Current transport mechanism in trapped oxides: A generalized trap-assisted tunneling model

Mau Phon Houng, Yeong Her Wang, Wai Jyh Chang

Research output: Contribution to journalArticlepeer-review

123 Citations (Scopus)

Abstract

A generalized trap-assisted tunneling (GTAT) model is proposed in this work, where an effective tunneling barrier of trapezoidal shape is considered, instead of the triangular barrier utilized in the conventional trap-assisted tunneling (TAT) model. It is demonstrated that trapezoidal barrier tunneling dominates at low electric fields (E<4 MV/cm), while triangular barrier tunneling contributes the main part of the tunneling current at high electric fields (E=6-8 MV/cm). The comparisons of this improved model and the results of the conventional TAT model at high and low electric fields are discussed. It is concluded that GTAT can more accurately model the current density-electric field (J-E) curves for the conduction enhancement of a trapped oxide film under various deposition conditions over a wider range of electric fields. This is confirmed by the comparative use of both TAT and GTAT models on experimental data obtained from existing reports. Furthermore, a simple method for determining the trap energy level is derived from the J-E relationship. This method provides a convenient way to characterize the trap levels inside the oxide layers, without the need of other complicated measurements. The developed GTAT model can be applied to the investigations of gate oxide reliability, especially the stress-related effects and impurity incorporated oxide films (i.e., SiOF or SiON).

Original languageEnglish
Pages (from-to)1488-1491
Number of pages4
JournalJournal of Applied Physics
Volume86
Issue number3
DOIs
Publication statusPublished - 1999 Aug

All Science Journal Classification (ASJC) codes

  • Physics and Astronomy(all)

Fingerprint

Dive into the research topics of 'Current transport mechanism in trapped oxides: A generalized trap-assisted tunneling model'. Together they form a unique fingerprint.

Cite this