Electron charge densities at conduction-band edges of semiconductors

Abstract

%'e demonstrate that both the empirical pseudopotential method (EPM) and the linear combination of atomiclike orbitals (LCAO) approach are capable of producing consistent electronic charge distributions in a compound semiconductor. Since the EPM approach is known to produce total valence electron charge densities which compare well with experimental x-ray data (e.g. , Si), this work serves as a further test for the LCAO method. In particular, the EPM scheme, which uses an extended plane-wave basis, and the LCAO scheme, which employs a localized Gaussian basis, are used, with the same empirical potential as input, to analyze both the total valence electron charge density and the charge density of the first conduction band at the I, L, and X k. points of the Brillouin zone. These charge densities are decomposed into their s-, p-, and d-orbital contributions, and this information is used to interpret the differences in the topologies of the conduction bands at I, L, and X. Such differences are crucial for a comprehensive understanding of interstitial impurities and the response of specific band states to perturbations in compound semiconductors.

Spin–orbit coupling in bulk GaAs

“…These couplings give the splitting linear to the first order of momentum around the valley bottoms. The corresponding splitting coefficients β L/X = ∆E(k)/2k for different momentum are plotted in Figs been revealed that the inclusion of the d orbit greatly improves the accuracy of the effective mass in L-valley [9,28,29,30,31]. Therefore, in our opinion the spin splitting determined by sp 3 s * d 5 model in L-valley is also more reliable than that by sp 3 s * model.…”

Calculation of Electronic States in Semiconductor Heterostructures with an Empirical spds* Tight-Binding Model

“…Analyzing the degenerate sets of reciprocal lattice vectors close to the direct gap, the decomposition into irreducible representations h1Y 1Y 1i 3 2G 1 2G 4 and h2Y 0Y 0i 3 G 1 G 3 G 4 allows to decide that the inclusion of a full set of five atomic d-states is the better choice, as these cover d3 G 3 G 4 . This completeness argument for the occupied states coincides with the necessity to use d-symmetric basis orbitals known from the projection of conduction band wavefunctions on the atomic symmetries [13,14].…”

“…Most of them use the precise experimental knowledge of band positions for well-controlled input parameters determining the quantities entering the model. From interpretations of the electronic wavefunctions calculated with pseudopotential methods in terms of atomic symmetries [13,14], it became obvious that d-symmetric contributions play a crucial role both for the valence band maximum at G and for the conduction states at X and L. This has led to basis extensions of earlier sp 3 [11] and sp 3 s * [15] TB models including parts of the d-states [16]. Finally, it turned out that the inclusion of all five atomic d-states becomes necessary for a reliable description of the valence bands and the first two conduction bands around the Brillouin zone [17], defining the empirical sp 3 d 5 s * TB model applied in the present work.…”

Effect of electron-nuclear spin interactions for electron-spin qubits localized in InGaAs self-assembled quantum dots

“…The same trend is reported in psudopotential studies and another tight-binding calculation. [48][49][50] The effective wave vector of the electron wave function in the quantum dot is roughly…”