The Combined Effect of Spin-transfer Torque and Voltage-controlled Strain Gradient on Magnetic Domain-wall Dynamics: Toward Tunable Spintronic Neuron

Abstract

Magnetic domain wall (DW), as one of the promising information carriers in spintronic devices, have been widely investigated owing to their nonlinear dynamics and tunable properties. Here, we have theoretically and numerically demonstrated the DW dynamics driven by the synergistic interaction between current-induced spintransfer torque (STT) and voltage-controlled strain gradient (VCSG) in multiferroic heterostructures. Through electromechanical and micromagnetic simulations, we have shown that a desirable strain gradient can be created and further modulated the equilibrium position and velocity of the current-driven DW motion. Meanwhile, an analytical Thiele’s model is developed to describe the steady motion of DW and the analytical results are quite consistent with the simulation one. Finally, we find that this combination effect can be leveraged to design DW-based biological neurons where the synergistic interaction between STT and VCSG-driven DW motion as integrating and leaking motivates mimicking leaky-integrate-and-fire (LIF) and self-reset function. Importantly, the firing response of the LIF neuron can be efficiently modulated, facilitating the exploration of tunable activation function generators, which can further help improve the computational capability of the neuromorphic system.