東北大学
学際科学フロンティア研究所

Recent advances in spintronics demand new ways to control spin dynamics across scales—from atomic structures to GHz-frequency dynamics. A key challenge is developing efficient, electric-field-driven spin control without relying solely on structural asymmetry. This research explores a novel approach based on quantum geometry, a property of quantum wavefunctions, to drive nonlinear spin-orbit torques (SOTs) in two-dimensional magnetic materials.

While conventional SOTs rely on broken inversion symmetry (e.g., at Rashba interfaces), recent theoretical studies have shown that the quantum metric—a component of the quantum geometric tensor—can generate second-order (nonlinear) torques even in inversion-symmetric materials. This represents a shift from symmetry-breaking-based mechanisms to geometry-driven spin control. However, the role of quantum geometry in spin torque generation has yet to be experimentally clarified.

We aim to develop an experimental platform to detect quantum-metric-driven SOTs using GHz-frequency excitation and nonlinear spin pumping in CrSBr, a van der Waals magnet with strong magnetic anisotropy. CrSBr exhibits A-type antiferromagnetic order, with ferromagnetic alignment within each layer and antiferromagnetic coupling between layers—making it ideal for probing quantum geometric effects. Devices with heavy-metal contacts will be used to detect spin currents and angular-dependent ferromagnetic resonance will be performed under high-power excitation to extract nonlinear voltage signals. These results will be compared with theoretical predictions from first-principles calculations of the quantum metric.

This research positions quantum geometry as a meaningful experimental design axis for spintronic materials, opening new directions for spin current generation in both low-symmetry and inversion-symmetric magnets. The outcomes will contribute to the development of spintronic devices for neuromorphic and quantum information technologies.