Abstract The dynamics of magnetization in thin ferromagnetic films play a central role in the advancement of nanotechnology, particularly in spintronic applications and high-density magnetic storage. In this work, we explore the dynamic regime of magnetization described by reduced micromagnetic models derived from the Landau-Lifshitz-Gilbert equation. These nonlinear vector-valued partial differential equations incorporate key interactions, including exchange, anisotropy, Zeeman, and magnetostatic contributions. Special attention is given to the impact of the external magnetic field with varying spatial profiles and the anisotropy constant of different materials. We develop and implement robust numerical schemes that preserve the geometric constraints of the magnetization vector field. The simulations are visualized using the ParaView platform to provide insight into the spatiotemporal evolution of magnetization. Our results contribute to a deeper understanding of how external fields influence magnetization dynamics, offering valuable perspectives for designing efficient magnetic devices at the nanoscale.

Abstract Modeling of pyrolytic carbon (PyC) coating on spheronized graphite particles was performed with the goal of developing a high-performance anode material for lithium-ion batteries. The model takes into account convective and radiative heat transfer, particle movement in a fluidized bed, benzene pyrolysis kinetics and carbon deposition in view of the adhesion coefficient. It has been shown that efficient benzene decomposition for PyC coating formation occurs at temperature of 1000–1100℃. A nitrogen pulse feed rate of 10 pulses min-1 maximizes particle concentration in the benzene active decomposition zone. A semi-empirical model predicting PyC coating thickness as a function of temperature and particle diameter was presented. The calculated layer thickness varies from 20 to 45 nm and is in good agreement with the experiment. The obtained anode material exhibits ≤ 5% loss in specific discharge capacity at 2C and recovers up to 99% capacity at 0.1C.

Abstract Grain boundary engineering has been a key topic in addressing the challenges of hydrogen embrittlement in iron. Particularly, the Σ3 symmetric tilt grain boundary (STGB) of iron inherits the resistance to hydrogen embrittlement. However, the effect of carbon as a steel solute on the hydrogen embrittlement mechanism remains elusive. Herein, we utilize classical molecular dynamics simulation and density functional theory calculations to investigate the role of carbon atoms on the mechanical properties and the H-atom diffusion mechanism of hydrogenated ∑3 (11 ̅2 ̅)[110] STGB Fe. The results reveal a significant decrease in ultimate tensile strength even when only 0.1 wt% of carbon atoms are added. Local lattice structural changes demonstrate a coupling effect between carbon and hydrogen atoms, which accelerates the formation of deformed areas and phase transformation. The addition of a carbon atom is also found to significantly reduce the diffusion barrier, thereby facilitating hydrogen diffusion.

Abstract Shale is a heterogeneous material at the mesoscale composed of multiple mineral constituents, and its macroscopic mechanical properties are strongly influenced by the proportion, spatial distribution, and mechanical characteristics of each component. Traditional finite element methods for investigating the effective modulus of shale require high-resolution models, leading to substantial computational cost. To overcome this challenge, this study proposes two prediction models based on the Fourier neural operator (FNO) for rapid estimation of the effective modulus of mesoscale shale samples. The first model directly maps shale scanning electron microscopy (SEM) images to effective modulus values, while the second model, a physics-informed FNO (PI-FNO), predicts the internal stress field of shale samples and derives the effective modulus by incorporating physical constraints such as equilibrium equations. Both models are trained and tested on 5000 mesoscale shale volume elements, with 1000 additional real SEM samples used for validation. Results show that the direct FNO model achieves a mean prediction error of only 0.27%, while the PI-FNO model yields 0.94% error but provides additional insights into internal stress distributions. Furthermore, the inference time per sample is about 17 ms for the FNO and 24 ms for the PI-FNO, demonstrating their potential for large-scale applications. These findings indicate that the proposed methodology not only ensures accurate and efficient prediction of the effective modulus of shale but also offers a generalizable framework for evaluating the effective mechanical parameters of other heterogeneous material systems.

Abstract Hierarchical nanoporous (HNP) metals demonstrate a strength transition from "smaller is stronger" to "smaller is weaker" below critical ligament sizes (10 nm), governed by surface-to-volume ratio evolution. By integrating Bond Order-Length-Strength (BOLS) theory, this work develops a predictive framework addressing the limitations of conventional scaling laws in capturing size effect reversal. Systematic analysis reveals that hierarchical architectures regulate surface-dominated mechanical behavior through coupled multiscale parameters, including nanoscale joint dimensions, submicron ligament geometry, hierarchical complexity (n ≥ 2), and density redistribution across structural levels. Experimental validation shows hierarchical designs mitigate surface effects while enabling strength compensation through optimized density allocation. These results establish a multiscale design paradigm bridging atomic-scale surface physics to tunable macroscopic mechanical properties in HNP systems.

Abstract Rapid estimation of Curie temperature is essential for discovering advanced functional materials like ABO 3 type ferromagnetic perovskites used in spintronics, sensors, and multiferroics. This study applies machine learning to predict Curie temperatures directly from chemical formulae, eliminating the need for computationally intensive simulations. A manually curated dataset of 300 perovskites with known Curie temperatures is used. Features are generated by separately considering A-site and B-site cation properties and also by treating the compound as a whole. Three strategies, feature selection, recursive elimination, and Principal Component Analysis are employed for feature engineering. Multiple models are trained and evaluated using cross-validation, with coefficient of determination R 2 as the performance metric. Shapley Additive Explanations highlights mean absolute deviation of A-site ground state volume and B-site space-group number as key factors influencing Curie temperature. The best model achieves an R 2 of ~0.8 and is applied to predict Curie temperatures for 21,467 ABO 3 compositions, demonstrating the potential of machine learning in accelerating magnetic materials discovery.

Abstract Deformation twins significantly influence the plastic deformation of TiAl alloys. However, it is difficult to obtain full information about the interaction between deformation twins and interfaces through the most advanced experimental methods. Here, we analyzed the interaction between deformation twins and three distinct γ/γ interfaces (true twin, 120° rotational twin, and pseudo twin) using classic molecular dynamics simulation. The simulations indicate that the activated slip systems vary across the three distinct γ/γ interfaces under two different tension loading directions. The Schmid factors corresponding to these slip systems were calculated, and they consist well with the operative slip systems in γ-TiAl. The simulation further shows that type-I intersection occurs under the interface-perpendicular tensile loading, while type-Ⅱ intersection occurs under the interface-parallel tensile loading. These results offer distinct perspectives on the plastic deformation modes involving deformation twinning in TiAl alloys, which can help us understand the relationships between microstructure and mechanical properties and enable us to design and optimize the performance of TiAl alloys.