Computational modeling and simulation study of electronic and thermal properties in semiconductor nanostructures

Abhijeet Paul, Purdue University

Abstract

The technological progress in dimensional scaling has not only kept Silicon CMOS industry on Moore's law for the past five decades but has also benefited many other areas such as thermoelectricity, photo-voltaics, and energy storage. Extending CMOS beyond Si (More Moore, MM) and adding functional diversity to CMOS (More Than Moore, MTM) requires a thorough understanding of the basic electron and heat flow in semiconductors. Along with experiments computer modeling and simulation are playing an increasingly vital role in exploring the numerous possibilities in materials, devices and systems. With these aspects in mind the present work applies computational physics modeling and simulations to explore the, (i) electronic, (ii) thermal, and (iii) thermoelectric properties in nano-scale semiconductors. The electronic structure of zinc-blende and lead-chalcogenide nano-materials is calculated using an atomistic Tight-Binding model. The phonon dispersion in zinc-blende materials is obtained using the Modified Valence Force Field model. Electronic and thermal transport at the nano-scale is explored using Green's function method and Landauer's method. Thermoelectric properties of semiconductor nanostructures are calculated using Landauer's method. Using computer modeling and simulations the variation of the three physical properties (i-iii) are explored with varying size, transport orientation, shape, porosity, strain and alloying of nanostructures. The key findings are, (a) III-Vs and Ge with optimized strain and orientation can improve transistors' and thermoelectric performance, (b) porous Si nanowires provide a lucrative idea for enhancing the thermoelectric efficiency at room temperature, and (c) Si/Ge superlattice nanowires can be used for nano-scale tuning of lattice thermal conductivity by period control. The present work led to the development of two new interface trap density extraction methods in ultra-scaled FinFETs and correlation of the phonon shifts in Si nanowires to their shape, size and orientation benchmarked against experimental Raman spectroscopy data, thereby enabling nano-scale metrology. Contribution of two research and six educational tools on nanoHUB.org forms an integral part of the work for global dissemination of semiconductor knowledge. Atomic level manipulation holds the key to engineer material properties at the nano-scale. The findings of this work will hopefully open and guide new ways of engineering the electronic and thermal properties for better performance.

Degree

Ph.D.

Advisors

Klimeck, Purdue University.

Subject Area

Computer Engineering|Electrical engineering|Nanotechnology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS