Computational Materials Science最新文献

ReaxFF parameter optimization for β-Ga₂O₃ MD simulations using Gaussian process Bayesian optimization 基于高斯过程贝叶斯优化的β-Ga₂O₃MD模拟ReaxFF参数优化

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-07 DOI: 10.1016/j.commatsci.2026.114577

Stepan Savka, Andriy Serednytski, Dmytro Popovych

β-Gallium oxide (β-Ga₂O₃) is a promising wide-bandgap semiconductor for power electronics, requiring accurate molecular dynamics (MD) simulations to understand its atomic-scale behavior. This work presents the first automated optimization of ReaxFF parameters for β-Ga₂O₃ using Gaussian Process (GP) Bayesian optimization with a multi-objective framework incorporating pressure matching, force matching, and NVE stability testing. We optimized 22 critical ReaxFF parameters including bond energies, bond lengths, angle parameters, van der Waals interactions, and electronic properties. Reference data were obtained from MACE-MP-0, a universal machine learning potential trained on >150,000 DFT calculations. The multi-objective optimization achieved validated NVE ensemble stability at 0.1 fs timestep, with equilibrium pressure matching within 1.2% of MACE-MP-0 predictions (6.75 vs 6.67 GPa). The optimized parameters accurately reproduce experimental structural properties (lattice parameters within 0.3–2.6%, GaO bonds within 1%) and elastic constants within 2% of DFT values. Systematic timestep testing at 0.1, 0.25, and 0.5 fs confirmed that 0.1 fs is optimal for stable dynamics, characteristic of ReaxFF potentials with stiff bond terms. Parameter importance analysis revealed that van der Waals interactions and bond energies are most critical for accurate Ga₂O₃ modeling. The GP-Bayesian framework with multi-objective optimization successfully produced production-ready ReaxFF parameters for β-Ga₂O₃ MD simulations, demonstrating an efficient approach for developing reactive force fields with validated dynamic stability.

β-氧化镓（β-Ga₂O₃）是一种很有前途的用于电力电子的宽带隙半导体，需要精确的分子动力学（MD）模拟来理解其原子尺度的行为。这项工作首次使用高斯过程（GP）贝叶斯优化对β-Ga₂O₃的ReaxFF参数进行了自动优化，该优化具有多目标框架，包括压力匹配、力匹配和NVE稳定性测试。我们优化了ReaxFF的22个关键参数，包括键能、键长、角参数、范德华相互作用和电子性质。参考数据来自MACE-MP-0， MACE-MP-0是一种通用机器学习潜力，经过150,000次DFT计算训练。多目标优化在0.1 fs时间步长下实现了有效的NVE集成稳定性，平衡压力匹配在MACE-MP-0预测的1.2%以内（6.75 vs 6.67 GPa）。优化后的参数准确再现了实验结构性能（晶格参数在0.3-2.6%之间，GaO键在1%之间）和弹性常数在DFT值的2%以内。在0.1、0.25和0.5 fs的系统时间步长测试证实，0.1 fs是稳定动力学的最佳选择，具有刚性键项的ReaxFF电位的特征。参数重要性分析表明，范德华相互作用和键能对于精确的Ga₂O₃建模是最关键的。基于多目标优化的GP-Bayesian框架成功地为β-Ga₂O₃MD模拟生成了生产就绪的ReaxFF参数，展示了一种有效的方法来开发具有动态稳定性的反作用力场。

{"title":"ReaxFF parameter optimization for β-Ga₂O₃ MD simulations using Gaussian process Bayesian optimization","authors":"Stepan Savka, Andriy Serednytski, Dmytro Popovych","doi":"10.1016/j.commatsci.2026.114577","DOIUrl":"10.1016/j.commatsci.2026.114577","url":null,"abstract":"<div><div>β-Gallium oxide (β-Ga₂O₃) is a promising wide-bandgap semiconductor for power electronics, requiring accurate molecular dynamics (MD) simulations to understand its atomic-scale behavior. This work presents the first automated optimization of ReaxFF parameters for β-Ga₂O₃ using Gaussian Process (GP) Bayesian optimization with a multi-objective framework incorporating pressure matching, force matching, and NVE stability testing. We optimized 22 critical ReaxFF parameters including bond energies, bond lengths, angle parameters, van der Waals interactions, and electronic properties. Reference data were obtained from MACE-MP-0, a universal machine learning potential trained on >150,000 DFT calculations. The multi-objective optimization achieved validated NVE ensemble stability at 0.1 fs timestep, with equilibrium pressure matching within 1.2% of MACE-MP-0 predictions (6.75 vs 6.67 GPa). The optimized parameters accurately reproduce experimental structural properties (lattice parameters within 0.3–2.6%, Ga<img>O bonds within 1%) and elastic constants within 2% of DFT values. Systematic timestep testing at 0.1, 0.25, and 0.5 fs confirmed that 0.1 fs is optimal for stable dynamics, characteristic of ReaxFF potentials with stiff bond terms. Parameter importance analysis revealed that van der Waals interactions and bond energies are most critical for accurate Ga₂O₃ modeling. The GP-Bayesian framework with multi-objective optimization successfully produced production-ready ReaxFF parameters for β-Ga₂O₃ MD simulations, demonstrating an efficient approach for developing reactive force fields with validated dynamic stability.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114577"},"PeriodicalIF":3.3,"publicationDate":"2026-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186227","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

Active learning for predicting the enthalpy of mixing in binary liquids based on ab initio molecular dynamics 基于从头算分子动力学的二元液体混合焓预测的主动学习

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-07 DOI: 10.1016/j.commatsci.2026.114568

Quentin Bizot , Ryo Tamura , Guillaume Deffrennes

The enthalpy of mixing in the liquid phase is an important property for predicting phase formation in alloys. In multicomponent metallic liquids, it can be estimated from the binary interactions using a geometrical model, but data are available in less than a third of the binary systems. The prediction of this property in binary liquids is therefore important, and machine learning has recently achieved the highest accuracy. Further improvements requires acquiring high-quality data in liquids where models are poorly constrained. In this study, we propose an active learning approach to identify in which liquids additional data are most needed to improve an initial dataset that covers over 400 binary liquids. We identify a critical need for new data on liquids containing refractory elements, which we address by performing ab initio molecular dynamics simulations for 29 equimolar alloys of Ir, Os, Re and W. This enables more accurate predictions of the enthalpy of mixing, and we discuss the trends obtained for refractory elements of period 6. We use clustering analysis to interpret the results of active learning and to explore how our features can be linked to Miedema’s semi-empirical theory.

液相混合焓是预测合金相形成的一个重要性质。在多组分金属液体中，它可以使用几何模型从二元相互作用中估计出来，但在不到三分之一的二元系统中可以获得数据。因此，二元液体的这种性质的预测是重要的，机器学习最近达到了最高的精度。进一步的改进需要在模型约束较差的液体中获取高质量的数据。在这项研究中，我们提出了一种主动学习方法来确定哪些液体最需要额外的数据，以改进覆盖400多种二元液体的初始数据集。我们确定了对含有难熔元素的液体的新数据的迫切需求，我们通过对29种Ir， Os， Re和w等摩尔合金进行从头算分子动力学模拟来解决这一问题。这可以更准确地预测混合焓，我们讨论了6期难熔元素的趋势。我们使用聚类分析来解释主动学习的结果，并探索如何将我们的特征与Miedema的半经验理论联系起来。

{"title":"Active learning for predicting the enthalpy of mixing in binary liquids based on ab initio molecular dynamics","authors":"Quentin Bizot , Ryo Tamura , Guillaume Deffrennes","doi":"10.1016/j.commatsci.2026.114568","DOIUrl":"10.1016/j.commatsci.2026.114568","url":null,"abstract":"<div><div>The enthalpy of mixing in the liquid phase is an important property for predicting phase formation in alloys. In multicomponent metallic liquids, it can be estimated from the binary interactions using a geometrical model, but data are available in less than a third of the binary systems. The prediction of this property in binary liquids is therefore important, and machine learning has recently achieved the highest accuracy. Further improvements requires acquiring high-quality data in liquids where models are poorly constrained. In this study, we propose an active learning approach to identify in which liquids additional data are most needed to improve an initial dataset that covers over 400 binary liquids. We identify a critical need for new data on liquids containing refractory elements, which we address by performing ab initio molecular dynamics simulations for 29 equimolar alloys of Ir, Os, Re and W. This enables more accurate predictions of the enthalpy of mixing, and we discuss the trends obtained for refractory elements of period 6. We use clustering analysis to interpret the results of active learning and to explore how our features can be linked to Miedema’s semi-empirical theory.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114568"},"PeriodicalIF":3.3,"publicationDate":"2026-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186228","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

FBformer: A four-body feature enhanced periodic graph transformer for crystal property prediction FBformer：一种用于晶体性能预测的四体特征增强周期图变压器

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114566

Yang Li , Zhihui Wang , Wei Zhou , Rui Wang , Haiyan Zhang , Shu Zhan , Jiajia Xu

Driven by the rapid progress of high-throughput DFT calculations and the expansion of materials databases, machine learning has become increasingly central to the prediction of materials properties. Traditional descriptor-driven models, though physically interpretable, often fail to comprehensively capture the high-order geometric characteristics of complex crystals. To address this limitation, this study proposes FBformer, a crystal property prediction model based on periodic graph encoding. Built upon the Matformer framework, FBformer introduces four-body features, including bond angles and dihedral angles, to explicitly model crystal periodicity and multi-body interactions. By constructing a dual-graph architecture that integrates atomic and angular representations, FBformer effectively fuses atomic types, bond lengths, bond angles, and dihedral angles across multi-level node and edge embeddings, thereby enhancing the model's structural representation capability. Across the eight prediction tasks on the Materials Project and JARVIS-DFT databases, except for formation energy on the Materials Project, FBformer significantly outperforms existing models in predicting Ehull, formation energy on JARVIS-DFT, bandgap, total energy, bulk moduli, and shear moduli. Ablation experiments show that progressively incorporating three-body and four-body features consistently enhances model performance, underscoring the crucial importance of high-order geometric information in crystal property modeling. This study presents novel conceptual and methodological contributions that drive the deeper convergence of AI and materials science, and lays a solid foundation for the efficient prediction and design of novel crystalline materials. The source code can be accessed at: https://github.com/YangLi2025/FBformer.

在高通量DFT计算的快速发展和材料数据库的扩展的推动下，机器学习在预测材料特性方面变得越来越重要。传统的描述符驱动模型虽然在物理上是可解释的，但往往不能全面地捕捉复杂晶体的高阶几何特征。为了解决这一问题，本研究提出了一种基于周期图编码的晶体性能预测模型FBformer。FBformer基于Matformer框架，引入了四体特征，包括键角和二面角，来明确地模拟晶体周期性和多体相互作用。FBformer通过构建原子表示和角度表示相结合的双图架构，有效融合了原子类型、键长、键角和二面角，跨越多层次节点和边缘嵌入，增强了模型的结构表示能力。在材料项目和JARVIS-DFT数据库的8个预测任务中，除了材料项目的地层能量预测任务外，FBformer在预测Ehull、JARVIS-DFT上的地层能量、带隙、总能量、体模量和剪切模量方面明显优于现有模型。烧蚀实验表明，逐步纳入三体和四体特征可以持续提高模型性能，强调了高阶几何信息在晶体性质建模中的重要性。本研究提出了新颖的概念和方法贡献，推动了人工智能与材料科学的更深层次融合，为新型晶体材料的有效预测和设计奠定了坚实的基础。源代码可以在https://github.com/YangLi2025/FBformer上访问。

{"title":"FBformer: A four-body feature enhanced periodic graph transformer for crystal property prediction","authors":"Yang Li , Zhihui Wang , Wei Zhou , Rui Wang , Haiyan Zhang , Shu Zhan , Jiajia Xu","doi":"10.1016/j.commatsci.2026.114566","DOIUrl":"10.1016/j.commatsci.2026.114566","url":null,"abstract":"<div><div>Driven by the rapid progress of high-throughput DFT calculations and the expansion of materials databases, machine learning has become increasingly central to the prediction of materials properties. Traditional descriptor-driven models, though physically interpretable, often fail to comprehensively capture the high-order geometric characteristics of complex crystals. To address this limitation, this study proposes FBformer, a crystal property prediction model based on periodic graph encoding. Built upon the Matformer framework, FBformer introduces four-body features, including bond angles and dihedral angles, to explicitly model crystal periodicity and multi-body interactions. By constructing a dual-graph architecture that integrates atomic and angular representations, FBformer effectively fuses atomic types, bond lengths, bond angles, and dihedral angles across multi-level node and edge embeddings, thereby enhancing the model's structural representation capability. Across the eight prediction tasks on the Materials Project and JARVIS-DFT databases, except for formation energy on the Materials Project, FBformer significantly outperforms existing models in predicting Ehull, formation energy on JARVIS-DFT, bandgap, total energy, bulk moduli, and shear moduli. Ablation experiments show that progressively incorporating three-body and four-body features consistently enhances model performance, underscoring the crucial importance of high-order geometric information in crystal property modeling. This study presents novel conceptual and methodological contributions that drive the deeper convergence of AI and materials science, and lays a solid foundation for the efficient prediction and design of novel crystalline materials. The source code can be accessed at: <span><span>https://github.com/YangLi2025/FBformer</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114566"},"PeriodicalIF":3.3,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186233","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

Mechanisms of multiple V-doping in tuning mechanical and hydrogen storage properties of ZrCo alloys 多重v掺杂调整ZrCo合金力学性能和储氢性能的机理

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114562

Qin Qin , Yawen Hua , Luyao Hai , Meidie Wu , Siqi Jiang , Rongxing Ye , Jiangfeng Song , Yiliang Liu , Linsen Zhou

Zirconium‑cobalt (ZrCo) alloy is a promising candidate for replacing uranium in tritium storage, yet its practical application is limited by disproportionation-induced capacity decay. This study explores the effect of multi-V doping on the configurations, mechanical properties, and hydrogen storage behavior of ZrCo alloys. Specifically, V dopants induce lattice contraction owing to their smaller atomic radius and exhibit an energetically preferred homogeneous dopant dispersion. Mechanistically, the substitutional strengthening effect is highly sensitive to the concentration and configuration of V dopants, with an optimal concentration of ∼11.1%. Furthermore, multiple-V doping can enhance the thermodynamic stability of hydrogen at OCT1 interstitial sites and lower the migration barrier for hydrogen diffusion, thereby facilitating hydriding/dehydriding kinetics in ZrCo alloys. For β-phase hydrides, it significantly improves the anti-disproportionation performance through a synergistic mechanism involving structural reduction of 8e site volume, thermodynamic destabilization of H(8e) occupation, and kinetic facilitation of H(8e) egress. These findings provide a theoretical basis for designing high-performance ZrCoalloys for advanced tritium storage applications.

锆钴（ZrCo）合金是替代氚储存中的铀的有前途的候选材料，但其实际应用受到歧化诱导的容量衰减的限制。本研究探讨了多v掺杂对ZrCo合金结构、力学性能和储氢行为的影响。具体来说，V掺杂剂由于其较小的原子半径而诱导晶格收缩，并表现出能量优先的均匀掺杂色散。从机理上看，取代强化效应对V掺杂剂的浓度和结构高度敏感，最佳浓度为~ 11.1%。此外，多v掺杂可以增强氢在OCT1间隙位置的热力学稳定性，降低氢扩散的迁移势垒，从而促进ZrCo合金的氢化/脱氢动力学。对于β相氢化物，它通过减少8e位点体积的结构、H（8e）占据的热力学不稳定性和H（8e）排出的动力学促进等协同机制显著提高了抗歧化性能。这些发现为设计高性能的zrco合金用于先进的氚储存提供了理论基础。

{"title":"Mechanisms of multiple V-doping in tuning mechanical and hydrogen storage properties of ZrCo alloys","authors":"Qin Qin , Yawen Hua , Luyao Hai , Meidie Wu , Siqi Jiang , Rongxing Ye , Jiangfeng Song , Yiliang Liu , Linsen Zhou","doi":"10.1016/j.commatsci.2026.114562","DOIUrl":"10.1016/j.commatsci.2026.114562","url":null,"abstract":"<div><div>Zirconium‑cobalt (ZrCo) alloy is a promising candidate for replacing uranium in tritium storage, yet its practical application is limited by disproportionation-induced capacity decay. This study explores the effect of multi-V doping on the configurations, mechanical properties, and hydrogen storage behavior of ZrCo alloys. Specifically, V dopants induce lattice contraction owing to their smaller atomic radius and exhibit an energetically preferred homogeneous dopant dispersion. Mechanistically, the substitutional strengthening effect is highly sensitive to the concentration and configuration of V dopants, with an optimal concentration of ∼11.1%. Furthermore, multiple-V doping can enhance the thermodynamic stability of hydrogen at OCT1 interstitial sites and lower the migration barrier for hydrogen diffusion, thereby facilitating hydriding/dehydriding kinetics in ZrCo alloys. For <em>β</em>-phase hydrides, it significantly improves the anti-disproportionation performance through a synergistic mechanism involving structural reduction of 8e site volume, thermodynamic destabilization of H(8e) occupation, and kinetic facilitation of H(8e) egress. These findings provide a theoretical basis for designing high-performance ZrCoalloys for advanced tritium storage applications.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114562"},"PeriodicalIF":3.3,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186231","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

Corrigendum to “Tunable magnetic and topological phases in EuMnXBi₂ (X=Mn, Fe, co, Zn) pnictides” [Comput. Mater. Sci. 264 (2026) 114481] “eumnxbi2 (X=Mn, Fe, co, Zn) pnictides中的可调谐磁性和拓扑相”的更正[computer]。板牙。科学通报。264 (2026)114481]

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114556

Deep Sagar , Abhishek Sharma , Arti Kashyap

引用次数: 0

Comparative analysis of plasticity-based GND density estimation methods in crystal plasticity finite element models 晶体塑性有限元模型中基于塑性的GND密度估算方法的比较分析

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114567

Michael Pilipchuk , Chaitali Patil , Veera Sundararaghavan

In crystal plasticity finite element (CPFE) simulations, accurately quantifying geometrically necessary dislocations (GNDs) is critical for capturing strain gradients in polycrystals. We compare different methods for quantifying GNDs, all of which originate from the Nye tensor, which is computed as the curl of the plastic deformation gradient. The projection technique directly decomposes the Nye tensor onto individual screw and edge dislocation components to compute GNDs. This approach requires converting a nine-component Nye tensor into densities for a larger number of dislocation systems, a fundamentally underdetermined (non-unique) process, which is resolved using

L 2

minimization. In contrast, when employing CPFE analysis, one could directly compute dislocation densities on each slip system using shear gradients. Projection and slip gradient methods are compared with respect to their prediction of GNDs with changing grain size, strain, and grain neighborhoods, including multigrain junctions. Although these techniques match analytical GND densities for single slip, single crystal deformation, and are consistent with anticipated overall GND trends, we find that the GND densities from projection techniques are significantly lower than those predicted from CPFE-based slip gradients in polycrystals. A suggested improvement of only using the active dislocation systems in the projection technique almost entirely resolved this mismatch.

在晶体塑性有限元（CPFE）模拟中，精确量化几何必要位错（GNDs）是捕获多晶体应变梯度的关键。我们比较了量化gds的不同方法，所有这些方法都源于Nye张量，它被计算为塑性变形梯度的旋度。投影技术直接将Nye张量分解为单独的螺旋位错和边缘位错分量来计算GNDs。这种方法需要将一个九分量的Nye张量转换成大量位错系统的密度，这是一个根本不确定（非唯一）的过程，可以使用L2最小化来解决。相比之下，当采用CPFE分析时，可以直接使用剪切梯度计算每个滑移系统上的位错密度。比较了投影法和滑动梯度法在晶粒尺寸、应变和晶粒邻域（包括多晶粒结）变化情况下对GNDs的预测。虽然这些技术与单滑移、单晶变形的分析地地密度相匹配，并且与预期的总体地地趋势一致，但我们发现，投影技术的地地密度明显低于基于cpfe的多晶滑移梯度的预测地地密度。建议在投影技术中只使用主动位错系统的改进几乎完全解决了这种不匹配。

{"title":"Comparative analysis of plasticity-based GND density estimation methods in crystal plasticity finite element models","authors":"Michael Pilipchuk , Chaitali Patil , Veera Sundararaghavan","doi":"10.1016/j.commatsci.2026.114567","DOIUrl":"10.1016/j.commatsci.2026.114567","url":null,"abstract":"<div><div>In crystal plasticity finite element (CPFE) simulations, accurately quantifying geometrically necessary dislocations (GNDs) is critical for capturing strain gradients in polycrystals. We compare different methods for quantifying GNDs, all of which originate from the Nye tensor, which is computed as the curl of the plastic deformation gradient. The projection technique directly decomposes the Nye tensor onto individual screw and edge dislocation components to compute GNDs. This approach requires converting a nine-component Nye tensor into densities for a larger number of dislocation systems, a fundamentally underdetermined (non-unique) process, which is resolved using <span><math><mrow><mi>L</mi><mn>2</mn></mrow></math></span> minimization. In contrast, when employing CPFE analysis, one could directly compute dislocation densities on each slip system using shear gradients. Projection and slip gradient methods are compared with respect to their prediction of GNDs with changing grain size, strain, and grain neighborhoods, including multigrain junctions. Although these techniques match analytical GND densities for single slip, single crystal deformation, and are consistent with anticipated overall GND trends, we find that the GND densities from projection techniques are significantly lower than those predicted from CPFE-based slip gradients in polycrystals. A suggested improvement of only using the active dislocation systems in the projection technique almost entirely resolved this mismatch.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114567"},"PeriodicalIF":3.3,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186232","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

Molecular dynamics insights into energy barrier modulation by thiol-mixed co-surfactants in surfactant-mediated gold nanocrystal growth 在表面活性剂介导的金纳米晶体生长中，巯基混合共表面活性剂对能量势垒调制的分子动力学见解

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114537

Mona Vishwakarma, Debdip Bhandary

How do thiol-based co-surfactants influence micelle morphology and, in turn, regulate gold nucleation dynamics? To answer this question, we investigated a ternary surfactant system comprising cetyltrimethylammonium bromide (CTAB), oleylamine (OLA), and hexadecanethiol (HT) using all-atom molecular dynamics simulations at different molar ratios. Among the compositions studied, a CTAB/OLA/HT ratio of 3:1:1 leads to the well-defined and stable cylindrical micelle, creating a confined environment conducive to anisotropic gold nucleation. This geometry promoted directional growth, leading to the formation of an elliptical gold nucleate, with the alignment of surfactant tails guiding preferential nucleation along a single axis. To track the structural evolution during nucleation, we monitored micelle behaviour as gold atoms were incorporated over time. Initially dispersed randomly in solution, gold atoms progressively aggregated into larger nucleates that penetrated the micelle core. Their incorporation caused a gradual decrease in the surfactant order parameter, indicating a transition towards bilayer-like organization and highlighting the micelle’s adaptive response to nucleate growth. The free energy barrier associated with the migration of a gold nucleate from the micelle interior to its surface was quantified using umbrella sampling combined with the Weighted Histogram Analysis Method (WHAM). The inclusion of thiol molecules significantly lowered this barrier to 2.83

\pm

0.12 kcal/mol, demonstrating a more favourable pathway for nucleation and growth within the micelle.

巯基共表面活性剂如何影响胶束形态，进而调节金的成核动力学？为了回答这个问题，我们研究了由十六烷基三甲基溴化铵（CTAB）、油胺（OLA）和十六烷硫醇（HT）组成的三元表面活性剂体系，采用不同摩尔比下的全原子分子动力学模拟。在所研究的组合物中，CTAB/OLA/HT比例为3:1:1时，形成了轮廓清晰且稳定的柱状胶束，形成了有利于各向异性金成核的密闭环境。这种几何形状促进了定向生长，导致椭圆金核的形成，表面活性剂尾部的排列沿着单轴引导优先成核。为了跟踪成核过程中的结构演变，我们监测了随着时间的推移，金原子被纳入胶束的行为。最初在溶液中随机分散，金原子逐渐聚集成更大的核，穿透胶束核心。它们的掺入导致表面活性剂有序参数逐渐降低，表明胶束向双层状组织过渡，并突出了胶束对核生长的适应性响应。利用伞形采样和加权直方图分析法（WHAM）对金核从胶束内部向表面迁移的自由能势垒进行了定量分析。巯基分子的加入显著降低了这一势垒至2.83±0.12 kcal/mol，表明这是一个更有利的胶束成核和生长途径。

{"title":"Molecular dynamics insights into energy barrier modulation by thiol-mixed co-surfactants in surfactant-mediated gold nanocrystal growth","authors":"Mona Vishwakarma, Debdip Bhandary","doi":"10.1016/j.commatsci.2026.114537","DOIUrl":"10.1016/j.commatsci.2026.114537","url":null,"abstract":"<div><div>How do thiol-based co-surfactants influence micelle morphology and, in turn, regulate gold nucleation dynamics? To answer this question, we investigated a ternary surfactant system comprising cetyltrimethylammonium bromide (CTAB), oleylamine (OLA), and hexadecanethiol (HT) using all-atom molecular dynamics simulations at different molar ratios. Among the compositions studied, a CTAB/OLA/HT ratio of 3:1:1 leads to the well-defined and stable cylindrical micelle, creating a confined environment conducive to anisotropic gold nucleation. This geometry promoted directional growth, leading to the formation of an elliptical gold nucleate, with the alignment of surfactant tails guiding preferential nucleation along a single axis. To track the structural evolution during nucleation, we monitored micelle behaviour as gold atoms were incorporated over time. Initially dispersed randomly in solution, gold atoms progressively aggregated into larger nucleates that penetrated the micelle core. Their incorporation caused a gradual decrease in the surfactant order parameter, indicating a transition towards bilayer-like organization and highlighting the micelle’s adaptive response to nucleate growth. The free energy barrier associated with the migration of a gold nucleate from the micelle interior to its surface was quantified using umbrella sampling combined with the Weighted Histogram Analysis Method (WHAM). The inclusion of thiol molecules significantly lowered this barrier to 2.83 <span><math><mo>±</mo></math></span> 0.12 kcal/mol, demonstrating a more favourable pathway for nucleation and growth within the micelle.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114537"},"PeriodicalIF":3.3,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186234","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

Deep learning interatomic potential for metal-doped silicon carbide nanotubes: Development, validation, and mechanical response 金属掺杂碳化硅纳米管的深度学习原子间势：开发、验证和机械响应

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-06 DOI: 10.1016/j.commatsci.2026.114531

Shin-Pon Ju , Dong-Yeh Wu , Chun-Wen Cheng , Hsing-Yin Chen

We present the development and validation of a deep learning interatomic potential (DLP) for metal-doped silicon carbide nanotubes (SiCNTs), trained on extensive density functional theory (DFT) data. The potential was constructed using an active-learning workflow (DPGEN + DeePMD-kit) and reproduces reference DFT energies, forces, and energy–area relations with near-ab initio fidelity. This transferable model enables large-scale molecular dynamics simulations of pristine and Al-, Fe-, Mn-, and Pt-doped (6) SiCNTs at dopant fractions from 0.5% to 8.3%. The DLP-MD simulations reveal that dilute Fe and Pt doping largely preserve tensile strength, whereas Al and Mn induce early lattice disorder and brittleness; at higher concentrations, all dopants degrade strength due to dopant–dopant interactions within the short-range cutoff. Analysis of bond-length statistics, polygonal ring evolution, and angular distribution functions connects local electronic structure to fracture mechanisms. The results demonstrate that data-driven potentials can extend ab initio accuracy to complex doped nanostructures, providing both methodological advances and design guidelines for robust SiCNT-based nanomaterials.

我们提出了基于广泛密度泛函理论（DFT）数据训练的金属掺杂碳化硅纳米管（SiCNTs）的深度学习原子间势（DLP）的开发和验证。使用主动学习工作流（DPGEN + DeePMD-kit）构建势，并以接近从头开始的保真度再现参考DFT能量、力和能量-面积关系。这种可转移的模型使原始和Al-， Fe-， Mn-和pt掺杂(6)SiCNTs在掺杂分数从0.5%到8.3%的大规模分子动力学模拟成为可能。DLP-MD模拟结果表明，稀Fe和Pt的掺杂在很大程度上保持了材料的抗拉强度，而Al和Mn的掺杂则导致材料的早期晶格无序和脆性；在较高的浓度下，所有的掺杂剂由于在短截止范围内的掺杂剂相互作用而降低了强度。键长统计、多边形环演化和角分布函数的分析将局部电子结构与断裂机制联系起来。结果表明，数据驱动势可以将从头算精度扩展到复杂的掺杂纳米结构，为稳健的sicnt基纳米材料提供了方法上的进步和设计指导。

{"title":"Deep learning interatomic potential for metal-doped silicon carbide nanotubes: Development, validation, and mechanical response","authors":"Shin-Pon Ju , Dong-Yeh Wu , Chun-Wen Cheng , Hsing-Yin Chen","doi":"10.1016/j.commatsci.2026.114531","DOIUrl":"10.1016/j.commatsci.2026.114531","url":null,"abstract":"<div><div>We present the development and validation of a deep learning interatomic potential (DLP) for metal-doped silicon carbide nanotubes (SiCNTs), trained on extensive density functional theory (DFT) data. The potential was constructed using an active-learning workflow (DPGEN + DeePMD-kit) and reproduces reference DFT energies, forces, and energy–area relations with near-ab initio fidelity. This transferable model enables large-scale molecular dynamics simulations of pristine and Al-, Fe-, Mn-, and Pt-doped (6) SiCNTs at dopant fractions from 0.5% to 8.3%. The DLP-MD simulations reveal that dilute Fe and Pt doping largely preserve tensile strength, whereas Al and Mn induce early lattice disorder and brittleness; at higher concentrations, all dopants degrade strength due to dopant–dopant interactions within the short-range cutoff. Analysis of bond-length statistics, polygonal ring evolution, and angular distribution functions connects local electronic structure to fracture mechanisms. The results demonstrate that data-driven potentials can extend ab initio accuracy to complex doped nanostructures, providing both methodological advances and design guidelines for robust SiCNT-based nanomaterials.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114531"},"PeriodicalIF":3.3,"publicationDate":"2026-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186230","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

Surface-informed active learning prediction of thermophysical properties for liquid refractory multicomponent alloy 液态难熔多组分合金热物性的表面主动学习预测

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-05 DOI: 10.1016/j.commatsci.2026.114565

Kelun Liu, Yingchao Hai, Ying Ruan, Bingbo Wei

The thermophysical properties of liquid alloys such as density and surface tension are indispensable to explore heat and mass transport, interfacial dynamics and phase transitions in metallic systems. In refractory multicomponent alloys, the large disparities in thermophysical properties among constituent elements, combined with pronounced chemical and configurational disorder, lead to inherent complexity that challenges the accurate prediction of surface properties. Here, we proposed a surface-informed active learning approach that incorporated surface atomic environments into interatomic potential development. Applying the obtained potential, the liquid properties of W₂₅Nb₂₅Hf₂₅Zr₂₅ alloy were predicted and its surface atomic structure was resolved. Electrostatic and electromagnetic levitation measurements were used to validate the predicted surface tension and density. The predictions were in excellent agreement with the measurements, with deviations of only 3.6% for surface tension and 4.6% for density. Atomic-scale analysis showed pronounced Zr enrichment at the liquid surface, consistent with its lowest surface tension relative to the other constituent elements. A kinetic-energy peak was observed near the surface, suggesting enhanced atomic activity at elevated temperatures. The coordination number of surface atoms was lower than that of bulk atoms, indicating reduced local packing at the surface.

液态合金的热物理性质，如密度和表面张力，对于探索金属系统中的热、质传递、界面动力学和相变是必不可少的。在难熔多组分合金中，组成元素之间热物理性质的巨大差异，加上明显的化学和构型紊乱，导致其固有的复杂性，对表面性质的准确预测提出了挑战。在这里，我们提出了一种表面知情的主动学习方法，将表面原子环境纳入原子间电位的发展。利用所得电位预测了W25Nb25Hf25Zr25合金的液相性能，并对其表面原子结构进行了分析。采用静电和电磁悬浮测量来验证预测的表面张力和密度。预测结果与测量结果非常吻合，表面张力偏差仅为3.6%，密度偏差为4.6%。原子尺度分析表明，液体表面明显富集Zr，其表面张力相对于其他组成元素最低。在表面附近观察到一个动能峰值，表明在高温下原子活动增强。表面原子的配位数低于体原子的配位数，表明表面局部堆积减少。

{"title":"Surface-informed active learning prediction of thermophysical properties for liquid refractory multicomponent alloy","authors":"Kelun Liu, Yingchao Hai, Ying Ruan, Bingbo Wei","doi":"10.1016/j.commatsci.2026.114565","DOIUrl":"10.1016/j.commatsci.2026.114565","url":null,"abstract":"<div><div>The thermophysical properties of liquid alloys such as density and surface tension are indispensable to explore heat and mass transport, interfacial dynamics and phase transitions in metallic systems. In refractory multicomponent alloys, the large disparities in thermophysical properties among constituent elements, combined with pronounced chemical and configurational disorder, lead to inherent complexity that challenges the accurate prediction of surface properties. Here, we proposed a surface-informed active learning approach that incorporated surface atomic environments into interatomic potential development. Applying the obtained potential, the liquid properties of W<sub>25</sub>Nb<sub>25</sub>Hf<sub>25</sub>Zr<sub>25</sub> alloy were predicted and its surface atomic structure was resolved. Electrostatic and electromagnetic levitation measurements were used to validate the predicted surface tension and density. The predictions were in excellent agreement with the measurements, with deviations of only 3.6% for surface tension and 4.6% for density. Atomic-scale analysis showed pronounced Zr enrichment at the liquid surface, consistent with its lowest surface tension relative to the other constituent elements. A kinetic-energy peak was observed near the surface, suggesting enhanced atomic activity at elevated temperatures. The coordination number of surface atoms was lower than that of bulk atoms, indicating reduced local packing at the surface.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114565"},"PeriodicalIF":3.3,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186229","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

Atomistic insights into the behaviors of helium in W-Ni-Fe alloys with a new quaternary interatomic potential 氦在W-Ni-Fe合金中的原子行为与新的四元原子间势

IF 3.3 3区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

Computational Materials Science

Pub Date : 2026-02-04 DOI: 10.1016/j.commatsci.2026.114554

Xichuan Liao , Haipan Xiang , Rongyang Qiu , Yangchun Chen , Yong Liu , Ning Gao , Fei Gao , Wangyu Hu , Huiqiu Deng

Helium bubbles are a critical type of radiation damage that can greatly degrade the mechanical properties of W-Ni-Fe ductile-phase toughened tungsten (DPT-W) composites. Understanding helium-induced embrittlement requires a detailed investigation of helium behavior in DPT-W composites. The availability of suitable interatomic potentials would enable atomistic simulations, which are essential to elucidate the complex mechanisms underlying bubble formation. In this work, a new W-Ni-Fe-He quaternary potential, constructed within the Finnis–Sinclair framework, has been developed to facilitate the modeling of helium bubbles in the DPT-W composite. We then demonstrate that our potential closely reproduces available density functional theory results on important properties relevant to the helium behavior, including the formation and migration energies of single helium defects, the binding energies of helium-solute pairs, and small helium-defect clusters. In addition, molecular dynamics tests establish that our potential leads to the nucleation of helium bubbles from an initial random distribution of helium interstitial atoms. Our study shows that considerable helium accumulation near the interphase boundaries induces morphological changes in the misfit dislocations. The newly developed interatomic potential and the simulation data provide valuable insights into the behavior of helium in W-Ni-Fe tungsten heavy alloys.

氦气泡是一种重要的辐射损伤类型，会严重降低W-Ni-Fe韧性相增韧钨（DPT-W）复合材料的力学性能。了解氦致脆需要对DPT-W复合材料中的氦行为进行详细的研究。适当的原子间电位的可用性将使原子模拟成为可能，这对于阐明气泡形成的复杂机制至关重要。在这项工作中，在Finnis-Sinclair框架内构建了一个新的W-Ni-Fe-He季位，以促进DPT-W复合材料中氦气泡的建模。然后，我们证明了我们的势能与现有的密度泛函理论结果密切地再现了与氦行为相关的重要性质，包括单个氦缺陷的形成和迁移能、氦-溶质对的结合能和小氦缺陷团簇。此外，分子动力学测试证实，我们的潜力导致氦气泡从初始随机分布的氦间隙原子成核。我们的研究表明，在相间边界附近大量的氦积累引起了错配位错的形态变化。新开发的原子间势和模拟数据为氦在W-Ni-Fe重钨合金中的行为提供了有价值的见解。

{"title":"Atomistic insights into the behaviors of helium in W-Ni-Fe alloys with a new quaternary interatomic potential","authors":"Xichuan Liao , Haipan Xiang , Rongyang Qiu , Yangchun Chen , Yong Liu , Ning Gao , Fei Gao , Wangyu Hu , Huiqiu Deng","doi":"10.1016/j.commatsci.2026.114554","DOIUrl":"10.1016/j.commatsci.2026.114554","url":null,"abstract":"<div><div>Helium bubbles are a critical type of radiation damage that can greatly degrade the mechanical properties of W-Ni-Fe ductile-phase toughened tungsten (DPT-W) composites. Understanding helium-induced embrittlement requires a detailed investigation of helium behavior in DPT-W composites. The availability of suitable interatomic potentials would enable atomistic simulations, which are essential to elucidate the complex mechanisms underlying bubble formation. In this work, a new W-Ni-Fe-He quaternary potential, constructed within the Finnis–Sinclair framework, has been developed to facilitate the modeling of helium bubbles in the DPT-W composite. We then demonstrate that our potential closely reproduces available density functional theory results on important properties relevant to the helium behavior, including the formation and migration energies of single helium defects, the binding energies of helium-solute pairs, and small helium-defect clusters. In addition, molecular dynamics tests establish that our potential leads to the nucleation of helium bubbles from an initial random distribution of helium interstitial atoms. Our study shows that considerable helium accumulation near the interphase boundaries induces morphological changes in the misfit dislocations. The newly developed interatomic potential and the simulation data provide valuable insights into the behavior of helium in W-Ni-Fe tungsten heavy alloys.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"266 ","pages":"Article 114554"},"PeriodicalIF":3.3,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186290","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