Computational Materials Science最新文献_第5页

Efficient material model parameter optimization in finite element analysis with differentiable physics

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-12 DOI: 10.1016/j.commatsci.2025.113828

Sultan Al Hassanieh , Wesley F. Reinhart , Allison M. Beese

{"title":"Efficient material model parameter optimization in finite element analysis with differentiable physics","authors":"Sultan Al Hassanieh , Wesley F. Reinhart , Allison M. Beese","doi":"10.1016/j.commatsci.2025.113828","DOIUrl":"10.1016/j.commatsci.2025.113828","url":null,"abstract":"<div><div>In this study, an efficient finite element model parameter optimization method is proposed by integrating differentiable physics into an optimization scheme for faster convergence with fewer function evaluations than finite difference (FD) gradient–based and gradient–free methods. The method is demonstrated using constitutive material model calibration and stress-field homogenization problems. The method leverages the efficiency of commercial finite element solvers by integrating them into a differentiable programming framework and applying automatic differentiation (AD) to the stress return-mapping algorithm, enabling the direct computation of loss function gradients. This approach circumvents the need for finite differences in gradient-based methods, while outperforming gradient-free methods. The performances of AD-enhanced and gradient-free methods are compared across problems ranging in dimensionality from 1-D to 24-D. In a 3-D problem, Bayesian optimization and Nelder-Mead required over 50 additional objective function evaluations on average and took ∼ 13 times longer in wall-clock time to converge than the AD-enhanced methods. For the 24-D problem, it took FD over 15 times longer to compute gradients than AD. AD-enhanced methods maintained their efficiency with increasing dimensionality, making them especially powerful for complex materials problems with high dimensional parameter spaces.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113828"},"PeriodicalIF":3.1,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143601448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

BN/PSZ composite polymer: A molecular dynamics study and experimental characterization

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-12 DOI: 10.1016/j.commatsci.2025.113820

Qin Tang , Yujie Guan , Liming Tan , Chengzong Zeng , Jun Shen

{"title":"BN/PSZ composite polymer: A molecular dynamics study and experimental characterization","authors":"Qin Tang , Yujie Guan , Liming Tan , Chengzong Zeng , Jun Shen","doi":"10.1016/j.commatsci.2025.113820","DOIUrl":"10.1016/j.commatsci.2025.113820","url":null,"abstract":"<div><div>The excellent insulation, stability, and processability of Polysilazane (PSZ) and its composites have attracted attention. However, due to complex crosslinking and pyrolysis processes and its amorphous nature, numerical simulations on the crosslinking and properties of PSZ and its composites are scarce. This study focuses on a BN/PSZ polymer composite by blending Boron Nitride (BN) with PSZ. Molecular dynamics (MD) investigate crosslinking and thermal conductivity. Results show PSZ forms a network structure during crosslinking, enhancing thermal conductivity pathways with BN, resulting in a significant increase in composite thermal conductivity. With 100 % crosslinking and 79.03 wt% BN, thermal conductivity reaches 4.130 W/(m·K). At the same time, BN/PSZ mixture was prepared by the blending method. After heating and curing the mixture at 250 °C with Nitrogen atmosphere, the obtained cured product was pressed into tablets using a tablet press to observe its thermal conductivity. Comparing the simulation results with the actual test results, the maximum error rate was only 6.05 % at a BN of 60 vol%, which provides highly valuable references for the research on nanofiller/PSZ composites.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113820"},"PeriodicalIF":3.1,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143601569","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Double lone pair electrons driving polar semiconductors and metals

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-11 DOI: 10.1016/j.commatsci.2025.113835

Lulu Zhao , YiXuan Li , RuiFeng Zhang, Hu Zhang

引用次数: 0

Electron-dislocation interactions in electroplastic effects of pure aluminum: Thermal fluctuation-assisted electron wind mechanism

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-11 DOI: 10.1016/j.commatsci.2025.113818

Xiuwen Ren , Zhongjin Wang

{"title":"Electron-dislocation interactions in electroplastic effects of pure aluminum: Thermal fluctuation-assisted electron wind mechanism","authors":"Xiuwen Ren , Zhongjin Wang","doi":"10.1016/j.commatsci.2025.113818","DOIUrl":"10.1016/j.commatsci.2025.113818","url":null,"abstract":"<div><div>The electroplastic effect refers to the intrinsic mechanism by which high-density electric currents significantly enhance the plasticity and mechanical properties of metals. The electron wind force (EWF) mechanism is one of the fundamental principles underlying this effect; however, the interplay of multiple concurrent phenomena has hindered precise elucidation of electron-dislocation interactions. In this paper, we employ molecular dynamics (MD) simulations to investigate electron-driven dislocation behavior in pure aluminum at the atomic level, explicitly incorporating thermal fluctuations—an inherent atomic property—into the analysis. The evolution of dislocation configuration introduced by deformation was further explored based on this model. The results show that EWF induces the directional movement of atoms, and the kink nucleation decreases the critical EWF required for edge dislocation slip from 46.3fN to 0.046fN due to thermal fluctuation. For the metal with high density dislocations, electric current reduces the density of mobile dislocations while leaving immobile dislocation unchanged. This work can help clarify certain controversies surrounding the electron wind mechanism.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113818"},"PeriodicalIF":3.1,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143591995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Machine learning approach for predicting orientation-dependent elastic properties of 2D materials

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-10 DOI: 10.1016/j.commatsci.2025.113819

Shahram Yalameha

{"title":"A Machine learning approach for predicting orientation-dependent elastic properties of 2D materials","authors":"Shahram Yalameha","doi":"10.1016/j.commatsci.2025.113819","DOIUrl":"10.1016/j.commatsci.2025.113819","url":null,"abstract":"<div><div>Orientation-dependent mechanical properties, such as Young’s modulus (E), shear modulus (G), and Poisson’s ratio (ν), play a crucial role in characterizing the anisotropic behavior of two-dimensional (2D) materials. Conventionally, these properties are determined through tensorial transformations of second-order elastic stiffness tensors (C<sub>ij</sub>) as a function of angle, followed by analysis of the resulting elastic surfaces to identify extrema (E<sub>max</sub>, E<sub>min</sub>, G<sub>max</sub>, G<sub>min</sub>, ν<sub>max</sub>, ν<sub>min</sub>). The ratio of E<sub>max</sub> /E<sub>min</sub> serves as a key indicator of elastic anisotropy, while the occurrence of negative ν<sub>min</sub> identifies <em>auxetic</em> behavior. This work presents a machine learning approach, specifically employing a neural network, to directly predict these extrema from the elastic constants (C<sub>11</sub>, C<sub>12</sub>, C<sub>22</sub>, C<sub>66</sub>). A comprehensive dataset of over 6300 2D materials, extracted from the computational 2D materials database (C2DB), was used to train and validate the model. The developed model demonstrates exceptional predictive accuracy, exceeding 99 % for all predicted extrema, thereby bypassing the computationally intensive process of explicit tensorial transformations and orientation-dependent calculations. This efficient and accurate methodology enables rapid screening of 2D materials for specific mechanical properties, facilitating the identification of <em>auxetic</em> materials and the quantification of elastic anisotropy. This efficient approach enables the rapid screening of 2D materials for desired mechanical properties, facilitating the identification of auxetic materials and the quantification of elastic anisotropy. The developed methodology has the potential to accelerate materials discovery and design for a range of applications, including flexible electronics, mechanical metamaterials, and nano-scale devices.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113819"},"PeriodicalIF":3.1,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579976","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Wetting and spreading of NiCrFeSiB on iron-based alloy substrate at different temperatures: A molecular dynamics study

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-09 DOI: 10.1016/j.commatsci.2025.113826

Liu Yang, Rui Xue, Yibo Kang, Mingwei Zou

引用次数: 0

Quasiperiodic [110] symmetric tilt FCC grain boundaries

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-08 DOI: 10.1016/j.commatsci.2025.113811

Wenwen Zou , Juan Zhang , Jie Xu , Kai Jiang

引用次数: 0

Impacts of alkali and alkaline earth defects on the Electronic, magnetic properties and work function of 2H-CrS2

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-07 DOI: 10.1016/j.commatsci.2025.113822

Xuejiao Wang, Xintao Li, Longhua Li

{"title":"Impacts of alkali and alkaline earth defects on the Electronic, magnetic properties and work function of 2H-CrS2","authors":"Xuejiao Wang, Xintao Li, Longhua Li","doi":"10.1016/j.commatsci.2025.113822","DOIUrl":"10.1016/j.commatsci.2025.113822","url":null,"abstract":"<div><div>The substitutional doping method offers a unique opportunity to engineer the electronic and magnetic properties of 2D materials, possible to design new materials for nanoelectronics. Herein, by means of density functional theory (DFT), we systematically studied the structure, electronic, magnetic properties and work function of ten alkali and alkaline earth metals (X) substitution for S doping CrS<sub>2</sub> (X<sub>S</sub>-CrS<sub>2</sub>). The formation and binding energy calculations indicate that the X<sub>S</sub>-CrS<sub>2</sub> may be realized in experiments. The ab-initio molecular dynamics (AIMD) and elastic constants further suggest the stability of the doping structures. The electronic structures show that the electronic band gap is determined by the splitting energy of impurity states induced by X, creating pairs of electron and hole traps close to the Fermi level. The local magnetic moments of Cr are significantly enhanced by the X<sub>S</sub> doping. In particular, it is found that the work functions of X<sub>S</sub>-CrS<sub>2</sub> are asymmetric and linearly dependent on the atomic number of X. The alkali and alkaline earth metals reduce the work function on the top of the X<sub>S</sub>-CrS<sub>2</sub> nanosheets by 0.4–1.7 eV compared to the pristine 2H-CrS<sub>2</sub>. Moreover, the work function exhibits a linear relationship with the dipole moment of the nanosheet. All these findings provide insights in the defect behavior of alkali and alkaline earth metals and provide possibilities for electronics applications for X<sub>S</sub>-CrS<sub>2</sub>.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113822"},"PeriodicalIF":3.1,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Short-range order based ultra fast large-scale modeling of high-entropy alloys

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-07 DOI: 10.1016/j.commatsci.2025.113792

Caimei Niu , Lifeng Liu

{"title":"Short-range order based ultra fast large-scale modeling of high-entropy alloys","authors":"Caimei Niu , Lifeng Liu","doi":"10.1016/j.commatsci.2025.113792","DOIUrl":"10.1016/j.commatsci.2025.113792","url":null,"abstract":"<div><div>High-Entropy Alloys (HEAs) exhibit complex atomic interactions, with short-range order (SRO) playing a critical role in determining their properties. Traditional methods, such as Monte Carlo generator of Special Quasirandom Structures within the Alloy Theoretic Automated Toolkit (ATAT-mcsqs), Super-Cell Random APproximates (SCRAPs), and hybrid Monte Carlo-Molecular Dynamics (MC-MD)—are often hindered by limited system sizes and high computational costs. In response, we introduce PyHEA, a Python-based toolkit with a high-performance C++ core that leverages global and local search algorithms, incremental SRO computations, and GPU acceleration for unprecedented efficiency. When constructing random HEAs, PyHEA achieves speedups exceeding 333,000<span><math><mo>×</mo></math></span> and 13,900<span><math><mo>×</mo></math></span> over ATAT-mcsqs and SCRAPs, respectively, while maintaining high accuracy. PyHEA also offers a flexible workflow that allows users to incorporate target SRO values from external simulations (e.g., LAMMPS or density functional theory (DFT)), thereby enabling more realistic and customizable HEA models. As a proof of concept, PyHEA successfully replicated literature results for a 256,000-atom Fe–Mn–Cr–Co system within minutes—an order-of-magnitude improvement over hybrid MC-MD approaches. This dramatic acceleration opens new possibilities for bridging theoretical insights and practical applications, paving the way for the efficient design of next-generation HEAs.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113792"},"PeriodicalIF":3.1,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143563629","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Predicting the heat of formation and energy above convex hull of 2D MXenes using machine-learning methods

IF 3.1 3区材料科学

Computational Materials Science Pub Date : 2025-03-06 DOI: 10.1016/j.commatsci.2025.113790

Umair Haider , Gul Rahman , Imen Kebaili , Norah Alomayrah

{"title":"Predicting the heat of formation and energy above convex hull of 2D MXenes using machine-learning methods","authors":"Umair Haider , Gul Rahman , Imen Kebaili , Norah Alomayrah","doi":"10.1016/j.commatsci.2025.113790","DOIUrl":"10.1016/j.commatsci.2025.113790","url":null,"abstract":"<div><div>MXenes provide a high degree of compositional flexibility that can be used to provide adjustable mechanical, optical, and electrical properties. We provide a set of machine learning algorithms that are designed to forecast the heat of formation of MXenes and the energy above the convex hull. Our model is trained on 300 entries from the Computational 2D Materials Database (C2DB) using the fundamental chemical properties of the elements that make up MXene as features. The neural network model predicts the heat of formation using 12 different MXenes characteristics, with a mean absolute error (MAE) of 0.18 eV on training data and 0.21 eV on testing data. Characteristics of atoms terminating the MXene surface, including electronegativity, are important, according to feature importance analysis. Additionally, we employ a neural network model to estimate energy above the convex hull based on 14 characteristics. With a MAE of 0.03 eV on training data and 0.08 eV on testing data, the neural network model predicts the energy above the convex hull. We introduce reduced-order models comprising seven and four features. These reduced-order models exhibit easier transferability while exhibiting the same mean average error.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"253 ","pages":"Article 113790"},"PeriodicalIF":3.1,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143549338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0