M. J. Cardillo

Extraction of kinetic parameters in temperature programmed desorption: A comparison of methods

An investigation of the temperature programmed desorption (TPD) of CO and D2 from Ni(111) has been carried out. It has been shown that a differential method for the extraction of the kinetic parameters, threshold temperature programmed desorption (TTPD), can be applied with accuracy near the limit of zero coverage. In this limit, agreement is found between integral and differential methods for kinetic parameter evaluation. The factors which limit the applicability of TTPD are explored and a method to verify its proper application is presented.

Hard‐cube model analysis of gas‐surface energy accommodation

Two simple hard‐cube models of gas‐surface collisions are re‐examined in light of recent argon–tungsten atomic beam scattering experiments. Both models provide a good description of the average energy exchange. The inclusion of a square well attraction to the hard‐wall potential results in an accurate two parameter fit to the data. The derived well depth is in agreement with previous measurements of the heat of desorption.

A molecular beam study of the NO interaction with Pt(111)

Using angle and time resolved molecular beam techniques, an investigation of the low coverage adsorption‐desorption kinetics of NO on Pt(111) is made. The experiments are carried out over a crystal temperature range of 300 K<Ts<900 K. For Ts≳500 K, the sticking probability s≳0.9. NO adsorbs molecularly with little dissociation (<5%). The desorption rate is found to be strongly dependent on the incident beam flux and trace amounts of chemisorbed oxygen (<1%) on the surface. These findings suggest that steps play the dominant role in low coverage kinetics for a nominally flat crystal. We present a model which incorporates the effect of steps, explains the nonlinearity of the desorption kinetics, and reconciles the disagreement between these results and previous molecular beam studies.

Atom-surface scattering dynamics at hyperthermal energies

We report molecular beam scattering of hyperthermal Xe atoms over an energy range 1<Ei(eV)<10 from single crystal surfaces of GaAs(110), Ag(100), and Ge(100). The angular distributions from the corrugated surfaces show sharp backscattered rainbow maxima related to the topography of the crystal surface. In contrast the smooth surfaces yield quasispecular lobes suggestive of structure scattering. The large energy loss for all surfaces scales on average with the energy of local normal motion. A simple binary interaction model is developed which accounts for many of the phenomena observed from corrugated surfaces. With the aid of a comparison classical trajectory study, these results provide some understanding of the mechanism by and extent to which a solid can dissipate the energy of a hyperthermal collision.

The influence of steps on the desorption kinetics of NO from Pt(111)

A kinetic model is presented which describes thermal desorption of molecules from surfaces and incorporates trapping at steps. The model is applied to desorption of NO from Pt(111). The results of three molecular beam experiments, which differ substantially from each other, are accounted for quantitatively by the model using one set of rate parameters. It is demonstrated that steps, even when present in small concentrations on accurately cut crystals, can play a dominant role in thermal desorption. The magnitude of the effect depends on the difference between the binding energies on the terrace and at steps, which for NO on Pt(111) is large (25±2 kcal/mol and 34±2 kcal/mol, respectively). Furthermore, it is shown that the rate parameters governing terrace desorption, escape from steps, and surface diffusion can all be extracted from modulated molecular beam studies carried out for a suitable range of conditions.

Classical trajectory calculations of the dissociation of hydrogen on copper: III. The effect of surface roughness

Abstract Further calculations have been performed for the dissociation of H 2 on copper in which the empirical potential surface has been increased in roughness as suggested by our previous studies. This change is effected through an increase in the barriers to adatom surface diffusion, keeping the minimum energy requirements for dissociation about the same. The result is to bring the calculation into satisfactory agreement with molecular beam experiments. We examine the effect of changing the internal energy (vibration and rotation), mass, (D 2 ), the adatom binding site, and evaluate the validity of the correlating parameter E ⊥ = E cos 2 θ .

Classical trajectory study of the dissociation of hydrogen on copper single crystals: II. Cu(100) and Cu(110)

Abstract We report a second series of classical trajectory calculations for the activated dissociation of hydrogen on copper single crystals. We have modified the empirical interaction potential surface used previously by raising the energy requirements for dissociation, thereby lowering the dissociation probabilities and bringing their magnitudes in line with experiment. Dissociation probabilities on both a Cu(100) and Cu(110) surface for H2 and D2 are computed. The results, while in reasonable magnitude agreement with experiment, still do not fit the experimental translational energy dependence well and do not predict the correct isotope effect for Cu(110). A detailed comparison of the differences between the ground and first vibrationally excited states for the (100) and (110) surfaces suggest that within the approximations of this model calculation, the potential surfaces employed have underestimated the effects of surface roughness in determining the incident energy and angle dependence of the dissociation probabilities.

Trajectory studies of hyperthermal Xe scattering from GaAs(110)

We present the results of stochastic classical trajectory simulations of the scattering of a nonreactive gas, xenon, from a semiconductor surface of known structure and electronic and vibrational properties, GaAs(110). The range of incident energies considered is 1 to 8 eV, in order to make direct comparison with results of molecular beam experiments reported in the accompanying paper. We employed a 48‐atom three‐layer slab of GaAs, with periodic boundary conditions in two dimensions and frictional and stochastic forces in the third (surface normal) dimension. Pairwise additive Lennard‐Jones potentials describe the gas–surface interaction. The calculations reproduce the large energy exchange and surprising structural sensitivity observed experimentally. Energy transfer is dominated by an initial binary interaction of the Xe with a single Ga or As atom. The repetitive collision nature of this binary encounter produces angular scattering patterns very similar to those obtained for scattering of light atoms,...

The diffraction of He atoms at the GaAs(110) surface

Abstract We present an experimental study of the diffraction of He beams at the GaAs(110) surface. Diffraction angular scans and specular intensity scans are obtained for a wide range of incident angles, azimuths, and at two energies ( E = 0.063 eV and E = 0.021 eV ). The data have been qualitatively analyzed from a classical scattering viewpoint and the hard wall eikonal scattered wave approximation. A hardwall corrugation function of the form ζ(x, y) = 1 2 d x cos( 2πx L x ) + 1 2 d y cos( 2πy L y ) yields a qualitative fit to the data with d x ~ 1.1 A and d y ~ 0.3 A, w deep corrugation corresponds to going across the surface troughs. This value of d x is approximately half the corrugation of the nuclear positions. Specular intensity scans are analyzed in terms of interference in the normal momentum transfer, Δk ⊥ , which yields structural information about vertical displacements. We discuss what is required in order to appropriately describe scattering from strongly corrugated surfaces.

He atom diffraction from a silicide precipitate on a Pt(100) surface

Abstract We have observed the formation of an ordered silicide at the surface of a nominally pure Pt(100) crystal. The silicide appears after thermally activated migration of a trace Si impurity ( −4 at%) and subsequent cooling. The silicide is about one monolayer thick and forms a large area unit mesh, (6 2 ×6 2 )R45° , which is observed with both He diffraction and LEED. A thermal attenuation analysis of the specular and a diffracted He beam yields a consistent value for the attractive depth of the He-surface potential D =13 meV . From a simple classical analysis of the He diffraction pattern we have obtained an estimate for the corrugation of the He-surface potential, ξ=0.07 A . The well depth D is shown to be consistent with the wavelength dependence of the angular distribution of the diffraction pattern.