Jia-CHEN, Sungjun Kim, Ying‐Chen Chen, Min-Hwi Kim, Yi Li, X. Miao, Yao‐Feng Chang, Byung-Gook Park, Jack C. Lee
{"title":"考虑温度效应的hfox基忆阻器的突触特性-热突触行为的演示","authors":"Jia-CHEN, Sungjun Kim, Ying‐Chen Chen, Min-Hwi Kim, Yi Li, X. Miao, Yao‐Feng Chang, Byung-Gook Park, Jack C. Lee","doi":"10.1109/VLSI-TSA.2018.8403853","DOIUrl":null,"url":null,"abstract":"In this work, we fabricated HfOx-based RRAM-type memristors that are CMOS-compatible and analyzed its synapse characteristics. Ni/HfOx/p++Si filament-type device exhibits abrupt resistive changes in both SET and RESET operations in the same pulse response as the DC sweep, while interface-type Ni/HfOx/n++Si devices show gradual changes. Interface-type devices exhibit more dynamic states as the temperature rises, while the recognition rate of neural network using memristive synapses did not degrade for T < 145 °C. Our results suggest that neuromorphic chips fabricated using RRAM-type memristors can operate under wide temperature range.","PeriodicalId":209993,"journal":{"name":"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)","volume":"16 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Synaptic properties considering temperature effect in HfOx-based memristor - Demonstration of homo-thermal synaptic behaviors\",\"authors\":\"Jia-CHEN, Sungjun Kim, Ying‐Chen Chen, Min-Hwi Kim, Yi Li, X. Miao, Yao‐Feng Chang, Byung-Gook Park, Jack C. Lee\",\"doi\":\"10.1109/VLSI-TSA.2018.8403853\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this work, we fabricated HfOx-based RRAM-type memristors that are CMOS-compatible and analyzed its synapse characteristics. Ni/HfOx/p++Si filament-type device exhibits abrupt resistive changes in both SET and RESET operations in the same pulse response as the DC sweep, while interface-type Ni/HfOx/n++Si devices show gradual changes. Interface-type devices exhibit more dynamic states as the temperature rises, while the recognition rate of neural network using memristive synapses did not degrade for T < 145 °C. Our results suggest that neuromorphic chips fabricated using RRAM-type memristors can operate under wide temperature range.\",\"PeriodicalId\":209993,\"journal\":{\"name\":\"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)\",\"volume\":\"16 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/VLSI-TSA.2018.8403853\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 International Symposium on VLSI Technology, Systems and Application (VLSI-TSA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VLSI-TSA.2018.8403853","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Synaptic properties considering temperature effect in HfOx-based memristor - Demonstration of homo-thermal synaptic behaviors
In this work, we fabricated HfOx-based RRAM-type memristors that are CMOS-compatible and analyzed its synapse characteristics. Ni/HfOx/p++Si filament-type device exhibits abrupt resistive changes in both SET and RESET operations in the same pulse response as the DC sweep, while interface-type Ni/HfOx/n++Si devices show gradual changes. Interface-type devices exhibit more dynamic states as the temperature rises, while the recognition rate of neural network using memristive synapses did not degrade for T < 145 °C. Our results suggest that neuromorphic chips fabricated using RRAM-type memristors can operate under wide temperature range.