Two methods for calculating transport parameters in semiconductor superlattices by applying Green’s functions are compared in the paper. For one of the methods, the Wannier functions method, where computations in the complex space and Wannier functions base are required, the Hamiltonian matrix is small in size and its elements depend solely on the energy. For the real space method, as it operates in the floating point domain and uses the Hamiltonian containing the elements dependent both on energy and position, the Hamiltonian matrix is larger in size. The size makes the method computationally challenging. To find the consequences of choosing one of the methods, a direct comparison between the computations, obtained for both methods with the same input parameters, was undertaken. The differences between the results are shown and explained. Selected simulations allowed us to discuss advantages and disadvantages of both methods. The calculations include transport parameters such as the density of states and the occupation functions, with regard to scattering processes where the self-consistent Born approximation was used, as well as the spatial distribution of electron concentration for two superlattices structures. The numerical results are obtained within the non-equilibrium Green’s functions formalism by solving the Dyson and the Keldysh equations.
In this work we discuss 3D selfconsistent solution of Poisson and Schrödinger equations for electrostatically formed quantum dot. 3D simulations give detailed insight into the energy spectrum of the device and allow us to find values of respective voltages ensuring given number of electrons in the dot. We performed calculations for fully 3D potential and apart from that calculations for the same potential separated into two independent parts, i.e. regarding to the plane of 2DEG and to the direction perpendicular to the meant plane. We found that calculations done for the two independent parts of the potential give good information about quantum dot properties and they are much faster compared to fully 3D simulations.