Skip to main navigation Skip to search Skip to main content

Study of bulk and elementary screw dislocation assisted reverse breakdown in low-voltage (<250 V) 4H-SiC p+n junction diodes - Part I: DC properties

  • IEEE
  • NASA Glenn Research Center

Research output: Contribution to journalArticlepeer-review

184 Scopus citations

Abstract

Given the high-density (̃104 cm -2) of elementary screw dislocations (Burgers vector -le with no hollow core) in commercial SiC wafers and epilayers, all large current (>1 A) SiC power devices will likely contain elementary screw dislocations for the foreseeable future. It is therefore important to ascertain the electrical impact of these defects, particularly in high-field vertical power device topologies where SiC is expected to enable large performance improvements in solid-state high-power systems. This paper compares the de-measured reverse-breakdown characteristics of low-voltage (<250 V) small-area (<5 × 10-4 cm2) 4H-SiC p+n diodes with and without elementary screw dislocations. Diodes containing elementary screw dislocations exhibited higher pre-breakdown reverse leakage currents, softer reverse breakdown current-voltage (I-V) knees, and highly localized microplasmic breakdown current filaments compared to screw dislocation-free devices. The observed localized 4H-SiC breakdown parallels microplasmic breakdown observed in silicon and other semiconductors, in which space-charge effects limit current conduction through the local microplasma as reverse bias is increased.

Original languageEnglish
Pages (from-to)478-484
Number of pages7
JournalIEEE Transactions on Electron Devices
Volume46
Issue number3
DOIs
StatePublished - 1999

Keywords

  • PN junctions
  • Power semiconductor diodes
  • Semiconductor defects
  • Semiconductor device breakdown
  • Silicon carbide

Fingerprint

Dive into the research topics of 'Study of bulk and elementary screw dislocation assisted reverse breakdown in low-voltage (<250 V) 4H-SiC p+n junction diodes - Part I: DC properties'. Together they form a unique fingerprint.

Cite this