Unveiling the hybrid filaments-induced forming-free resistive switching dynamics in Cu-doped oxygenated amorphous carbon-based memristorsopen access
- Authors
- Woo, Dae-Seong; Jin, Soo-Min; Kim, Jae-Kyeong; Jung, Uijin; Park, Gwang-Ho; Lee, Woo-Guk; Han, Min-Jong; Shim, Tae-Hun; Park, Jinsub; Park, Jea-Gun
- Issue Date
- May-2025
- Publisher
- Nature Publishing Group
- Citation
- NPG Asia Materials, v.17, no.1, pp 1 - 16
- Pages
- 16
- Indexed
- SCIE
SCOPUS
- Journal Title
- NPG Asia Materials
- Volume
- 17
- Number
- 1
- Start Page
- 1
- End Page
- 16
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/207404
- DOI
- 10.1038/s41427-025-00604-9
- ISSN
- 1884-4049
1884-4057
- Abstract
- Oxygenated amorphous carbon (alpha-C:O-x) media in resistive memories has gained attention due to their cost-effectiveness, high resilience to external stimuli, and versatility in various applications. However, the forming process at high voltages and the low durability for alpha-C:O-x-based resistive memories impose limitations on their use in memory-centric computing systems. We report reliable forming-free Cu-doped alpha-C:O-x resistive memories (CCRMs) with multi-level properties, where resistive switching occurs via a hybrid conducting path of sp(2) covalent bonds and Cu filaments. To unveil the possible forming-free dynamics, we conducted in-depth studies using bias-dependent time-of-flight secondary ion mass spectroscopy and X-ray photoelectron spectroscopy for ion depth profiles and chemical bonding states analysis, respectively. We scaled down CCRMs to similar to 37 nm, achieving over 10(7) write/read endurance cycles and exceptional non-volatility of about 10.7 years at 85 degrees C. By varying reset voltage amplitudes, we achieved stable multi-level states. We demonstrated stable resistive switching in one-selector and one-resistor (1S1R) crossbar arrays with vertically stacked CCRMs and chalcogenide-based super-linear-threshold-switching selectors, confirming a readout margin of similar to 98.9% at similar to 1 terabit size. Finally, we demonstrated outstanding inference performance in binarized neural networks using 1S1R cell-based binary synapses, comparable to ideal cases. Our research is poised to provide groundbreaking advancements in carbon-based electronics.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 융합전자공학부 > 1. Journal Articles

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.