High performance resistive random access memory based on Ag/TiO 2 Nanorods/FTO for image recognition applications
Kanyu Yang,Chaojie Shi,Ruizhao Tian,Haoyue Deng,Jie He,Yangyang Qi,Zhengchun Yang,Jinshi Zhao,Zhen Fan,Jun Liu
DOI: https://doi.org/10.1016/j.mssp.2023.107998
IF: 4.1
2023-11-24
Materials Science in Semiconductor Processing
Abstract:This study investigates the electrical and synaptic properties of Ag/TiO 2 nanorod/FTO-based resistive random access memory (RRAM) devices, focusing on the impact of different seed layer thicknesses on nanorod thickness and RRAM performance. The devices exhibit notable achievements, including a high DC endurance of up to 150,000 cycles, a self-compliance function of 10 −2 A, and a resistance switching ratio exceeding 100-fold. Notably, the device with a 3.5 μm TiO 2 nanorod thickness demonstrates exceptional stability. Analysis of the IV curve reveals that the switching mechanism is attributed to space-charge-limited conduction (SCLC) resulting from electron trapping in oxygen vacancy traps. Furthermore, the device maintains stable synaptic properties even after undergoing 20 cycles of long-term potentiation and depression. When employed in the context of artificial neural networks for image recognition, these devices attain high accuracy rates in recognizing handwritten digits, achieving approximately 90% accuracy for 8 × 8 pixel images and approximately 88% for 28 × 28 pixel images, demonstrating their suitability for practical applications.
engineering, electrical & electronic,materials science, multidisciplinary,physics, condensed matter, applied