Gordon Research Conference: Thin Film & Crystal Growth Mechanism, Mount Holyoke College, 6/26 - 7/01/05 (poster)

Atomically and Time Resolved Pattern Formation in Strained Metal Films: S / Ag / Ru(0001)

Bogdan Diaconescu and Karsten Pohl

Self-ordering growth of nanoarrays on strained metallic interfaces is an attractive option for preparing highly ordered nano-templates of specific feature size densities.1,2 The great potential of this natural templating approach is that the feature sizes and densities are predicted to depend on the interfacial stress in these strained layer. The growth of the 2D sulfur clusters on submonolayer Ag on Ru(0001) transforms the herringbone like dislocation network of Ag with a 6 nm x 4 nm unit cell into a large scale ordered triangular array of S filled vacancy islands 5 nm apart, as shown by STM1 and LEED.

Variable Temperature STM studies reveal that S growth takes place in two regimes: (1) In the dilute S cluster phase for low S coverage, Ag threading dislocation sites are etched one by one, forming vacancy islands with an average size of 1.5 nm2 corresponding to two S atoms per cluster until the complete ordered S cluster array is formed. In this phase the mobility of the S clusters is investigated as a function of temperature; an activation temperature of 265 K is found for the isolated S clusters. For dimers and trimers of S clusters higher temperatures are required to activate their surface diffusion. (2) Above 0.018 ML S we observe the highly ordered solid cluster phase characterized by a p(2x2) S structure and a coverage dependence of the cluster size. In the growth process S partially relieves the strain in Ag film as seen by the relaxation of the threading dislocation network. As the S coverage increases beyond 0.33 ML on the Ru(0001) terrace, the ordering of the 2D S island array is partially destroyed, and the compressed S phase pushes Ag onto a second layer.

1 K. Pohl et al., Nature 397, 238 (1999)
2 K. Thürmer et al., Phys. Rev. Lett. 92, 106101 (2004)
This work is supported by NSF-CAREER-DMR-0134933 and ACS-PRF-37999-G5