Generation and optimization of tight binding (TB) parameters using genetic algorithms and their validation using NEMO3D
Abstract
It is often the case in engineering problems that a desired output from a particular process is known, but the inputs that are fed into the process to get this output are not. Optimization, commonly referred to as fitting, reverse engineering, or mathematical programming is a well developed class of problem solving techniques that is used to tackle the above scenario. This thesis intends to discuss the application of one such scheme, the Genetic Algorithm, to specific problems in nano-electronics. We initially intend to motivate the thesis by outlining the main problem at hand: The Tight Binding (TB) parametrization of materials at low temperature (LT). We then show why the Genetic Algorithm is well suited to tackle this non-trivial problem effectively and discuss some of the unique features of Genetic Algorithms. Finally, we describe in detail the procedure used to generate, optimize and validate TB parameters and discuss some results we have for materials like InAs and GaAs. We also include a description of some of the attempts to create a general purpose optimization engine for the nanoHUB (www.nanoHUB.org).
Degree
M.S.E.C.E.
Advisors
Klimeck, Purdue University.
Subject Area
Electrical engineering|Condensed matter physics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.