Mimicking Synaptic Behaviors with Cross-Point Structured TiOx/TiOy-Based Filamentary RRAM for Neuromorphic Applications

Jongtae Kim, Sanghoon Cho, Taeheon Kim, James Jungho Pak

Research output: Contribution to journalArticle

Abstract

This paper presents the fabrication and characterization of the cross-point structure 20 × 20 μm2 RRAM with TiOx/TiOy bi-layer insulator for synaptic application in neuromorphic systems. The measured oxygen concentration of the TiOx/TiOy switching layers of the fabricated devices using X-ray photoelectron spectroscopy analysis showed that the oxygen concentration ratio between TiOx and TiOy is ~ 1.5. After electroforming at ~ 5.62 V, the on/off ratio was ~ 76 at 0.2 V with the DC sweep voltage scheme. Synaptic behaviors including long-term potentiation (LTP) and long-term depression (LTD) were performed with 50 identical pulses for the implementation of RRAM into neuromorphic systems based on convolutional neural networks. Also, linearly increased (or decreased) 25 pulses were applied to the device so that the conductance changes linearly. The resulting linear LTP and LTD characteristics were mirror-symmetric, which could maximize the accuracy. For Hebbian learning, the device also mimicked the spike-timing-dependent plasticity properties with a conductance change from − 77.79% to 96.07% using a time-division multiplexing approach.

Original languageEnglish
Pages (from-to)869-875
Number of pages7
JournalJournal of Electrical Engineering and Technology
Volume14
Issue number2
DOIs
Publication statusPublished - 2019 Mar 1

Fingerprint

Electroforming
Time division multiplexing
Oxygen
Plasticity
Mirrors
X ray photoelectron spectroscopy
Neural networks
Fabrication
Electric potential
RRAM

Keywords

  • Cross-point
  • Neuromorphic
  • RRAM
  • Synaptic application
  • TiO/TiO bi-layer

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Mimicking Synaptic Behaviors with Cross-Point Structured TiOx/TiOy-Based Filamentary RRAM for Neuromorphic Applications. / Kim, Jongtae; Cho, Sanghoon; Kim, Taeheon; Pak, James Jungho.

In: Journal of Electrical Engineering and Technology, Vol. 14, No. 2, 01.03.2019, p. 869-875.

Research output: Contribution to journalArticle

@article{521ef09a9113474e817288b44ac481b5,
title = "Mimicking Synaptic Behaviors with Cross-Point Structured TiOx/TiOy-Based Filamentary RRAM for Neuromorphic Applications",
abstract = "This paper presents the fabrication and characterization of the cross-point structure 20 × 20 μm2 RRAM with TiOx/TiOy bi-layer insulator for synaptic application in neuromorphic systems. The measured oxygen concentration of the TiOx/TiOy switching layers of the fabricated devices using X-ray photoelectron spectroscopy analysis showed that the oxygen concentration ratio between TiOx and TiOy is ~ 1.5. After electroforming at ~ 5.62 V, the on/off ratio was ~ 76 at 0.2 V with the DC sweep voltage scheme. Synaptic behaviors including long-term potentiation (LTP) and long-term depression (LTD) were performed with 50 identical pulses for the implementation of RRAM into neuromorphic systems based on convolutional neural networks. Also, linearly increased (or decreased) 25 pulses were applied to the device so that the conductance changes linearly. The resulting linear LTP and LTD characteristics were mirror-symmetric, which could maximize the accuracy. For Hebbian learning, the device also mimicked the spike-timing-dependent plasticity properties with a conductance change from − 77.79{\%} to 96.07{\%} using a time-division multiplexing approach.",
keywords = "Cross-point, Neuromorphic, RRAM, Synaptic application, TiO/TiO bi-layer",
author = "Jongtae Kim and Sanghoon Cho and Taeheon Kim and Pak, {James Jungho}",
year = "2019",
month = "3",
day = "1",
doi = "10.1007/s42835-019-00107-y",
language = "English",
volume = "14",
pages = "869--875",
journal = "Journal of Electrical Engineering and Technology",
issn = "1975-0102",
publisher = "Korean Institute of Electrical Engineers",
number = "2",

}

TY - JOUR

T1 - Mimicking Synaptic Behaviors with Cross-Point Structured TiOx/TiOy-Based Filamentary RRAM for Neuromorphic Applications

AU - Kim, Jongtae

AU - Cho, Sanghoon

AU - Kim, Taeheon

AU - Pak, James Jungho

PY - 2019/3/1

Y1 - 2019/3/1

N2 - This paper presents the fabrication and characterization of the cross-point structure 20 × 20 μm2 RRAM with TiOx/TiOy bi-layer insulator for synaptic application in neuromorphic systems. The measured oxygen concentration of the TiOx/TiOy switching layers of the fabricated devices using X-ray photoelectron spectroscopy analysis showed that the oxygen concentration ratio between TiOx and TiOy is ~ 1.5. After electroforming at ~ 5.62 V, the on/off ratio was ~ 76 at 0.2 V with the DC sweep voltage scheme. Synaptic behaviors including long-term potentiation (LTP) and long-term depression (LTD) were performed with 50 identical pulses for the implementation of RRAM into neuromorphic systems based on convolutional neural networks. Also, linearly increased (or decreased) 25 pulses were applied to the device so that the conductance changes linearly. The resulting linear LTP and LTD characteristics were mirror-symmetric, which could maximize the accuracy. For Hebbian learning, the device also mimicked the spike-timing-dependent plasticity properties with a conductance change from − 77.79% to 96.07% using a time-division multiplexing approach.

AB - This paper presents the fabrication and characterization of the cross-point structure 20 × 20 μm2 RRAM with TiOx/TiOy bi-layer insulator for synaptic application in neuromorphic systems. The measured oxygen concentration of the TiOx/TiOy switching layers of the fabricated devices using X-ray photoelectron spectroscopy analysis showed that the oxygen concentration ratio between TiOx and TiOy is ~ 1.5. After electroforming at ~ 5.62 V, the on/off ratio was ~ 76 at 0.2 V with the DC sweep voltage scheme. Synaptic behaviors including long-term potentiation (LTP) and long-term depression (LTD) were performed with 50 identical pulses for the implementation of RRAM into neuromorphic systems based on convolutional neural networks. Also, linearly increased (or decreased) 25 pulses were applied to the device so that the conductance changes linearly. The resulting linear LTP and LTD characteristics were mirror-symmetric, which could maximize the accuracy. For Hebbian learning, the device also mimicked the spike-timing-dependent plasticity properties with a conductance change from − 77.79% to 96.07% using a time-division multiplexing approach.

KW - Cross-point

KW - Neuromorphic

KW - RRAM

KW - Synaptic application

KW - TiO/TiO bi-layer

UR - http://www.scopus.com/inward/record.url?scp=85067681812&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067681812&partnerID=8YFLogxK

U2 - 10.1007/s42835-019-00107-y

DO - 10.1007/s42835-019-00107-y

M3 - Article

AN - SCOPUS:85067681812

VL - 14

SP - 869

EP - 875

JO - Journal of Electrical Engineering and Technology

JF - Journal of Electrical Engineering and Technology

SN - 1975-0102

IS - 2

ER -