Abstract:
Many heteroepitaxial metal-on-metal systems reconstruct into patterns of alternating domains of stacking faults separated by partial misfit dislocations. Here, we use two approaches to investigate the question of whether these can be predicted and controlled: (i) We map the system onto a one-dimensional Frenkel-Kontorova model, and then obtain a simple criterion to determine whether or not the surface will reconstruct; this had earlier been done for homoepitaxial systems, but is here generalized to the heteroepitaxial case. (ii) The two-dimensional Frenkel-Kontorova model is solved numerically by performing quenched molecular dynamics simulations. The necessary microscopic parameters are obtained by performing ab initio density functional theory calculations on the unreconstructed systems. The systems considered are overlayers of Fe, Co, Pt, Ag, Au, and Pb on a Ru(0001) substrate, as well as clean Ru(0001). The predictions of the two approaches agree with one another as well as with experiment. Both the presence and periodicity of the reconstruction are very sensitive to the value of “chemical potential” Γ. Accordingly, we suggest that Γ can be used to tune the periodicity so as to obtain a desired nanotemplate for subsequent growth of self-organized nanostructures.
