First-principles study of point defects in an fcc Fe-10Ni-20Cr model alloy

Abstract

The influence of the local environment on vacancy and self-interstitial formation energies has been investigated in a face-centered-cubic (fcc) Fe-10Ni-20Cr model alloy by analyzing an extensive set of first-principle calculations based on density functional theory. Chemical disorder has been considered by designing special quasirandom structures and four different collinear magnetic structures have been investigated in order to determine a relevant reference state to perform point defect calculations at 0 K. Two different convergence methods have also been used to characterize the importance of the method on the results. Although our fcc Fe-10Ni-20Cr would be better represented in terms of applications by the paramagnetic state, we found that the antiferromagnetic single-layer magnetic structure was the most stable at 0 K and we chose it as a reference state to determine the point defect properties. Point defects have been introduced in this reference state, i.e., vacancies and Fe-Fe, Fe-Ni, Fe-Cr, Cr-Cr, Ni-Ni, and Ni-Cr dumbbell interstitials oriented either parallel or perpendicular to the single layer antiferromagnetic planes. Each point defect studied was introduced at different lattice sites to consider a sufficient variety of local environments and analyze its influence on the formation energy values. We have estimated the point defect formation energies with linear regressions using variables which describe the local environment surrounding the point defects. The number and the position of Ni and Cr first nearest neighbors to the point defects were found to drive the evolution of the formation energies. In particular, Ni is found to decrease and Cr to increase the vacancy formation energy of the model alloy, while the opposite trends are found for the dumbbell interstitials. This study suggested that, to a first approximation, the first nearest atoms to point defects can provide reliable estimates of point defect formation energies.

Average-atom interatomic potential for random alloys

“…Large fluctuations of some point defect properties with the local chemical environnment -without attempt to reach the average propertyhave also been observed by some authors by DFT calculations in concentrated alloys 55,56 . Attempts to obtain interaction energies between "solutes" X 0 = Fe, Ni and Cr and a stacking fault E int X0−γ were made for the equicomposition fcc Fe 33 Ni 33 Cr 33 alloy.…”

Defect energetics of concentrated solid-solution alloys from ab initio calculations: Ni0.5Co0.5, Ni0.5Fe0.5, Ni0.8Fe0.2and Ni0.8Cr0.2

“…Indeed, a recent study of the formation energies of intrinsic point defects in a Fe-10Ni-20Cr model alloy reveals that the variables describing the local environment surrounding a point defect can be used to fit the formation energies of defects. 12 The dominant factor is shown to be the first nearest neighbors around the defect site. However, the supercell size is limited in this method.…”

“…[9][10][11][12][13][14] However, for concentrated solid-solution alloys, the chemical disorder makes it difficult to calculate the defect energies. Usually, there are two methods that can be used to address this problem.…”

Ab initioprediction of vacancy properties in concentrated alloys: The case of fcc Cu-Ni

“…[20,21]), and from the theoretical side also, there have been rather few studies that deal with specific alloy systems. Initially using empirical potentials [22][23][24][25], and later through ab initio approaches, both through supercell calculations [26][27][28][29] and ab initio based cluster expansions [30][31][32][33], it has been established that the local atomic environment around a vacancy plays a significant role. While the influence of vacancies on phase stability [30,34] and kinetics [31][32][33] received some attention for Al-Ni [30], Sc-S [34], and Al-Li [31][32][33] alloys, the actual vacancy properties in specific alloys were mostly neglected with the exception of an empirical potential study of Cu-Ni alloys by Zhao et al [24].…”