Bruce McCarroll

Trapping and Energy Transfer in Atomic Collisions with a Crystal Surface

The condensation of atoms on a solid is examined by calculating the critical energy for trapping of a particle of arbitrary mass and force constant colliding with a linear lattice. In the harmonic approximation and using classical mechanics, the maximum kinetic energy for trapping is found to depend strongly upon the well depth, in qualitative agreement with experiment. For a gas atom colliding with its own lattice (that is, force constant ratio β=1 and mass ratio μ=1) capture occurs at translational energies up to 25 times the binding energy. For β=0.2, this critical energy has diminished to 1.3 times the dissociation energy Q of the homogeneous lattice. The dependence of the critical energy on force constant is not monotonic, however—there is a maximum at β∼0.75. At β=0.75 and μ=1, for example, the incident energy must exceed ∼31Q to prevent trapping.Most of the energy exchange occurs during the repulsive part of the collision. For this part of the collision interval, the presence of a trapping potentia...

A molecular orbital approach to chemisorption: I. Atomic hydrogen on graphite

Abstract We examine the use of Extended Huckel Theory, a semi-empirical molecular orbital approach, in treating chemisorption. The calculations include all valence electrons and overlap integrals. Adsorbate energy levels are self consistently determined. The substrate is represented by a relatively small number of atoms which is shown to be adequate for obtaining semi-quantitative results. The small representation permits investigation of adsorbate interactions with both ideal and imperfect surfaces. The binding energy, binding sites, and barrier to surface mobility exhibited by atomic hydrogen on a graphite basal surface are obtained. The experimentally observed formation of CH4 on adsorption of hydrogen is also considered. Calculations on electrophilic adsorbates reveal the necessity for a more completely self consistent treatment in which all energy levels are adjusted.

Chemisorption and Oxidation: Oxygen on Tungsten

Oxygen chemisorption and the extent of oxidation on a tungsten ribbon at room temperature were studied by flash desorption using line‐of‐sight detection with a time‐of‐flight mass spectrometer. The only chemical species in the adlayer are oxygen atoms and the monotungstic oxides: WO, WO2, and WO3. These oxides are desorbed with an energy of ∼100 kcal. In agreement with earlier predictions, oxygen desorbs as atoms. Neither molecular oxygen nor polytungstic oxides were ever observed from a ribbon flashed after a room‐temperature exposure to oxygen. However, flowing oxygen on tungsten at surface temperatures > 1400°K does produce the polytungstic oxides. An order‐of‐magnitude estimate of the adlayer composition at room temperature suggests that at saturation there is approximately one monolayer of oxygen atoms in a random configuration with a few tenths of a per cent each of the monotungstic oxides. Consequently, oxidation is not a significant surface process on clean tungsten at room temperature.

Analysis of Thermal Desorption Spectra

The behavior of detector signals corresponding to the thermal desorption of gas from a solid surface for integral‐order kinetics is investigated with respect to variations of the parameters describing (1) the kinetics of the desorption process itself, and (2) the interaction of the desorbed gas with the vacuum system. A computational approach is used because the equations describing the simple model used are, generally, analytically intractable. A computer‐driven display and a data‐logging system facilitate not only the analysis indicated above, but also curve fitting so that the kinetic parameters can be extracted from experimental desorption data containing distortions caused by gas scattering from structures in the vacuum system.

The reactivity of graphite surfaces with atoms and molecules of hydrogen, oxygen and nitrogen

Abstract The topographical changes resulting from the attack of gaseous hydrogen, oxygen and nitrogen on the basal plane surfaces of natural graphite crystals have been studied optically. Undissociated hydrogen or nitrogen appear unreactive toward heated graphite in the temperature range 300°–1200°C. The hexagonal etch pits produced by atomic hydrogen on graphite at 700–800°C have two sides oriented perpendicular to the {1121} twin bands, whereas nitrogen atoms impinging on the crystals at temperatures above 1000°C give rise to hexagonal pits with two sides parallel to the twin bands. These latter pits are similar to those found with molecular oxygen at lower temperatures. By contrast, atomic oxygen causes random unoriented pitting over the whole basal plane surface. These results are discussed in terms of the possible reaction products and estimates of the behavior of the adsorbate species on the graphite surface.

Trapping and Energy Transfer in Atomic Collisions with a Crystal Surface. II. Impurities

The influence of surface impurities and internal impurities on (1) atomic condensation on a solid, (2) equilibration of the newly captured atom with the solid, and (3) the accommodation coefficient, are examined through calculations with a one‐dimensional semi‐infinite analog in the harmonic approximation. The general equation for the semi‐infinite lattice (arbitrary composition) and impinging gas atom are specialized to (1) a pure lattice with a surface impurity (variable mass and bonding) and (2) a pure lattice with a defective third atom (variable mass and bonding).For the surface impurity: a decrease in coupling K1 decreases the condensation efficiency. Decreasing either the mass M1 or K1 decreases the rate of thermalization of a captured atom. There is a maximum in the condensation efficiency when M1 is varied. Light surface atoms do increase the accommodation coefficient.For the internal impurity: a weakening of impurity bonding decreases both trapping efficiency and rate of thermalization. Light im...

Molecular Orbital Approach to Chemisorption. IV. LCAO Band Structures and the Molecular Unit Cell

The interaction of atomic species with simple surfaces has been effectively simulated previously by us, using molecular orbital treatments at two levels of approximation to the Hartree-Fock equations. These earlier efforts allowed examination of binding sites, bonding arrangements, and charge transfer in terms of orbital and spatial geometries of the systems simulated. However, these techniques do not describe the electronic structure of the solid substrate as well as they do for small molecules, for which they were initially designed. We present the results of an LCAO-band structure calculation of graphite that compares well with the results of an ab initio calculation. A crucial step for a balanced description of chemisorption is a band structure calculation applied to a properly defined molecular unit cell, which is described in detail. The advantages of this approach are discussed, and a brief, qualitative description of the results of preliminary calculations is presented.

A molecular orbital approach to chemisorption: III. Atomic H and O on graphitic BN

Abstract The chemisorption behavior of a hydrogen atom and of an oxygen atom on graphitic boron nitride (BN) is calculated with the CNDO/2 molecular orbital procedure. The substrate BN representation is isoelectronic and isostructural to that used previously to study atomic adsorption on graphitic carbon, and thus allows an elementary quantum mechanical examination of adsorption site sensitivity to substrate composition. Calculated adsorbate-substrate charge transfer behavior again demonstrates the inadequacy of atomic electronegativity reasoning for chemisorption. A critique of the suitability of the CNDO/2 technique for chemisorption calculations is included.

Surface Interactions of Iodine Vapor with Tungsten

The interactions of iodine vapor with a clean tungsten surface were studied by first adsorbing the iodine on the tungsten ribbon and then thermally desorbing the adlayer. The composition of the desorption products was determined with a time‐of‐flight mass spectrometer. No evidence is found for the formation of surface compounds. Only iodine atoms and, at higher coverages, iodine molecules populate the surface. Both desorptions are first order, and the most tightly bound atom state has a desorption energy of ∼80 kcal. A second, lower‐energy atom state is populated at higher coverages. An estimate of the iodine molecule desorption energy gives ∼40 kcal.