Zinc-Alloyed HFO2 Synaptic RRAM with Operating Voltage and Switching Energy Enhancement

Jun Lan1; Quanzhou Zhu1; Yiyang Zhang1; Wenhui Wang1; Muhammad Zaheer1; Jinxuan Liang1; Mei Shen1; Zhixiong Li2; Zhen Chen2; Hongxue Wei2; Guobiao Zhang1; Yida Li1

10.1109/CSTIC55103.2022.9856852

2022

2022 IEEE CSTIC

978-1-6654-9759-6

2022 China Semiconductor Technology International Conference (CSTIC)

1-3

20-21 June 2022

Shanghai, China

In this work, we demonstrate the performance engineering of a HfO2-based analog RRAM device based on a Zn alloying approach using ALD. The Zn-alloyed HfO2 RRAM results in a significant smaller operating voltages and lower switching energy due to the increase in oxygen vacancies elucidated via XPS measurements. Meanwhile, its retention at 85 °C can exceed that of 105 s. The LTP and LTD of the Zn-alloyed RRAM shows good linearity for the implementation of artificial neural network in handwriting recognition. Our results provide a pathway for RRAM to meet the requirements of emerging embedded memory applications for analog computing.

Performance Evaluation Handwriting Recognition Semiconductor Device Measurement Voltage Measurement Neuromorphic Engineering Memory Management Switches

第一 ; 通讯

EI

20223712740163

Character Recognition ; Hafnium Oxides ; Neural Networks ; Zinc Alloys

Zinc And Alloys:546.3 ; Data Storage, Equipment And Techniques:722.1 ; Chemical Products Generally:804

IEEE

1.Southern University of Science and Technology, Shenzhen, China
2.Shenzhen Longsys Electronics Co., Ltd, Shenzhen, China

Jun Lan,Quanzhou Zhu,Yiyang Zhang,et al. Zinc-Alloyed HFO2 Synaptic RRAM with Operating Voltage and Switching Energy Enhancement[C],2022:1-3.