Advertisement

Non-Linear Charge Dynamics in Semiconductor Superlattices

  • D. Sánchez
  • M. Moscoso
  • R. Aguado
  • G. Platero
  • L. L. Bonilla
Conference paper
Part of the Lecture Notes in Physics book series (LNP, volume 547)

Abstract

In the last few years, vertical transport in n-doped weakly coupled semiconductor superlattices has been shown to exhibit non-linear phenomena such as domain formation, multistability, self-sustained current oscillations, and driven and undriven chaos [1]. The most succesful theoretical models combine discrete rate equations for the carrier density and the electric field in the different wells, and use reasonable boundary conditions which mimic the experimental setup [2]. Although these models explain rather well the observed phemomena, the boundary conditions turn out to have a strong effect on the dynamics of electric-field domains.

We study self-sustained current oscillations in weakly-doped superlattices by means of a self-consistent microscopic model of vertical sequential tunneling which includes boundary conditions in a natural way [3]. For highly doped injecting contacts, self-oscillations arise due to recycling and motion of domain walls which are charge monopoles. As the contact doping decreases, a new oscillatory mode due to recycling and motion of charge dipole waves appears. This mode has not been observed so far because the contact doping density is too high in the usual experimental setups. We predict that dipole-mediated oscillations dominate at low doping for which monopole-mediated oscillations disappear. There is an intermediate doping range where both oscillation modes coexist as stable solutions, and hysteresis between them is possible [4].

In addition, our model reproduce experimentally observed current spikes. They are due to the well-to-well hopping of domain walls and appear as a high-frequency oscillation superimposed to the natural current oscillation due to monopole dynamics. Our model makes a distinction between the average potential drops at barriers and wells which (together with the backward tunneling current present only at low electric fields) causes spikes. Thus spikes should also be present at high bias, as experimentally observed. Several preceding models needed to introduced disorder in the doping to obtain current spikes, in a less natural way than ours. [4].

References

  1. 1.
    H. T. Grahn et al, Phys. Rev. Lett. 67, 1618 (1991); J. Kastrup et al, Phys. Rev. B 55, 2476 (1997); Y. Zhang et al, Phys. Rev. Lett. 77, 3001 (1996). K.J. Luo et al, Phys. Rev. Lett. 81, 1290 (1998).CrossRefGoogle Scholar
  2. 2.
    F. Prengel et al, Phys. Rev. B 50, 1705 (1994); L. L. Bonilla et al, Phys. Rev. B 50, 8644 (1994); O. M. Bulashenko and L. L. Bonilla, Phys. Rev. B 52, 7849 (1995); L. L. Bonilla et al, SIAM J. Appl. Math. 57, 1588 (1997).CrossRefGoogle Scholar
  3. 3.
    R. Aguado et al, Phys. Rev. B 55, R16053 (1997).CrossRefGoogle Scholar
  4. 4.
    D. Sánchez et al, (unpublised).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • D. Sánchez
    • 1
  • M. Moscoso
    • 2
  • R. Aguado
    • 1
  • G. Platero
    • 1
  • L. L. Bonilla
    • 2
  1. 1.Instituto de Ciencia de Materiales (CSIC)MadridSpain
  2. 2.Universidad Carlos IIILeganésSpain

Personalised recommendations