Si tight-binding parameters from genetic algorithm fitting

Klimeck, G Bowen, RC Boykin, TB Salazar-Lazaro, C Cwik, TA Stoica, A

G. Klimeck et al., Si tight-binding parameters from genetic algorithm fitting, SUPERLATT M, 27(2-3), 2000, pp. 77-88

26

INGLESE

Article

Apllied Physucs/Condensed Matter/Materiales Science

SUPERLATTICES AND MICROSTRUCTURES

0749-6036 → ACNP

27

2-3

2000

77 - 88

ISI

0749-6036(200002/03)27:2-3<77:STPFGA>2.0.ZU;2-5

Quantum mechanical simulations of carrier transport in Si require an accura te model of the complicated Si bandstructure. Tight-binding models are an a ttractive method of choice since they bear the full electronic structure sy mmetry within If hem and can discretize a realistic device on an atomic sca le. However, tight-binding models are not simple to pa rameterize and chara cterize. This work addresses two issues: (1) the need for an automated fitt ing:procedure that maps tight-binding orbital interaction-energies to physi cal observables such as effective masses and band edges, and (2) the capabi lities and accuracy of the nearest and second-nearest neighbor tight-bindin g sp3s* models with respect to carrier transport in indirect bandgap materi als. A genetic algorithm approach is used to fit orbital interaction energi es of these tight-binding models in a nine-; and 20-dimensional global opti mization problem for Si. A second-nearest neighbor sp3s* parameter set that fits all relevant conduction and valence band properties: with a high degr ee of accuracy is presented. No such global fit was found for the nearest n eighbor sp3s* model and two sets, one heavily weighed for electron properti es and the other for hole properties, are presented. Bandstructure properti es relevant for electron and hole transport in Si derived from these three sets are compared with the seminal Vogl et al. [Journal of the Physics and Chemistry of Solids 44, 365 (1983)] parameters. (C) 2000 Academic Press.