Multiscale modeling of doping processes in advanced semiconductor devices

The development of advanced semiconductor devices relies heavily on technology computer-aided design. Front-end process simulators model the fabrication of devices including different process steps and effects such as ion implantation, dopant and defect diffusion and interaction, epitaxial growth, and stress effects. Continuum process simulators continue to be the main tool; however, kinetic Monte Carlo simulators with and without lattice have become important as well. With the need to explore new materials other than silicon, such as SiGe, germanium, and III-V, more fundamental modeling is required to compensate for missing experimental information. Ab initio methods based on density functional theory and molecular dynamics can help address open questions of material characteristics that cannot be extracted explicitly by experiments. In this paper, we give an overview of a multiscale approach for advanced process modeling and discuss the latest progress.