Antisite defects at oxide interfaces

Hanghui Chen and Andrew Millis

Phys. Rev. B 93, 104111 – Published 31 March 2016

Abstract

We use ab initio calculations to estimate the formation energies of cation (transition-metal) antisite defects at oxide interfaces and to understand the basic physical effects that drive or suppress the formation of these defects. Antisite defects are found to be favored in systems with substantial charge transfer across the interface, while Jahn-Teller distortions and itinerant ferromagnetism can prevent antisite defects and help stabilize atomically sharp interfaces. Our results enable identification of classes of systems that may be more and less susceptible to the formation of antisite defects, and they motivate experimental studies and further theoretical calculations to elucidate the local structure and stability of oxide interface systems.

Received 21 June 2015
Revised 3 March 2016

DOI:https://doi.org/10.1103/PhysRevB.93.104111

Physics Subject Headings (PhySH)

Defects

Interfaces

Ab initio calculations

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

Hanghui Chen^1,2 and Andrew Millis¹

¹Department of Physics, Columbia University, New York, New York 10027, USA
²Department of Applied Physics and Applied Math, Columbia University, New York, New York 10027, USA

Article Text (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 93, Iss. 10 — 1 March 2016

Reuse & Permissions

Access Options

Article Available via CHORUS

Download Accepted Manuscript

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Panel A: $2 \times 2 \times 2$ supercell. Panel B: $\sqrt{2} \times \sqrt{2} \times 8$ supercell. Column 1: ideal interfaces with no antisite defects. Column 2: one antisite defect in the supercell. Column 3: two antisite defects in the supercell. Antisite defects are highlighted by the black ellipses.
Reuse & Permissions
Figure 2
Solid symbols (solid lines are there to guide the eye): total energies (with respect to the total energy of ideal interfaces) of various antisite defect configurations (see Fig. 1) calculated (a) in a $2 \times 2 \times 2$ supercell (40 atoms) and (b) in a $\sqrt{2} \times \sqrt{2} \times 8$ supercell (80 atoms). Dashed lines: estimation of defect formation energies calculated in a $\sqrt{2} \times \sqrt{2} \times 2$ supercell (20 atoms). $N$ is the number of antisite defects in the supercell.
Reuse & Permissions
Figure 3
Sketch of atomic and magnetic structures considered in this work. Panels A: rocksalt configuration. Panels B: layered configuration. Large orange balls denote $A$ -site ions La or Sr; small red balls denote O ions; shaded octahedra denote octahedra containing transition-metal $M$ (darker shade) and $M^{'}$ (lighter shade) ions, respectively. A1 and B1 are ferromagnetic ordering; A2 and B2 are checkerboard $G$ -type antiferromagnetic ordering; B3 is $A$ -type antiferromagnetic ordering (spins are parallel in each layer and antiparallel between adjacent layers). Magnetic moments are schematically indicated by green/yellow arrows.
Reuse & Permissions
Figure 4
A: $La M O_{3} / La M^{'} O_{3}$ . (A1) DFT-calculated defect formation energy $Δ E^{DFT}$ for each combination. The unit is eV per defect. (A2) Structural parameters $λ$ and $Q$ for each combination with the most favorable magnetic structure. $λ$ is shown in blue in the lower-right triangle and $Q$ is shown in red in the upper-left triangle. The unit on the color bar for $λ$ and $Q$ is %. (A3) Comparison between the fitted defect formation energy $Δ E^{fit}$ and the DFT-calculated defect formation energy $Δ E^{DFT}$ . $Δ E^{fit}$ are obtained by minimizing $Ω$ of Eq. (4). B: $Sr M O_{3} / Sr M^{'} O_{3}$ . (B1) Same as A1. (B2) Same as A2. The red (blue) backslash denotes those combinations in which $S_{M M'}$ is $- 1$ (+1). The other combinations without a backslash have a zero value of $S_{M M'}$ . (B3) Same as A3. The solid squares denote $Δ E^{fit}$ that are obtained by minimizing $Ω_{Sr}$ of Eq. (5). The open squares denote $Δ E^{fit}$ that are obtained by minimizing $Ω$ of Eq. (4). The combinations with a red or blue slash are explicitly labeled.
Reuse & Permissions
Figure 5
(A1) Top view of two vertically adjacent layers in the $R$ configuration with a large oxygen octahedron (blue) and a small oxygen octahedron (purple). (A2) Top view of two vertically adjacent layers in the $L$ configuration with a naturally large oxygen octahedron (blue) and a naturally small oxygen octahedron (purple). Compatibility with the geometry of the $L$ configuration imposes strains (green arrows) on the system, compressing the large octahedra and expanding the small ones. (B1) Top view of two vertically adjacent layers in the $L$ configuration with one anisotropic oxygen octahedron (purple) and one isotropic oxygen octahedron (blue). (B2) Top view of two vertical adjacent layers in the $R$ configuration with one naturally anisotropic oxygen octahedron (purple) and one naturally isotropic oxygen octahedron (blue). Compatibility with the geometry of the $R$ configuration imposes strains (green arrows) reducing the bond disproportionation of the anisotropic material and inducing a disproportionation in the isotropic one. Rotations and tilts of oxygen octahedra are suppressed for clarity.
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics