Searcharxiv⌕ Search

arXiv subjects

Ryosho Nakane

Publications and source records attributed to Ryosho Nakane.

At least 19 recordsLinked to original sources
arXiv2026

Spin-related transport in a polycrystalline NiCo2O4 film: Drastic current-induced change in resistivity-temperature characteristics via spin injection

We have studied spin-related transport in a polycrystalline NiCo2O4 (NCO) film on a MgAl2O4/Si(001) substrate, motivated by potential applications of the theoretical half-metallicity of NCO to Si-based high-performance spin-transport devices. Our approach is to systematically measure and analyze the temperature dependence of the film's resistivity ($\rho-T$) with various in-plane currents (100 nA$-$1 mA) and temperatures (4$-$290 K). With increasing current, the $\rho-T$ curve changes drastically from semiconducting ($d\rho/dT<0$) to non-monotonic and eventually toward metallic ($d\rho/dT>0$). A distinctive feature is that the single NCO film exhibits a $\rho-T$ characteristic of polycrystalline defective NCO at 100 nA, whereas it exhibits a $\rho-T$ characteristic of epitaxial less-defective NCO over a wide temperature range at 1 mA. This current-induced evolution of $\rho-T$ reflects the enhancement of the Curie temperature of defective regions near grain boundaries, accompanied by enhanced spin alignment there. We proposed a spin-related transport model that extends conventional hopping conduction models by incorporating the temperature- and current-dependent degree of spin alignment, as well as its spatial dependence inherent to polycrystalline NCO. This model comprehensively explains the interplay between the spin-alignment profile and transport mechanism. The analysis reveals that spin injection from grain bodies to grain boundaries enhances the spin alignment there and strengthens double-exchange interactions, facilitating conduction. This phenomenon strongly depends on both temperature and current. Our findings provide evidence of spin-polarized electrons inside the grain bodies, highlighting the potential of our polycrystalline NCO film as an efficient spin source. The present model is further supported by current$-$voltage and magnetoresistance features.

cond-mat.mtrl-sci↗
arXiv2026

Substrate-Voltage-Controlled Temporal Nonlinearity in Ferroelectric FET-based Reservoir Computing

Physical reservoir computing exploits inherent nonlinearity and short-term memory of physical dynamics to achieve efficient processing of time-series data with extremely-low training cost. In this study, we demonstrate a ferroelectric field-effect transistor (FeFET)-based reservoir computing system with augmented temporal and spatial nonlinearity by utilizing both gate and substrate terminals as inputs. The ferroelectric polarization state in the next time step can additionally be controlled by modifying the electric field distribution in the gate stack of FeFET through a substrate input, enabling more diverse internal states compared with the case where inputs are applied only to the gate. To introduce a nonlinearity in the time domain, we introduce a delay between a gate input and a substrate input, which facilitates efficient nonlinear mixing between the current and past inputs. As a result, both the short-term memory and nonlinearity of the FeFET reservoir computing system are enhanced with an improved capability of feature extraction of complex input time-series. These findings demonstrate that introducing substrate input provides an additional degree of freedom for controlling ferroelectric polarization dynamics, enabling a flexible, energy-efficient, and highly integrable FeFET-based reservoir computing platform suitable for diverse time-series processing applications.

physics.app-ph↗
arXiv2025

Spin-transport characteristics in a Si-based spin metal-oxide-semiconductor field-effect transistor (spin MOSFET): Bias dependence of the spin polarization in Si and magnetoresistance in spin-valve signals

We have studied the spin transport characteristics of a spin metal-oxide-semiconductor field-effect transistor (spin MOSFET), particularly the bias voltage dependence of the electron spin polarization P_S in Si and the magnetoresistance ratio MR in spin-valve signals, to optimize the device performance. The spin MOSFET device has an 8-nm-thick p-Si channel with a back gate (G) and ferromagnetic source / drain (S/D) junctions consisting of Fe/Mg/MgO/SiOx/n+-Si. In addition to transistor characteristics with an on-off ratio of 104, clear spin-valve signals and Hanle spin precession signals were observed at 4 K in a wide range of the source-to-gate V_GS and source-to-drain V_DS bias voltages. We achieved a high P_S of 50% and a high MR of 0.35% as the maximum values in their single-peaked curves plotted as a function of the junction voltage V_J, mainly because the ferromagnetic S/D junction can generate high P_S and the spin diffusion length is very long in the Si inversion channel. These P_S and MR values are the highest ever reported in spin-MOSFETs. Our spin transport model for our spin MOSFET structure was improved in this study by taking into account the electron distribution and band profile of the n+-Si regions in the ferromagnetic S/D junctions, which enables the accurate estimation of P_S. Detailed analyses with various V_GS and V_J clarified that P_S is determined only by V_J. Our analyses also revealed that the main parameters for determining MR, such as P_S and the resistance-area products of the S/D ferromagnetic junctions, have different V_J dependences, leading to the finding that the present device does not exploit the full potential of the ferromagnetic S/D junctions to maximize MR. Based on the results, we discuss the device physics and engineering for further enhancement of MR, with a focus on the electrical and spin-related properties of the ferromagnetic S/D junctions.

cond-mat.mes-hall↗
arXiv2025

Spin injection in Si-based ferromagnetic tunnel junctions with MgO/MgAl2O4 barriers:Experimental and theoretical investigation of barrier thickness-dependent spin tunneling efficiency

We have experimentally and theoretically investigated the spin transport in Fe/Mg/MgO/MgAl2O4/n+-Si(001) ferromagnetic tunnel junctions on a Si substrate, by systematically varying the thickness combination of amorphous MgO and MgAl2O4 tunnel barrier layers with a sliding shutter between the evaporation sources and substrate during electron-beam evaporation. A technical advantage of MgAl2O4 is that a continuous and flat thin film is realized on a Si substrate even when the MgAl2O4 thickness is as thin as 0.5 nm, unlike MgO, which enables us to examine the spin transport in a thinner range of the tunnel barrier thickness. Our distinct finding is as follows: When the Fe/Mg/MgO interface is used on the top side, the spin polarization PS of tunneling electrons increases at 10 K as the total MgO/MgAl2O4 tunnel barrier thickness (tox = 0.47 - 1.4 nm) is increased, regardless of different thickness combinations, and PS shows saturation-like behavior when tox is above 1.1 nm. Since this feature cannot be explained by the well-known conductivity mismatch in semiconductor-based ferromagnetic tunnel junctions, we propose a simple phenomenological tunneling model based on two different direct tunneling paths, which have higher/lower spin polarizations with longer/shorter decay lengths. Our numerical calculation reproduces the relationship between the spin polarization PS and total tunnel barrier thickness tox in the experiments, indicating that the dominant mechanism is an increasing contribution of the lower spin polarization path as tox is decreased. We discuss possible origins for this phenomenon including intrinsic and extrinsic tunneling mechanisms. Our analysis method provides an insight into the detailed spin transport physics in semiconductor-based ferromagnetic junctions, particularly, with a very thin tunnel barrier layer.

physics.app-ph↗
arXiv2024

Hardware-Friendly Implementation of Physical Reservoir Computing with CMOS-based Time-domain Analog Spiking Neurons

This paper introduces an analog spiking neuron that utilizes time-domain information, i.e., a time interval of two signal transitions and a pulse width, to construct a spiking neural network (SNN) for a hardware-friendly physical reservoir computing (RC) on a complementary metal-oxide-semiconductor (CMOS) platform. A neuron with leaky integrate-and-fire is realized by employing two voltage-controlled oscillators (VCOs) with opposite sensitivities to the internal control voltage, and the neuron connection structure is restricted by the use of only 4 neighboring neurons on the 2-dimensional plane to feasibly construct a regular network topology. Such a system enables us to compose an SNN with a counter-based readout circuit, which simplifies the hardware implementation of the SNN. Moreover, another technical advantage thanks to the bottom-up integration is the capability of dynamically capturing every neuron state in the network, which can significantly contribute to finding guidelines on how to enhance the performance for various computational tasks in temporal information processing. Diverse nonlinear physical dynamics needed for RC can be realized by collective behavior through dynamic interaction between neurons, like coupled oscillators, despite the simple network structure. With behavioral system-level simulations, we demonstrate physical RC through short-term memory and exclusive OR tasks, and the spoken digit recognition task with an accuracy of 97.7% as well. Our system is considerably feasible for practical applications and also can be a useful platform for studying the mechanism of physical RC.

cs.NE↗
arXiv2024

Spin injection and detection in a Si-based ferromagnetic tunnel junction: A theoretical model based on the band diagram and experimental demonstration

We have experimentally and theoretically investigated the spin injection/detection polarization in a Si-based ferromagnetic tunnel junction with an amorphous MgO layer, and demonstrated that the experimental features of the spin polarization in a wide bias range can be well explained using our theoretical model based on the band diagram of the junction and the direct tunneling mechanism. It is shown that the spin polarization originates from the band diagrams of the ferromagnetic Fe layer and n+-Si channel in the junction, while the spin selectivity of the MgO tunnel barrier is not necessary. Besides, we clarified the mechanism of the reduction in spin polarization when the bias is high and nonlinear properties are prominent, where the widely-used spin injection/detection model proposed by Valet and Fert is not applicable. The dominant mechanism of such reduction is found to be spin accumulation saturation (SAS) at the n+-Si interface in contact with the MgO layer as the bias is increased in the spin extraction geometry, which is inevitable in semiconductor-based ferromagnetic tunnel junctions. We performed numerical calculations on a two-terminal spin transport device with a n+-Si channel using the junction properties extracted from the experiments, and revealed that the magnetoresistance (MR) ratio is suppressed mainly by SAS in a higher bias range. Furthermore, we proposed methods for improving the MR ratio in two-terminal spin transport devices. Our experiments and theoretical model provide a deep understanding of the spin injection/detection phenomena in semiconductor-based spin transport devices, toward the realization of high performance under reasonably high bias conditions for practical use.

physics.app-ph↗
arXiv2022

Performance enhancement of a spin-wave-based reservoir computing system utilizing different physical conditions

The authors have numerically studied how to enhance reservoir computing performance by thoroughly extracting their spin-wave device potential for higher-dimensional information generation. The reservoir device has a 1-input exciter and 120-output detectors on the top of a continuous magnetic garnet film for spin-wave transmission. For various nonlinear and fading-memory dynamic phenomena distributing in the film space, small in-plane magnetic fields were used to prepare stripe domain structures and various damping constants at the film sides and bottom were explored. The ferromagnetic resonant frequency and relaxation time of spin precession clearly characterized the change in spin dynamics with the magnetic field and damping constant. The common input signal for reservoir computing was a 1 GHz cosine wave with random 6-valued amplitude modulation. A basic 120-dimensional reservoir output vector was obtained from time-series signals at the 120 output detectors under each of the three magnetic field conditions. Then, 240- and 360-dimensional reservoir output vectors were also constructed by concatenating two and three basic ones, respectively. In nonlinear autoregressive moving average (NARMA) prediction tasks, the computational performance was enhanced as the dimension of the reservoir output vector becomes higher and a significantly low prediction error was achieved for the 10th-order NARMA using the 360-dimensional vector and optimum damping constant. The results are clear evidence that the collection of diverse output signals efficiently increases the dimensionality effective for reservoir computing, i.e., reservoir-state richness. This paper demonstrates that performance enhancement through various configuration settings is a practical approach for on-chip reservoir computing devices with small numbers of real output nodes.

physics.comp-ph↗
arXiv2022

Simulation platform for pattern recognition based on reservoir computing with memristor networks

Memristive systems and devices are potentially available for implementing reservoir computing (RC) systems applied to pattern recognition. However, the computational ability of memristive RC systems depends on intertwined factors such as system architectures and physical properties of memristive elements, which complicates identifying the key factor for system performance. Here we develop a simulation platform for RC with memristor device networks, which enables testing different system designs for performance improvement. Numerical simulations show that the memristor-network-based RC systems can yield high computational performance comparable to that of state-of-the-art methods in three time series classification tasks. We demonstrate that the excellent and robust computation under device-to-device variability can be achieved by appropriately setting network structures, nonlinearity of memristors, and pre/post-processing, which increases the potential for reliable computation with unreliable component devices. Our results contribute to an establishment of a design guide for memristive reservoirs toward a realization of energy-efficient machine learning hardware.

cs.ET↗
arXiv2021

Reduced magnetocrystalline anisotropy of CoFe$_2$O$_4$ thin films studied by angle-dependent x-ray magnetic circular dichroism

Spinel-type CoFe$_2$O$_4$ is a ferrimagnetic insulator with the Néel temperature exceeding 790 K, and shows a strong cubic magnetocrystalline anisotropy (MCA) in bulk materials. However, when a CoFe$_2$O$_4$ film is grown on other materials, its magnetic properties are degraded so that so-called magnetically dead layers are expected to be formed in the interfacial region. We investigate how the magnetic anisotropy of CoFe$_2$O$_4$ is modified at the interface of CoFe$_2$O$_4$/Al$_2$O$_3$ bilayers grown on Si(111) using x-ray magnetic circular dichroism (XMCD). We find that the thinner CoFe$_2$O$_4$ films have significantly smaller MCA values than bulk materials. The reduction of MCA is explained by the reduced number of Co$^{2+}$ ions at the $O_h$ site reported by a previous study [Y. K. Wakabayashi $\textit{et al.}$, Phys. Rev. B $\textbf{96}$, 104410 (2017)].

cond-mat.mtrl-sci↗
arXiv2021

A numerical exploration of signal detector arrangement in a spin-wave reservoir computing device

This paper studies numerically how the signal detector arrangement influences the performance of reservoir computing using spin waves excited in a ferrimagnetic garnet film. This investigation is essentially important since the input information is not only conveyed but also transformed by the spin waves into high-dimensional information space when the waves propagate in the film in a spatially distributed manner. This spatiotemporal dynamics realizes a rich reservoir-computational functionality. First, we simulate spin waves in a rectangular garnet film with two input electrodes to obtain spatial distributions of the reservoir states in response to input signals, which are represented as spin vectors and used for a machine-learning waveform classification task. The detected reservoir states are combined through readout connection weights to generate a final output. We visualize the spatial distribution of the weights after training to discuss the number and positions of the output electrodes by arranging them at grid points, equiangularly circular points or at random. We evaluate the classification accuracy by changing the number of the output electrodes, and find that a high accuracy ($>$ 90\%) is achieved with only several tens of output electrodes regardless of grid, circular or random arrangement. These results suggest that the spin waves possess sufficiently complex and rich dynamics for this type of tasks. Then we investigate in which area useful information is distributed more by arranging the electrodes locally on the chip. Finally, we show that this device has generalization ability for input wave-signal frequency in a certain frequency range. These results will lead to practical design of spin-wave reservoir devices for low-power intelligent computing in the near future.

cs.ET↗
arXiv2021

Spin waves propagating through a stripe magnetic domain structure and their applications to reservoir computing

Spin waves propagating through a stripe domain structure and reservoir computing with their spin dynamics have been numerically studied with focusing on the relation between physical phenomena and computing capabilities. Our system utilizes a spin-wave-based device that has a continuous magnetic garnet film and 1-input/72-output electrodes on the top. To control spatially-distributed spin dynamics, a stripe magnetic domain structure and amplitude-modulated triangular input waves were used. The spatially-arranged electrodes detected spin vector outputs with various nonlinear characteristics that were leveraged for reservoir computing. By moderately suppressing nonlinear phenomena, our system achieves 100$\%$ prediction accuracy in temporal exclusive-OR (XOR) problems with a delay step up to 5. At the same time, it shows perfect inference in delay tasks with a delay step more than 7 and its memory capacity has a maximum value of 21. This study demonstrated that our spin-wave-based reservoir computing has a high potential for edge-computing applications and also can offer a rich opportunity for further understanding of the underlying nonlinear physics.

physics.app-ph↗
arXiv2020

Spin transport in Si-based spin metal-oxide-semiconductor field-effect transistors: Spin drift effect in the inversion channel and spin relaxation in the n+-Si source/drain regions

We have experimentally and theoretically investigated the electron spin transport and spin distribution at room temperature in a Si two-dimensional (2D) inversion channel of back-gate-type spin metal-oxide-semiconductor field-effect transistors (spin MOSFETs). The magnetoresistance ratio of the spin MOSFET with a channel length of 0.4$μ$m was increased by a factor of 6 from that in our previous paper [Phys. Rev. B 99, 165301 (2019)] by lowering the parasitic resistances at the source/drain junctions with highly-phosphorus-doped n+-Si regions and by increasing the lateral electric field in the channel along the electron transport, called "spin drift". Clear Hanle signals with some oscillation peaks were observed for the spin MOSFET with a channel length of 10 $μ$ m under the lateral electric field, indicating that the effective spin diffusion length is dramatically enhanced by the spin drift. By taking into account the n+-Si regions and the spin drift in the channel, one-dimensional analytic functions were derived for analyzing the effect of the spin drift on the spin transport through the channel and these functions were found to explain almost all the experimental results. From the calculated spin current and spin distribution, it was revealed that almost all the spins are unflipped during the spin-drift-assisted transport through the 0.4-$μ$m-long inversion channel, but the most part of the injected spins from the source electrode are relaxed in the n+-Si regions of both the source and drain junctions. This means that the spin drift is useful and precise design of the device structure is essential to obtain a higher magnetoresistance ratio. Furthermore, we showed that the effective spin resistances that are introduced in this study are very helpful to understand how to improve the magnetoresistance ratio of spin MOSFETs for practical use.

physics.app-ph↗
arXiv2019

Characterization of in-gap states in epitaxial CoFe2O4(111) layers grown on Al2O3(111)/Si(111) by resonant inelastic x-ray scattering

We have studied in-gap states in epitaxial CoFe2O4(111), which potentially acts as a perfect spin filter, grown on a Al2O3(111)/Si(111) structure by using ellipsometry, Fe L2,3-edge x-ray absorption spectroscopy (XAS), and Fe L2,3-edge resonant inelastic x-ray scattering (RIXS), and revealed the relation between the in-gap states and chemical defects due to the Fe2+ cations at the octahedral sites (Fe2+ (Oh) cations). The ellipsometry measurements showed the indirect band gap of 1.24 eV for the CoFe2O4 layer and the Fe L2,3-edge XAS confirmed the characteristic photon energy for the preferential excitation of the Fe2+ (Oh) cations. In the Fe L3-edge RIXS spectra, a band-gap excitation and an excitation whose energy is smaller than the band-gap energy (Eg = 1.24 eV) of CoF2O4, which we refer to as "below-band-gap excitation (BBGE)" hereafter, were observed. The intensity of the BBGE was strengthened at the preferential excitation energy of the Fe2+ (Oh) cations. In addition, the intensity of the BBGE was significantly increased when the thickness of the CoFe2O4 layer was decreased from 11 to 1.4 nm, which coincides with the increase in the site occupancy of the Fe2+ (Oh) cations with decreasing the thickness. These results indicate that the BBGE comes from the in-gap states of the Fe2+ (Oh) cations whose density increases near the heterointerface on the bottom Al2O3 layer. We have demonstrated that RIXS measurements and analyses in combination with ellipsometry and XAS are effective to provide an insight into in-gap states in thin-film oxide heterostructures.

cond-mat.mtrl-sci↗
arXiv2019

Recent Advances in Physical Reservoir Computing: A Review

Reservoir computing is a computational framework suited for temporal/sequential data processing. It is derived from several recurrent neural network models, including echo state networks and liquid state machines. A reservoir computing system consists of a reservoir for mapping inputs into a high-dimensional space and a readout for pattern analysis from the high-dimensional states in the reservoir. The reservoir is fixed and only the readout is trained with a simple method such as linear regression and classification. Thus, the major advantage of reservoir computing compared to other recurrent neural networks is fast learning, resulting in low training cost. Another advantage is that the reservoir without adaptive updating is amenable to hardware implementation using a variety of physical systems, substrates, and devices. In fact, such physical reservoir computing has attracted increasing attention in diverse fields of research. The purpose of this review is to provide an overview of recent advances in physical reservoir computing by classifying them according to the type of the reservoir. We discuss the current issues and perspectives related to physical reservoir computing, in order to further expand its practical applications and develop next-generation machine learning systems.

cs.ET↗
arXiv2018

Systematic study of the electronic structure and the magnetic properties of a few-nm-thick epitaxial (Ni1-xCox)Fe2O4 (x = 0 - 1) layers grown on Al2O3(111)/Si(111) using soft X-ray magnetic circular dichroism: effects of cation distribution

We study the electronic structure and the magnetic properties of epitaxial (Ni1-xCox)Fe2O4(111) layers (x = 0 - 1) with thicknesses d = 1.7 - 5.2 nm grown on Al2O3(111)/Si(111) structures, to achieve a high value of inversion parameter y, which is the inverse-to-normal spinel-structure ratio, and hence to obtain good magnetic properties even when the thickness is thin enough for electron tunneling as a spin filter. We revealed the crystallographic (octahedral Oh or tetrahedral Td) sites and the valences of the Fe, Co, and Ni cations using experimental soft X-ray absorption spectroscopy and X-ray magnetic circular dichroism spectra and configuration-interaction cluster-model calculation. In all the (Ni1-xCox)Fe2O4 layers with d = about 4 nm, all Ni cations occupy the Ni2+ (Oh) site, whereas Co cations occupy the three different Co2+ (Oh), Co2+ (Td), and Co3+ (Oh) sites with constant occupancies. According to these features, the occupancy of the Fe3+ (Oh) cations decreases and that of the Fe3+ (Td) cations increases with decreasing x. Consequently, we obtained a systematic increase of y with decreasing x and achieved the highest y value of 0.91 for the NiFe2O4 layer with d = 3.5 nm. From the d dependences of y and magnetization in the d range of 1.7 - 5.2 nm, a magnetically dead layer is present near the NiFe2O4/Al2O3 interface, but its influence on the magnetization was significantly suppressed compared with the case of CoFe2O4 layers reported previously [Y. K. Wakabayasi et al., Phys. Rev. B 96, 104410 (2017)], due to the high site selectivity of the Ni cations. Since our epitaxial NiFe2O4 layer with d = 3.5 nm has a high y values (0.91) and a reasonably large magnetization (180 emu/cc), it is expected to exhibit a strong spin filter effect, which can be used for efficient spin injection into Si.

cond-mat.mtrl-sci↗
arXiv2018

Impurity band conduction in group-IV ferromagnetic semiconductor Ge1-xFex with nanoscale fluctuations in Fe concentration

We study the carrier transport and magnetic properties of group-IV-based ferromagnetic semiconductor Ge1-xFex thin films (Fe concentration x = 2.3 - 14 %) with and without boron (B) doping, by measuring their transport characteristics; the temperature dependence of resistivity, hole concentration, mobility, and the relation between the anomalous Hall conductivity versus conductivity. At relatively low x (= 2.3 %), the transport in the undoped Ge1-xFex film is dominated by hole hopping between Fe-rich hopping sites in the Fe impurity band, whereas that in the B-doped Ge1-xFex film is dominated by the holes in the valence band in the degenerated Fe-poor regions. As x increases (x = 2.3 - 14 %), the transport in the both undoped and B-doped Ge1-xFex films is dominated by hole hopping between the Fe-rich hopping sites of the impurity band. The magnetic properties of the Ge1-xFex films are studied by various methods including magnetic circular dichroism, magnetization and anomalous Hall resistance, and are not influenced by B-doping. We show band profile models of both undoped and B-doped Ge1-xFex films, which can explain the transport and the magnetic properties of the Ge1-xFex films.

cond-mat.mtrl-sci↗
arXiv2017

Spin injection into Si in three-terminal vertical and four-terminal lateral devices with Fe/Mg/MgO/Si tunnel junctions having an ultrathin Mg insertion layer

We demonstrated that the spin injection/extraction efficiency is enhanced by an ultrathin Mg insertion layer (<= 2 nm) in Fe/Mg/MgO/n+-Si tunnel junctions. In diode-type vertical three-terminal devices fabricated on a Si substrate, we observed the narrower three-terminal Hanle (N-3TH) signals indicating true spin injection into Si, and estimated the spin polarization in Si to be 16% when the thickness of the Mg insertion layer is 1 nm, whereas no N-3TH signal was observed without Mg insertion. This means that the spin injection/extraction efficiency is enhanced by suppressing the formation of a magnetically-dead layer at the Fe/MgO interface. We have also observed clear spin transport signals, such as non-local Hanle signals and spin-valve signals, in a lateral four-terminal device with the same Fe/Mg/MgO/n+-Si tunnel junctions fabricated on a Si-on-insulator substrate. It was found that both the intensity and linewidth of the spin signals are affected by the geometrical effects (device geometry and size). We have derived analytical functions taking into account the device structures, including channel thickness and electrode size, and estimated important parameters; spin lifetime and spin polarizations. Our analytical functions well explain the experimental results. Our study shows the importance of suppressing a magnetically-dead layer, and provides a unified understanding of spin injection/detection signals in different device geometries.

physics.app-ph↗