I. Materials

Unconventional exchange interactions

In the field of spintronics, research has focused on engineering couplings between magnetic bilayers, driven by symmetric and anti-symmetric exchange interactions. Symmetric interactions yield ferromagnetic (FM) and antiferromagnetic (AFM) couplings, including the famous Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction. The anti-symmetric Dzyaloshinskii-Moriya interaction (DMI) has gained attention due to its role in noncollinear chiral spin configurations, requiring inversion symmetry breaking and strong spin-orbit coupling (SOC). Applications include skyrmions and Néel domain wall logics. Our recent work extended DMI effects from 2D to 3D systems, emphasizing the importance of in-plane symmetry breaking. We demonstrated interlayer DMI's potential for current-induced field-free switching in orthogonally magnetized systems, offering a promising path for practical magnetic memory devices and materials growth approaches.

II. Devices

Enhancing spin-orbit torque efficiency for next generation MRAM

The spin Hall effect (SHE) in heavy transition metals has been experimentally proved to be strong enough to induce magnetization switching and magnetic oscillations in the adjacent ferromagnetic layer via spin-orbit torque (SOT) mechanism. The spin Hall angle, which roughly describes the sign and the magnitude of the charge-to-spin conversion efficiency, is about 0.06, -0.15, and -0.30 for Pt, Ta, and W, respectively. Other emergent materials systems such as topological insulators have also been proved to possess giant spin Hall effect, with spin Hall angle possibly greater than unity (>100%). More recently,  we showed that orbital Hall effect (OHE) from light transition metals can also be used for enhancing the overall SOT efficiency. Our research aims for materials systems with high spin plus orbital Hall effects and their potential applications, for instance in the magnetic random access memory (MRAM) industry.

III. Beyond Moore

Unconventional computing based on spintronics

Unconventional computing based on spintronics encompasses a wide range of innovative approaches, including neuromorphic device, compute-in-memory architectures, and quantum-inspired machines, which have the potential to reshape the landscape of data processing and optimization tasks. Spintronic compute-in-memory devices integrate memory and processing functions within a single unit, leveraging the intrinsic properties of electron spin to perform both tasks simultaneously. This paradigm shift has the potential to dramatically reduce the data movement bottleneck, significantly enhancing the efficiency of computation, particularly for AI and machine learning applications. By embedding spintronic memory elements like magnetic tunnel junctions into the processing units, compute-in-memory architectures promise to accelerate complex calculations and reduce energy consumption.