Spintronic logic design methodology based on spin Hall effect–driven magnetic tunnel junctions. (14th January 2016)

• Record Type: Journal Article
• Authors: Kang, Wang; Wang, Zhaohao; Zhang, Youguang; Klein, Jacques-Olivier; Lv, Weifeng; Zhao, Weisheng
• Is Part Of: Journal of physics. Volume 49:Number 6(2016)
• Publication Date: 2016-01-14
• Subjects: direct cascading -- magnetic tunnel junction -- spin Hall effect -- spintronic logic; Physics -- Periodicals; 530
• DOI: 10.1088/0022-3727/49/6/065008
• Languages: English
• ISSNs: 0022-3727

Abstract

Abstract: Conventional complementary metal-oxide-semiconductor (CMOS) technology is now approaching its physical scaling limits to enable Moore's law to continue. Spintronic devices, as one of the potential alternatives, show great promise to replace CMOS technology for next-generation low-power integrated circuits in nanoscale technology nodes. Until now, spintronic memory has been successfully commercialized. However spintronic logic still faces many critical challenges (e.g. direct cascading capability and small operation gain) before it can be practically applied. In this paper, we propose a standard complimentary spintronic logic (CSL) design methodology to form a CMOS-like logic design paradigm. Using the spin Hall effect (SHE)-driven magnetic tunnel junction (MTJ) device as an example, we demonstrate CSL implementation, functionality and performance. This logic family provides a unified design methodology for spintronic logic circuits and partly solves the challenges of direct cascading capability and small operation gain in the previously proposed spintronic logic designs. By solving a modified Landau–Lifshitz–Gilbert equation, the magnetization dynamics in the free layer of the MTJ is theoretically described and a compact electrical model is developed. With this electrical model, numerical simulations have been performed to evaluate the functionality and performance of the proposed CSL design. Simulation results demonstrate that the proposed CSL design paradigm is rather promising for low-power logic computing.