Electrical annealing and stochastic resonance in superparamagnets for oscillatory networks with dynamic connectivity

ORAL

Abstract

In this work, we report electrical annealing and stochastic resonance in thermally unstable nanomagnets with weak perpendicular anisotropy made from Ta/CoFeB/MgO stack. The magnetization of such nanomagnets shows random telegraphic fluctuation with a mean that is tunable by the giant spin Hall effect (GSHE) current generated in the Ta underlayer. These nanomagnets form stochastic oscillators with a natural frequency f0, given by f0 = (tup+tdn)-1, where tup and tdn are the average dwell time of the magnetization in the “up” and “down” state. We demonstrate tuning of f0 of individual nanomagnets by electrical feedback of the magnetization state to the GSHE underlayer. Depending on the polarity and strength of the feedback, the magnetization fluctuation becomes slower or faster, analogous to temperature annealing. We further demonstrate the phenomenon of stochastic resonance in our nanomagnets, where the magnetization follows a weak external periodic drive through the GSHE underlayer only when the drive frequency matches with f0. These novel phenomena can be used to realize coupled oscillatory nanomagnet networks with dynamic connectivity, which alleviates the full connectivity requirement in a network of N oscillators from N2 weighted connections to N homogenous connections.

Presenters

  • Punyashloka Debashis

    Electrical and Computer Engineering, Purdue University

Authors

  • Punyashloka Debashis

    Electrical and Computer Engineering, Purdue University

  • Pramey Upadhyaya

    Electrical and Computer Engineering, Purdue University, Purdue University

  • Zhihong Chen

    Electrical and Computer Engineering, Purdue University