Proceedings of TECHCON 2008, Austin, TX


As semiconductor devices scale down, the role of surfaces and interfaces becomes increasingly important. This effort seeks to develop methods for analysis of nano-structures, including the effect of surfaces, which represent a broader family of non-ideal bonding environments in the material, and evaluate the effects of such environments on device behavior. The presented work focuses on surfaces in 1-D silicon nanostructures. While a lot of theory and computational methods, particularly the empirical sp3d5s* tightbinding (TB) and the non-equilibrium Green's function (NEGF) approach, already exist for predicting electronic structure and transport through nano-devices, the lack of knowledge of the modified Hamiltonian that includes the effect of changes in the bonding at surfaces and interfaces inhibit the use of such theory for a full and precise analysis of devices. Even worse, sometimes the actual atomic configuration might be unknown. Hence, we first present a viable technique to predict atomic configuration for structures where nothing or very little is known a priori and show novel low energy structures of silicon nano-tubes and their properties. We then present how the use of density functional theory (DFT) helps one construct the proper TB Hamiltonian including the effects of non-ideal bonds. We show that the bulk empirical parameters give an electronic structure that qualitatively matches well with DFT-LDA and GW results, but the band gap is lower than the GW corrected gap. We also present the electronic structure of small dimension silicon nanowires and show how surfaces can play an important role in futuristic small dimension devices.

Date of this Version

September 2008