Improved memory and synaptic device performance of HfO2-based multilayer memristor by inserting oxygen gradient TiOx layer

Jihee Park, Heeseong Jang, Yongjin Byun, Hyesung Na, Hyeonseung Ji and Sungjun Kim

Chaos, Solitons & Fractals, 2025, vol. 191, issue C

Abstract: Pt/Al/TiOx/HfO2/AlN/Pt devices demonstrate abrupt filamentary resistive switching (RS) that is strongly dependent on the bias polarity due to variations of the defect states in the TiOx layer, which is deposited using pulse direct current (DC) sputtering. In this study, we compare a device with a TiOx layer that serves as an oxygen reservoir layer (which has the same oxygen flow rate) with another device that has three TiOx layers with different oxygen flow rates. Both devices form an AlOx layer that acts as an overshoot suppression layer (OSL) due to a natural oxidation reaction between the TiOx oxygen reservoir layer and the Pt/Al top electrode. Additionally, the ultrathin AlN layer serves as an oxygen barrier layer (OBL) at the interface between the HfO2 layer and the Pt/Ti bottom electrode, inhibiting the movement of oxygen ions. The device with three oxygen layers achieves low current characteristics and low-power operation by gradually increasing the breakdown voltage. The optimized device demonstrates excellent linearity in terms of both potentiation and depression based on analog RS characteristics. Moreover, when the change in conductivity is employed as a weight in the neural network, neuromorphic system simulation can achieve a pattern recognition accuracy exceeding 91 %. Essential synaptic functions, including spike-rate-dependent plasticity (SRDP), spike-number-dependent plasticity (SNDP), spike-duration-plasticity (SDDP), and spike amplitude dependent plasticity (SADP), are also demonstrated to mimic biological synapses for neuromorphic computing applications. These results suggest that variations in oxygen flow in the Pt/Al/TiOx/HfO2/AlN/Pt structure could serve as a viable memory device for integration into neuromorphic systems.

Keywords: Neuromorphic; Artificial neural network; Artificial synapses; Synaptic devices; Oxygen gradient (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0960077924014620
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:chsofr:v:191:y:2025:i:c:s0960077924014620

DOI: 10.1016/j.chaos.2024.115910

Access Statistics for this article

Chaos, Solitons & Fractals is currently edited by Stefano Boccaletti and Stelios Bekiros

More articles in Chaos, Solitons & Fractals from Elsevier
Bibliographic data for series maintained by Thayer, Thomas R. ().