Multistate Compound Magnetic Tunnel Junction Synapses for Digital Recognition

Multistate Compound Magnetic Tunnel Junction Synapses for Digital Recognition

ACS Applied Materials & Interfaces

10.1021/acsami.3c17195

http://dx.doi.org/10.1021/acsami.3c17195

Anuj Kumar, Dennis J. X. Lin, Debasis Das, Lisen Huang, Sherry L. K. Yap, Hui Ru Tan, Hang Khume Tan, Royston J. J. Lim, Yeow Teck Toh, Shaohai Chen, Sze Ter Lim, Xuanyao Fong, Pin Ho

20 February 2024

Kumar, A., Lin, D. J. X., Das, D., Huang, L., Yap, S. L. K., Tan, H. R., Tan, H. K., Lim, R. J. J., Toh, Y. T., Chen, S., Lim, S. T., Fong, X., & Ho, P. (2024). Multistate Compound Magnetic Tunnel Junction Synapses for Digital Recognition. ACS Applied Materials & Interfaces, 16(8), 10335–10343. https://doi.org/10.1021/acsami.3c17195

The quest to mimic the multistate synapses for bio-inspired computing has triggered nascent research that leverages the well-established magnetic tunnel junction (MTJ) technology. Early works on spin transfer torque MTJ-based ANN are susceptible to poor thermal reliability, high latency, and high critical current densities. Meanwhile, works on spin-orbit torque (SOT) MTJ-based ANN mainly utilized domain wall motion, which yields negligibly small readout signals differentiating consecutive states and has designs that are incompatible with technological scale-up. Here, we propose a multistate device concept built upon a compound MTJ consisting of multiple SOT-MTJs (number of MTJs, n = 1 – 4) on a shared write channel, mimicking the spin-based ANN. The n+1 resistance states representing varying synaptic weights can be tuned by varying the voltage pulses (±1.5 ‒ 1.8 V), pulse duration (100 – 300 ns), and applied in-plane fields (5.5 ‒ 10.5 mT). A large TMR difference of more than 13.6% is observed between two consecutive states for the 4-cell compound MTJ, a four-fold improvement from reported state-of-the-art spin-based synaptic devices. The ANN built upon the compound MTJ shows high learning accuracy for digital recognition tasks with incremental states and retraining, achieving test accuracy as high as 95.75% in the 4-cell compound MTJ. These results provide an industry-compatible platform to integrate these multistate SOT-MTJ synapses directly into neuromorphic architecture for in-memory and unconventional computing applications.

Publisher Copyright

This research / project is supported by the A*STAR - SpOTLITE programme
Grant Reference no. : A18A6b0057

This research / project is supported by the A*STAR - Singapore’s RIE2020
Grant Reference no. : C210812017

This document is the Accepted Manuscript version of a Published Work that appeared in final form in ACS Applied Materials & Interfaces, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see doi.org/10.1021/acsami.3c17195

https://oar.a-star.edu.sg/communities-collections/articles/19726

1944-8252
1944-8244

Institute of Materials Research and Engineering