Partial Spin Absorption Induced Magnetization Switching and Its Voltage-Assisted Improvement in an Asymmetrical All Spin Logic Device at the Mesoscopic Scale

Yue Zhang,Zhizhong Zhang,Lezhi Wang,Jiang Nan,Zhenyi Zheng,Xiang Li,Kin Wong,Yu Wang,Jacques-Olivier Klein,Pedram Khalili Amiri,Youguang Zhang,Kang L. Wang,Weisheng Zhao
DOI: https://doi.org/10.1063/1.4997422
IF: 4
2017-01-01
Applied Physics Letters
Abstract:Beyond memory and storage, future logic applications put forward higher requirements for electronic devices. All spin logic devices (ASLDs) have drawn exceptional interest as they utilize pure spin current instead of charge current, which could promise ultra-low power consumption. However, relatively low efficiencies of spin injection, transport, and detection actually impede high-speed magnetization switching and challenge perspectives of ASLD. In this work, we study partial spin absorption induced magnetization switching in asymmetrical ASLD at the mesoscopic scale, in which the injector and detector have the nano-fabrication compatible device size (>100 nm) and their contact areas are different. The enlarged contact area of the detector is conducive to the spin current absorption, and the contact resistance difference between the injector and the detector can decrease the spin current backflow. Rigorous spin circuit modeling and micromagnetic simulations have been carried out to analyze the electrical and magnetic features. The results show that, at the fabrication-oriented technology scale, the ferromagnetic layer can hardly be switched by geometrically partial spin current absorption. The voltage-controlled magnetic anisotropy (VCMA) effect has been applied on the detector to accelerate the magnetization switching by modulating magnetic anisotropy of the ferromagnetic layer. With a relatively high VCMA coefficient measured experimentally, a voltage of 1.68 V can assist the whole magnetization switching within 2.8 ns. This analysis and improving approach will be of significance for future low-power, high-speed logic applications.
What problem does this paper attempt to address?