Published May 1, 2012 | Version v1
Journal article

Numerical method for hydrodynamic modulation equations describing Bloch oscillations in semiconductor superlattices

  • 1. Gregorio Millán Institute for Fluid Dynamics, Nanoscience and Industrial Mathematics, Universidad Carlos III de Madrid, Avenida de la Universidad 30, 28911 Leganés (Spain)

Description

We present a finite difference method to solve a new type of nonlocal hydrodynamic equations that arise in the theory of spatially inhomogeneous Bloch oscillations in semiconductor superlattices. The hydrodynamic equations describe the evolution of the electron density, electric field and the complex amplitude of the Bloch oscillations for the electron current density and the mean energy density. These equations contain averages over the Bloch phase which are integrals of the unknown electric field and are derived by singular perturbation methods. Among the solutions of the hydrodynamic equations, at a 70 K lattice temperature, there are spatially inhomogeneous Bloch oscillations coexisting with moving electric field domains and Gunn-type oscillations of the current. At higher temperature (300 K) only Bloch oscillations remain. These novel solutions are found for restitution coefficients in a narrow interval below their critical values and disappear for larger values. We use an efficient numerical method based on an implicit second-order finite difference scheme for both the electric field equation (of drift-diffusion type) and the parabolic equation for the complex amplitude. Double integrals appearing in the nonlocal hydrodynamic equations are calculated by means of expansions in modified Bessel functions. We use numerical simulations to ascertain the convergence of the method. If the complex amplitude equation is solved using a first order scheme for restitution coefficients near their critical values, a spurious convection arises that annihilates the complex amplitude in the part of the superlattice that is closer to the cathode. This numerical artifact disappears if the space step is appropriately reduced or we use the second-order numerical scheme.

Availability note (English)

Available from http://dx.doi.org/10.1016/j.jcp.2012.02.024

Additional details

Identifiers

DOI
10.1016/j.jcp.2012.02.024;
arXiv
arXiv:1205.1292v1;
PII
S0021-9991(12)00133-7;

Publishing Information

Journal Title
Journal of Computational Physics
Journal Volume
231
Journal Issue
13
Journal Page Range
p. 4499-4514
ISSN
0021-9991
CODEN
JCTPAH

Optional Information

Copyright
Copyright (c) 2012 Elsevier Science B.V., Amsterdam, The Netherlands, All rights reserved.