High-temperature memristors enabled by interfacial engineering
Non-volatile memories (NVM) that operate reliably at high temperature are essential for electronics in extreme environments. Here, we reported graphene (Gra)/HfO<sub>x</sub>/tungsten (W) memristors that operated reliably up to 700 °C with an ON/OFF current ratio >10<sup>3</sup>, data retention >50 hours and endurance >10<sup>9</sup> switching cycles. Transmission electron microscopy (TEM) revealed significant W diffusion into the inert platinum (Pt) electrode of conventional Pt/HfO<sub>x</sub>/W memristors after high-temperature annealing, which was an effect responsible for the thermal failure of conventional devices but not observed in Gra/HfO<sub>X</sub>/W devices. First-principles calculations attributed the enhanced thermal stability to weaker W adsorption and higher surface diffusion barriers on graphene than metals like Pt. These results underscore the critical role of interfacial engineering and potential of 2D materials in enabling reliable high-temperature NVM technologies.
Authors
- Kohei Shimamura (ORCID: https://orcid.org/0000-0003-3235-2599)
- Cameron S. Jorgensen (ORCID: https://orcid.org/0000-0002-0064-5920)
- K. Mahalingam (ORCID: https://orcid.org/0000-0003-0917-3656)
- Priya D. Vashishta (ORCID: https://orcid.org/0000-0003-4683-429X)
- Fuyuki Shimojo (ORCID: https://orcid.org/0000-0002-4025-0069)
- Wataru Sugimoto (ORCID: https://orcid.org/0000-0003-3868-042X)
- Sabyasachi Ganguli (ORCID: https://orcid.org/0000-0002-1993-7173)
- Y.-H Xu
- J. L. Zhao (ORCID: https://orcid.org/0000-0002-7102-0979)
- Ajit K. Roy
- Kai Ito
- Han-Ting Liao
- Qiangfei Xia
- Ruoyu Zhao
- Aiichiro Nakano
- Seung Ju Kim
- Ning Ge
- Cynthia Bowers
- J. Joshua Yang
- Rajiv K. Kalia
- Miao Hu
- R. Stanley Williams
Institutions
- University of Southern California (US)
- United States Air Force Research Laboratory (US)
- University of Massachusetts Amherst (US)
- Kumamoto University (JP)
Publication Details
- Journal
- Science
- Published
- 2026-03-26
- DOI
- https://doi.org/10.1126/science.aeb9934
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00