Skip to main navigation Skip to search Skip to main content

Analysis of radiation defects in gallium nitride using deep level transient spectra and first principles methods

  • Harbin Institute of Technology

Research output: Contribution to journalArticlepeer-review

Abstract

Radiation defects in gallium nitride devices were analyzed by combining experiments with simulation methods. Various forms of native defects, impurities, and these hydrogenated defect complexes in GaN were studied with first-principles calculations employing hybrid functionals which exhibit a reference data base for experimental analysis. We consider hydrogen complexes made of the combinations of single point defects (VGa, VN), di-vacancies (VGa-VGa, VN-VN, VGa-VN) and impurities (C, O, Si). Formation energies were computed for configurations with different charge states after full geometry optimizations. From the calculated formation energies, thermodynamic transition levels were evaluated, which are related to the thermal activation energies observed in experimental techniques such as deep level transient spectroscopy (DLTS). Hydrogen interstitial atoms were identified by the combination of DLTS and first-principles calculations. The transition level measured by DLTS was EC-0.5 eV, which is consistent with the calculation value. The signal of EC-0.9 eV peak in DLTS may be consisting of [VGa]-3/-2 and [VN-VN]+1/+2. According to the degradation of the electrical performance of the AlGaN/GaN high electron mobility transistors and the calculation results, the hydrogen passivation may be caused by [VGa]-3, [VGa-VGa]-2 and [VGa-VN]-1.

Original languageEnglish
Article number165120
JournalNuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms
Volume545
DOIs
StatePublished - Dec 2023
Externally publishedYes

Keywords

  • DLTS
  • GaN
  • HEMTs
  • Hydrogen
  • Transition levels

Fingerprint

Dive into the research topics of 'Analysis of radiation defects in gallium nitride using deep level transient spectra and first principles methods'. Together they form a unique fingerprint.

Cite this