Electrode-dependent electrical switching characteristics of InGaZnO memristor

To improve the electrical characteristics of an analog indium–gallium–zinc oxide (IGZO) memristor, two kinds of IGZO memristive device with different electrode materials were fabricated, and their electrical characteristics and switching mechanisms were analyzed. It was confirmed that the endurance...

Full description

Saved in:
Bibliographic Details
Published inChaos, solitons and fractals Vol. 158; p. 112106
Main Authors Choi, Woo Sik, Kim, Donguk, Yang, Tae Jun, Chae, Inseok, Kim, Changwook, Kim, Hyungjin, Kim, Dae Hwan
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 01.05.2022
Subjects
Analog switching
Endurance
InGaZnO memristor
Linearity
Neuromorphic system
Retention
Neuromorphic system
Endurance
InGaZnO memristor
Retention
Analog switching
Linearity
Online AccessGet full text

Cover

More Information
Summary:To improve the electrical characteristics of an analog indium–gallium–zinc oxide (IGZO) memristor, two kinds of IGZO memristive device with different electrode materials were fabricated, and their electrical characteristics and switching mechanisms were analyzed. It was confirmed that the endurance and retention characteristics of the IGZO memristive device were improved, which was explained by the analysis of operation mechanism. Analog IGZO memristive devices operate through interfacial switching based on the Schottky barrier modulation resulting from the generation and recombination of oxygen vacancies, but the devices with silicon electrode perform analog switching based on conductive filaments through the forming process. The retention time of both devices was measured and estimated with exponential decay functions, and the pulse-induced learning characteristics were also verified. The electrical and synaptic characteristics were considered in the modified national institute of standards and technology (MNIST) database pattern recognition test, and the effects of retention and linearity of weight-update on the recognition rate were analyzed. •InGaZnO memristor devices with different electrode materials for neuromorphic system•Analyzing conduction and switching mechanisms through energy band diagram•Electrode-dependent synaptic behaviors including retention and learning characteristics•Pattern recognition test considering synaptic learning and retention characteristics
ISSN:0960-0779
1873-2887
DOI:10.1016/j.chaos.2022.112106