Abstract

The electronic structure of nanostructures has been almost universally addressed by the “standard model” of effective-mass k·p envelope function approach. While eminently successful for quantum wells, this model breaks down for small structures, in particular, for small dots and wires[l]. Until recently, it was impractical to test the “standard model” against more general approaches that allow many-band (Γ-X-L) coupling. However, it is now possible, due to special tricks[2], to apply the all-band pseudopotential method to 10³ - 10⁴ atom nanostructures. This shows (i) how the “standard model” fails, for thin superlattices, [3], (ii) how size effect lead to a reduction in dielectric constants[3] and to band gaps that differ from what is expected in effective-mass theory, (iii) the emergence of a “zero-confinement state” in 2D films [4], (iv) that small dots of III-V materials have an indirect gap that converts to direct above a critical size[5], (v) how the spectra of CdSe dots evolve from the bulk[6] and (vi) how the spectra of dots of Si, GaAs, InP and CdSe compare with experiment, and (vii) how the use of pseudopotential wavefunctions leads to very different electron-hole coulomb and exchange energies relative to the “standard model”.

© 1997 Optical Society of America