Searcharxiv⌕ Search

arXiv subjects

Ryo Kobayashi

Publications and source records attributed to Ryo Kobayashi.

8 recordsLinked to original sources
arXiv2026

Multiscale Modelling of Ferroelectrics using a Physics-Informed Neural Network Driven by Molecular Dynamics Data: Parameter Identification and Field Reconstruction

In multiscale modeling of ferroelectrics, combining atomistic simulation with continuum-scale phase-field models (PFM) remains a fundamental challenge. A key difficulty lies in faithfully capturing discrete atomic-level information within a continuum modeling framework, while accurately representing material behavior at the mesoscale. In this paper, a Physics-Informed Neural Network (PINN) driven by molecular dynamics (MD) data is used. The loss function of the network consists of a supervised term that fits the discrete spatial polarization distributions obtained from MD simulations of systems containing domain walls, and a physics-based term that incorporates the residuals of partial differential equations (PDEs) of steady-state PFM. To ensure stable and balanced training among the different loss components, adaptive gradient normalization (GradNorm) is used to dynamically adjust the task weights. By minimizing the total loss, the model not only reconstructs the polarization field along with the associated strain, stress, and energy landscape at the continuum scale, but also identifies critical physical parameters of the phase-field model, including the characteristic energy density, characteristic length factor, gradient energy anisotropy factor, and Landau polynomial coefficients. By using the PINN-predicted physical parameters in COMSOL Multiphysics to solve the corresponding PDEs within a finite element framework, we demonstrate that these parameters enable accurate reproduction of the ferroelectric domain structure and the associated material response, including stress/strain distributions and energy landscape. This framework provides an effective methodology for establishing multiscale connections between atomistic and continuum descriptions, and holds the potential to infer underlying physical properties directly from polarization distributions for a wide range of materials.

cond-mat.mtrl-sci↗
arXiv2023

Molecular dynamics study of electronic temperature effects on the laser ablation of silicon

The molecular dynamics (MD) approach is an effective tool for investigating atomistic dynamical phenomena at the surface of materials under strong laser irradiation. Therefore, numerous laser ablation MD simulation studies have been conducted to date. However, in most MD studies, non-thermal and entropic effects via hot electrons on interatomic interactions that could cause significant differences in the simulation results are not considered. In this study, the MD simulation of the laser ablation of the Si surface was conducted using an interatomic potential whose parameters depended on the electronic temperature. Moreover, the results obtained with and without electronic temperature dependence were compared. The electronic temperature dependence resulted in an approximately four-times-greater compressive pressure near the surface, enhanced evaporation of atomic or smaller clusters, and slightly longer melt depth. Compared to the strong compressive pressure near the surface, the tensile pressure, which originated from the reflection of the compressive pressure wave at the surface, and ablation phenomena were less dependent on the electronic temperature.

cond-mat.mtrl-sci↗
arXiv2015

Enhancement of critical current density and mechanism of vortex pinning in H$^+$-irradiated FeSe single crystal

In this report, we comprehensively study the effect of H$^+$ irradiation on the critical current density, $J_c$, and vortex pinning in FeSe single crystal. It is found that the value of $J_c$ for FeSe is enhanced more than twice after 3-MeV H$^+$ irradiation. The scaling analyses of the vortex pinning force based on the Dew-Hughes model reveal that the H$^+$ irradiation successfully introduce point pinning centers into the crystal. We also find that the vortex creep rates are strongly suppressed after irradiation. Detailed analyses of the critical current dependent pinning energy based on the collective creep theory and extend Maley's method show that the H$^+$ irradiation enhances the value of $J_c$ before the flux creep, and also reduces the size of flux bundle, which will further reduce the field dependence of $J_c$ due to vortex motion.

cond-mat.supr-con↗
arXiv2015

Critical current density, vortex dynamics, and phase diagram of FeSe single crystal

We present a comprehensive study of the vortex pinning and dynamics in a high-quality FeSe single crystal, which is free from doping introduced inhomogeneities and charged quasi-particle-scattering because of its innate superconductivity. Critical current density, $J_c$, is found to be almost isotropic, and reaches a value $\sim$ 3 $\times$ 10$^4$ A/cm$^2$ at 2 K (self-field) for both $H$ $\|$ $c$ and $ab$. The normalized magnetic relaxation rate $S$ (= $\mid$dln$M$/dln$t$$\mid$) shows a temperature insensitive plateau behavior in the intermediate temperature range with a relatively high creep rate ($S$ $\sim$ 0.02 under zero field), which is interpreted in the framework of the collective creep theory. A crossover from the elastic to plastic creep is observed, while the fish-tail effect is absent for both $H$ $\|$ $c$ and $ab$. Based on this observation, the origin of the fish-tail effect is also discussed. Combining the results of $J_c$ and $S$, vortex motion in FeSe single crystal is found to be dominated by sparse strong point-like pinning from nm-sized defects or imperfections. The weak collective pinning is also observed and proved in the form of large bundles. Besides, the vortex phase diagram of FeSe is also constructed and discussed.

cond-mat.supr-con↗
arXiv2015

Sparse selection of bases in neural-network potential for crystalline and liquid Si

The neural-network interatomic potential for crystalline and liquid Si has been developed using the forward stepwise regression technique to reduce the number of bases with keeping the accuracy of the potential. This approach of making the neural-network potential enables us to construct the accurate interatomic potentials with less and important bases selected systematically and less heuristically. The evaluation of bulk crystalline properties, and dynamic properties of liquid Si show good agreements between the neural-network potential and ab-initio results.

cond-mat.mtrl-sci↗
arXiv2013

Interlimb neural connection is not required for gait transition in quadruped locomotion

Quadrupeds transition spontaneously to various gait patterns (e.g., walk, trot, pace, gallop) in response to the locomotion speed. The generation of these gait patterns has been the subject of debate for a long time. We propose a coupled oscillator model that is coupled with the physical interactions of the body. The results of this study showed that the gait pattern transitions spontaneously to walking/trotting/pacing/bounding in manner similar to that of real quadruped animals when the resonating portion of the body is changed according to the speed of leg movement. We also observed that pacing is expressed exclusively instead of trotting by changing the physical characteristics. In addition to leading to understanding of the principles of locomotion in living things, the coupled oscillator model proposed in this study is expected to lead to the creation of a legged robot that can select an energy-efficient gait and transition to it spontaneously.

q-bio.QM↗
arXiv1998

Modeling Grain Boundaries using a Phase Field Technique

We propose a two dimensional frame-invariant phase field model of grain impingement and coarsening. One dimensional analytical solutions for a stable grain boundary in a bicrystal are obtained, and equilibrium energies are computed. We are able to calculate the rotation rate for a free grain between two grains of fixed orientation. For a particular choice of functional dependencies in the model the grain boundary energy takes the same analytic form as the microscopic (dislocation) model of Read and Shockley.

cond-mat.mtrl-sci↗