Faculty Mentor – Dr. Joseph G. Harrison
The core of one part of this program is the introduction of undergraduates to the world of electronic structure calculations as a tool in studying the physical properties of molecules and solids. The 10-week program will begin with a two-week intensive introduction to molecular orbital (MO) and energy-band calculations using a user-friendly package (BICON-CEDiT) developed by the Calzaferri group at the University of Berne based in part on programs developed by the Hoffmann group at Cornell. This package is based on extended Huckel theory (EHT) and is strongly intuitive yet, because of its semiempirical basis, capable of sufficient accuracy to correctly predict material properties of a very wide range of molecular and solid-state systems. The introductory material will survey some of the quantum mechanics behind each program but only to the extent needed to establish the terminology used in some of the program input. Since the EHT is largely based on concepts that students with an introductory chemistry background can readily grasp (atomic ionization energies, orbital overlap, etc.), the mode of instruction will be heavily oriented towards visual presentations and tutorial calculations on simple molecules and solids. 1 week of library research and definition of a problem within the student's capability will follow the introduction. The next 4-5 weeks would be devoted to carrying through with necessary computation to solve the problem. The last 2 weeks will be spent on organizing the results for presentation in American Physical Society style before other students and faculty. The use of standardized codes has the advantage in giving the student experience greater portability. Furthermore because EHT is a widely used approximation the student will be able to recognize other work in the literature carried out using such codes. Because of the simplicity of the EHT program, it has tremendous flexibility in dealing with systems ranging from molecules, to clusters, to solids, and even to surface/interface systems. Some project areas are: 1) Effect of ligands on vibrational spectra, spin resonance parameters, optical and nonlinear optical properties, etc. Each of the perturbations may be visualized in terms of local-orbital modifications. 2) Electronic and vibronic features of BCS superconductors such as LiBC and MgB2. Solid state features such as band-width, band gap, effective mass, density of states, etc. may be studied, again with emphasis on visual presentation over detailed theoretical treatment. For a student who is more inclined toward software development, a project will involve the development of a tool which uses the output from the EHT code to produce derived quantities such as electrical conductivity, bulk modulus, phonon spectra, etc.
The second program deals with the computer simulation of the MPCVD process. This problem subdivides along two research lines. The first deals with the chemical kinetics of the gas phase and the heterogeneous gas/surface reactions. For this the student will learn to use a standard package (CHEMKIN) and to modify existing model reaction networks to study the effects of key parameters like temperature, pressure, feed-gas composition and substrate on the growth of diamond or graphite. The second line of research is the investigation of the coupling of the microwave energy and its role in 1) promotion of atomic hydrogen production and 2) heating of the gas and substrate. For this we will use a program called ARGUS, which was developed in part for such applications. The goal is to eventually couple both lines of research into a package that more fully treats the overall MPCVD process. REU students Steven Baker, Richard Lee, Brian Geislinger, Danny Green, MaQuita Warren and Will Shanks have worked on similar computational materials research projects and made several presentations at the Alabama Materials Research Conferences [17-18].