HfO2-based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware
(2026) In Science Advances 12(12).- Abstract
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)O2 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 >105 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... (More)
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)O2 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 >105 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)O2 and n-type TiOxNy, resulting in a fully depleted space-charge layer asymmetrically extended into Hf(Sr,Ti)O2, 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.
(Less)
- author
- organization
- publishing date
- 2026-03
- type
- Contribution to journal
- publication status
- published
- subject
- in
- Science Advances
- volume
- 12
- issue
- 12
- article number
- eaec2324
- publisher
- American Association for the Advancement of Science (AAAS)
- external identifiers
-
- scopus:105034079927
- pmid:41861000
- ISSN
- 2375-2548
- DOI
- 10.1126/sciadv.aec2324
- language
- English
- LU publication?
- yes
- id
- bc493314-aed8-419d-a512-e5febf449570
- date added to LUP
- 2026-05-11 11:51:54
- date last changed
- 2026-06-08 13:39:25
@article{bc493314-aed8-419d-a512-e5febf449570,
abstract = {{<p>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<sub>2</sub> thin films for energy-efficient, resistive switching–based neuromorphic devices. We demonstrate interfacial memristors with ultralow switching currents (≤~10<sup>−8</sup> A), exceptional cycle-to-cycle and device-to-device uniformities, and retention >10<sup>5</sup> 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<sub>2</sub> and n-type TiO<sub>x</sub>N<sub>y</sub>, resulting in a fully depleted space-charge layer asymmetrically extended into Hf(Sr,Ti)O<sub>2</sub>, 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.</p>}},
author = {{Bakhit, Babak and Xie, Xiao and Fairclough, Simon M. and Jan, Atif and Persson, Ingemar and Martino, Giuliana Di and Zhu, Bonan and Ducati, Caterina and Jia, Quanxi and Yildiz, Bilge and Flewitt, Andrew J. and MacManus-Driscoll, Judith L.}},
issn = {{2375-2548}},
language = {{eng}},
number = {{12}},
publisher = {{American Association for the Advancement of Science (AAAS)}},
series = {{Science Advances}},
title = {{HfO<sub>2</sub>-based memristive synapses with asymmetrically extended p-n heterointerfaces for highly energy-efficient neuromorphic hardware}},
url = {{http://dx.doi.org/10.1126/sciadv.aec2324}},
doi = {{10.1126/sciadv.aec2324}},
volume = {{12}},
year = {{2026}},
}
