Skip to main navigation Skip to search Skip to main content

Structural correlations at Si/Si<inf>3</inf>N<inf>4</inf> interface and atomic stresses in Si/Si<inf>3</inf>N<inf>4</inf> nanopixel-10 million-atom molecular dynamics simulation on parallel computers

  • Martina E. Bachlechner
  • , Rajiv K. Kalia
  • , Aiichiro Nakano
  • , Andrey Omeltchenko
  • , Priya Vashishta
  • , Ingvar Ebbsjö
  • , Anupam Madhukar
  • , Guang Lin Zhao

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

We have combined first-principle calculations of charge transfer at the Si/Si3N4 interface with the interaction potential models for bulk Si and Si3N4 to produce a model for the Si/Si3N4 interface. Using these interatomic potentials, million atom molecular dynamics simulations have been performed to characterize the structure of Si(111)/Si3N4(0001) and the Si(111)/a-Si3N4 interfaces. Ten million-atom simulations are performed using multiresolution molecular-dynamics method on parallel computers. Atomic stress distributions are determined in a 54 nm nanopixel on a 0·1 μm silicon substrate. Effects of surfaces, edges, and lattice mismatch at the Si(111)/Si3N4(0001) interface on the stress distributions are also investigated. Stresses are found to be highly inhomogeneous in the nanopixel - the top surface of silicon nitride has a compressive stress of +3 GPa and the stress is tensile, -1 GPa, in silicon below the inter-face. These simulation methods can also be applied to other semiconductor/ceramic interfaces as well as to metal/ceramic and ceramic/ceramic interfaces. © 1999 Elsevier Science Ltd.
Original languageEnglish
Pages (from-to)2265-2272
Number of pages8
JournalJournal of the European Ceramic Society
Volume19
Issue number13-14
DOIs
StatePublished - Jan 1 1999
Externally publishedYes

Fingerprint

Dive into the research topics of 'Structural correlations at Si/Si<inf>3</inf>N<inf>4</inf> interface and atomic stresses in Si/Si<inf>3</inf>N<inf>4</inf> nanopixel-10 million-atom molecular dynamics simulation on parallel computers'. Together they form a unique fingerprint.

Cite this