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

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 (similar to 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 (>10(3)), 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 similar to 93.8% on fashion product images is obtained, which is very close to the result of similar to 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.

Automated summary

Ferroelectric tunnel junctions (FTJs) have the capability to scale down and exhibit high-performance and multi-state storage characteristics, enabling them to emulate the fundamental functions of biological synapses. A convolutional neural network simulation based on experimental results achieves high recognition accuracy on fashion product images and exhibits robustness to input image noises, demonstrating the potential of FTJs for information storage and neural network computing.