Emerging synaptic devices: from two-terminal memristors to multiterminal neuromorphic transistors. (December 2019)

• Record Type: Journal Article
• Authors: Jiang, S.; Nie, S.; He, Y.; Liu, R.; Chen, C.; Wan, Q.
• Is Part Of: Materials today nano. Volume 8(2019)
• Publication Date: 2019-12
• Subjects: Neuromorphic device -- Resistive-random-access memory -- Phase-change memory -- Electrolyte-gated thin-film transistor -- Ferroelectric thin-film transistor; Nanoscience -- Periodicals; Nanotechnology -- Periodicals; Nanotechnology; Nanoscience; Nanotechnology -- Periodicals; Periodicals; Periodical; Electronic journals; Electronic journals; 620.5
• DOI: 10.1016/j.mtnano.2019.100059
• Languages: English
• ISSNs: 2588-8420

Abstract

Abstract: The conventional von-Neumann computing architecture is facing challenges including the heat wall, the memory wall, and the end of Moore's law, which become even more severe in today's era of big-data. Therefore, researchers are committed to exploit a new computing paradigm to address these issues and satisfy our ever-growing appetite for data and information. Inspired by our brain, new-concept synaptic devices and systems have aroused great research interests. Up to now, diverse neuromorphic devices have been proposed, such as resistive random-access memories, phase-change memories, electrolyte-gated thin-film transistors, and ferroelectric thin-film transistors. This review presents the structures, the operation mechanisms, and the advantages and disadvantages of each kind of synaptic devices. Moreover, recent advances of those synaptic devices for the hardware implementation of neuromorphic computing have also been reviewed in detail, as well as the challenges that need to be addressed in future research.