ON-state retention of Atom Switch eNVM for IoT/AI Inference Solution

An ON-state retention of a 40nm-node atom switch embedded nonvolatile memory (eNVM) has been carefully investigated for IoT/AI inference solution. Based on ON-conductance (G on ) tuning model of atom switch, one order of magnitude lower programming power is achieved while keeping the same G on . Sma...

Full description

Saved in:
Bibliographic Details
Published in2020 IEEE International Reliability Physics Symposium (IRPS) pp. 1 - 4
Main Authors Okamoto, Koichiro, Nebashi, Ryusuke, Banno, Naoki, Bai, Xu, Numata, Hideaki, Iguchi, Noriyuki, Miyamura, Makoto, Hada, Hiromitsu, Funahashi, Kazunori, Sugibayashi, Tadahiko, Sakamoto, Toshitsugu, Tada, Munehiro
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.04.2020
Subjects
Atom switch
CBRAM
Nonvolatile memory
ON-state retention
Programming
Reliability
Switches
Temperature
Temperature dependence
Tuning
Online AccessGet full text

Cover

More Information
Summary:An ON-state retention of a 40nm-node atom switch embedded nonvolatile memory (eNVM) has been carefully investigated for IoT/AI inference solution. Based on ON-conductance (G on ) tuning model of atom switch, one order of magnitude lower programming power is achieved while keeping the same G on . Smaller ON-state retention dependences on temperature (E a = 0.2eV) and time (n = 0.11) are experimentally clarified and the lifetime is predicted to be more than 10 years at 150°C under +20% shift criteria of G on .
ISSN:1938-1891
DOI:10.1109/IRPS45951.2020.9128967