Search

The adsorption of NO and NO2 was studied on the ( 2 1 ¯ 1 ¯ 0 ) crystal face of zinc oxide (ZnO). Binding energies, workfunction changes, vibrational frequencies, electron localisation functions, charge density differences and Bader charges were calculated. More than one stable structure was found for both molecules on this surface and it was elucidated that adsorbed NO was more stable than NO2. The relatively small binding energy values indicate that the interaction of NO and NO2 with the surface is weak and was found to cause minimal distortion to the surface geometry. The most stable configurations of NO were those with the nitrogen atom of the adsorbate interacting with surface oxygen and zinc atoms. Similarly, stable NO2 configurations were hallmarked by interactions between the adsorbate oxygen atoms and surface zinc atoms. The calculated workfunction changes were negative for all NO systems, and most NO2 systems. Positive workfunction changes were calculated for geometries where the nitrogen atom of the NO2 adsorbate was positioned closer to the surface than the oxygen atoms which were both pointing away from the surface. The Bader charges and charge density differences indicate that the interaction between the adsorbate molecule and the respective surface atoms is weak. Despite showing a weak interaction, there is still a calculated transfer of charge from the surface to the adsorbate for the most stable NO2 and NO structures. Even though this transfer of charge is opposite to what is predicted by some of the calculated negative work function changes, small features in the planar averaged charge density differences may explain this discrepancy.

The adsorption of H2S on Fe(1 0 0) is examined using ab initio molecular dynamics at 298 and 1808 K. Dissociation of H2S occurs at both temperatures simulated, to leave adsorbed S and two H atoms. The dissociation occurs via a two step process and the mechanism is found to be different depending on the temperature of the reaction. At 1808 K, diffusion of the dissociated H atoms into the subsurface region is also observed.

Density functional theory was employed to examine the interactions of atomic sulfur with the Fe(1 0 0) and Fe(1 1 0) surfaces. Vibrational frequency calculations were performed to determine the nature of stationary points at the high symmetry atop, bridge and hollow adsorption sites and to indicate the direction the adsorbate would move across the surface. The values were also used in the determination of the rate constant for hopping of S from one energy minimum site to another. Ab initio molecular dynamics (MD) simulations were then performed to monitor the mobility of the S atom on the (1 0 0) surface at different temperatures up to the melting point of Fe (1808 K) and were compared to our previously obtained ab initio MD results for S/Fe(1 1 0) [N. Todorova, M.J.S. Spencer, I. Yarovsky, Australian Institute of Physics 16th Biennial Congress, 2005, Canberra, Australia, ISBN 0-9598064-8-2].

The effect of S contamination on the properties of Fe(1 0 0) is examined using density functional theory (DFT) calculations. S is adsorbed at 1/2 monolayer coverage in atop, bridge and hollow sites in a c(2 × 2) arrangement. The effect of S on the clean surface properties is first examined for the three adsorption sites and compared with experimental and other theoretical data. S is found to adsorb preferentially in hollow sites on the isolated surface in agreement with experiment. The adhesion energy at different interfacial separations is then calculated and the effect of S on the interfacial properties of Fe(1 0 0) is characterised quantitatively and qualitatively. S is found to enhance adhesion at larger separations though at the equilibrium interfacial separation the maximum interfacial strength is reduced.

The adsorption of silane and methylsilane on the (1 1 0) and polycrystalline surfaces of gold is examined using vibrational electron energy loss spectroscopy (VEELS), angle-resolved ultraviolet photoelectron spectroscopy (ARUPS) and X-ray photoelectron spectroscopy (XPS). Adsorption of silane onto the Au(1 1 0) surface at low temperatures is dissociative and yields an SiH 2 and possibly also SiH 3 surface species. Further dissociation occurs at room temperature to yield adsorbed SiH, which is tilted on the surface, with complete dissociation to Si occurring by 110 °C. The similarity in the UP spectra for silane adsorbed on the polycrystalline sample suggests that the same surface species are present over that temperature range. Above 200 °C, spectral changes suggest rearrangement of the Si atoms, which, by 350 °C, have diffused into the bulk. Adsorption of methylsilane onto the (1 1 0) surface at low temperatures initially produces adsorbed CH 3 SiH or CH 3 SiH 2 , with undissociated methylsilane physisorbing at higher exposures. By room temperature, desorption and decomposition leaves (or direct adsorption yields) only adsorbed CH 3 Si. After further heating, the hydrogen–carbon bonds of the CH 3 group break to leave an adsorbed SiC species. On the polycrystalline surface, methylsilane adsorption is the same at low temperatures as on (1 1 0). In contrast to the latter, though, the UP spectra indicate that direct exposures at room temperature yield adsorbed Si or SiC initially, with CH 3 Si again adsorbing at higher exposures. Upon further heating to 330 °C, little if any methyl-groups remain on the surface and the Si has started to diffuse into the bulk.

The adsorption of methylsilane on the (1 1 1), (1 1 0) and polycrystalline surfaces of copper is examined using vibrational electron energy loss spectroscopy, angle-resolved ultraviolet photoelectron spectroscopy and X-ray photoelectron spectroscopy. On all surfaces, low exposures of methylsilane below −140 °C yield chemisorbed CH3Si species, while undissociated methylsilane physisorbs at higher exposures. This physisorbed layer then dissociates at higher temperatures to leave CH3Si on the surface by room temperature. Above room temperature, the VEELS and ARUP spectra suggest that Si–C bond cleavage occurs on the (1 1 1) surface, whereas the C–H bonds break to leave an SiC molecular species on the (1 1 0) surface. A surface–molecule bonding model is constructed to describe the adsorbate–substrate interactions and suggests that CH3Si adsorbs in a hollow site on the (1 1 1) surface and in a long-bridge site on the (1 1 0) surface.

Adhesion between ideal bulk-terminated bcc Fe(1 1 1) match and mismatch interfaces was simulated using density functional theory (DFT) within the plane-wave pseudopotential representation. Interfaces were modelled using the supercell approach where the interfacial separation was varied by changing the size of the vacuum spacer between image cells in the z-direction. The adhesive energy values were calculated for discrete interfacial separations and the data was fitted to the universal binding energy relation (UBER) [Rose et al., Phys. Rev. B 28 (1983) 1835]. The parameters obtained from these fits allowed the work of separation (Wsep) to be determined and a comparison to be made of the adhesion properties of the match and mismatch interfaces. The results were also compared to those obtained previously for the (1 0 0) and (1 1 0) surfaces.

The relaxations and energies of the (1 0 0), (1 1 0) and (1 1 1) surfaces of (bcc) Fe have been calculated using density functional theory. A plane-wave pseudopotential method was employed. The results demonstrate that for the (1 0 0) surface a contraction of the first (outer) layer is observed while the second and third layers expand perpendicular to the surface plane; for the (1 1 0) surface, the displacements are the same, however, the magnitude of the relaxations is much smaller, showing the surface to be basically bulk terminated; for the (1 1 1) surface the first two layers contract while the third expands, with the magnitude of the relaxations being much larger than for the other surfaces. The surface energy values for the relaxed and unrelaxed surfaces were determined showing the (1 0 0) and (1 1 0) surfaces to have almost identical surface energies with the (1 0 0) being slightly higher followed by the (1 1 1) surface. Our results are compared to other experimental and computational studies and generally show good agreement with experiment; an explanation is provided for any differences. The surface models used in this study provide a good basis for future work examining the adsorption of impurities on the surface and the effect of relaxation on adhesion.

The adsorption of silane on Cu(1 1 1) and Cu(1 1 0) is examined using vibrational electron energy loss spectroscopy, angle-resolved ultraviolet photoelectron spectroscopy and X-ray photoelectron spectroscopy. At room temperature, SiH adsorbs on Cu(1 1 1), whereas complete dissociation occurs on Cu(1 1 0) to leave adsorbed silicon. Below room temperature, an SiHx (x=2 or 3) species exists on both surfaces. A surface-molecule bonding model is constructed to describe the adsorbate–substrate interactions and suggests that both Si and SiH adsorb in a substitutional or hollow site on the (1 1 1) surface and Si adsorbs in one of the same two sites on the (1 1 0) surface.