Ferroelectric artificial synapses for high-performance neuromorphic computing: Status, prospects, and challenges

Abstract

Neuromorphic computing provides alternative hardware architectures with high computational efficiencies and low energy consumption by simulating the working principles of the brain with artificial neurons and synapses as building blocks. This process helps overcome the insurmountable speed barrier and high power consumption from conventional von Neumann computer architectures. Among the emerging neuromorphic electronic devices, ferroelectric-based artificial synapses have attracted extensive interest for their good controllability, deterministic resistance switching, large output signal dynamic range, and excellent retention. This Perspective briefly reviews the recent progress of two- and three-terminal ferroelectric artificial synapses represented by ferroelectric tunnel junctions and ferroelectric field effect transistors, respectively. The structure and operational mechanism of the devices are described, and existing issues inhibiting high-performance synaptic devices and corresponding solutions are discussed, including the linearity and symmetry of synaptic weight updates, power consumption, and device miniaturization. Functions required for advanced neuromorphic systems, such as multimodal and multi-timescale synaptic plasticity, are also summarized. Finally, the remaining challenges in ferroelectric synapses and possible countermeasures are outlined.

Performance Projection of Vacuum Gate Dielectric Doping-Free Carbon Nanoribbon/Nanotube Field-Effect Transistors for Radiation-Immune Nanoelectronics

“…In addition to the aforementioned insights, optimizing vacuum-based devices and electronic systems using bio-inspired optimization approaches (e.g., Ant Colony Optimization (ACO), Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Simulated Annealing (SA)) holds considerable promise [66,67]. Since the vacuum gate dielectric degrades the coupling capacitance, leading to worse carrier control, leveraging negative capacitance ferroelectric materials [68][69][70] to improve such carbon nanodevices can be a Simulation of such devices and their derivatives (considering different gate configurations) while taking into account scattering mechanisms within the NEGF framework can provide a deeper insight into the behavior and performance projection of carbon-based nanotransistors in radiative environments. Additionally, exploring the analog/RF performance of these devices presents an intriguing avenue for further investigation.…”

“…In addition to the aforementioned insights, optimizing vacuum-based devices and electronic systems using bio-inspired optimization approaches (e.g., Ant Colony Optimization (ACO), Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Simulated Annealing (SA)) holds considerable promise [ 66 , 67 ]. Since the vacuum gate dielectric degrades the coupling capacitance, leading to worse carrier control, leveraging negative capacitance ferroelectric materials [ 68 , 69 , 70 ] to improve such carbon nanodevices can be a matter for further investigation. Furthermore, very interesting results are presented in [ 71 , 72 , 73 ].…”

Low-power and robust ternary SRAM cell with improved noise margin in CNTFET technology

“…The design of ternary logic using CNTFET is more prolific than other well technologies due to the ease of adjusting the threshold voltage [54,55]. The CNTFET technology offers ballistic transport [56,57]…”

Ferroelectricity-Defects Synergistic Artificial Synapses for High Recognition Accuracy Neuromorphic Computing

“…Among them, ferroelectric memristors consist of a ferroelectric film sandwiched between two electrodes. They offer advantages such as rapid reading and writing speeds, a continuously adjustable resistance state, a substantial switching ratio, and the absence of a need for an electroforming process. − It is well-known that the occurrence of defects such as oxygen vacancies is unavoidable during the deposition of ferroelectric thin films. Therefore, many researchers have turned to the resistive switching (RS) effect of ferroelectric materials for drift and charge trapping in oxygen vacancy defects. − Ferreyra et al revealed the significance of oxygen vacancy migration and polarization direction flipping in the RS and underscored the crucial role played by oxygen vacancies .…”