Pub Date : 2026-10-01Epub Date: 2026-01-09DOI: 10.1016/j.commatsci.2025.114476
Haolun Yuan , Jun Zeng , Jie Zuo , Xin Wang , Dingguo Xu
In this paper, we present a general framework for automating the information extraction process from materials science literature. Our aim is to meet the increasing demand for large-scale databases in both research and engineering. The text mining part consists of three continuous stages: labeling, extraction, and post-processing, which are all powered by large language models (LLMs). Through these successive stages, the framework enables the extraction of material data from both text and tables. It supports the generation of high-quality databases with only a moderate level of prior knowledge about the extraction targets and minimal coding effort, thereby facilitating the rapid development of data-driven models from the ground up. The Framework was applied to high entropy alloys (HEAs) research papers and constructed a comprehensive database of 5393 records encompassing mechanical properties, phase information, and processing histories. Such a database provides a valuable foundation for investigating process–structure–property relationships in alloys, which may support both mechanistic understanding and data-driven design. To assess the quality of the database, we also trained machine learning models to accurately predict phase and yield strength. Our database of HEAs provides a rich resource for future data-driven design of new alloy materials.
{"title":"A general LLM-powered text mining framework: Applied to extract high entropy alloys","authors":"Haolun Yuan , Jun Zeng , Jie Zuo , Xin Wang , Dingguo Xu","doi":"10.1016/j.commatsci.2025.114476","DOIUrl":"10.1016/j.commatsci.2025.114476","url":null,"abstract":"<div><div>In this paper, we present a general framework for automating the information extraction process from materials science literature. Our aim is to meet the increasing demand for large-scale databases in both research and engineering. The text mining part consists of three continuous stages: labeling, extraction, and post-processing, which are all powered by large language models (LLMs). Through these successive stages, the framework enables the extraction of material data from both text and tables. It supports the generation of high-quality databases with only a moderate level of prior knowledge about the extraction targets and minimal coding effort, thereby facilitating the rapid development of data-driven models from the ground up. The Framework was applied to high entropy alloys (HEAs) research papers and constructed a comprehensive database of 5393 records encompassing mechanical properties, phase information, and processing histories. Such a database provides a valuable foundation for investigating process–structure–property relationships in alloys, which may support both mechanistic understanding and data-driven design. To assess the quality of the database, we also trained machine learning models to accurately predict phase and yield strength. Our database of HEAs provides a rich resource for future data-driven design of new alloy materials.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114476"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145923415","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}
Pub Date : 2026-10-01Epub Date: 2026-01-03DOI: 10.1016/j.commatsci.2025.114478
Shota Arai, Takashi Yoshidome
This work presents a novel algorithm for generating porous structures as an alternative to the PoreSpy program suite. Unlike PoreSpy, which often produces structures whose porosity deviates from the target value, our proposed algorithm generates structures whose porosity closely matches the specified input, within a defined error margin. Furthermore, parallel computation enables efficient generation of large-scale structures, while memory usage is reduced compared to PoreSpy. To evaluate performance, structures were generated using both PoreSpy and the proposed method with parameters corresponding to X-ray ptychography experiments. The porosity mismatch in PoreSpy led to a relative error exceeding 20 % in the computed gas diffusion coefficients, whereas our method reproduced the experimental values within 5 %. These results demonstrate that the proposed method provides an efficient, high-precision approach for generating porous structures and supports reliable prediction of material properties. The program called “PorousGen” is publicly available under the MIT License from https://github.com/YoshidomeGroup-Hydration/PorousGen.
{"title":"PorousGen: An efficient algorithm for generating porous structures with accurate porosity and uniform density distribution","authors":"Shota Arai, Takashi Yoshidome","doi":"10.1016/j.commatsci.2025.114478","DOIUrl":"10.1016/j.commatsci.2025.114478","url":null,"abstract":"<div><div>This work presents a novel algorithm for generating porous structures as an alternative to the <em>PoreSpy</em> program suite. Unlike <em>PoreSpy</em>, which often produces structures whose porosity deviates from the target value, our proposed algorithm generates structures whose porosity closely matches the specified input, within a defined error margin. Furthermore, parallel computation enables efficient generation of large-scale structures, while memory usage is reduced compared to <em>PoreSpy</em>. To evaluate performance, structures were generated using both <em>PoreSpy</em> and the proposed method with parameters corresponding to X-ray ptychography experiments. The porosity mismatch in <em>PoreSpy</em> led to a relative error exceeding 20 % in the computed gas diffusion coefficients, whereas our method reproduced the experimental values within 5 %. These results demonstrate that the proposed method provides an efficient, high-precision approach for generating porous structures and supports reliable prediction of material properties. The program called “PorousGen” is publicly available under the MIT License from <span><span>https://github.com/YoshidomeGroup-Hydration/PorousGen</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114478"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145923417","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}
Pub Date : 2026-10-01Epub Date: 2026-01-05DOI: 10.1016/j.commatsci.2025.114475
Elaheh Kazemi-Khasragh , Rocío Mercado , Carlos Gonzalez , Maciej Haranczyk
Copolymers are highly versatile materials with a vast range of possible chemical compositions. By using computational methods for property prediction, the design of copolymers can be accelerated, allowing for the prioritization of candidates with favorable properties. In this study, we utilized two distinct representations of molecular ensembles to predict the seven different physical polymer properties copolymers using machine learning: we used a random forest (RF) model to predict polymer properties from molecular descriptors, and a graph neural network (GNN) to predict the same properties from 2D polymer graphs under both a single- and multi-task setting. To train and evaluate the models, we constructed a data set from molecular dynamic simulations for 140 binary copolymers with varying monomer compositions and configurations. Our results demonstrate that descriptors-based RFs excel at predicting density and specific heat capacities at constant pressure (Cp) and volume (Cv) because these properties are strongly tied to specific molecular features captured by molecular descriptors. In contrast, graph representations better predict expansion coefficients (, ) and bulk modulus (K), which depend more on complex structural interactions better captured by graph-based models. This study underscores the importance of choosing appropriate representations for predicting molecular properties. Our findings demonstrate how machine learning models can expedite copolymer discovery with learnable structure–property relationships, streamlining polymer design and advancing the development of high-performance materials for diverse applications.
{"title":"Descriptor and graph-based molecular representations in prediction of copolymer properties using machine learning","authors":"Elaheh Kazemi-Khasragh , Rocío Mercado , Carlos Gonzalez , Maciej Haranczyk","doi":"10.1016/j.commatsci.2025.114475","DOIUrl":"10.1016/j.commatsci.2025.114475","url":null,"abstract":"<div><div>Copolymers are highly versatile materials with a vast range of possible chemical compositions. By using computational methods for property prediction, the design of copolymers can be accelerated, allowing for the prioritization of candidates with favorable properties. In this study, we utilized two distinct representations of molecular ensembles to predict the seven different physical polymer properties copolymers using machine learning: we used a random forest (RF) model to predict polymer properties from molecular descriptors, and a graph neural network (GNN) to predict the same properties from 2D polymer graphs under both a single- and multi-task setting. To train and evaluate the models, we constructed a data set from molecular dynamic simulations for 140 binary copolymers with varying monomer compositions and configurations. Our results demonstrate that descriptors-based RFs excel at predicting density and specific heat capacities at constant pressure (C<sub>p</sub>) and volume (C<sub>v</sub>) because these properties are strongly tied to specific molecular features captured by molecular descriptors. In contrast, graph representations better predict expansion coefficients (<span><math><mi>γ</mi></math></span>, <span><math><mi>α</mi></math></span>) and bulk modulus (K), which depend more on complex structural interactions better captured by graph-based models. This study underscores the importance of choosing appropriate representations for predicting molecular properties. Our findings demonstrate how machine learning models can expedite copolymer discovery with learnable structure–property relationships, streamlining polymer design and advancing the development of high-performance materials for diverse applications.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114475"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145923041","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}
Pub Date : 2026-10-01Epub Date: 2026-01-07DOI: 10.1016/j.commatsci.2025.114467
Ákos Szabó
This study investigates the ring-opening multibranching polymerization (ROMBP) of glycidol using stochastic simulation. We analyzed the graph diameter of virtually generated macromolecules and examined how this parameter, denoted as dmathn, responds to variations in the initial composition of protected (monofunctional) and unprotected (bifunctional) monomers. The results uncover a distinct mathematical relationship between dmathn and the average degree of branching (DBₐᵥ). It was demonstrated that dmathn serves as a powerful indicator of the topological features of hyperbranched polymers obtained under different feed conditions. Unlike DBₐᵥ, dmathn more accurately reflects changes in macromolecular size. These findings establish dmathn as a reliable topological descriptor, offering new insights into the complex structure-property relationships of hyperbranched polymers.
{"title":"Graph diameter as a topological descriptor for hyperbranched polymers: insights from stochastic simulation of ring-opening multibranching polymerization of glycidol","authors":"Ákos Szabó","doi":"10.1016/j.commatsci.2025.114467","DOIUrl":"10.1016/j.commatsci.2025.114467","url":null,"abstract":"<div><div>This study investigates the ring-opening multibranching polymerization (ROMBP) of glycidol using stochastic simulation. We analyzed the graph diameter of virtually generated macromolecules and examined how this parameter, denoted as <em>d</em><sup>math</sup><sub>n</sub>, responds to variations in the initial composition of protected (monofunctional) and unprotected (bifunctional) monomers. The results uncover a distinct mathematical relationship between <em>d</em><sup>math</sup><sub>n</sub> and the average degree of branching (<em>DB</em>ₐᵥ). It was demonstrated that <em>d</em><sup>math</sup><sub>n</sub> serves as a powerful indicator of the topological features of hyperbranched polymers obtained under different feed conditions. Unlike <em>DB</em>ₐᵥ, <em>d</em><sup>math</sup><sub>n</sub> more accurately reflects changes in macromolecular size. These findings establish <em>d</em><sup>math</sup><sub>n</sub> as a reliable topological descriptor, offering new insights into the complex structure-property relationships of hyperbranched polymers.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114467"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145923416","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}
Pub Date : 2026-10-01Epub Date: 2026-01-15DOI: 10.1016/j.commatsci.2026.114503
Chaitanya Bhave, Somayajulu L.N. Dhulipala, Mathew Swisher, Jacob A. Hirschhorn, Ryan Terrence Sweet, Stephen R. Novascone
TRistructural ISOtropic (TRISO) particle fuel relies on a silicon carbide (SiC) layer as the primary structural material and barrier to metallic fission products (FPs) release. Accurate prediction of palladium (Pd) transport and penetration into the SiC is therefore critical for qualifying TRISO fuels for advanced reactors. The empirical correlation for Pd penetration in BISON is derived from historical particle-fuel data, but it cannot explain the large scatter in the experimental data that arises from varying experimental conditions. To aid fuel qualification, we previously developed a mechanistic reduced order model (ROM) using BISON that resolves these dependencies (Bhave et al., 2025). In this work we built on that mechanistic ROM, validated it, and quantified its uncertainty using Bayesian uncertainty quantification (UQ). We calibrated against a suite of in-pile and out-of-pile experiments spanning particle compositions, geometries, and operating conditions, and benchmarked the mechanistic ROM against the empirical correlation. We used Bayesian UQ to identify influential parameters and calibrate them to data, which yielded predictive intervals. Results show that while the empirical correlation can be tuned to fit a single experiment type, it transfers poorly; the mechanistic ROM sustains accuracy with credible uncertainty across disparate conditions. This process demonstrates a practical path — via Bayesian UQ applied to mechanistic ROMs — to leverage single-effect experiments for inferring in-reactor behavior and supporting TRISO fuel qualification.
三结构各向同性（TRISO）粒子燃料依靠碳化硅（SiC）层作为主要结构材料和金属裂变产物（FPs）释放的屏障。因此，准确预测钯（Pd）在碳化硅中的传输和渗透对于先进反应堆的TRISO燃料的资格至关重要。BISON中Pd渗透的经验相关性来源于历史颗粒-燃料数据，但它不能解释实验数据中由于不同实验条件而产生的大分散。为了帮助燃料鉴定，我们之前使用BISON开发了一种机制降order模型（ROM）来解决这些依赖关系（Bhave et al., 2025）。在这项工作中，我们建立了机械ROM，验证了它，并使用贝叶斯不确定性量化（UQ）量化了它的不确定性。我们根据一套桩内和桩外实验进行了校准，涵盖了颗粒组成、几何形状和操作条件，并根据经验相关性对机械ROM进行了基准测试。我们使用贝叶斯UQ来识别有影响的参数，并将其校准为数据，从而产生预测区间。结果表明，虽然经验相关性可以调整到适合单一实验类型，但它的转移性很差；机械式只读存储器在不同的条件下保持具有可靠的不确定性的准确性。该过程展示了一种实用的途径——通过将贝叶斯UQ应用于机械rom——利用单效应实验来推断反应堆内行为并支持TRISO燃料鉴定。
{"title":"Bayesian discovery of optimal reduced order models from mechanistic and experimental data: A case study of Pd penetration in TRISO fuels using BISON","authors":"Chaitanya Bhave, Somayajulu L.N. Dhulipala, Mathew Swisher, Jacob A. Hirschhorn, Ryan Terrence Sweet, Stephen R. Novascone","doi":"10.1016/j.commatsci.2026.114503","DOIUrl":"10.1016/j.commatsci.2026.114503","url":null,"abstract":"<div><div>TRistructural ISOtropic (TRISO) particle fuel relies on a silicon carbide (SiC) layer as the primary structural material and barrier to metallic fission products (FPs) release. Accurate prediction of palladium (Pd) transport and penetration into the SiC is therefore critical for qualifying TRISO fuels for advanced reactors. The empirical correlation for Pd penetration in BISON is derived from historical particle-fuel data, but it cannot explain the large scatter in the experimental data that arises from varying experimental conditions. To aid fuel qualification, we previously developed a mechanistic reduced order model (ROM) using BISON that resolves these dependencies (Bhave et al., 2025). In this work we built on that mechanistic ROM, validated it, and quantified its uncertainty using Bayesian uncertainty quantification (UQ). We calibrated against a suite of in-pile and out-of-pile experiments spanning particle compositions, geometries, and operating conditions, and benchmarked the mechanistic ROM against the empirical correlation. We used Bayesian UQ to identify influential parameters and calibrate them to data, which yielded predictive intervals. Results show that while the empirical correlation can be tuned to fit a single experiment type, it transfers poorly; the mechanistic ROM sustains accuracy with credible uncertainty across disparate conditions. This process demonstrates a practical path — via Bayesian UQ applied to mechanistic ROMs — to leverage single-effect experiments for inferring in-reactor behavior and supporting TRISO fuel qualification.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114503"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974140","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}
Pub Date : 2026-10-01Epub Date: 2026-01-12DOI: 10.1016/j.commatsci.2026.114491
Sabrina Weber , Benedikt Prifling , Ravi Kumar Jeela , Andreas Prahs , Daniel Schneider , Britta Nestler , Volker Schmidt
Solid-oxide fuel cells (SOFCs) are a promising energy conversion technology, offering a low environmental impact, low costs and high flexibility regarding the choice of the fuel. However, electrochemical performance of SOFCs decreases with time as a result of complex structural aging mechanisms of their anodes that are not yet fully understood. An option to quantitatively investigate this aging behavior could be tomographic imaging of the 3D microstructure of SOFC anodes for different aging durations, which is expensive and time-consuming. To overcome this issue, physics-based aging simulations resolving the 3D microstructural evolution can be exploited, which use tomographic image data of pristine SOFC anodes consisting of nickel, gadolinium-doped ceria (GDC) and pore space, as initial state. This microstructure simulation method is based on a grand-chemical potential multi-phase-field approach including surface diffusion. Computations conducted with the simulation framework are capable to predict the coarsening of the multiphase polycrystalline electrode. A promising approach to further accelerate the quantitative investigation of SOFC degradation is to combine physics-based aging simulation with data-driven stochastic 3D microstructure modeling, which is typically less computationally intensive compared to phase-field simulations. More precisely, an excursion set model based on Gaussian random fields is used to characterize the 3D microstructure of SOFC anodes by means of a small number of interpretable model parameters. Moreover, the evolution of the parameter vector of the calibrated stochastic 3D model over time is modeled by analytical functions that make fast predictive simulations possible. The prediction robustness is investigated by first assuming that the evolution of the 3D microstructure is known up to a certain point in time. Then, in a second step, the 3D microstructure of SOFC anodes is predicted for further future points in time and, through geometrical descriptors, compared with the results of physics-based aging simulation.
{"title":"A time-continuous approach to analyzing anode aging in solid-oxide fuel cells via stochastic 3D microstructure modeling and physics-based simulations","authors":"Sabrina Weber , Benedikt Prifling , Ravi Kumar Jeela , Andreas Prahs , Daniel Schneider , Britta Nestler , Volker Schmidt","doi":"10.1016/j.commatsci.2026.114491","DOIUrl":"10.1016/j.commatsci.2026.114491","url":null,"abstract":"<div><div>Solid-oxide fuel cells (SOFCs) are a promising energy conversion technology, offering a low environmental impact, low costs and high flexibility regarding the choice of the fuel. However, electrochemical performance of SOFCs decreases with time as a result of complex structural aging mechanisms of their anodes that are not yet fully understood. An option to quantitatively investigate this aging behavior could be tomographic imaging of the 3D microstructure of SOFC anodes for different aging durations, which is expensive and time-consuming. To overcome this issue, physics-based aging simulations resolving the 3D microstructural evolution can be exploited, which use tomographic image data of pristine SOFC anodes consisting of nickel, gadolinium-doped ceria (GDC) and pore space, as initial state. This microstructure simulation method is based on a grand-chemical potential multi-phase-field approach including surface diffusion. Computations conducted with the simulation framework are capable to predict the coarsening of the multiphase polycrystalline electrode. A promising approach to further accelerate the quantitative investigation of SOFC degradation is to combine physics-based aging simulation with data-driven stochastic 3D microstructure modeling, which is typically less computationally intensive compared to phase-field simulations. More precisely, an excursion set model based on Gaussian random fields is used to characterize the 3D microstructure of SOFC anodes by means of a small number of interpretable model parameters. Moreover, the evolution of the parameter vector of the calibrated stochastic 3D model over time is modeled by analytical functions that make fast predictive simulations possible. The prediction robustness is investigated by first assuming that the evolution of the 3D microstructure is known up to a certain point in time. Then, in a second step, the 3D microstructure of SOFC anodes is predicted for further future points in time and, through geometrical descriptors, compared with the results of physics-based aging simulation.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114491"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974220","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}
Pub Date : 2026-10-01Epub Date: 2026-01-12DOI: 10.1016/j.commatsci.2026.114509
Shi Chen , Aijun Hong , Junming Liu
Tackling the intertwined and often contradictory nature of thermoelectric (TE) transport parameters remains a central challenge and opportunity in TE research. Herein, we designed a one-dimensional (1D) stacked material Ta2Pd3Te8 to decouple the strong coupling of TE parameters. Its thermal, mechanical, and dynamic stabilities were confirmed by molecular dynamics simulations, elastic constants, and phonon spectrum calculations, respectively. Combined first-principles calculations with phonon and electron Boltzmann transport equations reveal that Ta2Pd3Te8 is a compelling candidate for TE applications, owing to its high power factor () and low lattice thermal conductivity, both resulting from strong anisotropic characteristics. By rigidly widening the band gap to suppress the bipolar effect, we further decoupled the TE parameters, achieving a significantly enhanced value of up to 1.20 along the -axis. These findings not only stimulate further theoretical investigations into one-dimensional van der Waals stacked TE materials but also provide valuable insights for the experimental advancement of high-performance TE materials.
{"title":"High power factor and low thermal conductivity from strong anisotropy in 1D van der Waals stacked Ta2Pd3Te8 crystal","authors":"Shi Chen , Aijun Hong , Junming Liu","doi":"10.1016/j.commatsci.2026.114509","DOIUrl":"10.1016/j.commatsci.2026.114509","url":null,"abstract":"<div><div>Tackling the intertwined and often contradictory nature of thermoelectric (TE) transport parameters remains a central challenge and opportunity in TE research. Herein, we designed a one-dimensional (1D) stacked material Ta<sub>2</sub>Pd<sub>3</sub>Te<sub>8</sub> to decouple the strong coupling of TE parameters. Its thermal, mechanical, and dynamic stabilities were confirmed by molecular dynamics simulations, elastic constants, and phonon spectrum calculations, respectively. Combined first-principles calculations with phonon and electron Boltzmann transport equations reveal that Ta<sub>2</sub>Pd<sub>3</sub>Te<sub>8</sub> is a compelling candidate for TE applications, owing to its high power factor (<span><math><mrow><mi>P</mi><mi>F</mi></mrow></math></span>) and low lattice thermal conductivity, both resulting from strong anisotropic characteristics. By rigidly widening the band gap to suppress the bipolar effect, we further decoupled the TE parameters, achieving a significantly enhanced <span><math><mrow><mi>Z</mi><mi>T</mi></mrow></math></span> value of up to 1.20 along the <span><math><mi>a</mi></math></span>-axis. These findings not only stimulate further theoretical investigations into one-dimensional van der Waals stacked TE materials but also provide valuable insights for the experimental advancement of high-performance TE materials.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114509"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145974221","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}
Pub Date : 2026-10-01Epub Date: 2026-01-09DOI: 10.1016/j.commatsci.2025.114474
Stephen P. Fluckey, Christopher N. Sterling, Blas P. Uberuaga, Xiang-Yang Liu
In materials, point defects often control or modify functional properties. To predict the performance of materials intended for application in optoelectronic devices, it is imperative to understand the properties of those point defects. For the first time, all six intrinsic defects of GaAs, a key optoelectronics material, and their charge transition levels are calculated using density functional theory with the HSE06 functional. For comparison, both PBE and calculations are also carried out. The HSE06 results are found to be in better agreement with experimental data than previous calculations. The importance of using the exact electron exchange present in hybrid functionals and larger supercells to accurately determine defect levels and ground state defect configurations is demonstrated.
{"title":"Point defect energetics in gallium arsenide, a comprehensive density functional theory study","authors":"Stephen P. Fluckey, Christopher N. Sterling, Blas P. Uberuaga, Xiang-Yang Liu","doi":"10.1016/j.commatsci.2025.114474","DOIUrl":"10.1016/j.commatsci.2025.114474","url":null,"abstract":"<div><div>In materials, point defects often control or modify functional properties. To predict the performance of materials intended for application in optoelectronic devices, it is imperative to understand the properties of those point defects. For the first time, all six intrinsic defects of GaAs, a key optoelectronics material, and their charge transition levels are calculated using density functional theory with the HSE06 functional. For comparison, both PBE and <span><math><mrow><msup><mrow><mi>r</mi></mrow><mrow><mn>2</mn></mrow></msup><mi>SCAN</mi></mrow></math></span> calculations are also carried out. The HSE06 results are found to be in better agreement with experimental data than previous calculations. The importance of using the exact electron exchange present in hybrid functionals and larger supercells to accurately determine defect levels and ground state defect configurations is demonstrated.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114474"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145923039","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}
Pub Date : 2026-10-01Epub Date: 2026-01-02DOI: 10.1016/j.commatsci.2025.114458
Alex Keilmann , Claudia Redenbach , François Willot
In fields such as material design or biomedicine, fiber materials play an important role. Fiber simulations, also called digital twins, provide a basis for testing and optimizing the material’s physical behavior digitally. Inter-fiber contacts can influence the thermal and mechanical behavior of a fiber system; to our knowledge, however, there exist no parametric fiber models allowing for explicit modeling of the number of inter-fiber contacts. Therefore, this paper proposes an extension of the iterative force-biased fiber packing by Altendorf & Jeulin. In this extension, we model the inter-fiber contacts explicitly and add another force to the force-biased packing to increase the number of contacts. We successfully validate the packing with respect to its parameter accuracy. Moreover, we show that the extension indeed increases the number of contacts, even exceeding theoretical values. Hence, this packing scheme has the potential to achieve higher accuracy in physical simulations.
{"title":"Increasing inter-fiber contact in the Altendorf-Jeulin model","authors":"Alex Keilmann , Claudia Redenbach , François Willot","doi":"10.1016/j.commatsci.2025.114458","DOIUrl":"10.1016/j.commatsci.2025.114458","url":null,"abstract":"<div><div>In fields such as material design or biomedicine, fiber materials play an important role. Fiber simulations, also called digital twins, provide a basis for testing and optimizing the material’s physical behavior digitally. Inter-fiber contacts can influence the thermal and mechanical behavior of a fiber system; to our knowledge, however, there exist no parametric fiber models allowing for explicit modeling of the number of inter-fiber contacts. Therefore, this paper proposes an extension of the iterative force-biased fiber packing by Altendorf & Jeulin. In this extension, we model the inter-fiber contacts explicitly and add another force to the force-biased packing to increase the number of contacts. We successfully validate the packing with respect to its parameter accuracy. Moreover, we show that the extension indeed increases the number of contacts, even exceeding theoretical values. Hence, this packing scheme has the potential to achieve higher accuracy in physical simulations.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"264 ","pages":"Article 114458"},"PeriodicalIF":3.3,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145876717","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}
Pub Date : 2026-03-10Epub Date: 2026-02-16DOI: 10.1016/j.commatsci.2026.114593
Xiongwei He , Fan-Shun Meng , Yanjing Su , Lijie Qiao , Shigenobu Ogata , Lei Gao
Hydrogen is known to reduce grain boundary (GB) toughness, but how it couples to plasticity remains debated. Using a high accuracy neural network interatomic potential, we combine grand canonical Monte Carlo and molecular dynamics (GCMC-MD) to simulate in-situ H charging and tensile loading of Σ9{10} twist GB (Σ9 GB) in α-Fe across H concentrations and strain rates. Hydrogen segregates at GB, reduces grain boundary energy, lowers barrier for dislocation emission, and advances GB mediated plasticity. Increasing the , yielding occurs earlier, less elastic energy accumulates and the peak dislocation density declines. Meanwhile, hydrogen lowers the effective surface energy of cavity/GB facets, promoting premature cavity nucleation and growth. As a result, the coupled outcomes of increasing : earlier yielding but diminished plastic accommodation and cavity generation boost intergranular fracture at reduced toughness. Lower strain rates elevate boundary and accentuate these trends, clarifying a pathway by which hydrogen-accelerated GB-mediated plasticity ultimately undermines GB cohesion and toughness.
{"title":"Unravelling the interplay between hydrogen and grain boundary of α-Fe under different concentration and strain rate via neural network interatomic potential","authors":"Xiongwei He , Fan-Shun Meng , Yanjing Su , Lijie Qiao , Shigenobu Ogata , Lei Gao","doi":"10.1016/j.commatsci.2026.114593","DOIUrl":"10.1016/j.commatsci.2026.114593","url":null,"abstract":"<div><div>Hydrogen is known to reduce grain boundary (GB) toughness, but how it couples to plasticity remains debated. Using a high accuracy neural network interatomic potential, we combine grand canonical Monte Carlo and molecular dynamics (GCMC-MD) to simulate in-situ H charging and tensile loading of Σ9<strong>{</strong>1<span><math><mover><mn>1</mn><mo>¯</mo></mover></math></span>0<strong>}</strong> twist GB (Σ9 GB) in α-Fe across H concentrations <span><math><msub><mi>C</mi><mi>H</mi></msub></math></span> and strain rates. Hydrogen segregates at GB, reduces grain boundary energy, lowers barrier for dislocation emission, and advances GB mediated plasticity. Increasing the <span><math><msub><mi>C</mi><mi>H</mi></msub></math></span>, yielding occurs earlier, less elastic energy accumulates and the peak dislocation density declines. Meanwhile, hydrogen lowers the effective surface energy of cavity/GB facets, promoting premature cavity nucleation and growth. As a result, the coupled outcomes of increasing <span><math><msub><mi>C</mi><mi>H</mi></msub></math></span>: earlier yielding but diminished plastic accommodation and cavity generation boost intergranular fracture at reduced toughness. Lower strain rates elevate boundary <span><math><msub><mi>C</mi><mi>H</mi></msub></math></span> and accentuate these trends, clarifying a pathway by which hydrogen-accelerated GB-mediated plasticity ultimately undermines GB cohesion and toughness.</div></div>","PeriodicalId":10650,"journal":{"name":"Computational Materials Science","volume":"267 ","pages":"Article 114593"},"PeriodicalIF":3.3,"publicationDate":"2026-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147403391","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号