Ab Initio Studies of the Diffusion of Intrinsic Defects and Silicon Dopants in Bulk InAs

  • Mardochee Reveil
  • , Hsien Lien Huang
  • , Huang Ta Chen
  • , Jason Liu
  • , Michael O. Thompson
  • , Paulette Clancy

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

We expose the predominant diffusional pathways for In and As in InAs, as well as dopant Si atoms in InAs, using Nudged Elastic Band calculations in conjunction with accurate Density Functional Theory calculations of the energy of defective systems. Our results show that As is a very fast diffuser compared to In and Si for both vacancy-assisted and interstitially mediated mechanisms. Larger indium atoms, on the other hand, are very slow diffusers and strongly prefer to remain on the In sublattice. Silicon also prefers to stay in substitutional sites in the In sublattice, in agreement with the fact that Si is used to create n-doped InAs. We find that the mechanism by which Si diffuses within the InAs lattice is very unlikely to proceed via vacancy-assisted jumps, since these routes encounter energy barriers above 2 eV. In contrast, silicon can readily make interstitial jumps since they occur with energy barriers as small as 0.23 eV. This suggests that an interstitial diffusion mechanism is strongly preferred for Si diffusion in InAs which challenges the common presumption made for another similar III-V compound, namely GaAs, that Si diffusion takes place via a vacancy-assisted mechanism.

Original languageEnglish
Pages (from-to)11484-11489
Number of pages6
JournalLangmuir
Volume33
Issue number42
DOIs
StatePublished - Oct 24 2017
Externally publishedYes

Funding

The authors wish to acknowledge the following current and former members of the Clancy group for useful discussions: Dr. Binit Lukose, Dr. Jonathan Saathoff, Dr. Victoria Sorg, Jingyang Wang, and Yacet Acevedo. Mardochee Reveil thanks the Colman family for generous funding through the Colman Fellowship at Cornell. This work benefitted from computing resources provided by the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number ACI-1053575 and from the Cornell Institute of Computational Science and Engineering (ICSE).

Fingerprint

Dive into the research topics of 'Ab Initio Studies of the Diffusion of Intrinsic Defects and Silicon Dopants in Bulk InAs'. Together they form a unique fingerprint.

Cite this