS) and transition state (TS), that is, the co-adsorption energy difference between IS and TS is − 0.04 eV on fcc Mo2C (100), while it is as high as 0.68 eV on hcp Mo2C (101), and thus leading to the lower activation barrier for the reaction of CH3 + H  CH4 on fcc-Mo2C (100) compared to that of hcp-Mo2C (101).

Introduction

Molybdenum carbide (Mo2C) has exhibited excellent catalytic properties similar to those of more expensive noble metals, and has been widely used in hydrodesulfurization (HDS) and hydrodenitrogenation (HDN) processes and also exhibits high activity for methanation, alkane isomerization, and CO hydrogenation [1], [2], [3], [4], [5]. In particular, CO hydrogenation on Mo2C catalysts has been studied extensively because of the tolerance of these catalysts to sulfur poisoning. There are two main crystalline structures Mo2C, i.e., the orthorhombic phase (fcc-Mo2C) and the hexagonal phase (hcp-Mo2C). Ranhotra and Reimer [6] identified the main product of CO hydrogenation on the two phase (fcc and hcp) Mo2C catalysts that is methane, interestingly, the activity of fcc-Mo2C for CO hydrogenation is twofold higher than that of hcp-Mo2C. Patterson at el. [7] demonstrated that methane and CO2 were found to be the main products of CO hydrogenation over fcc-Mo2C catalyst. Christensen at el. [8] reported that the use of the bulk Mo2C as catalyst is the effective conversion of syngas (CO and H2) into higher alcohols. Furthermore, other workers also studied the CO hydrogenation over unsupported and supported Mo2C catalysts. Lee at el. [9], [10] studied the CO hydrogenation on alkali metal modified Mo2C, the products are C1–5 alkanes without the assistance of potassium, in which methane is the main product. On potassium modified surface, mixture of lower alcohols, especially the selectivity toward C2–7 alcohol is significantly improved. Woo et al. [11] and Xiang et al. [12] reported that the synthesis for mixed alcohol from CO hydrogenation on Mo2C produced hydrocarbons and by adding potassium as a promoter, the selectivity shifts from hydrocarbons to alcohols, especially methanol and ethanol. Although the interactions of CO and H on MoC2 were well-studied [13], [14], [15] and there are many experimental investigations that focus on hydrogenation reactions from syngas on Mo2C surfaces that has been done, [16], [17] there are very few systematic theoretical studies on CO hydrogenation mechanism over Mo2C surfaces that have been reported, [18] CO hydrogenation mechanism on Mo2C catalysts is still unclear.

Identifying the mechanism of CO hydrogenation is essential for the design of better transition metal carbides catalyst that they might be used as a cheaper substitute for noble metal catalysts. Although Fischer–Tropsch synthesis has been widely studied since 1923, [19] there is still controversy about the CO hydrogenation mechanism, which involves formation and scission of Csingle bondH, Osingle bondH, and Csingle bondO bonds. The hydrogenation mechanism is rather complicated and various reaction schemes have been proposed. For instance, using density functional theory (DFT) methods, Choi and Liu [20] investigated the reaction of CO with H on Rh (111) and proposed that the reaction pathway is CO hydrogenation to form CH3O, then Csingle bondO bond of CH3O breaks, and finally forms CH4 by CH3 hydrogenation. Huang and Cho [21] calculated CO hydrogenation on MoS2 surface and found that CO hydrogenation firstly produces CHxO, followed by Csingle bondO bond breaking and CH4 formation. On the non-stoichiometric MoSx surfaces, Shi et al. [22] simulated CO hydrogenation on Mo-termination with 42% S coverage and S-termination with 50% S coverage, and identified CH2OH as the prior intermediate for Csingle bondO bond breaking and CH4 formation.

On the basis of above discussed reaction mechanisms, all intermediates and the elementary steps of CH4 formation from CO hydrogenation on periodic slab models of hcp and fcc Mo2C has been calculated by DFT methods. Based on the calculated results, we intend to answer the following questions: (1) What is the order for CO hydrogenation process on the two phase Mo2C surfaces? (2) When is the Csingle bondO bond breaking of CHxOy on the two surfaces? (3) Why does fcc-Mo2C have the higher catalytic activity for CO hydrogenation than hcp-Mo2C? This paper is organized as follows: Section 2 describes model geometries and computational methods while Section 3 presents the calculated results and discussion. Finally, we summarize our conclusions in Section 4.

Section snippets

Computational method and models

The DFT calculations are performed by using the Vienna Ab initio Simulation Package (VASP) [23], [24]. The exchange-correlation energy and potential are described by generalized gradient approximation (PW91) [25]. The electron–ion interaction is described by the projector-augmented wave (PAW) method, [26], [27] and the electronic wave functions are expanded by plane waves up to a kinetic energy of 350 eV. The surface Brillouin zone is sampled using a 4 × 4 × 1 Monkhorst–Pack mesh [28]. Our

Adsorption properties of CO and possible reaction intermediates

In this section, we present a detailed investigation of stable adsorption of all possible intermediates involved in the processes of CO hydrogenation on fcc and hcp Mo2C surfaces. The corresponding configurations are shown in Fig. 2, and the geometric and energetic information is given in Table 1.

Thermodynamical trends in CO hydrogenation

Brönsted–Evans–Polanyi (BEP) correlation is useful in the estimation of activation energy based on the thermodynamical properties. This relationship is previously used for a given reaction like O2 dissociation on different metals, [42], [43], [44], [45], [46], [47], [48] and later is applied to the case of series bonds cleavage reactions on a given metal, such as Csingle bondH/Osingle bondH/Csingle bondO bond scission of ethanol on Pt (111) [49]. Additionally, the BEP-relation is usually applied to the case of bond cleavage

Conclusions

The present theoretical studies give a clear conclusion of the adsorption of reaction intermediates appearing during CO hydrogenation on fcc-Mo2C (100) and hcp-Mo2C (101) as well as of the elementary reaction steps. The calculated adsorption energies of the species on the two surfaces show highly localized strong adsorbate–molybdenum interaction for all adsorbates. Furthermore, DFT studies on possible reaction steps during CO hydrogenation at the two phase Mo2C surfaces are used to identify

References (57)

  • M. Saito et al.

    J. Catal.

    (1980)
  • J.S. Lee et al.

    Catal. Lett.

    (1993)
  • G.S. Ranhotra et al.

    J. Catal.

    (1987)
  • P.M. Patterson et al.

    Appl. Catal. A Gen.

    (2003)
  • H.C. Woo et al.

    Appl. Catal.

    (1991)
  • M.L. Xiang et al.

    Catal. Today

    (2008)
  • H. Shou et al.

    J. Catal.

    (2013)
  • C. Pistonesi et al.

    Catal. Today

    (2012)
  • H. Tominaga et al.

    Appl. Catal. A Gen.

    (2012)
  • H. Shou et al.

    J. Catal.

    (2011)
  • A.J. Medford et al.

    J. Catal.

    (2012)
  • X.R. Shi et al.

    J. Mol. Catal. A

    (2009)
  • G. Kresse et al.

    Comput. Mater. Sci.

    (1996)
  • J.R. Kitchin et al.

    Catal. Today

    (2005)
  • C. Pistonesi et al.

    Surf. Sci.

    (2008)
  • A. Michaelides et al.

    Surf. Sci.

    (1999)
  • V. Pallassana et al.

    J. Catal.

    (2000)
  • A. Logadottir et al.

    J. Catal.

    (2001)
  • J.K. Norskov et al.

    J. Catal.

    (2002)
  • J.J. Mortensen et al.

    J. Catal.

    (1999)
  • R. Alcalá et al.

    J. Catal.

    (2003)
  • B. Hammer

    Surf. Sci.

    (2000)
  • M. Boudart et al.

    Science

    (1973)
  • J.C. Schlatter et al.

    Ind. Eng. Chem. Res.

    (1988)
  • S. Ramanathan et al.

    J. Phys. Chem.

    (1995)
  • J.M. Christensen et al.

    Ind. Eng. Chem. Res.

    (2012)
  • J.S. Lee et al.

    J. Mol. Catal.

    (1990)
  • K.Y. Park et al.

    Catal. Lett.

    (1991)
  • Cited by (46)

    • Catalytic mechanism and selectivity prediction for syngas conversion over pure and K-promoted Mo<inf>2</inf>C catalysts

      2021, Applied Catalysis A: General
      Citation Excerpt :

      Competing steps from different reaction pathways exist for all hydrogenation products, and elucidation of the reaction network is important for the evaluation of product selectivity and yield. Our calculations show that the initial reaction step of CO activation is the formation of the HCO intermediate, where CO direct dissociation is unfavoured on the Mo2C surface due to its very high energy barrier, which agrees well with previous studies in the literature [24,55]. Furthermore, it appears that CH2 conversion is key to the selectivity of alcohols over Mo2C catalysts, as CH2 can be sequentially hydrogenated to form methane, or coupled with another CH2 to form ethylene, or hydrogenated and then coupled with HCO to form C2 oxygenates.

    View all citing articles on Scopus
    View full text