DocumentCode
1240025
Title
An efficient 3-D spectral-element method for Schrödinger equation in nanodevice simulation
Author
Lee, Joon-Ho ; Liu, Qing Huo
Author_Institution
Dept. of Electr. & Comput. Eng., Duke Univ., Durham, NC, USA
Volume
24
Issue
12
fYear
2005
Firstpage
1848
Lastpage
1858
Abstract
A three-dimensional (3-D) spectral-element method (SEM) based on Gauss-Lobatto-Legendre (GLL) polynomials is proposed to solve the Schrödinger equation in nanodevice simulation. Galerkin´s method is employed to obtain the system equation. The high-order basis functions employed are orthogonal and the numerical quadrature points are the same as the GLL integration points, leading to a diagonal mass matrix and a more sparse stiffness matrix. Thus, the proposed method leads to a regular eigenvalue problem, rather than a generalized eigenvalue problem, greatly reducing the computer-memory requirement and central-processing-unit (CPU) time in comparison with the conventional finite-element method (FEM). Furthermore, the SEM is implemented for high geometrical orders, where curved structures can be modeled up to the accuracy comparable to the interpolation accuracy afforded by the basis functions. Numerical examples verify a spectral accuracy with the interpolation orders, and confirm that higher geometrical orders are essential for curved structures to achieve overall spectral accuracy. Examples of quantum dots in various structures, including a waveguide, are analyzed with mixed boundary conditions. Numerical results show that the SEM is an efficient alternative to conventional FEM and to the finite-difference method (FDM) for nanodevice simulation.
Keywords
Galerkin method; Legendre polynomials; Schrodinger equation; nanoelectronics; quantum dots; spectral analysis; 3D spectral element method; Galerkin method; Gauss-Lobatto-Legendre polynomials; Schrodinger equation; diagonal mass matrix; eigenvalue problem; high geometrical orders; high order basis functions; nanodevice simulation; Eigenvalues and eigenfunctions; Equations; Finite element methods; Gaussian processes; Interpolation; Moment methods; Numerical analysis; Polynomials; Sparse matrices; Transmission line matrix methods; Galerkin´s method; Gauss–Lobatto–Legendre (GLL) interpolation; SchrÖdinger equation; nanodevices; quantum dot; spectral-element method (SEM);
fLanguage
English
Journal_Title
Computer-Aided Design of Integrated Circuits and Systems, IEEE Transactions on
Publisher
ieee
ISSN
0278-0070
Type
jour
DOI
10.1109/TCAD.2005.852675
Filename
1542240
Link To Document