Formation of a passivation layer via oxygen plasma oxidation for GaN-on-GaN regrowth

Research output: Contribution to journalResearch articleContributedpeer-review

Contributors

  • Marie Louise Bilke - , Chair of Nanoelectronics, NaMLab - Nanoelectronic materials laboratory gGmbH (Author)
  • Jan Gärtner - , NaMLab - Nanoelectronic materials laboratory gGmbH (Author)
  • Andreas Großer - , NaMLab - Nanoelectronic materials laboratory gGmbH (Author)
  • Thomas Mikolajick - , Chair of Nanoelectronics, NaMLab - Nanoelectronic materials laboratory gGmbH (Author)
  • Alexander Ruf - , NaMLab - Nanoelectronic materials laboratory gGmbH (Author)
  • Stefan Schmult - , Chair of Nanoelectronics (Author)

Abstract

The adhesion of atmospheric silicon (Si) to the highly polar gallium nitride (GaN) regrowth surface leads to the formation of a parasitic, conductive channel in lateral GaN-on-GaN devices. This parasitic channel can result in high transistor off-state leakage currents, impeding the use of GaN as substrate material for certain applications. Ex-situ cleaning methods to remove the surficial Si effectively were unsuccessful, because of fast re-adsorption from the ambient atmosphere. Therefore, establishing a passivation for GaN after an in-situ cleaning step, which can be desorbed prior to GaN/AlGaN heterostructure growth, is considered to be a technical solution. The formation of a thin passivating gallium oxide layer on the GaN surface is demonstrated by oxygen plasma oxidation. While an amorphous character of the passivation layer is needed for subsequent desorption, the oxygen plasma treatment resulted in a stable crystalline phase preventing desorption. Moreover, the plasma treatment causes significant damage to the GaN surface and introduces additional impurities. Specifically, these two aspects – the introduction of additional impurities and defects at the GaN surface and the crystalline character of the oxide – prevented removal of the Si contamination at the GaN/GaN regrowth interface. Nevertheless, the experiments provide valuable insights into one possible oxidation mechanism and particularly point at the necessity of establishing an amorphous gallium oxide passivation layer, which will be the focus of future work.

Details

Original languageEnglish
Article number140910
Number of pages5
JournalThin solid films
Volume839
Publication statusPublished - 12 Mar 2026
Peer-reviewedYes

External IDs

ORCID /0000-0003-3814-0378/work/211721444

Keywords

Keywords

  • Gallium nitride, Gallium oxide, Oxygen plasma oxidation, Parasitic, conductive channel, Passivation layer, Silicon contamination