Wide-band Lorentzian media in the FDTD algorithm

Marina Y. Koledintseva*, James L. Drewniak, David J. Pommerenke, Giulio Antonini, Antonio Orlandi, Konstantin N. Rozanov

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

This paper considers the case of a wide-band Lorentzian (WBL) algorithm in the finite-difference time-domain (FDTD) modeling of dispersive media. It is shown herein that the WBL model is a physically meaningful and practically useful case of the frequency behavior of materials along with the Debye and narrow-band Lorentzian (NBL). The recursive convolution algorithms for the finite-difference time-domain technique for NBL and WBL models differ. The Debye model, which is suitable for comparatively low-frequency dispersive materials, may not have sufficient number of parameters for describing the wide-band material, especially if this material exhibits pronounced absorption at higher frequencies. It is shown that the Debye model can be used, if the Q-factor of the linear circuit analog corresponding to the Lorentzian model of the material is less than approximately 0.8. If the quality factor is in the limits of about 0.8 < Q ≤ 1, then the WBL model is appropriate. For Q > 1, the NBL model must be applied. The NBL model is suitable for dielectrics exhibiting resonance effects in the microwave frequency range. The WBL model is typical for composites filled with conducting fibers.

Original languageEnglish
Pages (from-to)392-399
Number of pages8
JournalIEEE Transactions on Electromagnetic Compatibility
Volume47
Issue number2
DOIs
Publication statusPublished - 1 May 2005
Externally publishedYes

Keywords

  • Debye model
  • Dispersive media
  • Finite-difference time-domain (FDTD) technique
  • Lorentzian model
  • Recursive convolution

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Condensed Matter Physics
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Wide-band Lorentzian media in the FDTD algorithm'. Together they form a unique fingerprint.

Cite this