MESH Nano-Oscillator: All electrical doubly tunable spintronic oscillator

Spin-Torque Nano-Oscillators (STNO) are of great research interest due to their compactness and scalability, GHz operation, frequency tunability, and integrability in existing CMOS process flow [1]. However, they have some short-comings that prevents commercial uptake. These are, broadly, the need of an external magnetic field for operation, low output power, poor Q-factor and high phase noise which can be mitigated by injection locking [2]. We propose a new design for STNO that combines the physics of the Magneto-Electric (ME) effect of single phase multiferroics for an electrically tunable, in-situ magnetic field with Giant Spin Hall Effect (GSHE) to provide a separate current terminal and a Magnetic Tunnel Junction for read-out and additional spin current. In a recent experiment, the ME effect was demonstrated to fully switch a magnet along its easy axis [3], and therefore can be used to generate tunable magnetic fields required in STNO operation in conjunction with the established GSHE physics and MTJs. The proposed design addresses the drawbacks of the STNOs by eliminating the need for a global external magnetic field. The dual tunability of operating frequency by changing both the spin-currents and magnetic field electrically provides flexibility for circuit design. Possibility of voltage control of magnetism allows injection locking by an rf voltage source that can be used for synchronizing large array of oscillators, circumventing existing difficulties associated with impedance mismatch of injection locking through spin-currents. Note that the ME phenomenon is very different from Voltage-Controlled Magnetic Anisotropy (VCMA) or compound multiferroics based proposals as it does not change the intrinsic anisotropy of the magnet but creates an external magnetic field that can be tuned by an electric field. The device has been modeled (fig. 1b) and all the results are generated using the multi-physics spin-circuit framework described in [6]. We believe the proposed design provides critical remedies towards STNOs' adoption in modern architectures.