Ab initio characterization of B, C, N, and O in bcc iron: Solution and migration energies and elastic strain fields

Maaouia Souissi, Ying Chen, Marcel H.F. Sluiter, Hiroshi Numakura

Research output: Contribution to journalArticlepeer-review

31 Citations (Scopus)


Practical and reliable methods for theoretically determining the properties of B, C, N, and O in bcc iron have been explored by systematic DFT calculations. The energies of solution and migration, and the elastic strain fields due to the solute atom have been evaluated by supercell calculations under various conditions. By applying correction for spurious elastic interaction of the solute atom with its images in the periodic supercells, reasonable estimates of the solution energy have been obtained without employing very large supercells. The correction turned out unimportant for the migration energy, as it modifies the energies of the stable position and the saddle-point similarly. The lambda tensor, which uniquely characterizes the strain field induced by a solute atom, has been evaluated for the four species of various configurations, first through the force-dipole tensor obtained in zero-strain calculations, and second from changes in supercell dimensions in zero-stress calculations. The two procedures give similar results that typically differ by a few per cent from each other. The computed values for C and N in octahedral interstitial sites are comparable to experimental values. When experimental data become available for B and O, these evaluations will resolve the as-yet contentious location of these atomic species in bcc iron.

Original languageEnglish
Pages (from-to)249-258
Number of pages10
JournalComputational Materials Science
Publication statusPublished - 2016 Nov 1


  • Density functional theory
  • Diffusion
  • Iron
  • Point defects
  • Solute atoms
  • Thermodynamics


Dive into the research topics of 'Ab initio characterization of B, C, N, and O in bcc iron: Solution and migration energies and elastic strain fields'. Together they form a unique fingerprint.

Cite this