Premium
Facile Approach for Improving Synaptic Modulation of Analog Resistive Characteristics by Embedding Oxygen Vacancies‐Rich Sub‐TaO x in Pt/Ta 2 O 5 /Ti Device Stacks
Author(s) -
Jin Soeun,
Lee Donguk,
Kwak Myonghoon,
Kim Ah Ra,
Hwang Hyunsang,
Cheon Hyuknyeong,
Kwon Jung-Dae,
Kim Yonghun
Publication year - 2021
Publication title -
physica status solidi (a)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.532
H-Index - 104
eISSN - 1862-6319
pISSN - 1862-6300
DOI - 10.1002/pssa.202170012
Subject(s) - neuromorphic engineering , mnist database , nanoclusters , atomic layer deposition , materials science , resistive touchscreen , modulation (music) , synaptic weight , embedding , optoelectronics , resistive random access memory , nanotechnology , computer science , voltage , layer (electronics) , physics , artificial neural network , electrical engineering , artificial intelligence , engineering , acoustics , computer vision
Neuromorphic Synaptic Devices In article number 2000534 , Jung‐Dae Kwon, Yonghun Kim, and co‐workers report a reliable Ta 2 O 5 ‐based resistive synaptic device inserted with sub‐TaOx nanoclusters, enabling that the concentration of nanoclusters is precisely controlled using plasma enhanced atomic layer deposition (PE‐ALD). As a result, the MNIST pattern recognition accuracy improves by 11.45%, which can demonstrate highly reliable analog synaptic devices for neuromorphic hardware systems.