Pattern recognition with neuromorphic computing using magnetic field-induced dynamics of skyrmions

Topological Spin Textures: Basic Physics and Devices

In the face of escalating modern data storage demands and the constraints of Moore's Law, exploring spintronic solutions, particularly the devices based on magnetic skyrmions, has emerged as a promising frontier in scientific research. Since the first experimental observation of skyrmions, topological spin textures have been extensively studied for their great potential as efficient information carriers in spintronic devices. However, significant challenges have emerged alongside this progress. This review aims to synthesize recent advances in skyrmion research while addressing the major issues encountered in the field. Additionally, current research on promising topological spin structures in addition to skyrmions is summarized. Beyond 2D structures, exploration also extends to 1D magnetic solitons and 3D spin textures. In addition, a diverse array of emerging magnetic materials is introduced, including antiferromagnets and 2D van der Waals magnets, broadening the scope of potential materials hosting topological spin textures. Through a systematic examination of magnetic principles, topological categorization, and the dynamics of spin textures, a comprehensive overview of experimental and theoretical advances in the research of topological magnetism is provided. Finally, both conventional and unconventional applications are summarized based on spin textures proposed thus far. This review provides an outlook on future development in applied spintronics.

Proto-neural networks from thermal proteins

Proteinoids are synthetic polymers that have structural similarities to natural proteins, and their formation is achieved through the application of heat to amino acid combinations in a dehydrated environment. The thermal proteins, initially synthesised by Sidney Fox during the 1960s, has the ability to undergo self-assembly, resulting in the formation of microspheres that resemble cells. These microspheres have fascinating biomimetic characteristics. In recent studies, substantial advancements have been made in elucidating the electrical signalling phenomena shown by proteinoids, hence showcasing their promising prospects in the field of neuro-inspired computing. This study demonstrates the advancement of experimental prototypes that employ proteinoids in the construction of fundamental neural network structures. The article provides an overview of significant achievements in proteinoid systems, such as the demonstration of electrical excitability, emulation of synaptic functions, capabilities in pattern recognition, and adaptability of network structures. This study examines the similarities and differences between proteinoid networks and spontaneous neural computation. We examine the persistent challenges associated with deciphering the underlying mechanisms of emergent proteinoid-based intelligence. Additionally, we explore the potential for developing bio-inspired computing systems using synthetic thermal proteins in forthcoming times. The results of this study offer a theoretical foundation for the advancement of adaptive, self-assembling electronic systems that operate using artificial bio-neural principles.

Brownian Electric Bubble Quasiparticles

Recent works on electric bubbles (including the experimental demonstration of electric skyrmions) constitute a breakthrough akin to the discovery of magnetic skyrmions some 15 years ago. So far research has focused on obtaining and visualizing these objects, which often appear to be immobile (pinned) in experiments. Thus, critical aspects of magnetic skyrmions-e.g., their quasiparticle nature, Brownian motion-remain unexplored (unproven) for electric bubbles. Here we use predictive atomistic simulations to investigate the basic dynamical properties of these objects in pinning-free model systems. We show that it is possible to find regimes where the electric bubbles can present long lifetimes (∼ns) despite being relatively small (diameter <2  nm). Additionally, we find that they can display stochastic dynamics with large and highly tunable diffusion constants. We thus establish the quasiparticle nature of electric bubbles and put them forward for the physical effects and applications (e.g., in token-based probabilistic computing) considered for magnetic skyrmions.

Reconfigurable Skyrmion-Based Logic Gates: Versatile Design and Full-Scale Implementation

Herein, we investigate the behavior of skyrmions within a racetrack design incorporating voltage-controlled magnetic anisotropy (VCMA) gates. Our analysis encompassed multiple forces, including spin currents and anisotropy gradients induced by bias voltages. As a result, the efficient control of skyrmion dynamics was achieved across various VCMA gate configurations. Building upon these findings, we propose an efficient approach to reconfigurable skyrmion logic (RSL) in a thin antiferromagnetic (AFM) film through a versatile design. Our RSL harnesses the selective integration of VCMA, spin-polarized currents, and skyrmion-skyrmion (sky-sky) interactions to implement multiple logic gates, including AND, OR, XOR, NOT, NAND, XNOR, and NOR. The design brings a significant advantage with its simplified fabrication process, making the implementation of the RSL practical and accessible for various applications. Furthermore, the RSL enables seamless dynamic switching between logic gates, thereby enhancing its multifunctionality. Additionally, the strategic incorporation of sky-sky interactions and skyrmion-edge repulsion prominently facilitates the realization of complex gates, such as NAND, XNOR, and NOR gates, that typically require intricate design efforts. Hence, this streamlined integration of RSL, coupled with its adaptability to changing computational needs, underscores its potential as a practical solution for implementing high-functionality skyrmion-based logic gates.

Task-adaptive physical reservoir computing

Reservoir computing is a neuromorphic architecture that may offer viable solutions to the growing energy costs of machine learning. In software-based machine learning, computing performance can be readily reconfigured to suit different computational tasks by tuning hyperparameters. This critical functionality is missing in 'physical' reservoir computing schemes that exploit nonlinear and history-dependent responses of physical systems for data processing. Here we overcome this issue with a 'task-adaptive' approach to physical reservoir computing. By leveraging a thermodynamical phase space to reconfigure key reservoir properties, we optimize computational performance across a diverse task set. We use the spin-wave spectra of the chiral magnet Cu2OSeO3 that hosts skyrmion, conical and helical magnetic phases, providing on-demand access to different computational reservoir responses. The task-adaptive approach is applicable to a wide variety of physical systems, which we show in other chiral magnets via above (and near) room-temperature demonstrations in Co8.5Zn8.5Mn3 (and FeGe).

Dynamic Motion of Polar Skyrmions in Oxide Heterostructures

Polar skyrmions have been widely investigated in oxide heterostructures due to their exotic properties and intriguing physical insights. However, the field-driven motion of polar skyrmions, akin to that of the magnetic counterpart, remains elusive. Herein, using phase-field simulations, we demonstrate the dynamic motion of polar skyrmions with integrated external thermal, electrical, and mechanical stimuli. External heating reduced the spontaneous polarization, while an applied electric field decreased the skyrmion size and weakened the interactions between the skyrmions. Together, the skyrmion motion barrier is significantly reduced from 40 to 2 eV under 9 V at 500 K. An applied mechanical force transformed the skyrmions into a c-domain region near the indenter center under the electric field, providing the space and driving force needed for the motion of the skyrmions. This study confirms that polar skyrmions can move like particles and provides concrete design principles for polar skyrmion-based electronic devices.

Room-Temperature Zero-Field kπ-Skyrmions and Their Field-Driven Evolutions in Chiral Nanodisks

Target skyrmion, characterized by a central skyrmion surrounded by a series of concentric cylinder domains known as kπ-skyrmions (k ≥ 2), holds promise as a novel storage state in next-generation memories. However, target skyrmions comprising one or more concentric cylindrical domains have not been observed in chiral magnets, particularly at room temperature. In this study, we experimentally achieved kπ-skyrmions (k = 2, 3, and 4) with diameters of ∼220, 320, and 410 nm, respectively, and room-temperature stability under zero magnetic field by tightly confining these topological spin textures in β-Mn-type Co8Zn10Mn2 nanodisks. The magnetic configurations and their field-driven evolutions were simultaneously investigated by using in situ off-axis electron holography. In combination with numerical simulations, we further investigated the dependence of kmax on the nanodisk diameter. These findings highlight the potential of kπ-skyrmions as information carriers and offer insights into manipulation of kπ-skyrmions in the future.

Handwritten digit recognition by spin waves in a Skyrmion reservoir

By performing numerical simulations for the handwritten digit recognition task, we demonstrate that a magnetic skyrmion lattice confined in a thin-plate magnet possesses high capability of reservoir computing. We obtain a high recognition rate of more than 88%, higher by about 10% than a baseline taken as the echo state network model. We find that this excellent performance arises from enhanced nonlinearity in the transformation which maps the input data onto an information space with higher dimensions, carried by interferences of spin waves in the skyrmion lattice. Because the skyrmions require only application of static magnetic field instead of nanofabrication for their creation in contrast to other spintronics reservoirs, our result consolidates the high potential of skyrmions for application to reservoir computing devices.

Thermally-robust spatiotemporal parallel reservoir computing by frequency filtering in frustrated magnets

Physical reservoir computing is a framework for brain-inspired information processing that utilizes nonlinear and high-dimensional dynamics in non-von-Neumann systems. In recent years, spintronic devices have been proposed for use as physical reservoirs, but their practical application remains a major challenge, mainly because thermal noise prevents them from retaining short-term memory, the essence of neuromorphic computing. Here, we propose a framework for spintronic physical reservoirs that exploits frequency domain dynamics in interacting spins. Through the effective use of frequency filters, we demonstrate, for a model of frustrated magnets, both robustness to thermal fluctuations and feasibility of frequency division multiplexing. This scheme can be coupled with parallelization in spatial domain even down to the level of a single spin, yielding a vast number of spatiotemporal computational units. Furthermore, the nonlinearity via the exchange interaction allows information processing among different frequency threads. Our findings establish a design principle for high-performance spintronic reservoirs with the potential for highly integrated devices.

Experimental demonstration of a skyrmion-enhanced strain-mediated physical reservoir computing system

Physical reservoirs holding intrinsic nonlinearity, high dimensionality, and memory effects have attracted considerable interest regarding solving complex tasks efficiently. Particularly, spintronic and strain-mediated electronic physical reservoirs are appealing due to their high speed, multi-parameter fusion and low power consumption. Here, we experimentally realize a skyrmion-enhanced strain-mediated physical reservoir in a multiferroic heterostructure of Pt/Co/Gd multilayers on (001)-oriented 0.7PbMg_1/3Nb_2/3O₃-0.3PbTiO₃ (PMN-PT). The enhancement is coming from the fusion of magnetic skyrmions and electro resistivity tuned by strain simultaneously. The functionality of the strain-mediated RC system is successfully achieved via a sequential waveform classification task with the recognition rate of 99.3% for the last waveform, and a Mackey-Glass time series prediction task with normalized root mean square error (NRMSE) of 0.2 for a 20-step prediction. Our work lays the foundations for low-power neuromorphic computing systems with magneto-electro-ferroelastic tunability, representing a further step towards developing future strain-mediated spintronic applications.

Solid-State Lithium Ion Supercapacitor for Voltage Control of Skyrmions

Ionic control of magnetism gives rise to high magnetoelectric coupling efficiencies at low voltages, which is essential for low-power magnetism-based nonconventional computing technologies. However, for on-chip applications, magnetoionic devices typically suffer from slow kinetics, poor cyclability, impractical liquid architectures, or strong ambient effects. As a route to overcoming these problems, we demonstrate a LiPON-based solid-state ionic supercapacitor with a magnetic Pt/Co₄₀Fe₄₀B₂₀/Pt thin-film electrode which enables voltage control of a magnetic skyrmion state. Skyrmion nucleation and annihilation are caused by Li ion accumulation and depletion at the magnetic interface under an applied voltage. The skyrmion density can be controlled through dc applied voltages or through voltage pulses. The skyrmions are nucleated by single 60 μs voltage pulses, and devices are cycled 750000 times without loss of electrical performance. Our results demonstrate a simple and robust approach to ionic control of magnetism in spin-based devices.

Brownian reservoir computing realized using geometrically confined skyrmion dynamics

Reservoir computing (RC) has been considered as one of the key computational principles beyond von-Neumann computing. Magnetic skyrmions, topological particle-like spin textures in magnetic films are particularly promising for implementing RC, since they respond strongly nonlinearly to external stimuli and feature inherent multiscale dynamics. However, despite several theoretical proposals that exist for skyrmion reservoir computing, experimental realizations have been elusive until now. Here, we propose and experimentally demonstrate a conceptually new approach to skyrmion RC that leverages the thermally activated diffusive motion of skyrmions. By confining the electrically gated and thermal skyrmion motion, we find that already a single skyrmion in a confined geometry suffices to realize nonlinearly separable functions, which we demonstrate for the XOR gate along with all other Boolean logic gate operations. Besides this universality, the reservoir computing concept ensures low training costs and ultra-low power operation with current densities orders of magnitude smaller than those used in existing spintronic reservoir computing demonstrations. Our proposed concept is robust against device imperfections and can be readily extended by linking multiple confined geometries and/or by including more skyrmions in the reservoir, suggesting high potential for scalable and low-energy reservoir computing.