AVS1997 Session SS1-ThM: Reaction Kinetics at Surfaces

Thursday, October 23, 1997 8:20 AM in Room A3/4

Thursday Morning

Start	Invited?	Item
8:20 AM		SS1-ThM-1 Direct Observation of C₂N₂ Uptake and Dissociation on Pd(110) by Fast High Resolution XPS A. Baraldi, S. Lizzit (Sincrotrone Trieste, Italy); M.G. Ramsey, F.P. Netzer (Karl-Franzens-Universität Graz, Austria) The full parameter space of C₂N₂ adsorption and reaction on Pd(110) from 80 to 800K has been mapped out by fast high-resolution XPS performed at the Super-ESCA beamline of the ELETTRA synchrotron radiation facility. Both C 1s and N 1s core level emissions have been followed continuously during exposure and on ramping the temperature, thus allowing direct observation of the dissociation of adsorbed species. The fraction of C₂N₂ dissociated to CN increases from zero at 80K to 100% at 220K. Between these two extremes dissociation is favoured at low coverages with the amount of coexisting intact C₂N₂ species decreasing with temperature. The uptake kinetics appear different at the various substrate temperatures, mirroring the competition between dissociative and non-dissociative adsorption of the parent molecule. The dissociation in the saturated C₂N₂ monolayer occurs in a narrow temperature range, 190-220K. The CN surface species are stable up to the point of complete desorption at 800K. the growth of C₂N₂ molecular layers was followed at various substrate temperatures showing a different evolution of the multilayers, which reflects the rich temperature programmed desorption previously reported ¹. The elucidation of the complex behaviour of this dissociated system was made possible by the very fast data acquisition which allowed a continuous monitoring of the surface species and thus vividly demonstrates the possibilities of third generation light sources ¹M.G.Ramsey, G.Rosina, F.P.Netzer, H.B.Saalfeld and D.R.Lloyd, Surf. Sci. 217 (1989) 140
8:40 AM		SS1-ThM-2 Carbon Monoxide Adsorption and Oxidation on Ir(110) U. Burghaus, J, Ding, H.W. Weinberg (University of California, Santa Barbara) We report measurements of the initial adsorption probability S₀ and its coverage dependence S(Θ_CO) of CO on clean and oxygen-precovered Ir(110) for adsorption temperatures T < 300 K, i.e., below the onset of CO₂ formation. Additionally, measurements of CO oxidation by CO molecular beam titration of the surface precovered with atomically bonded oxygen are presented (T > 300 K). The impact energy (1.3 < E < 24 kcal/mol), the oxygen precoverage Θ, and the surface temperature (85 < T < 800 K) were varied. For the initially clean surface S₀ decreases linearly with increasing E, and is independent of T. Additionally, the shape of S(Θ_CO,E₀) points to nonactivated precursor-mediated adsorption kinetics for low E₀; for E > 20 kcal/mol direct adsorption dominates the shape of S(Θ_CO,E₀). For E > 3 kcal/mol, S₀(Θ_O) decreases almost linearly with Θ_O. However, for E₀ < 3 kcal/mol, S₀(Θ_O) remains constant at 0.92 +- 0.03 independent of Θ_O. The results suggest a physical site blocking mechanism. During the course of the titration (T > 300 K) the reaction product CO₂ and the backscattered CO intensity are detected. The CO₂ formation rate increases with increasing titration time and depends strongly on T. The ratio of the initial to the maximum CO₂ formation rate decreases slightly with increasing impact energy. The self-accelerating CO₂ formation kinetics will therefore be discussed semi-quantitatively by an LH-type model with coverage dependent reaction rate coefficient and adsorption probabilities.
9:00 AM	Invited	SS1-ThM-3 Non-uniform Reaction Rates in Surface Chemical Reactions Studied by STM J. Wintterlin, S. Voelkening, T. Janssens, G. Ertl (Fritz-Haber-Institut der Max-Planck-Gesellschaft, Germany) The Langmuir-Hinshelwood kinetics of surface reactions is usually described in the mean-field approximation, assuming a uniform reaction rate over the surface. This neglects interactions between the adsorbates which, quite in contrast to the underlying assumption, cause non-uniform adsorbate distributions, mostly leading to island formation. We have undertaken studies by means of variable temperature STM to reveal possible consequences of these effects on the macroscopic reaction rate. For the oxidation of CO on a Pt(111) surface small attractive interactions between the oxygen atoms lead to an exclusive reaction at boundaries between oxygen and CO covered areas, causing qualitative deviations of the reaction rate from the mean-field description. Monte-Carlo simulations were perfomed in addition which reproduce the experimental findings.
9:40 AM		SS1-ThM-5 The NO-H₂ Reaction over Rh Surfaces with (111) Terraces and Differing (100) Step Densities: XPS Measurements using Synchrotron Radiation. P.D. Cobden, B.E. Nieuwenhuys (Leiden University, The Netherlands); A. Baraldi, G. Comelli, F. Esch, S. Lizzit, M. Kiskinova (Synchrotron Trieste, Italy) The hysteresis in surface species that occurs during a heat-cool cycle of the NO-H₂ reaction over Rh(111), Rh(533) and Rh(311) was studied using X ray Photoelectron Spectroscopy (XPS), and performance enhancing synchrotron radiation. The high resolution available to this fast XPS technique indicated that NO was not present on the surface during the hysteresis on any of the three surfaces, and that there was a dynamic balance between several N-species, and O_ads on the surface. The uptake of NO and NH₃ on these surfaces was measured for a range of different temperatures. Differences with respect to the number of surface species identified and form of the hysteresis on each surface will be discussed.
10:00 AM		SS1-ThM-6 Probing the Role of Oxygen Coordination in Hydrocarbon Oxidation: Reactions of Methyl Radicals and Alcohols on Oxygen Covered Mo(110) K.T. Queeney, D.A. Chen, M.-S. Chen, P.G. Clark, D. Jentz, C.M. Friend (Harvard University) Molybdena (MoO₃) has been shown to be an effective catalyst for partial oxidation of hydrocarbons, an essential process in the formation of chemical starting materials and fuels. However, the exact roles of different oxygen species in, for example, the conversion of methane to formaldehyde are not well understood. By preparing oxygen overlayers on Mo(110) in which distinct oxygen species are populated and identified by their vibrational spectra, we can isolate the reactivity of oxygen in different coordination sites. For instance, gas-phase methyl radicals add facilely to oxygen bound in quasi-threefold sites to form a methoxy intermediate, identified using high resolution electron energy loss spectroscopy. Reaction of methanol on different oxygen overlayers also indicates that multiply-coordinated oxygen, rather than terminally bound (Mo=O) species, is active in the formation of these partial oxidation intermediates. Reaction of longer-chain alcohols on oxygen overlayers also implicates multi-fold coordinated oxygen rather than Mo=O in alkoxide formation; furthermore, the ability of oxygen to migrate subsurface on Mo(110) is seen to sustain reactivity even on highly oxidized Mo(110) surfaces. Scanning tunneling microscopy studies are planned to probe for morphological changes associated with different degrees of oxidation and different reactivity. Density functional slab calculations are underway to address the differences in the bonding of different oxygen species which may lead to their observed reactivity, including a comparison of atomic oxygen species on Mo(110) and Mo(100), which exhibit significantly different reactivities toward methyl oxidation.
10:20 AM		SS1-ThM-7 Particle Size Dependent CO Dissociation on Alumina Supported Rh: A Model Study M. Baeumer, M. Frank, J. Libuda, S. Stempel, H.-J. Freund (Fritz-Haber-Institut der MPG, Germany); A. Sandell (Lund University, Sweden); S. Andersson, B. Brena, A. Giertz, P.A. Bruehwiler, N. Martensson (Uppsala University, Sweden) The growth and properties of small metal particles on oxide substrates have attracted a lot of attention in the last few years. One obvious reason for this is the possibility to use them as model systems for supported metal catalysts allowing a detailed investigation of the interplay between structure on the one hand and adsorption behaviour and reactivity on the other hand. In our model study, we have prepared a variety of alumina supported Rh particle systems via deposition of Rh from the gas phase onto a thin, well-ordered alumina film. By controlling metal exposure and substrate temperature, different morphologies and particle sizes were obtained and characterized with SPA-LEED (Spot Profile Analysis LEED) as well as with STM measurements. The probabilities for thermally induced dissociation of adsorbed CO taking place at temperatures between 350 K and 500 K could be determined from C1s photoelectron spectra which allowed clear differentiation between molecular and dissociated CO. We find low dissociation activities for the smallest aggregates which, however, increase with increasing average particle size. The dissociation probabilities reach a maximum when the particles contain an average of roughly 1000 Rh atoms. For larger particles the fraction of dissociated CO decreases and reaches values which have to be compared with those on well-prepared Rh single crystal surfaces where the dissociation probability is very low.
10:40 AM		SS1-ThM-8 Chemisorption of Br₂ onto As-rich GaAs(100) Y. Liu, A.J. Komrowski, A.C. Kummel (University of California, San Diego) The chemisorption of Br₂ onto As-rich GaAs(100) has been studied using monoenergetic molecular beams and scanning tunneling microscopy (STM). An arsenic capped GaAs(100) substrate was heated in vacuum to form a well-ordered As-rich surface. The unit cells consisted of three arsenic dimers (6 As atoms) and one missing arsenic dimer. The surface was dosed with a 0.89 eV beam of Br₂ to a coverage of 0.05 ML. STM images of the reacted sites appears as dark as the vacancy dimers. Two types of reacted sites are formed. (1) Large islands (10 nm) of reacted sites were observed in STM. These are probably due to precursor mediated chemisorption in which the Br₂ first traps into a mobile physisorption precursor state. These physisorbed molecules diffuse to existing reacted sites prior to dissociation. It is surprising to find precursor mediated chemisorption at these high translational energies, but this is consistent with our previous studies upon the sticking probability versus translational energy and surface temperature for Cl₂ on GaAs(100). (2) Isolated reacted sites were also observed in STM. These sites are probably the result of direct chemisorption. In direct chemisorption, the molecules chemisorb to the surface instantaneously upon collision with the substrate. For all these isolated sites, both atoms in each dimer were observed to react. We never observed dimers in which only one of the two As atoms had reacted. This could be explained by the Br atoms being bridge bonded between two As atoms at low coverage or by dissociative chemisorption of a Br₂ molecule always resulting in the Br atoms bonding to the As atoms within an As surface dimer.
11:00 AM		SS1-ThM-9 Nanometer Scale Selective Oxygen Etching of Si(111) Surface Using Silicon Nitride Islands J.S. Ha, K.-H. Park, W.S. Yun, E.-H. Lee (ETRI, Republic of Korea) We have investigated the oxidation of Si(111) surface covered with silicon nitride islands using scanning tunneling microscopy (STM). A monolayer of silicon nitride islands with sizes of 6 to 15 nm was formed by exposure of 100 eV positive nitrogen ions at room temperature followed by subsequent annealing at 980° C. In STM images, silicon nitride islands appeared as both bright and dark features surrounded by protruding walls. On the silicon nitride covered Si(111) surface, O₂ gas was dosed at the surface temperatures ranging from 700 to 800°C under O₂ pressures between 1x10^-7 and 1x10^-6 torr, where the selective etching of silicon surface was dominant over oxide formation. Silicon island pillars, covered with monolayer thick silicon nitride layer, as high as 2-3 nm appeared after reaction with O₂ gas as a result of selective etching of silicon surface area. Interestingly, the protruding walls surrounding the silicon nitride islands remained unreacted. Depending upon the extent of O₂ exposure, the spatial uniformity of etching changed. Accordingly, the LEED 7x7 pattern became dimmer as the etching proceeded. The I/V measurements on nitride showed a passivated characteristics even after etching of up to 10 layers of Si indicating the rigidity of silicon nitride layer upon O₂ exposure. In this paper, we will discuss the possibility of the formation of silicon nano-pillars and underlying etching mechanism of the silicon nitride covered Si(111) surface.
11:20 AM		SS1-ThM-10 Simulation of Diamond Film Growth in Premixed Low-Pressure Acetylene-Oxygen Flames B. Ruf, F. Behrendt (Universität Heidelberg, Germany); O. Deutschmann (University of Minnesota); J. Warnatz (Universität Heidelberg, Germany) Diamond synthesis from premixed low-pressure flames has the advantage that relatively big areas can be covered with a uniform diamond film of high quality. Optimizing the growth process requires models that incorporate gas-phase processes as well as processes on the surface. In the present model a stagnation-point flow is directed toward a chemical reactive surface. The gas phase is described by the Navier-Stokes equations in their one-dimensional form, while the surface is introduced as a set of boundary conditions. The surface reaction scheme describes growth on the reconstructed diamond (100)-surface. It is based on a model for the incorporation of CH₃ in the diamond lattice ¹. This scheme has been extended to include reactions with CH₂-, CH- and C-radicals opening additional growth channels. Reactions of O atoms and O₂ molecules are introduced leading to an oxidation of the diamond film. Model results are compared to experimental data of Kim and Cappelli ² and Goodwin et al. ³ The modeled growth rates as function of equivalence ratio, substrate temperature and distance between burner and substrate are in qualitative agreement with the experiments. It turns out that Carbon and O atoms play an important role in the diamond synthesis of acetylene-oxygen flames. ¹F. Behrendt, O. Deutschmann, B. Ruf and J. Warnatz, J. Vac. Sci. Technol. A 14(3), 1604 (1996). ²J. S. Kim and M. A. Cappelli, J. Mater. Res. 10(1), 149 (1995). ³D. G. Goodwin, N. G. Glumac and H. S. Shin, Twenty-Sixth Symposium(International) on Combustion/The Combustion Institute, Pittsburgh, 1817 (1996).