This paper reports on the synthesis and the characterization of magnetic and magneto-transport properties of electrically interconnected Ni1-xCox alloy crossed nanotube networks. The alloy nanotube networks have been fabricated by an adapted electrochemical technique based on a dealloying method previously used for the fabrication of pure Ni nanotubes. Here, we demonstrate that this synthesis approach can be adapted to fabricate NiCo alloy nanotubes with Co contents (x) up to 27%. Increasing the Co content within this range has revealed an increase in the nanotube walls that is accompanied by an increase in the effective magnetic anisotropy, which is attributed to a reinforcement of the magnetostatic field along the nanotube axis. Magneto-transport measurements are consistent with the typical behavior observed previously in magnetic nanotubes, where the dominant magnetization reversal process involves the propagation of vortex-like domain walls. Furthermore, the field position of the drop in resistance close to zero field is consistent with the coercive field, which depends on the nanotube material and wall thickness. The possibility of synthesizing complex electrically interconnected nanoarchitectures, such as crossed nanotube networks made of various ferromagnetic materials, using electrochemical deposition techniques is of particular interest for the development of potential applications in magnetic sensing devices, neuromorphic computing, and spintronics.

Thiele-based descriptions of magnetic vortex dynamics in thin ferromagnetic nanodots rely on magnetization ansätze that describe the equilibrium texture but cannot represent perturbation-induced deformations. We introduce the Deformed Image Vortex Ansatz (DIVA): a perturbation-aware ansatz in which the response to an external perturbation is built into the magnetization profile itself, rather than appended to the dynamics as a correction. Here, we demonstrate the concept for a uniform, stationary in-plane field applied to a Permalloy nanodot, for which the deformation is analytically tractable. A symmetry-based perturbative expansion identifies the leading deformation as a single m = 1 harmonic around the disk, while energy minimization and a dominant-balance analysis yield a closed-form interpolant for the radial profile. Benchmarked against micromagnetic simulations on Permalloy disks of aspect ratio t/R = 0.1 and 0.0125, this realization reduces the disk-averaged angular deviation by a factor of 3 to 6 relative to the two-vortex ansatz, depending on geometry and field, and reduces the total-energy deviation by about a factor of six in the thicker disk.

In-memory computing (IMC) is a paradigm that enables neural network inference by computing analog matrix-vector multiplications (MVM) directly in memory crossbar arrays, with the potential for energy efficiency gains over conventional von Neumann architectures. In this work we present a simulation framework for N-ary crossbar architectures that retrieves MVM results with minimal implementation assumptions. The XOR and MNIST classification tasks were successfully inferred using a simulated crossbar array of (4 × 4) 4-states magnetic tunnel junctions (MTJ). MNIST accuracy reached 93.56% (vs. 97.56% software baseline). PCA dimensionality reduction was shown to drastically lower the number of required operations and improve the software baseline, for only a modest reduction in crossbar inference accuracy. We identified weight quantization as the primary error source, and studied its impact alongside systematic non-idealities and random noise. We find that cell-specific random noise is less detrimental than systematic errors due to averaging across the array. Finally, we demonstrate an optimal number of states per cell that balances quantization error against resistance state resolution to minimize total MVM error.

In-memory computing (IMC) enables energy-efficient neural network inference by computing analog matrix-vector multiplications (MVM) in memory crossbar arrays. In this work we present a simulation framework for N-ary crossbar architectures that retrieves MVM results with minimal implementation assumptions. The XOR and MNIST classification tasks were successfully inferred using a simulated crossbar array of (4x4) 4-states magnetic tunnel junctions (MTJ). MNIST accuracy reached 94.48

Interconnected networks of Bi0.89Sb0.11 crossed nanowires (CNWs) with controlled diameters were fabricated by electrodeposition within track-etched polyimide (PI) membranes, following a single irradiation step from multiple angles, by fine-tuning the density and pore diameter. These PI membranes exhibit excellent thermal resistance, allowing annealing at temperatures close to the melting point of the bismuth–antimony (Bi–Sb) alloy. For Bi0.89Sb0.11 CNWs with a diameter of 200 nm annealed at 285 °C, a significant increase in the Seebeck coefficient is observed, with values identical to those of the bulk alloy for temperatures between 170 and 320 K. Furthermore, a significant magneto-thermoelectric effect is observed at relatively low magnetic fields, which until now had only been reported in Bi–Sb alloy single crystals. The thermopower decreases with the reduction in nanowire diameter, as does the mobility of charge carriers. The temperature variations in electrical resistance and Hall coefficient are consistent with those of n-type semiconductors with narrow bandgap. These results are promising for obtaining flexible thermoelectric films for the direct conversion of wasted heat into electrical power.

This study investigates the magnetic properties of Ni and Co nanowire (NW) arrays at room temperature, embedded in porous anodic aluminum oxide (Al2O3) membranes exhibiting amorphous and gamma crystalline phases. Characterizing the Al2O3 membrane morphology has revealed increased nanopore roughness and the appearance of mesopores within their walls with Al2O3 crystallization. As a result of these morphological changes, the morphology of the NWs is also modified, increasing their surface roughness due to the inverse replication of the membrane during the electrodeposition process. Ferromagnetic resonance experiments reveal that the magnetic anisotropy of Ni NWs in amorphous Al2O3 is the result of the superposition of magnetostatic (MS) and magnetoelastic contributions, while Ni NWs in crystalline Al2O3 exhibit only the MS contribution, highlighting the host membrane influence on the NW magnetic behavior. In contrast, Co NW arrays exhibit purely MS behavior across all membranes. The present results are of prime interest for developing magnetic nanocomposites with tunable properties for advanced spintronic and magnetic sensing applications.

Spin-torque vortex oscillators provide a model system for the nonlinear dynamics of a confined magnetic texture. Their motion is commonly described by the Thiele equation, but its standard constant-coefficient form relies on a rigid-texture approximation and becomes inaccurate at large gyration amplitudes. We introduce a data-driven Thiele equation approach (DD-TEA) that extracts effective, position-dependent Thiele coefficients from a single current-ramp micromagnetic simulation by interpolating the magnetization in vortex-core-position space. The extracted maps reveal a weak increase of the gyrovector magnitude and a pronounced separation of the radial and azimuthal dissipation as the orbit expands, providing quantitative signatures of confinement- and motion-induced vortex deformation. Because the gyrotropic, dissipative, conservative, and spin-transfer contributions retain their usual Thiele structure, the resulting description remains physically interpretable rather than acting as a black-box surrogate. Incorporating these state-dependent coefficients into the equation reproduces the nonlinear stable-orbit dynamics and accurately predicts the response to time-varying currents, whereas a conventional constant-coefficient description predicts vortex expulsion. The framework therefore provides a systematic route for deriving effective collective-coordinate dynamics from full micromagnetic states with high computational efficiency

Three-dimensional interconnected nanowire networks have recently attracted notable attention for the fabrication of new devices for energy harvesting/storage, sensing, catalysis, magnetic and spintronic applications, and for the design of new hardware neuromorphic computing architectures. However, the complex branching of these nanowire networks makes it challenging to investigate these 3D nanostructured systems theoretically. Here, we present a theoretical description and simulations of the geometric properties of these 3D interconnected nanowire networks with selected characteristics. Our analysis reveals that the nanowire segment length between two crossing zones follows an exponential distribution. This suggests that shorter nanowire segments have a more pronounced influence on the nanowire network properties compared to their longer counterparts. Moreover, our observations reveal a homogeneous distribution in the smallest distance between the cores of two crossing nanowires. The results are highly reproducible and unaffected by changes in the nanowire network characteristics. The density of crossing zones and interconnected nanowire segments is found to vary as the square of the nanowire density multiplied by their diameter, further multiplied by a factor dependent on the packing factor. Finally, densities of interconnected segments up to 1013cm−2 can be achieved for 22-µm-thick nanowire networks with high packing factors. This has important implications for neuromorphic computing applications, suggesting that the realization of 1014 interconnections, which corresponds to the approximate number of synaptic connections in the human brain, is achievable with a nanowire network of about 10cm2. Published by the American Physical Society 2024

Nonlinear random projections are a powerful tool to efficiently classify real-life data while requiring much less computational resources than conventional artificial neural networks. We showcase the implementation of an echo-state network (ESN) based on a single spin-torque vortex oscillator (STVO) delayed in time. This network achieves accuracy of over 98% on the MNIST handwritten digit recognition task. Ultrafast data-driven simulations based on the Thiele equation approach are used to show that the performance of our STVO-based network is equivalent to that of conventional software implementations. We demonstrate how hardware neural networks based on STVOs can be studied for specific tasks through data-driven simulation, hence speeding up the development of such hardware intelligent systems.

Three-dimensional interconnected nanowire networks have recently attracted notable attention for the fabrication of new devices for energy harvesting/storage, sensing, catalysis, magnetic and spintronic applications and for the design of new hardware neuromorphic computing architectures. However, the complex branching of these nanowire networks makes it challenging to investigate these 3D nanostructured systems theoretically. Here, we present a theoretical description and simulations of the geometric properties of these 3D interconnected nanowire networks with selected characteristics. Our analysis reveals that the nanowire segment length between two crossing zones follows an exponential distribution. This suggests that shorter nanowire segments have a more pronounced influence on the nanowire network properties compared to their longer counterparts. Moreover, our observations reveal a homogeneous distribution in the smallest distance between the cores of two crossing nanowires. The results are highly reproducible and unaffected by changes in the nanowire network characteristics. Finally, densities of interconnected segments up to 10$^{13}$ cm$^{-2}$ can be achieved for 22-$\mu$m-thick nanowire networks with high packing factors. This has important implications for neuromorphic computing applications, suggesting that the realization of 10$^{14}$ interconnections, which corresponds to the approximate number of synaptic connections in the human brain, is achievable with a nanowire network of about 10 cm$^{2}$.

Macroscopic-scale nanostructures, situated at the interface of nanostructures and bulk materials, hold significant promise in the realm of thermoelectric materials. Nanostructuring presents a compelling avenue for enhancing material thermoelectric performance as well as unlocking intriguing nanoscale phenomena, including spin-dependent thermoelectric effects. This is achieved while preserving high power output capabilities and ease of measurements related to the overall macroscopic dimensions. Within this framework, the recently developed three-dimensional interconnected nanowire and nanotube networks, integrated into a flexible polymer membrane, emerge as promising candidates for macroscopic nanostructures. The flexibility of these composites also paves the way for advances in the burgeoning field of flexible thermoelectrics. In this study, we demonstrate that the three-dimensional nanowire networks made of ferromagnetic metals maintain the intrinsic bulk thermoelectric power of their bulk constituent even for a diameter reduced to approximately 23 nm. Furthermore, we showcase the pioneering magneto-thermoelectric measurements of three-dimensional interconnected nickel nanotube networks. These macroscopic materials, comprising interconnected nanotubes, enable the development of large-area devices that exhibit efficient thermoelectric performance, while their nanoscale tubular structures provide distinctive magneto-transport properties. This research represents a significant step toward harnessing the potential of macroscopic nanostructured materials in the field of thermoelectrics.

Micromagnetic simulations are used to study a spin-torque vortex oscillator excited by an out-of-plane dc current. The vortex core gyration amplitude is confined between two orbits due to periodical vortex core polarity reversals. The upper limit corresponds to the orbit where the vortex core reaches its critical velocity triggering the first polarity reversal which is immediately followed by a second one. After this double polarity reversal, the vortex core is on a smaller orbit that defines the lower limit of the vortex core gyration amplitude. This double reversal process is a periodic phenomenon and its frequency, as well as the upper and lower limit of the vortex core gyration, is controlled by the input current density while the vortex chirality determines the apparition of this confinement regime. In this non-linear regime, the vortex core never reaches a stable orbit and thus, it can be of interest for neuromorphic application as a leaky integrate-and-fire neuron for example.

Arrays of thermoelectric nanowires embedded in organic films are attracting increasing interest to fabricate flexible thermoelectric devices with adjustable dimensions and complex shapes, useful for sustainable power sources of portable electronic devices and wireless sensor networks. Here, we report the electrochemical synthesis of interconnected bismuth-antimony (Bi1-xSbx) nanowires (with 0.06 < x < 0.15) within the branched cylindrical nanopores of polycarbonate membranes. The influence of temperature and magnetic field on the electrical and thermoelectric properties was studied by considering electric and thermal currents flowing in the plane of the films. We show that short annealing times with temperature up to 250 degrees C of the nanowire-based nanocomposite lead to a large increase in the thermoelectric power, reaching values up to -80 mu V K-1 at room temperature, which are comparable to those of bulk Bi-Sb alloys. In addition, we report Hall effect measurements on crossed nanowires, made possible for the first time by the remarkable electrical connectivity of the nanowire network. These measurements, combined with variations in temperature and under the magnetic field of the electrical resistance, indicate that the interconnected networks of Bi1-xSbx nanowires after thermal annealing behave like n-type, narrow band gap semiconductors. Overall, the electrical and thermoelectric properties near the ambient temperature of the heat-treated Bi1-xSbx nanowire networks are similar to those of bulk polycrystalline Bi-Sb alloys, which are well-known thermoelectric materials exhibiting optimal performance near and below room temperature.

AbstractResults of measurements on the thermoelectric power of 45 nm diameter interconnected nanowire networks consisting of pure Fe, dilute FeCu and FeCr alloys and Fe/Cu multilayers are presented. The thermopower values of Fe nanowires are very close to those found in bulk materials, at all temperatures studied between 70 and 320 K. For pure Fe, the diffusion thermopower at room temperature, estimated to be around − 15 $$\upmu$$ μ V/K from our data, is largely supplanted by the estimated positive magnon-drag contribution, close to 30 $$\upmu$$ μ V/K. In dilute FeCu and FeCr alloys, the magnon-drag thermopower is found to decrease with increasing impurity concentration to about 10 $$\upmu$$ μ V/K at 10$$\%$$ % impurity content. While the diffusion thermopower is almost unchanged in FeCu nanowire networks compared to pure Fe, it is strongly reduced in FeCr nanowires due to pronounced changes in the density of states of the majority spin electrons. Measurements performed on Fe(7 nm)/Cu(10 nm) multilayer nanowires indicate a dominant contribution of charge carrier diffusion to the thermopower, as previously found in other magnetic multilayers, and a cancellation of the magnon-drag effect. The magneto-resistance and magneto-Seebeck effects measured on Fe/Cu multilayer nanowires allow the estimation of the spin-dependent Seebeck coefficient in Fe, which is about − 7.6 $$\upmu$$ μ V/K at ambient temperature.

Understanding the dynamics of magnetic vortices has emerged as an important challenge regarding the recent development of spin-torque vortex oscillators. Either micromagnetic simulations or the analytical Thiele equation approach are typically used to study such systems theoretically. This work focuses on the precise description of the restoring forces exerted on the vortex when it is displaced from equilibrium. In particular, the stiffness parameters related to a modification of the magnetic potential energy terms are investigated. A method is proposed to extract exchange, magnetostatic and Zeeman stiffness expressions from micromagnetic simulations. These expressions are then compared to state-of-the-art analytical derivations. Furthermore, it is shown that the stiffness parameters depend not only on the vortex core position but also on the injected current density. This phenomenon is not predicted by commonly used analytical ans\"atze. We show that these findings result from a deformation of the theoretical magnetic texture caused by the current induced Amp\`ere-Oersted field.