8751
2008, 112, 8751–8753 Published on Web 05/27/2008
Oxidation of Hydrocarbons at Surface Defects: Unprecedented Confirmation of the Oxomethylidyne Pathway on a Stepped Rh Surface Tanushree Bhattacharjee,*,‡ Oliver R. Inderwildi,§ Stephen J. Jenkins,§ Uwe Riedel,‡ and Jürgen Warnatz†,‡ Interdisciplinary Center for Scientific Computing, UniVersity of Heidelberg, Im Neuenheimer Feld 368, D-69120, Heidelberg, Germany, and the Department of Chemistry, UniVersity of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom ReceiVed: March 27, 2008
The pathway of methylidyne (CH) oxidation on Rh{211} was studied using plane wave density functional theory (DFT) slab calculations and microkinetic simulations. We present ample evidence that the direct decomposition and oxidation mechanism is not the main reaction route to sythesis gas especially during light-off of catalysts. We characterize the transition state for the surface oxidation of CH to oxomethylidyne (CHO) species and subsequent decomposition to CO and H as products to establish the minimum energy pathway. Increasing concern about society’s dependence on energy provision by combustion of fossil fuels, the associated emission of CO2, and consequent climate change1 has generated interest in the use of natural gas as precursor for carbon-neutral biofuels.2 Natural gas (mainly methane) can now a days be converted into synthesis gas (carbon monoxide and hydrogen) mainly by steam reforming (SR) or catalytic partial oxidation (CPO) using transition metal based catalysts loaded, for instance with Rh.3–5 The CPO process for synthesis gas production is less energyintensive than SR since it is an exothermic process and results in a hydrogen to carbon monoxide ratio of 2:1, which is most suitable for most downstream processes.6,7 Owing to those reasons, the CPO is a noteworthy method that can drastically reduce the production costs of synthesis gas and consequently make one step in the synthesis of biofuels more efficient. Simulations of CPO give good agreement with experiment in the high-temperature steady state phase; however, in transient simulations, the light-off of the catalysts cannot be described as accurately.8 These simulations are based on the classic decomposition and subsequent oxidation mechanism, which is likely at high temperature but rather unlikely at lower temperature, i.e., during the light off of the catalyst. The present study provides a deeper insight into the mechanistic details for the catalytic oxidation of methylidyne (CH) on a stepped Rh{211} surface, following a study on the planar Rh{111} surface. Reactions of methylidyne are believed to be one of the rate determining steps in the catalytic oxidation and synthesis of hydrocarbons.9,10 The focus of this study is the CH oxidation process on a stepped rhodium surface, which is important because steps offer adsorption and reaction sites at undercoordinated surface atoms.11 Various theoretical and * To whom correspondence should be addressed. E-mail:
[email protected]. Fax: +49 6221 54 8884. † Deceased. ‡ University of Heidelberg. § University of Cambridge.
10.1021/jp8026594 CCC: $40.75
experimental findings suggest that reactants and products are more strongly bound at such sites resulting in a concomitantly lower activation barrier to reaction.12–14 We have used periodic DFT as implemented in CASTEP15 to determine accurate values for the kinetic barriers of reactions and diffusion processes and subsequently use these parameters in microkinetic simulation utilizing DETCHEM16. The precise setup used for the calculations can be found elsewhere.17 Previous studies of CH on Rh{111} surface18,19 considered CH decomposition and subsequent oxidation of atomic carbon by oxygen. A recent study indicates that there is an alternative pathway for the catalytic oxidation of methylidyne on Rh{111} surface.20 The fundamental question which still remains open is if defects such as steps alter the minimum energy pathway. We therefore carried out an analogous study on the Rh{211} surface. The calculated equilibrium structures of the initial and final states as well as the transition state structures of the oxidation of methylidyne are shown in Figure 1 as an energy diagram. CH adsorbing at a step edge is found to be more stable than on the terrace. The most stable adsorption site for CH is the 4-fold hollow site, 4f. Dissociation of CH to atomic carbon and hydrogen is unlikely from 4f due to a rather high dissociation barrier of 1.04 eV and after dissociation the products could additionally undergo subsurface penetration, as indicated by calculations on (1 × 2) elementary cell. The next most stable site for CH adsorption is fcc1 and hcp1 site. We find that the activation barrier for CH dissociation from fcc1 is 0.61 eV and from hcp1 is 1.07 eV. However, the diffusion barrier of CH from hcp1 to 4f is 0.65 eV, and surprisingly, the barrier from fcc1 to 4f is merely 0.05 eV. This low barrier will trigger fast diffusion of CH from the fcc1 to the 4f position and will consequently hinder the CH dissociation from fcc1 position. CH(4f) is the more stable species, and this site will therefore be more populated and hence there will be a dynamic interplay of reaction and diffusion processes. Oxidation of CH by atomic oxygen leading to an oxomethylidyne (CHO) species is activated 2008 American Chemical Society
8752 J. Phys. Chem. C, Vol. 112, No. 24, 2008
Letters
Figure 2. Equilibrated conversion of CH on Rh{211} as a function of temperature (top) and formation of CO as a function of time at 350 K.
Figure 1. (A) Adsorption sites of Rh{211}. The labels are as follows: 4f, 4-fold hollow; seb, step edge bridged; se, step bridged. (B) Energy diagram for surface oxidation of the CH species (reactant A) to CHO (product B) via transition state (TS) and decomposition of CHO to CO and H (product C) via TS. Rhodium is shown as turquoise, oxygen as red, hydrogen as white, and carbon as gray. (C) Transition state (TS) geometries. Hollow arrows indicate the direction of movement of atoms after TS.
by 0.75 eV. The reaction is also found to be thermochemically favorable with energy of reaction of -0.62 eV. It should be mentioned that the CHO species has been inferred experimentally as a combustion intermediate on Pt{110}.21 The CHO species formed by oxidation is bound to the surface via the carbon and oxygen atom. This species can subsequently undergo decomposition to adsorbed CO and H, and the barrier of this reaction is found to be 0.17 eV, considerably lower than the direct dissociation of CH. The reaction is exothermic (-0.90 eV) and so the reaction is kinetically as well as thermodynamically more favorable than decomposition of CH. Our DFT results strongly suggest that CH diffuses to the 4f position very rapidly rather than to dissociate from the fcc1 position and hence we assume that the oxidation of CH(4f) is the main pathway for oxidation via CHO species. At a step the dissociation barrier is merely 50% of the barrier on Rh{111}. The two main reasons for this is that (i) the step is able to bind the transition state geometry more strongly than a flat surface, hence lowering the barrier, and (ii) the atoms at the step are under coordinated which
also enhance the stabilization of the transition state, which also lowers the barriers. In order to determine the main reaction pathway, the conversion of CH on the Rh{211} surface was modeled including all reaction and diffusion barriers determined by DFT calculations, and pre-exponential factors were estimated using transition state theory. We assume a starting coverage of 0.25 ML CH and 0.50 ML oxygen on the surface. At low temperature (
