Dislocation-templated antiferromagnetic domains in epitaxial NiO
Abstract
Controlling antiferromagnetic domains is essential for spintronics, yet predictable manipulation remains challenging due to their lack of net magnetization. Here, we utilize scanning electron microscopy electron channeling contrast imaging (ECCI) to observe the formation of nanoscale antiferromagnetic domains in epitaxial NiO/MgO(001) at interface dislocations. Thermal cycling across the Néel temperature reveals that while domain contrast vanishes in the paramagnetic phase, the features re-emerge at identical spatial locations upon cooling. Diffraction-vector-dependent ECCI imaging, governed by a relative tilt contrast model, indicates that the domains stem from localized rhombohedral distortions of volumetric twin-type T-domains. These T-domains are stabilized by the long-range shear stress field of interface misfit dislocations (MDs). By tracking a film thickness series from 23 to 46~nm, we resolve a transition in oxide relaxation mechanics from primary 60∘ interface MDs along ⟨110⟩ to a cross-slipped network of pure edge 90∘ MDs deviated toward ⟨100⟩. We demonstrate that ECCI identifies the Burgers vector and line direction of these MDs via strict g⃗ · (b⃗ × u⃗) invisibility criteria, and an Eshelby interaction energy model accurately predicts the asymmetric nucleation of specific T-domains along MD flanks. Quantitative image analysis shows T-domain volume fractions track the spatial configuration of the evolving MD networks, providing a framework for defect-engineering antiferromagnetic textures using one-dimensional defects.
Related
- Temperature and Thickness Dependent Spin Current Transmission through Epitaxial Antiferromagnetic α-Fe2O3
- Light-matter interaction between templated molecular layers and surface lattice resonances
- Wavefunction-templated metamaterials
- Correcting for Evidence Uncertainty in the Templated Background Search for the Stochastic Gravitational-Wave Background
- Strain engineering of altermagnetic symmetry in epitaxial RuO 2 films
BibTeX
@misc{montenegro2026dislocationtemplated,
title = {{Dislocation-templated antiferromagnetic domains in epitaxial NiO}},
author = {Alexandra Fonseca Montenegro and Justin Michel and Fengyuan Yang and Roberto C. Myers},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2606.17302},
eprint = {2606.17302},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.17302},
}
RIS
TY - UNPB
TI - Dislocation-templated antiferromagnetic domains in epitaxial NiO
AU - Alexandra Fonseca Montenegro
AU - Justin Michel
AU - Fengyuan Yang
AU - Roberto C. Myers
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2606.17302
UR - https://arxiv.org/abs/2606.17302
AB - Controlling antiferromagnetic domains is essential for spintronics, yet predictable manipulation remains challenging due to their lack of net magnetization. Here, we utilize scanning electron microscopy electron channeling contrast imaging (ECCI) to observe the formation of nanoscale antiferromagnetic domains in epitaxial NiO/MgO(001) at interface dislocations. Thermal cycling across the Néel temperature reveals that while domain contrast vanishes in the paramagnetic phase, the features re-emerge at identical spatial locations upon cooling. Diffraction-vector-dependent ECCI imaging, governed by a relative tilt contrast model, indicates that the domains stem from localized rhombohedral distortions of volumetric twin-type T-domains. These T-domains are stabilized by the long-range shear stress field of interface misfit dislocations (MDs). By tracking a film thickness series from 23 to 46~nm, we resolve a transition in oxide relaxation mechanics from primary 60^∘ interface MDs along ⟨110⟩ to a cross-slipped network of pure edge 90^∘ MDs deviated toward ⟨100⟩. We demonstrate that ECCI identifies the Burgers vector and line direction of these MDs via strict g⃗ · (b⃗ × u⃗) invisibility criteria, and an Eshelby interaction energy model accurately predicts the asymmetric nucleation of specific T-domains along MD flanks. Quantitative image analysis shows T-domain volume fractions track the spatial configuration of the evolving MD networks, providing a framework for defect-engineering antiferromagnetic textures using one-dimensional defects.
ER -