First-principles study of point defects in Ni₃Al

Abstract

The energetics and structural properties of native, substitutional and interstitial defects in NiAl have been investigated by first-principles methods. In particular, we have determined the formation energies of composition conserving defects and established that the so-called penta defect, which consists of four vacancies on Ni sublattice and Ni antisite on the Al sublattice, is the main source of vacancies in NiAl. We show that this is due to the strong Ni-site preference of vacancies in NiAl. We have also calculated the site substitution behaviour of Cu, Pd, Pt, Si, Ti, Cr, V, Nb, Ta and Mo and their effect on the concentration expansion coefficient. We show the latter information can used for an indirect estimate of the site substitution behaviour of the alloying elements. The solution energy of carbon and its effect on the lattice constant of NiAl have been obtained in the dilute limit in the first-principles calculations. We have also determined the chemical and strain-induced carbon–carbon interactions in the interstitial positions of NiAl. These interactions have been subsequently used in the statistical thermodynamic simulations of carbon ordering in NiAl.

Keywords:

• defect formation energies
• site preference
• Ni₃Al
• carbon interstitials
• strain-induced interactions

Acknowledgments

AVR acknowledges the financial support of the support of the Swedish Research Council (VR project 15339-91505-33) and the European Research Council (ERC) grant. This work was partly performed within the VINNEX center Hero-m, financed by the Swedish Governmental Agency for Innovation Systems (VINNOVA), Swedish industry and the Royal Institute of Technology (KTH). This work was supported by Russian Federation Grant 12-03-00227. The computations were performed using the hardware of inter-agency supercomputer center of the Russian Academy of Science (MVS-100K supercomputer) and Research Computational Center of Moscow State University (Chebyshev SKIF supercomputer complex, Moscow State University). Computer resources for this study have been also provided by the Swedish National Infrastructure for Computing (SNIC) and MATTER Network, at the National Supercomputer Center (NSC), Linköping.