HfO 2 -based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware
Science Advances · 20.03.2026 · 12 (12)
DOI: 10.1126/sciadv.aec2324 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science Advances, erschienen am 20.03.2026. Verfasst von Babak Bakhit und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Materials Science and Metallurgy, University of Cambridge, Cambridge CB3 0FS, UK. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 199 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Royal Society, National Natural Science Foundation of China, Swedish Research Council Formas.
Abstract (Originalsprache)
The escalating energy consumption of existing artificial intelligence hardware has become a serious global issue that demands immediate action. Neuromorphic computing offers promises to drastically reduce this footprint. Here, we introduce multicomponent p-type Hf(Sr,Ti)O 2 thin films for energy-efficient, resistive switching–based neuromorphic devices. We demonstrate interfacial memristors with ultralow switching currents (≤~10 −8 A), exceptional cycle-to-cycle and device-to-device uniformities, and retention >10 5 s. They reveal hundreds of ultralow conductance levels with a modulation range of >50 (without reaching any saturation) and reproducibly satisfy unsupervised learning rules. This performance originates from incorporating a self-assembled p-n heterointerface between p-type Hf(Sr,Ti)O 2 and n-type TiO x N y , resulting in a fully depleted space-charge layer asymmetrically extended into Hf(Sr,Ti)O 2 , a large built-in potential, and extremely low saturation current density under reverse bias. Ultralow conductance modulation is controlled by tuning p-n heterointerface’s energy-barrier height through electro-ionic charge migration. This materials-engineering strategy addresses energy consumption and variability in existing memristors, opening a pathway toward energy-efficient neuromorphic computing systems.
Autor:innen und Institutionen
- Department of Materials Science and Metallurgy, University of Cambridge, Cambridge CB3 0FS, UK.
- Electrical Engineering, University of Cambridge, JJ Thomson Avenue, Cambridge CB3 0FA, UK.
- Thin Film Physics Division, Department of Physics (IFM), Linköping University, Linköping 58183, Sweden.
- School of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, China.
- Centre for Analysis and Synthesis, Lund University, Lund 22100, Sweden.
- Department of Materials Design and Innovation, University at Buffalo, 136 Bell Hall, Buffalo, NY 14260, USA.
ORCID-Profile: Babak Bakhit, Simon M Fairclough, Ingemar Persson, Giuliana Di Martino, Bonan Zhu, Caterina Ducati, Quanxi Jia, Bilge Yildiz, Andrew J Flewitt, Judith L MacManus-Driscoll
Förderung
- Royal Society
- National Natural Science Foundation of China
- Swedish Research Council Formas
- EPSRC
- Royal Academy of Engineering Chair in Emerging Technologies
- Winton Programme for Physics of Sustainability
- CAPE BlueSky Research Award
- Energy IRC Small Grant
Zitieren
APA
Bakhit, B., Xie, X., Fairclough, S. M., Jan, A., Persson, I., Di Martino, G., Zhu, B., Ducati, C., Jia, Q., Yildiz, B., Flewitt, A. J., & MacManus-Driscoll, J. L. (2026). HfO 2 -based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware. Science Advances, 12(12). https://doi.org/10.1126/sciadv.aec2324
Vancouver
Bakhit B, Xie X, Fairclough SM, Jan A, Persson I, Di Martino G, et al. HfO 2 -based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware. Science Advances. 2026;12(12). doi:10.1126/sciadv.aec2324
BibTeX
@article{Bakhit2026,
author = {Bakhit, Babak and Xie, Xiao and Fairclough, Simon M and Jan, Atif and Persson, Ingemar and Di Martino, Giuliana and Zhu, Bonan and Ducati, Caterina and Jia, Quanxi and Yildiz, Bilge and Flewitt, Andrew J and MacManus-Driscoll, Judith L},
title = {HfO 2 -based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware},
journal = {Science Advances},
year = {2026},
volume = {12},
number = {12},
doi = {10.1126/sciadv.aec2324}
}
EUVIMED