No abstract

The dynamics of the reconstruction of the Cu(110) surface induced by oxygen chemisorption has been studied by scanning tunneling microscopy. The nucleation and growth of the Cu(110)-(2x 1)O reconstructed phase shows up as "added rows" of Cu-0 atoms which grow preferentially in the [001] direction. The Cu atoms are supplied by diffusion from terrace edges.Oxygen-induced reconstruction of metal surfaces is a phenomenon which has received considerable attention in the past. This is partly due to its importance for processes like heterogeneous catalysis and corrosion, but it is equally interesting from a more fundamental point of view in the efforts to understand such processes on a microscopic scale.Oxygen on Cu(110) is one such system which has been studied extensively since the pioneering work of Ertl, who showed that molecular oxygen chemisorbs dissociatively on Cu(110), and that the low-energy electron diffraction (LEED) pattern shows a (2x 1) structure at an 0 coverage of 0.5 ML (monolayers). The halfand integer-order spots of the LEED pattern had comparable intensity indicating a surface reconstruction.However, after more than 20 years of research, the detailed atomic structure of this Cu (110)-(2&1)O surface is still open for debate. Low-energy ion scattering, impact-collision ion-scattering spectrometry, surface-extended x-ray absorption fine structure, He diffraction, and recent x-ray diffraction studies all favor the missing-row reconstruction, where every second [001] row on the surface is absent. On the other hand, the buckled-row model, where every second [001] row is shifted outward, has been suggested by photoemission, high-energy ion scatterinI%, s x-ray diffraction, and scanning-tunneling-microscope'(STM) data. A recent theoretical calculation based on the effective medium scheme also favors a missing-row-type structure as the most stable one. " However, all seem to agree that the oxygen is located at the long bridge positions along the [001] rows 2i4~8~1 1~12 It was the hope when initiating the present study that new information on the dynamics of the surface reconstruction might be obtained, which subsequently might help in understanding the static surface structure.The experiments were performed with a fully automated STM (Refs. 13 and 14) implemented in an UHV chamber with a base pressure (1x10 ' mbar. The bias voltage is applied to the W tip with the sample grounded, and normally the tip voltage is negative, implying that we are tunneling into empty states in the sample.The initial calibration of the lateral STM scans was obtained from the atomic resolution obtained on the reconstructed Si(111)-(7x7) surface, and the z calibration from a single-layer step on a Si(111)crystal.After cutting and mechanical polishing down to 1 pm, the Cu(110) crystal was electrolytically polished in dilute phosphoric acid and mounted in the UHV chamber. The surface was cleaned by repeated Ne sputtering and annealing cycles, until no impurities could be detected by Auger electron spectroscopy, and good quality LEED ...

No abstract

Efficient adhesion of gold thin films on dielectric or semiconductor substrates is essential in applications and research within plasmonics, metamaterials, 2D materials, and nanoelectronics. As a consequence of the relentless downscaling in nanoscience and technology, the thicknesses of adhesion layer and overlayer have reached tens of nanometers, and it is unclear if our current understanding is sufficient. In this report, we investigated how Cr and Ti adhesion layers influence the nanostructure of 2-20 nm thin Au films by means of high-resolution electron microscopy, complemented with atomic force microscopy and X-ray photoelectron spectroscopy. Pure Au films were compared to Ti/Au and Cr/Au bilayer systems. Both Ti and Cr had a striking impact on grain size and crystal orientation of the Au overlayer, which we interpret as the adhesion layer-enhanced wetting of Au and the formation of chemical bonds between the layers. Ti formed a uniform layer under the Au overlayer. Cr interdiffused with the Au layer forming a Cr-Au alloy. The crystal orientation of the Au layers was mainly [111] for all thin-film systems. The results showed that both adhesion layers were partially oxidized, and oxidation sources were scrutinized and found. A difference in bilayer electrical resistivity between Ti/Au and Cr/Au systems was measured and compared. On the basis of these results, a revised and more detailed adhesion layer model for both Ti/Au and Cr/Au systems was proposed. Finally, the implications of the results were analyzed, and recommendations for the selection of adhesion layers for nano-optics and nanoelectronics applications are presented.