AVS1996 Session SS-WeM: Reactions on Model Catalysts

Wednesday, October 16, 1996 8:20 AM in Room 203B

Wednesday Morning

Start	Invited?	Item
8:20 AM	Invited	SS-WeM-1 Kinetics and Mechanism of Acetylene Cyclotrimerization: LITD/FTMS and FTRAIRS Studies I. Abdelrehim, T. Caldwell, D. Hunka, D. Land (University of California, Davis) Acetylene cyclization reactions on Pd(111) in UHV will be reviewed with emphasis on recent results using laser-induced thermal desorption/FT mass spectrometry (LITD/FTMS) and FT reflection absorption infrared spectroscopy (FTRIRS) to elucidate the mechanism and factors controlling the rate of reaction. Laser-induced thermal desorption studies indicate that C\sub 4\H\sub 4\ dimers form at low temperatures with addition of a third acetylene to form adsorbed benzene being rate determining and becoming appreciable above 150 K for near-monolayer acetylene coverages. The activation barrier and preexponential factor for this process are strongly dependent on the initial acetylene coverage, with both parameters increasing dramatically with increased coverage. This argues against a diffusion-limited process or steric blocking and implies that adsorbate-metal bond breaking is the predominant factor affecting the reaction rate. This is supported by the picture of bonding between acetylene and Pt(111) developed by R. Hoffmann using an extended-H\um u\ckel approach, and work function change measurements by Ormerod and Lambert, where increased acetylene coverage should lead to stronger adsorbate-surface binding. The effects of coadsorbed S and O and island formation on the benzene formation kinetics will also be addressed, as will the formation of thiophene and furan. For example, oxygen islands cause acetylene to concentrate onto the clean Pd(111) patches, as evidenced by effects on the kinetic parameters consistent with the degree of O coverage.
9:00 AM		SS-WeM-3 Steady-State Catalytic C-C Bond Formation on Reduced TiO\sub 2\ Surfaces V. Lusvardi, K. Pierce, M. Barteau (University of Delaware) This work focuses on the cyclotrimerization of alkynes on reduced TiO\sub 2\ surfaces. TPD experiments on reduced TiO\sub 2\ surfaces have demonstrated that alkynes are converted to the corresponding aromatic products with high selectivity. Utilizing X-ray Photoelectron Spectroscopy (XPS) studies of reduced surfaces, the active site for the cyclotrimerization was identified as Ti cations in the +2 oxidation state. This reaction represents the first example of the catalytic assembly of carbon-carbon bonds on a metal oxide surface in ultra high vacuum. Although the catalytic formation of carbon-carbon bonds on single crystal surfaces at low pressures is a rarity, many important catalytic processes involve carbon-carbon bond formation, therefore it is worthwhile to consider how such reactions might be studied directly using the tools of surface science. Steady-state experiments involving the production of trimethylbenzene from methylacetylene at low pressure (10-9 - 10-8 mbar) conditions have demonstrated multiple turnovers of the catalyst and no significant catalyst deactivation at temperatures between 300 and 500 K. XPS was used to measure the C(1s) envelope during the steady-state reaction of methylacetylene, thus characterizing the reactant pool with an in-situ measurement of the adsorbates. The collection of XPS spectra under steady-state flow conditions over a single crystal is a novel application of a surface science tool, which can be applied in combination with mass spectroscopy, to characterize the populations of both adsorbed intermediates and desorbed products simultaneously. This information provided the opportunity to develop and test models of the reaction sequence and the resulting kinetics. A four-step model is proposed, which contains three non-activated steps for alkyne adsorption/insertion to form the aromatic, followed by the final step, aromatic desorption. This model captures the measured temperature and pressure dependences of the reaction rate, and is consistent with the adsorbate coverages measured with steady-state XPS.
9:20 AM		SS-WeM-4 Low Temperature STM Study of Adsorbed Molecules on Pd(111) S. Behler, M. Rose, F. Ogletree, M. Salmeron (Lawrence Berkeley National Laboratory) We have studied the adsorption of C\sub 4\H\sub 4\S, C\sub 2\H\sub 2\ and C\sub 6\H\sub 6\ on Pd(111) with a Scanning Tunneling Microscope (STM). The STM operates at variable sample temperature ranging from 20 to 300 K. Cryogenic cooling is needed to immobilize these small molecules on a substrate surface. This allows us to study single molecules at coverages of < 0.1% of a monolayer. The apparent shape of a molecule in an STM image gives us the ability to distinguish between the different adsorbates. Moreover, the apparent shape is also analyzed by theoretical image calculations.
10:00 AM		SS-WeM-6 Observation of Anomalous Surface Reactivities of Ni/Pt(111) Bimetallic Model Catalysts B. Fr\um u\hberger, J. Chen (Exxon Research and Engineering Company) The catalytic performance of a transition metal can often be improved by alloying with a second metal component. In order to understand the underlying relationship between alloy formation and reactivities, we have carried out a comparative investigation of the reactivities of Pt(111), Ni(111) and the Ni/Pt(111) bimetallic surfaces using HREELS, TPD, AES, XPS and LEED. The Ni/Pt surfaces were prepared by thermal evaporation of Ni onto Pt(111), which showed an epitaxial growth at 600 K. The different surface reactivities were compared by studying two probing reactions, the structure-insensitive dehydrogenation of cyclohexene and the structure-sensitive decomposition of ethylene. Similar to previous studies, we found that both Pt(111) and the epitaxial multilayer Ni(111)/Pt(111) surfaces were very reactive towards the decomposition of cyclohexene and ethylene. For example, the dehydrogenation of cyclohexene occurred on both Pt(111) and Ni(111)/Pt(111), producing benzene surface intermediates at 400 and 200 K, respectively. In addition, the decomposition of ethylene was observed on both Pt(111) and Ni(111)/Pt(111), with the dominant surface intermediates at 300 K being ethylidyne and acetylene, respectively. Very surprisingly, we found that the Ni/Pt(111) surfaces with submonolayer Ni coverages were essentially inert towards the decomposition of ethylene. We also found that the degree of cyclohexene dehydrogenation on the submonolayer Ni/Pt(111) bimetallic surfaces was at least five times less than that on either Pt(111) or Ni(111)/Pt(111). We will attempt to explain the unusual reactivities of Ni/Pt(111) surfaces by the formation of Ni-Pt bonds.
10:20 AM		SS-WeM-7 The Catalytic Chemistry of Small Hydrocarbons on Oxygen-modified Molybdenum. W. Tysoe (University of Wisconsin, Milwaukee) The relative activity for the metathesis of olefins is investigated using various model oxides prepared in an ultrahigh vacuum chamber and tested using an isolatable catalytic reactor where it is shown that both MoO\sub 2\ and MoO\sub 3\ provide the most active catalyst for reaction above ~650 K. It is also shown that the activation energy for reaction above this temperature is ~65 kcal/mol, much higher than that found for the commercial catalyst. However, another reaction regime is found below ~650 K where the activation energy decreases to ~6 kcal/mol and the absolute activity is close to that for alumina-supported molybdena. Reaction of ethylene at high temperatures shows the formation of higher hydrocarbons (up to C\sub 8\) where the product yield is well described by a Schulz-Flory distribution. The reaction is shown to proceed in the presence of a carbonaceous layer which is shown, using Raman spectroscopy, to consist primarily of adsorbed hydrocarbons. Oxygen overlayers also affect the catalytic activity and the chemistry of small hydrocarbons on oxygen-covered Mo(100) are used to gain insights into the catalytic activity at high pressures.
10:40 AM		SS-WeM-8 Tailoring Alumina Surface Chemistry for Efficient Use of Supported MoS\sub 2\ A. Sault (Sandia National Laboratories); J. Reardon, A. Datye (University of New Mexico) A wide variety of molybdenum oxide species can be formed on \gamma\-alumina [1]. The reducibility of these species varies widely and affects the formation of MoS\sub 2\ during sulfiding [2]. On \gamma\-alumina molybdates first react with the most acidic OH groups, associated with tetrahedral Al [3], to form isolated MoO\sub 4\\super 2-\2E. As loading increases, less acidic hydroxyls are consumed and polymeric molybdate species begin to form, ultimately resulting in MoO\sub 3\. We propose that increases in the reducibility of adsorbed molybdate with loading result in a maximum in the HDS specific activity of MoS\sub2\/\gamma\-alumina catalysts with loading. At low loadings Mo is present as isolated MoO\sub 4\\super 2-\ that is not reduced to MoS\sub 2\. As loading increases, a greater fraction of the Mo is bound in sites favorable to MoS\sub 2\ formation, and activity increases. At high loadings growth of MoS\sub 2\ platelets decreases the fraction of Mo atoms located at active edge sites and activity declines. If this hypothesis is true, then increasing the activity of MoS\sub 2\/alumina catalysts should be possible if the tetrahedral Al sites can be eliminated. Indeed, \alpha\-alumina contains no tetrahedral Al sites and HDS activity on MoS\sub 2\/\alpha\-alumina exhibits no maximum and greater activity at all loadings than MoS\sub 2\/\gamma\-alumina. Titania deposited on \gamma\-alumina also eliminates the activity maximum by titrating hydroxyl groups associated with tetrahedral Al. XPS, TEM, and FTIR results confirm that the activity trends are due to alteration of the chemical interaction of molybdates with the surface and not to morphological effects proposed elsewhere [4] (e.g., bookend vs. basal plane MoS\sub 2\ morphology). 1. D. S. Zingg, et al., J. Phys. Chem., 84, (1980) 2898. 2. C. P. Li and D. M. Hercules, J. Phys. Chem., 88, (1984) 456. 3. H. Knozinger and P. Ratnasamy, Catal. Rev. Sci. Eng., 17, (1978) 173. 4. K. C. Pratt, et al., J. Catal., 124, (1990) 416.
11:00 AM		SS-WeM-9 A Working Surface Science Model of CoMoS Hydrodesulfurization Catalysts A. De Jong, V. De Beer, J. Niemantsverdriet (Eindhoven University of Technology, The Netherlands) Model catalysts, consisting of a conducting substrate with a thin SiO\sub 2\ or Al\sub 2\O\sub 3\ layer on top of which the active catalytic phase is deposited, are advantageous in the study of catalyst preparation and fundamental processes on the catalyst surface. Such model systems are applied on the study of the sulfidation of CoMoS catalysts. CoMoS, the highly active cobalt promoted MoS\sub 2\ in which Co is thought to decorate the edges of MoS\sub 2\ slabs, can be synthesized by sulfiding a nitrilo triacetic acid (NTA) complex of cobalt and molybdenum. The Co-Mo-NTA complex was deposited on an SiO\sub 2\/Si(100) and Al\sub 2\O\sub 3\/Si(100) model support by spincoating. XPS measurements on these Co-Mo catalysts in different stages of sulfidation reveal that the molybdenum is the first to form sulfides, while cobalt starts to form sulfides when most of the molybdenum is in the sulfidic form. This is in contrast to measurements on Co catalysts, which indicate that cobalt forms sulfides at lower temperatures than molybdenum does. This picture agrees well with the concept of CoMoS being a phase in which cobalt decorates the edges of MoS\sub 2\ particles. Thiophene hydrodesulfurization studies of the CoMoS model catalysts yielded activities and product distributions consistent with those obtained from high surface area CoMoS catalysts, providing proof that these model catalysts are realistic as a model for the CoMoS catalysts. Hence, these model catalysts offer great potential for fundamental surface science studies of the catalytic phases and of adsorption, desorption and reactions of gases as well.