High-Speed Nanoscale Ferroelectric Tunnel Junction for Multilevel Memory and Neural Network Computing

Abstract

Ferroelectric tunnel junction (FTJ) is one promising candidate for next-generation nonvolatile data storage and neural network computing systems. In this work, the high-performance 50 nm-diameter Au/Ti/PbZr0.52Ti0.48O3 (∼3 nm, (111)-oriented)/Nb:SrTiO3 (Nb: 0.7 wt %) FTJs are achieved to demonstrate the scaling down capability of FTJ. As a nonvolatile memory, the FTJ shows eight distinct resistance states (3 bits) with a large ON/OFF ratio (>103), and these states can be switched at a fast speed of 10 ns. Intriguingly, the long-term potentiation/depression and spike timing-dependent plasticity, that is, fundamental functions of biological synapses, can be emulated in the nanoscale FTJ-based artificial synapse. A convolutional neural network (CNN) simulation is then carried out based on the experimental results, and a high recognition accuracy of ∼93.8% on fashion product images is obtained, which is very close to the result of ∼94.4% by a floating-point-based CNN software. In particular, the FTJ-based CNN simulation also exhibits robustness to input image noises. These results indicate the great potential of FTJ for high-density information storage and neural network computing.

Giant Electroresistance in Ferroelectric Tunnel Junctions via High-Throughput Designs: Toward High-Performance Neuromorphic Computing

“…(a) R – V loops, (b) resistances of HRS and LRS, and (c) corresponding ON/OFF ratios of all of the junctions. (d) ON/OFF ratios of FTJs in recent years. ,− (e) HRS and LRS retentions of the different BST-based devices and inset plots τ of different devices. (f) Retention time versus the retained ON/OFF ratio of FTJs in recent years. ,,,,,,,− …”

Ferroelectric Transistors for Memory and Neuromorphic Device Applications

“…Another type of ferroelectric memory is the FTJ. [18,[32][33][34] FTJ and ferroelectric capacitors exhibit similar structures, with an extremely thin ferroelectric layer used in the FTJ to facilitate the tunneling of charge carriers through the ferroelectric layer (Figure 2b). The accumulation or depletion of charge carriers at the ferroelectric layer-electrode interface depends on the polarization state of the ferroelectric layer.…”

Pure ZrO2 Ferroelectric Thin Film for Nonvolatile Memory and Neural Network Computing

“…Therefore, emerging neuromorphic electronic devices that can be utilized in efficient in-memory computing for AI tasks have been proposed and investigated . Among the various neuromorphic electronic mechanisms, , such as ferroelectric polarization, ferromagnetic magnetization, phase change, and conductive filament creation/disruption, the electric field-driven ferroelectric polarization is promising for realizing low write power consumption and nonvolatility. − In particular, ferroelectric fluorite structure HfO 2 -based thin films have drawn tremendous interest as a result of the good ferroelectricity at the nanometer scale and complementary metal oxide semiconductor (CMOS) compatibility. , Similar to HfO 2 , fluorite structure ZrO 2 is also CMOS compatible, in which the ferroelectricity was demonstrated very recently . More interestingly, the onset crystallization temperature of ZrO 2 is lower than that of HfO 2 ; for instance, the onset crystallization temperatures of ∼7 nm thick HfO 2 and ZrO 2 are ∼500 and ∼400 °C, respectively .…”