Synaptic Multi-Weight Wafer-Scale Domain Wall Magnetic Tunnel Junctions Driven by Spin Orbit Torque

Abstract

Spintronic synaptic devices based on magnetic random access memory (MRAM) are a rapidly evolving class within the analog neuromorphic architecture. However, obtaining the multi-weight or analog behavior desired for improved neural network inference accuracy in scaled MRAM remains a challenge. In this study, we demonstrate and tune up to eight distinct resistance levels in 300 mm wafer-scale magnetic tunnel junction artificial synapses driven by spin orbit torque. By varying the applied in-plane field, multi-weight resistance states emerge through domain wall nucleation and pinning. These multi-weight states appear stochastically and can be addressed with an iterative programming algorithm to program a specific resistance. Electromagnetic simulations indicate the multiple states are due to domain wall nucleation mediated by in-plane field and applied voltage. Implementation of the measured multi-weight devices in a neural network yields a 96.1% inference accuracy using the MNIST dataset if all devices provide the same number of multi-weight states. When experimental device-to-device variation is included, the accuracy is 48.6%. Overall, this work demonstrates a route towards multi-weight artificial synapse MRAM, establishing a path towards scalable neuromorphic systems based on spintronics.

Spin-Polarized Magnetic Metal Electrodes for Magnetic Tunnel Junctions
Zhiyuan Duan, Peixin Qin, Li Liu +7
arXiv24 Sep 2026PreprintCC0Full textApplied physics & instrumentation Condensed matterarXiv
Orbital and Spin–Orbit Torque Interplay in Ta/W-Based Magnetic Tunnel Junctions with Vertical Nonlocal Switching
Marco Biagi, Corrado Capriata, Corentin Bouchard +6
Nano Letters21 Sep 2026Peer-reviewedCC BY-NC-SA 4.0Full textCondensed matterHAL
Giant orbital torque-driven picosecond switching in magnetic tunnel junctions
Yuxuan Yao, Chen Xiao, Xiaobai Ning +23
中国科学通报：英文版9 Jul 2026Peer-reviewedCC BY8 citedAtomic, molecular & optical physicsOpenAlex
Magnetic tunnel junctions of twisted La2/3Sr1/3MnO3 freestanding bilayers
Kota Kanda, Shiho Irie, Kodai Niitsu +7
arXiv14 Oct 2026PreprintarXiv licenceFull textApplied physics & instrumentation Condensed matterarXiv
Robustness of Skyrmion-Mediated Magnetization Switching in Perpendicular Magnetic Tunnel Junctions (p-MTJs) to Material Disorder and Elevated Temperature
Prosenjit Das, Md Mahadi Rajib, Jayasimha Atulasimha
arXiv23 Oct 2026PreprintarXiv licenceFull textCondensed matterarXiv

Cite

BibTeX RIS

BibTeX

@misc{zogbi2026synaptic,
  title = {{Synaptic Multi-Weight Wafer-Scale Domain Wall Magnetic Tunnel Junctions Driven by Spin Orbit Torque}},
  author = {Nicholas Zogbi and Domenico Giuliano and Vaishnavi Kateel and Van Dai Nguyen and Jyotirmoy Chatterjee and Alvaro Palomino and Robert Carpenter and Siddharth Rao and Jean Anne C. Incorvia},
  howpublished = {arXiv},
  year = {2026},
  month = nov,
  doi = {10.48550/arxiv.2611.08405},
  eprint = {2611.08405},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2611.08405},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Synaptic Multi-Weight Wafer-Scale Domain Wall Magnetic Tunnel Junctions Driven by Spin Orbit Torque
AU  - Nicholas Zogbi
AU  - Domenico Giuliano
AU  - Vaishnavi Kateel
AU  - Van Dai Nguyen
AU  - Jyotirmoy Chatterjee
AU  - Alvaro Palomino
AU  - Robert Carpenter
AU  - Siddharth Rao
AU  - Jean Anne C. Incorvia
PY  - 2026
DA  - 2026/11/05
PB  - arXiv
DO  - 10.48550/arxiv.2611.08405
UR  - https://arxiv.org/abs/2611.08405
AB  - Spintronic synaptic devices based on magnetic random access memory (MRAM) are a rapidly evolving class within the analog neuromorphic architecture. However, obtaining the multi-weight or analog behavior desired for improved neural network inference accuracy in scaled MRAM remains a challenge. In this study, we demonstrate and tune up to eight distinct resistance levels in 300 mm wafer-scale magnetic tunnel junction artificial synapses driven by spin orbit torque. By varying the applied in-plane field, multi-weight resistance states emerge through domain wall nucleation and pinning. These multi-weight states appear stochastically and can be addressed with an iterative programming algorithm to program a specific resistance. Electromagnetic simulations indicate the multiple states are due to domain wall nucleation mediated by in-plane field and applied voltage. Implementation of the measured multi-weight devices in a neural network yields a 96.1% inference accuracy using the MNIST dataset if all devices provide the same number of multi-weight states. When experimental device-to-device variation is included, the accuracy is 48.6%. Overall, this work demonstrates a route towards multi-weight artificial synapse MRAM, establishing a path towards scalable neuromorphic systems based on spintronics.
ER  -

Abstract

Related

Cite