Synchronizing Spin-Torque Nano-Oscillators Using Low-Damping Pathways
ORAL
Abstract
Spin-torque nano-oscillators (STNOs) have been studied extensively as devices for neuromorphic computing, which aims to use brain-inspired hardware to achieve more energy-efficient computing. Particularly useful is the ability of several STNOs to phase-lock with each other when in close proximity, giving a greater combined power output from the devices [1]. However, this phase-locking behavior can be limited by the distance between two oscillators. Using the micromagnetic simulation software mumax3 [2], we investigate whether adding a low-damping region between two oscillators can enable phase-locking over greater distances. We find that the addition of low-damping pathways gives synchronization over distances where oscillators do not synchronize otherwise. Adding or removing pathways also enables the selective synchronization of particular oscillators in an array. The success of these low-damping regions in phase-locking several STNOs is promising for their application to larger arrays and more complex geometries of oscillators.
[1] Kaka et al., Nature 437, 389–392 (2005).
[2] Vansteenkiste et al., AIP Adv. 4, 107133 (2014).
–
Presenters
-
Katherine Matthews
University of California, San Diego
Authors
-
Katherine Matthews
University of California, San Diego
-
Biswajit Sahoo
University of California San Diego
-
Nakshatra Bansal
University of California, San Diego
-
Alex Frano
University of California, San Diego