Analog-Tunable Nonvolatile Regulator-Free Memristors for Neuromorphic Controlling
Source abstract
Abstract
Current analog-tunable memristors often suffer from poor retention, limiting their role as nonvolatile synaptic nodes in hardware-based neuromorphic networks. On the other hand, conventional nonvolatile memristors display abrupt switching, necessitating additional current-regulating devices for circuit function. To address these challenges, we developed vertically arranged Bi 2 Se 3 memristors suitable for crossbar array configuration via gold-assisted, plasma-etching-free physical vapor deposition (PVD), facilitating scalable production. These memristors exhibit notable analog conductance tuning (10–40%) with minimal postsetting relaxation and stable nonvolatile retention (<1% loss over 10 4 s) without the need for a current regulator (e.g., a current-limiting transistor). They operate at low switching voltages (−0.98 V SET, +0.7 V RESET). The superior switching attributes stem from the electromigration-driven filament formation in the Au/Bi 2 Se 3 /Ti structure. Utilizing these memristors, we built an all-hardware-based reservoir computing network where Bi 2 Se 3 memristors serve effectively in the readout layer, processing sensory signals and generating control commands with minimal digital electronics. This configuration achieves ultralow power consumption (∼7 μW), significantly lower than digital equivalents, while maintaining competitive performance, evidenced by a normalized root-mean-square error (NRMSE) of 0.094. These results position Bi 2 Se 3 memristors as robust, energy-efficient options for next-generation neuromorphic computing.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Advanced Memory and Neural Computing · Ferroelectric and Negative Capacitance Devices · Neural Networks and Reservoir Computing
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Mechanical Engineering Department
University of Michigan
Department of Electrical and Computer Engineering
Funding
Basic Energy Sciences · Division of Electrical, Communications and Cyber Systems
References (37)
Show source-provided references
1. 10.1038/s41578-022-00434-z10.1038/s41578-022-00434-z ↗
2. 10.1126/science.abj997910.1126/science.abj9979 ↗
3. 10.1038/s41467-019-11411-610.1038/s41467-019-11411-6 ↗
4. 10.1038/nature0693210.1038/nature06932 ↗
5. 10.1038/nnano.2008.16010.1038/nnano.2008.160 ↗
6. 10.1088/0957-4484/20/21/21520110.1088/0957-4484/20/21/215201 ↗
7. 10.1038/nmat307010.1038/nmat3070 ↗
8. 10.1063/1.362447210.1063/1.3624472 ↗
9. 10.1021/nn202983n10.1021/nn202983n ↗
10. 10.1038/nnano.2017.8310.1038/nnano.2017.83 ↗
11. 10.1002/admi.20160019210.1002/admi.201600192 ↗
12. 10.1002/aelm.20150029810.1002/aelm.201500298 ↗
13. 10.1002/aelm.20210112710.1002/aelm.202101127 ↗
14. 10.1039/c3nr01176c10.1039/c3nr01176c ↗
15. 10.1039/C9NR00747D10.1039/C9NR00747D ↗
16. 10.1088/1361-6528/aab6a310.1088/1361-6528/aab6a3 ↗
17. 10.1002/aelm.20180087610.1002/aelm.201800876 ↗
18. 10.1109/mnano.2023.329710510.1109/mnano.2023.3297105 ↗
19. 10.1021/acsaelm.3c0054410.1021/acsaelm.3c00544 ↗
20. 10.1038/s41563-018-0248-510.1038/s41563-018-0248-5 ↗
21. 10.1021/acsnano.8b0397710.1021/acsnano.8b03977 ↗
22. 10.1002/adma.20180679010.1002/adma.201806790 ↗
23. 10.1002/adma.20200209210.1002/adma.202002092 ↗
24. 10.1109/MNANO.2023.329710510.1109/MNANO.2023.3297105 ↗
25. 10.1021/acsami.1c0766510.1021/acsami.1c07665 ↗
26. 10.1116/6.000402210.1116/6.0004022 ↗
27. 10.1038/nmat305410.1038/nmat3054 ↗
28. 10.1038/s41467-021-26399-110.1038/s41467-021-26399-1 ↗
29. 10.1063/5.017520010.1063/5.0175200 ↗
30. 10.1021/acsanm.4c0018910.1021/acsanm.4c00189 ↗
31. 10.1021/nl500049g10.1021/nl500049g ↗
32. 10.1088/0957-4484/22/25/25400310.1088/0957-4484/22/25/254003 ↗
33. 10.1038/nmat202310.1038/nmat2023 ↗
34. 10.1016/0040-6090(75)90052-810.1016/0040-6090(75)90052-8 ↗
35. 10.1016/0167-5729(92)90007-X10.1016/0167-5729(92)90007-X ↗
36. 10.1126/sciadv.adu266310.1126/sciadv.adu2663 ↗
37. 10.1038/s41467-021-25801-210.1038/s41467-021-25801-2 ↗