Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform

Abstract Magnetic tunnel junctions (MTJs) offer a promising pathway toward energy-efficient neuromorphic computing due to their nanoscale footprint, nonvolatile switching, and intrinsic nonlinear dynamics that emulate synaptic behavior. However, generating large thermoelectric voltages with bias-tun...

Full description

Saved in:
Bibliographic Details
Main Authors: Felix Oberbauer, Tristan Joachim Winkel, Tim Böhnert, Clara C. Wanjura, Marcel S. Claro, Luana Benetti, Ihsan Çaha, Francis Leonard Deepak, Farshad Moradi, Ricardo Ferreira, Markus Münzenberg, Tahereh Sadat Parvini
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Communications Physics
Online Access:https://doi.org/10.1038/s42005-025-02257-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333167740157952
author Felix Oberbauer
Tristan Joachim Winkel
Tim Böhnert
Clara C. Wanjura
Marcel S. Claro
Luana Benetti
Ihsan Çaha
Francis Leonard Deepak
Farshad Moradi
Ricardo Ferreira
Markus Münzenberg
Tahereh Sadat Parvini
author_facet Felix Oberbauer
Tristan Joachim Winkel
Tim Böhnert
Clara C. Wanjura
Marcel S. Claro
Luana Benetti
Ihsan Çaha
Francis Leonard Deepak
Farshad Moradi
Ricardo Ferreira
Markus Münzenberg
Tahereh Sadat Parvini
author_sort Felix Oberbauer
collection DOAJ
description Abstract Magnetic tunnel junctions (MTJs) offer a promising pathway toward energy-efficient neuromorphic computing due to their nanoscale footprint, nonvolatile switching, and intrinsic nonlinear dynamics that emulate synaptic behavior. However, generating large thermoelectric voltages with bias-tunable nonlinearities for neuromorphic use remains largely unexplored. Here, we introduce a hybrid opto-electrical excitation scheme—combining pulsed laser heating with DC bias—to drive MTJs into the nonlinear bias-enhanced tunnel magneto-Seebeck regime. This regime yields thermoelectric voltages in the tens of millivolts with a strong contrast between magnetic states, while also revealing spiking and double-switching behavior linked to vortex dynamics and fixed-layer depinning. The thermovoltage exhibits cubic dependence on bias current, enabling tunable synaptic weights. We simulate a single-layer neuromorphic network using optically encoded inputs and achieve 93.7% classification accuracy on handwritten digits. These results establish hybrid-driven MTJs as a compact, CMOS-compatible platform for neuromorphic computing, integrating optical input with spintronic functionality.
format Article
id doaj-art-32a82555e2eb4735ab723f0faf43f3a7
institution Kabale University
issn 2399-3650
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Communications Physics
spelling doaj-art-32a82555e2eb4735ab723f0faf43f3a72025-08-20T03:45:57ZengNature PortfolioCommunications Physics2399-36502025-08-01811910.1038/s42005-025-02257-0Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platformFelix Oberbauer0Tristan Joachim Winkel1Tim Böhnert2Clara C. Wanjura3Marcel S. Claro4Luana Benetti5Ihsan Çaha6Francis Leonard Deepak7Farshad Moradi8Ricardo Ferreira9Markus Münzenberg10Tahereh Sadat Parvini11Institut für Physik, Universität GreifswaldInstitut für Physik, Universität GreifswaldINL - International Iberian Nanotechnology LaboratoryMax Planck Institute for the Science of LightINL - International Iberian Nanotechnology LaboratoryINL - International Iberian Nanotechnology LaboratoryINL - International Iberian Nanotechnology LaboratoryINL - International Iberian Nanotechnology LaboratoryElectrical and Computer Engineering Department, Aarhus UniversityINL - International Iberian Nanotechnology LaboratoryInstitut für Physik, Universität GreifswaldInstitut für Physik, Universität GreifswaldAbstract Magnetic tunnel junctions (MTJs) offer a promising pathway toward energy-efficient neuromorphic computing due to their nanoscale footprint, nonvolatile switching, and intrinsic nonlinear dynamics that emulate synaptic behavior. However, generating large thermoelectric voltages with bias-tunable nonlinearities for neuromorphic use remains largely unexplored. Here, we introduce a hybrid opto-electrical excitation scheme—combining pulsed laser heating with DC bias—to drive MTJs into the nonlinear bias-enhanced tunnel magneto-Seebeck regime. This regime yields thermoelectric voltages in the tens of millivolts with a strong contrast between magnetic states, while also revealing spiking and double-switching behavior linked to vortex dynamics and fixed-layer depinning. The thermovoltage exhibits cubic dependence on bias current, enabling tunable synaptic weights. We simulate a single-layer neuromorphic network using optically encoded inputs and achieve 93.7% classification accuracy on handwritten digits. These results establish hybrid-driven MTJs as a compact, CMOS-compatible platform for neuromorphic computing, integrating optical input with spintronic functionality.https://doi.org/10.1038/s42005-025-02257-0
spellingShingle Felix Oberbauer
Tristan Joachim Winkel
Tim Böhnert
Clara C. Wanjura
Marcel S. Claro
Luana Benetti
Ihsan Çaha
Francis Leonard Deepak
Farshad Moradi
Ricardo Ferreira
Markus Münzenberg
Tahereh Sadat Parvini
Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
Communications Physics
title Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
title_full Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
title_fullStr Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
title_full_unstemmed Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
title_short Magnetic tunnel junctions driven by hybrid optical-electrical signals as a flexible neuromorphic computing platform
title_sort magnetic tunnel junctions driven by hybrid optical electrical signals as a flexible neuromorphic computing platform
url https://doi.org/10.1038/s42005-025-02257-0
work_keys_str_mv AT felixoberbauer magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT tristanjoachimwinkel magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT timbohnert magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT claracwanjura magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT marcelsclaro magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT luanabenetti magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT ihsancaha magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT francisleonarddeepak magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT farshadmoradi magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT ricardoferreira magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT markusmunzenberg magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform
AT taherehsadatparvini magnetictunneljunctionsdrivenbyhybridopticalelectricalsignalsasaflexibleneuromorphiccomputingplatform