W. R. Graham

Electron transport of inhomogeneous schottky barriers : a numerical study

Numerical simulations are presented of the potential distribution and current transport associated with metal‐semiconductor (MS) contacts in which the Schottky barrier height (SBH) varies spatially. It is shown that the current across the MS contact may be greatly influenced by the existence of SBH inhomogeneity. Numerical simulations indicate that regions of low SBH are often pinched‐off when the size of these regions is less than the average depletion width. Saddle points in the potential contours in close proximity to the low‐SBH regions, which are shown to vary with the dimension and magnitude of the inhomogeneity as well as with bias, essentially determine the electron transport across the low‐SBH regions. It is these dependences of the saddle point which give rise to various abnormal behaviors frequently observed from SBH experiments, such as ideality factors greater than unity, various temperature dependences of the ideality factor, including the T 0 anomaly, and reverse characteristics which are strongly bias‐dependent. The results of these numerical simulations are shown to support the predictions of a recently developed analytic theory of SBH inhomogeneity.

Single atom self-diffusion on nickel surfaces

Abstract Results of a field ion microscope study of single atom self-diffusion on Ni(311), (331), (110), (111) and (100) planes are presented, including detailed information on the self-diffusion parameters on (311), (331), and (110) surfaces, and activation energies for diffusion on the (111), and (100) surfaces. Evidence is presented for the existence of two types of adsorption site and surface site geometry for single nickel atoms on the (111) surface. The presence of adsorbed hydrogen on the (110), (311), and (331) surfaces is shown to lower the onset temperature for self-diffusion on these planes.

Multilayer rippled structure of the NiAl(110) surface: A medium energy ion scattering study

The NiAl(110) surface structure has been studied with medium energy ion scattering in a channeling and blocking (MEIS/CB) experiment. A high sensitivity to the first layer ripple (10% ripple with the Al on top) is reported and a multilayer model is proposed. The first layer Ni atoms are contracted relative to the second layer Ni atoms by 7%, and the second layer Ni atoms are expanded relative to the third layer Ni atoms by 1%. The first layer Al atoms are expanded relative to the second layer Al atoms by 5%, and the second layer Al atoms are contracted relative to the third layer Al atoms by 1%. The model assumes bulk-like positions of the third layer for both Ni and Al atoms. The present results are in excellent agreement with a recent single layer LEED intensity analysis.

On the atomic structure of {001}W

Abstract Results of a field ion microscope study of the clean surface structure of the {001} surface plane of tungsten are presented. The major conclusions are that {001}W is reconstructed over the temperature range 15–580 K, and that the reconstructed surface contains an alternating vertical component to the displacements of the W surface atoms. There are two magnitudes of verticaldisplacement, each arrayed with p(2 × 2) symmetry, and if the differences in the magnitude are neglected, the vertical displacements are arrayed in the well known (√2 × √2) R 45° configuration. Details of this newly developed experimental approach for the study of surface reconstruction are reported, along with a number of control experiments which exclude the possibility that these results are artifacts due to the experimental technique. The discussion includes a comparison of the present results with those drawn from other experimental techniques, primarily low energy electron diffraction.

Pair potential calculations of single atom self-diffusion activation energies

Abstract The theoretical method most generally used to model single atom surface diffusion characteristics is evaluated in this paper. The method is based on the use of semi-empirical potentials to describe interatomic forces. In the present study calculations were performed for all bcc and fee transition metals. Comparison of these calculations with existing experimental data show that the final ordering of activation energies for single atom self diffusion on different surface orientations is critically dependent on the amount of lattice relaxation around the adatom, and that the existence of quantitative agreement between theory and experiment is fortuitous and strongly dependent on lattice relaxation procedures.

Formation of epitaxial yttrium silicide on (111) silicon

The growth of epitaxial yttrium silicide on Si(111) in ultrahigh vacuum is studied. Resistivity, epitaxial quality, and pinhole coverages are studied as a function of annealing temperature for each growth method used. The best films result from the growth of a thin, 30–40‐A template layer which is annealed to 700 °C, followed by a thicker film growth by depositing additional Y onto the substrate heated high enough to induce silicide formation (∼300 °C). Annealing to 900 °C results in a Rutherford backscattering minimum channeling yield χmin ∼3%, which is the same order of epitaxial quality previously achieved by only Ni‐ and Co‐silicide films on silicon. Films grown without templates have larger pinhole sizes with pronounced features indicative of the hexagonal nature of these structures. The deposition of Y metal onto a substrate held at room temperature, followed by annealing to 900 °C results in the lowest resistivities (48 μΩ cm for 425‐A films), but with a highly dislocated film structure featuring 1...

Coverage measurements of the Si(100) 2× 1 : Cs and Si(100) 2× 1 : K surfaces: resolution of structural models

In this study, medium energy ion scattering has been used in conjunction with Auger electron spectroscopy and LEED to measure the saturation coverage of Cs and K on Si(100) at room temperature, for which distinct 2 × 1 surface structures are observed. Defining one monolayer as the density of Si in a single (100) crystal plane (6.78 × 1014atomscm2), the overlayer coverages in these systems are 0.97 ± 0.05 ML for Si(100)2 × 1:Cs and 0.98 ± 0.05 ML for Si(100)2 × 1:K. These results clearly discriminate between proposed models containing 0.5 and 1.0 ML alkali coverage for the saturated surface.

Phosphorus interdiffusivity inα-Fe binary and alloy systems

The isothermal segregation kinetics of phosphorus to Fe and Fe-based alloy surfaces have been monitored from 783 to 923 K using Auger electron spectroscopy. The P segregation kinetics are consistent with a model which assumes bulk diffusion of P to be the rate controlling mechanism in the segregation process. The activation energy and preexponential for P diffusion in Fe calculated from the diffusion data are 314 ± 30 kJ · mol−1 and 8 × 105 cm2s in the temperature range studied. Similar results are found for the carbon free Fe-based alloys studied. Phosphorus segregation kinetics for a carbon containing alloy are more rapid, with evidence of a dependence on carbon concentration. The activation energy and pre-exponential calculated in this study for carbon-free alloys are considerably higher than published values measured at higher temperatures by conventional techniques. It is pointed out that an increase in activation energy with decreasing temperature is consistent with observations in several studies of Fe and other ferromagnetic alloys of an increase in activation energy for diffusion below the ferromagnetic transition temperature.

The segregation of phosphorus to the free surface of a ferritic iron alloy at 723 to 823 K

The isothermal segregation of phosphorus to the free surface of an annealed polycrystalline Fe-Ni-Cr-C alloy, containing 0.06 wt pct P, has been studied by Auger electron spectroscopy between 723 and 823 K using anin situ heating technique. It is found that after high temperature argon-ion bombardment the kinetics of isothermal segregation are consistent with bulk (lattice) diffusion of phosphorus as the rate determining mechanism. An approximate value for the adsorption isotherm constant is deduced to be 8×102 per wt pct by a limited analysis of the kinetic data in terms of linear bulk diffusion with a boundary condition given by an ideal adsorption isotherm.

Growth of pinhole‐free epitaxial yttrium silicide on Si(111)

This paper reports the growth of pinhole‐free epitaxial YSi2−x layers on Si(111) as thin as 30 A. This has been accomplished by depositing both Y and Si at room temperature and then annealing to 500–900 °C. Use of the template method allows for the growth of thicker films also free of pinholes. Deposition of yttrium metal only onto Si(111) requires a temperature ∼300 °C for nucleation of the silicide reaction between the Y overlayer and Si substrate. Such a process creates small pinholes ∼500 A in diameter, randomly distributed throughout the film. These pinholes increase in size with higher annealing temperature, resulting from a raised interface free energy intrinsic to the nucleation controlled growth.