EUVIMEDEUVIMED entdecken
Crossref

Ferroelectric‐Induced Phase Change Device with Polymorphic Mo 1‐x W x Te 2 for Neuromorphic Computing

Eunji Hwang iD, Dohyun Kim iD, Nayeon Kim, Heemyoung Hong iD, Chang Yun Heo, Woohyun Cho iD, Ho Yeon Yoo, Kenji Watanabe iD, Takashi Taniguchi, Heejun Yang iD

Small · 30.10.2025

DOI: 10.1002/smll.202505378 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Small, online erschienen am 30.10.2025. Verfasst von Eunji Hwang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Physics, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 92 Literaturangaben. Gefördert von: National Research Foundation of Korea, Samsung Science and Technology Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

While phase change devices have emerged as promising candidates for implementing artificial synapses, conventional phase change materials have faced challenges such as high‐power consumption and limited reliability, originating from their Joule heating‐driven phase transition mechanisms. Here, a phase change device based on a 2D material, Mo 0.95 W 0.05 Te 2 is demonstrated, which exhibits a phase transition between semiconducting 2H and semimetallic 1T’ structural phases, facilitated by a ferroelectric substrate. The structural phase transitions are confirmed by Raman spectroscopy under drain or gate voltage bias. These bias conditions allow two types of operation to be realized in a single device structure, resulting in gate voltage modulation and drain voltage modulation with a ferroelectric substrate. The ferroelectric‐induced phase change device exhibits key synaptic functions, including short‐term and long‐term plasticity, along with highly linear and symmetric multilevel conductance states. Furthermore, polymorphic Mo 0.95 W 0.05 Te 2 enables energy consumption as low as 5.3 pJ per switching event at monolayer thickness beyond conventional 3D phase change memory. These features highlight the potential of 2D material‐based phase change devices on ferroelectric substrates as energy‐efficient and high‐performance components for next‐generation neuromorphic computing systems.

Schlagwörter

Stichwörter der Autor:innen

ferroelectricitymemristormemtransistorneuromorphic computingphase changesynaptic devicetransition metal chalcogenides

Autor:innen und Institutionen

  • Department of Physics, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea.
  • Research Center for Electronic and Optical Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba, 305-0044, Japan.
  • Research Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, 305-0044, Japan.
  • Graduate School of Semiconductor Technology, School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of Korea.

ORCID-Profile: Eunji Hwang, Dohyun Kim, Heemyoung Hong, Woohyun Cho, Kenji Watanabe, Heejun Yang

Förderung

  • National Research Foundation of Korea
  • Samsung Science and Technology Foundation

Zitieren

APA

Hwang, E., Kim, D., Kim, N., Hong, H., Heo, C. Y., Cho, W., Yoo, H. Y., Watanabe, K., Taniguchi, T., & Yang, H. (2025). Ferroelectric‐Induced Phase Change Device with Polymorphic Mo 1‐x W x Te 2 for Neuromorphic Computing. Small. https://doi.org/10.1002/smll.202505378

Vancouver

Hwang E, Kim D, Kim N, Hong H, Heo CY, Cho W, et al. Ferroelectric‐Induced Phase Change Device with Polymorphic Mo 1‐x W x Te 2 for Neuromorphic Computing. Small. 2025. doi:10.1002/smll.202505378

BibTeX
@article{Hwang2025,
  author = {Hwang, Eunji and Kim, Dohyun and Kim, Nayeon and Hong, Heemyoung and Heo, Chang Yun and Cho, Woohyun and Yoo, Ho Yeon and Watanabe, Kenji and Taniguchi, Takashi and Yang, Heejun},
  title = {Ferroelectric‐Induced Phase Change Device with Polymorphic Mo 1‐x W x Te 2 for Neuromorphic Computing},
  journal = {Small},
  year = {2025},
  doi = {10.1002/smll.202505378}
}
Kostenlos starten →