Infrared Thermal Imaging and Electrical Characterization of Substrate-Dependence of Spin Hall Nano-Oscillators Characteristics
Oral-In-person · Withdrawn
Abstract
Spin Hall nano-oscillators (SHNO) are promising spintronic candidates for neuromorphic computing and Ising machine implementations. Controlling their characteristics is crucial for such applications, yet the role of substrate thermal conductivity, a key factor influencing device temperature, nonlinearity, output power, and linewidth, remains insufficiently explored. In this work, we systematically investigate identical CoFeB/Pt SHNOs fabricated on Si, Al₂O₃, and SiO₂/Si substrates, whose thermal conductivity decreases by roughly an order of magnitude from Si to Al₂O₃ to SiO₂/Si. By combining electrical measurements of auto-oscillations with infrared thermal imaging, we study how temperature changes driven by Joule heating affect the oscillator characteristics. Our results show that substrate-dependent temperature rise exerts a pronounced impact on SHNO behavior, significantly affecting their nonlinear dynamics, spectral linewidth, and oscillation power. Notably, auto-oscillations are completely suppressed in devices on SiO₂/Si due to excessive heating. In contrast, devices on Al₂O₃ exhibit stronger thermally induced nonlinearity and mode competition, and lower output power than devices on Si. These findings demonstrate the critical role of temperature in governing SHNO performance, underscoring the need to account for Joule heating to understand and control their characteristics.
–
Presenters
-
Padma Radhakrishnnan
- New York University