PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 3, 2026Applied Physics Letters0 citations

Vacuum barrier engineering tunneling magnetoresistance in van der Waals Fe5GeTe2 homojunctions

View Full Paper
HWHui WenSZShouguo ZhuWZWenkai Zhu

Key Points

  • This research aims to explore the effects of vacuum barrier thickness on tunneling magnetoresistance in Fe5GeTe2 homojunctions.
  • Developed an all-vdW magnetic tunnel junction device using Fe5GeTe2 homojunctions with a vacuum layer as the tunneling barrier.
  • Calculated tunneling magnetoresistance values while varying the vacuum barrier thickness.
  • Proposed an in-plane electron momentum resolved model for TMR evaluation.
  • Achieved a maximum tunneling magnetoresistance value of 68.4% as the vacuum barrier thickness varied.
  • Observed an initial increase followed by a decrease in TMR with growing vacuum barrier thickness.
  • Demonstrated effective calculation of TMR in ferromagnetic homojunctions through modeling.

Abstract

The atomically thin and flexible characteristics of 2D van der Waals (vdW) materials allow for high-quality, tunable multilayer stacking. This gives rise to pure physical phenomena and broad application prospects in spintronics. Recently, very high tunneling magnetoresistance (TMR) has been obtained in all-vdW magnetic tunnel junctions (MTJs), indicating the potential of vdW materials for non-volatile spintronic memory applications. Here, we report an all-vdW MTJ device based on an Fe5GeTe2 homojunction, using a vacuum layer as the tunneling barrier. The TMR value first increases and then decreases as the thickness of the vacuum barrier layer grows, with a maximum TMR value reaching up to 68.4%. We also propose an in-plane electron momentum (k∥) resolved tunneling model specifically for TMR calculation in ferromagnetic homojunctions, which can well explain our experimental results. The simplified bilayer-MTJ structure offers attractive possibilities for spin-based memory, logic, and neuromorphic computing.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Wen et al. (2026) studied this question.

synapsesocial.com/papers/6a1fc509dee9eb8c0dce6704https://doi.org/10.1063/5.0320868
Ask AI
Helpful
Bookmark
Share
View Full Paper