講演情報
[15p-P07-52]Efficient Optimization of Interfacial Structures in Lattice-Mismatched Fe/MgO Interfaces Using Machine Learning and DFT Integration
〇(D)Andi MuhNurFitrah Syamsul1, Kohji Nakamura1 (1.Mie Univ.)
キーワード：
Magnetic tunnel junctions (MTJs)、Interface optimization、Machine learning
Magnetic Tunnel Junctions (MTJs) with Fe/MgO interfaces are crucial for STT-MRAM technology due to their strong perpendicular magnetic anisotropy (PMA) and voltage-controlled magnetic anisotropy (VCMA) properties. However, interfacial defects, particularly those arising from lattice mismatch, significantly impact device performance. Using the Global Optimization with First-principles Energy Expressions (GOFEE) method, we investigate atomic-scale defect structures at Fe/MgO interfaces. Our analysis reveals two distinct optimized structures showing a relationship between Fe crystallinity and MgO layer stability. The first structure exhibits well-ordered bcc Fe with oxygen displacement, while the second shows compromised Fe crystallinity but better MgO stability. These findings demonstrate that interfacial oxygen dynamics critically influence structural integrity and reveal subtle defect mechanisms often overlooked by conventional methods.