Activation Energy For Surface Self-diffusion Research Articles

Bombardment-induced ripple topography on GaAs and InP

Bombardment-induced ripples have possible application in nano-technology as nano-wires. This paper reviews some of the published experimental data of ripple formation on GaAs and InP resulting from noble gas, oxygen, nitrogen and cesium ion bombardment. The dependences of the ripples and their corresponding wavelengths on substrate temperature, areic dose and (to a lesser extent) angle of incidence were considered. The experimental results were tested against the predictions of the Bradley–Harper theory [J. Vac. Sci. Technol. A 6 (1988) 2390]. Surprisingly good agreement was obtained between the experimental data and this theory. This is probably due to the amorphisation of the substrates under ion bombardment and, thereby, minimising the effect of crystallographic-orientated self-diffusion on metals, and of energy barriers (Ehrlich–Schwoebel barriers) to interlayer surface diffusion. From the data of MacLaren et al. [J. Vac. Sci. Technol. A 10 (1992) 469], the activation energy for surface self-diffusion of GaAs in the temperature range 60–100 °C was determined as 0.26 eV.

Low temperature deposition and characterization of polycrystalline Si films on polymer substrates

Polycrystalline Si films were deposited on poly(ethyleneterephthalate) (PET-Mylar®) and glass (7059 Corning) substrates at 140 and 200 °C, respectively. X-ray diffraction and Raman spectroscopy were used to confirm the polycrystalline nature of the films. The largest grain size obtained was 95 nm. Raman spectroscopy also showed a simultaneous presence of an amorphous phase. The relative fraction of this amorphous phase was controllable by adjusting the composition of the sputtering gas. Films deposited with only Ar or Ar+H2 sputtering gas show a very small polycrystalline Si peak. With the addition of up to 10% Kr to the gas mixture, a very strong polycrystalline peak appears in the Raman spectra. X-ray diffraction also confirmed the polycrystalline nature of the films. The Kr effect was related to the energetic condensation. The presence of Kr increased the energy of the sputtered atoms. Even after collision with the sputtering gas, these atoms impinge on the substrate surface with enough residual energy that the ad-atoms experience enhanced diffusion, which leads to polycrystalline film formation. For our system, the final energy of the ad-atoms was calculated to be 1.92 eV, which is more than the activation energy for surface self-diffusion for Si.

Calculation of the activation energy for surface self-diffusion of transition-metal atoms

The cohesion-based approach developed earlier for describing the adsorptive properties of d-metal atoms on d substrates was used to calculate the self-diffusion on the (111) and (100) planes of fcc metals, and the (110) plane of bcc metals. The results of the calculation are compared with the data obtained by other researchers and with available measurements.

Annealing study of gold films using scanning tunneling microscopy

We have studied thermal annealing effects on the surface morphology of 800 Å thick gold films, using scanning tunneling microscopy. The gold films were thermally evaporated onto glass substrates and were then measured with the scanning tunneling microscope at room temperature before and after annealing. The annealing treatments were done at temperatures between 200 and 500 °C and for periods ranging from 1 to 60 h. We present data showing the evolution of the average surface-grain size and rms roughness amplitude of the gold films as a function of annealing temperature and duration. Our data suggest that surface diffusion is the main process active at low annealing temperatures of 300 °C and below. The activation energy for surface self-diffusion of gold in our samples was around 1.1 eV. Other mechanisms seem to be dominant at higher temperatures.

Direct scanning tunneling microscope (STM) observation of Cr(III) complexes on hematite (001) surfaces

Scanning tunneling microscopy (STM) is used to observe, at the atomic scale, Cr(III) adsorbed to hematite (001) surfaces from aqueous solution. The Cr(III) adsorbates are relatively immobile, but estimated activation energies for surface self-diffusion are lower than those for water or hydroxyl substitution in aqueous Cr(III). Possible causes are effects of STM imaging (artifacts), high ligand-substitution rates for adsorbed species, or participation of substrate Fe (III) ligand exchange. STM imaging of suitable aqueous surface complexes is shown to be feasible, and constitutes a new way to study the relationships between microscopic and macroscopic chemical behavior of adsorbed species in aqueous systems.

Measurement of the activation energy for surface diffusion in gold by scanning tunneling microscopy

Measurements of the activation energy for surface self-diffusion in gold by the method of profile decay using a scanning tunneling microscope (STM) have been done on polycrystalline gold films deposited on glass substrate. The peak-to-peak surface roughness was measured as a function of annealing time after annealing at 125° C, 150° C and 170° C with a special pan-cake furnace in the STM. The resultant gold surface diffusion activation energy is found to be ∼21 kcal mol . These experiments demonstrate the unique ability of STM to study surface diffusion over macroscopic distances in the low temperature regime.

Influence of electric fields and gas adsorption on the surface self-diffusion of metals and on the growth of microcrystallites

A survey of the influence of electric fields and impurities upon surface rearrangements of the refractory metals W, Ta, and Nb is given. This includes the measurement of activation energies for surface self-diffusion at high electric fields ( ⋍ 2 × 10 7 V/cm ) and the surface structural changes at low fields ( ⋍ 2 V/cm , dc-effect). The influence of the electric field on crystallite (whisker) growth is discussed briefly.

Effect of impurities on surface self-diffusion and surface structure

Measurements using field emission techniques of the activation energy for surface selfdiffusion of several of the refractory transition metals when carbon or silicon is present on the surface show large increases which are dependent on the degree of surface coverage. Maximum values obtained were: 8.5 eV for carbon on tungsten, 7.0 eV for silicon on tungsten, 4.9 eV for carbon on tantalum, 4.5 eV for carbon on molybdenum and 2.8 eV for silicon on molybdenum. In addition, two anomalous effects have been observed in which surface changes occur at critical temperatures, (a) Sharp discontinuities occur in the plots of activation energy versus temperature for carbon on tungsten at about 2300 °K and for silicon on tungsten at about 2000 °K. In both cases the activation energy drops from the respective high value to that for the clean substrate material of 3.0 eV. Concomitant with this transition the emission patterns change in appearance from those typical of a contaminated surface to those typical of a clean surface, (b) For carbon on tungsten and silicon on tungsten, (433) planes are observed which decrease in size with temperature and suddenly disappear at a very sharp critical temperature. It is suggested that the presence of these impurities causes a restructuring of the surface layers even when present in much less than stoichiometric amounts and that surface phase changes occur independent of bulk changes.

The activation energy for surface self-diffusion of tantalum and the influence of residual gases on this quantity in the presence of high electric fields

When tantalum field emission microscope tips are heated in the presence of applied electric fields, the build-up of the 〈111〉 areas dominates. A linear relationship is indicated between the activation energy for surface self-diffusion, Q F, determined from this process and log p ( p average residual gas pressure during the measurement). The negative slope is interpreted in terms of the presence of a permanent dipole migration mechanism. At p=5×10 −10Torr and F apex=22 MV/cm, Q F≅1.4 eV/particle. From the build-up decay the corresponding activation energy for surface self-diffusion without applied field Qs=2.03±0.05 eV/particle could be determined. The higher values obtained by other authors are explained by the influence of contaminants.

The influence of residual gases and oxygen on the activation energy for surface self-diffusion of tungsten in the presence of high electric fields

Two build-up stages could be distinguished during the heating of tungsten field emitters at various residual gas pressures, p, while an electric field, F, was applied: (a) build-up around the 〈111〉 poles and (b) build-up around {100} planes. A plot of log p versus Q F, the activation energy for surface self-diffusion with field, indicated a linear relation with a positive slope for {100} build-up and a linear relation with a negative slope for 〈111〉 build-up. The value of Q F also, is a linear function of F, indicating a permanent dipole migration mechanism. In the presence of oxygen, 〈111〉 build-up occurred predominantly. The Q F values match the pressure dependence for residual gas 〈111〉 build-up. It could be determined: Q F (W-atom) = 1.9 eV/particle for F = 25 MV/cm , Q F = 0 (dipole) = 4.8 eV/particle at P residual = 4 × 10 −9 Torr, and an effective dipole moment of μ ⋍ 6 D . The increase of Q F {100} with increasing residual gas pressure is proposed to be due to pinning by carbon. The final build-up form ({110}, {100}, and {112}) is discussed in comparison with other single-crystal growth forms of tungsten.

NMR spin echo studies on mobility and diffusion of benzene adsorbed on silica

Diffusion coefficients of benzene in porous plugs compressed from Aerosil were measured by the spin echo technique with the use of constant and pulsed field gradients. In order to obtain surface self-diffusion coefficients D s ∗ the contribution from molecular mobility in the pore volume has been evaluated. D s ∗ shows a maximum near θ = 1 in agreement with earlier results obtained from rates of adsorption. Differences between the limiting experimental diffusion coefficients in the region of capillary condensation and the bulk liquid are attributed to the tortuosity of the pore system. Activation energies for surface self-diffusion are larger than for T 2.

A calculation of relaxation, migration and formation energies for surface defects in copper

The relaxation and migration energies of adatoms and terrace vacancies on copper surfaces have been calculated for (100), (110) and (111) surface orientations. The calculations were performed using a pairwise interaction model based on a modified Morse potential energy function and indicate a strong influence of orientation on the magnitude of the defect migration energies. Estimates of defect formation energies have also been made, using a simple bond-breaking model but including relaxation effects. Thus, it has been possible to estimate the activation energies for surface self-diffusion for three different surface orientations. The results obtained are discussed in terms of current mechanistic interpretations of the experimentally observed activation energy for surface diffusion.

Vapor-Phase Growth Kinetics of Potassium Whiskers by Field Emission

Using the method originated by Gomer, we have studied the growth kinetics of potassium whiskers, as they grow from the vapor phase in an electron field emission tube. To the best of our knowledge, this represents the first use of an alkali metal as a field emitter. The results for potassium whiskers show some features similar to those previously found for mercury, and some new features. The former include positive exponential growth, and the attainment of terminal lengths. New features include the absence, usually, of symmetric emission patterns, a very low activation energy for surface self-diffusion, the fact that the whiskers cannot be pulled off but instead appear to shorten gradually if the field is raised, and finally, an interesting sensitivity of the growth processes to illumination of the whiskers with visible light.