Modelling of Materials for Silicon-Compatible Microelectronics

Ab initio density functional theory (DFT) is an established method to model the behaviour of materials at the atomic scale. At the IHP, we use it to investigate materials systems that are of interest to the most popular and cost-efficient technology, by which electronics is made today: the silicon technology. Here we report on the results obtained for various materials: (a) for strictly 2D atomic sheets (graphene), (b) for heteroepitaxial layers (oxides and nitrides), their surfaces, and the interfaces between these films, and (c) for bulk crystals (defects in silicon). The graphene sheets are intended as components of chemical sensors, optical modulators, and high-speed and high-power transistors. The chemical reactions and diffusion processes governing the nucleation and growth of graphene on perfect (flat and stepped) and defected surfaces of germanium films were simulated, and the mechanisms responsible for the observed growth modes were elucidated. The Sc oxide and nitride films constitute the topmost part of a heterostructure on which GaN diodes, lasers, and high-power transistors can be assembled. The simulations provided insight into the intermixing of oxygen and nitrogen. The substrate on which all these films and other device structures are grown, is crystalline silicon. For numerous applications it is critical that the substrate getters (collects and binds) the impurities that are unintentionally introduced by the technological process. The formation of oxygen precipitates used as the gettering centres is associated with the presence of missing atoms (vacancies) in the Si bulk. We studied the process of vacancy clustering and oxidation, we extrapolated the clustering results to infinite separation between the defects, and we discussed the implications also for the interpretation of deep level transient spectroscopy (DLTS) or for the strategy to perform numerically expensive defect calculations (as done with hybrid potentials), among others