Controlling Si Surface Morphology at the Atomic Scale-Simulation and Experiments

K. Li, N. Pradeep, J. Fu, R. Silver
National Institute of Standards and Technology, US

Keywords: step-terrace morphology, surface evolution, kinetic monte carlo simulation, electromigration, anisotropic diffusion, step permeability, Ehrlich-Schwoebel effect, atomically flat


Controlling the evolution of atomic scale steps and terraces is central to atomic scale device fabrication. Experiments have shown that conventionally etched micro-scale features can significantly influence the step-terrace morphology during the high temperature annealing process. To effectively control the step-terrace morphology, it is critical to understand the dynamics of the surface atoms with respect to size and position of the micrometer size features. These macroscopic features act as pinning sites and directly affect the step flow dynamics. We have developed a kinetic Monte Carlo simulation algorithm using a solid-on-solid model to understand and predict the surface evolution at the atomic scale during a high temperature processing. The current-induced electromigration is one of the biggest driving forces to create ordered surface structure. Several other factors such as the anisotropic diffusion, the step permeability, and Ehrlich-Schwoebel effect have been accounted for to determine the atomic positions during the simulation. We forbid two-dimensional nucleation and steps overlapping, and neglected atom sublimation in the simulation. The simulation results show qualitative agreement with the experimental data. These results are used to optimize the micro-scale patterns to obtain atomic scale templates with large atomically flat terraces and preferred step-terrace morphology for nanoscale lithography.