Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Ferroelectric Tunnel Junction Memristors for In-Memory Computing Accelerators

Athle, Robin LU and Borg, Mattias LU orcid (2023) In Advanced Intelligent Systems
Abstract

Neuromorphic computing has seen great interest as leaps in artificial intelligence (AI) applications have exposed limitations due to heavy memory access, with the von Neumann computing architecture. The parallel in-memory computing provided by neuromorphic computing has the potential to significantly improve latency and power consumption. Key to analog neuromorphic computing hardware are memristors, providing non-volatile multistate conductance levels, high switching speed, and energy efficiency. Ferroelectric tunnel junction (FTJ) memristors are prime candidates for this purpose, but the impact of the particular characteristics for their performance upon integration into large crossbar arrays, the core compute element for both... (More)

Neuromorphic computing has seen great interest as leaps in artificial intelligence (AI) applications have exposed limitations due to heavy memory access, with the von Neumann computing architecture. The parallel in-memory computing provided by neuromorphic computing has the potential to significantly improve latency and power consumption. Key to analog neuromorphic computing hardware are memristors, providing non-volatile multistate conductance levels, high switching speed, and energy efficiency. Ferroelectric tunnel junction (FTJ) memristors are prime candidates for this purpose, but the impact of the particular characteristics for their performance upon integration into large crossbar arrays, the core compute element for both inference and training in deep neural networks, requires close investigation. In this work, a W/HfxZr1−xO2/TiN FTJ with 60 programmable conductance states, a dynamic range (DR) up to 10, current density >3 A m−2 at V read = 0.3 V and highly nonlinear current–voltage (I–V) characteristics (>1100) is experimentally demonstrated. Using a circuit macro-model, the system level performance of a true crossbar array is evaluated and a 92% classification accuracy of the modified nation institute of science and technology (MNIST) dataset is achieved. Finally, the low on conductance in combination with the highly nonlinear I–V characteristics enable the realization of large selector-free crossbar arrays for neuromorphic hardware accelerators.

(Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Contribution to journal
publication status
epub
subject
keywords
crossbar, ferroelectric tunnel junction, hafnium oxide, memristor, neuromorphic computing
in
Advanced Intelligent Systems
publisher
John Wiley & Sons Inc.
external identifiers
  • scopus:85180423799
ISSN
2640-4567
DOI
10.1002/aisy.202300554
project
Development and Implementation of Ferroelectric oxides
Ultra-fast thermal processing for next-generation ferroelectric hafnia
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2023 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH.
id
befaeaba-cdbd-41f7-a087-a9b3a1507dcc
date added to LUP
2024-01-07 21:12:56
date last changed
2024-02-06 09:55:42
@article{befaeaba-cdbd-41f7-a087-a9b3a1507dcc,
  abstract     = {{<p>Neuromorphic computing has seen great interest as leaps in artificial intelligence (AI) applications have exposed limitations due to heavy memory access, with the von Neumann computing architecture. The parallel in-memory computing provided by neuromorphic computing has the potential to significantly improve latency and power consumption. Key to analog neuromorphic computing hardware are memristors, providing non-volatile multistate conductance levels, high switching speed, and energy efficiency. Ferroelectric tunnel junction (FTJ) memristors are prime candidates for this purpose, but the impact of the particular characteristics for their performance upon integration into large crossbar arrays, the core compute element for both inference and training in deep neural networks, requires close investigation. In this work, a W/Hf<sub>x</sub>Zr<sub>1−x</sub>O<sub>2</sub>/TiN FTJ with 60 programmable conductance states, a dynamic range (DR) up to 10, current density &gt;3 A m<sup>−2</sup> at V <sub>read</sub> = 0.3 V and highly nonlinear current–voltage (I–V) characteristics (&gt;1100) is experimentally demonstrated. Using a circuit macro-model, the system level performance of a true crossbar array is evaluated and a 92% classification accuracy of the modified nation institute of science and technology (MNIST) dataset is achieved. Finally, the low on conductance in combination with the highly nonlinear I–V characteristics enable the realization of large selector-free crossbar arrays for neuromorphic hardware accelerators.</p>}},
  author       = {{Athle, Robin and Borg, Mattias}},
  issn         = {{2640-4567}},
  keywords     = {{crossbar; ferroelectric tunnel junction; hafnium oxide; memristor; neuromorphic computing}},
  language     = {{eng}},
  month        = {{12}},
  publisher    = {{John Wiley & Sons Inc.}},
  series       = {{Advanced Intelligent Systems}},
  title        = {{Ferroelectric Tunnel Junction Memristors for In-Memory Computing Accelerators}},
  url          = {{http://dx.doi.org/10.1002/aisy.202300554}},
  doi          = {{10.1002/aisy.202300554}},
  year         = {{2023}},
}