npj Computational Materials最新文献

Leveraging unlabeled SEM datasets with self-supervised learning for enhanced particle segmentation. 利用未标记的SEM数据集与自我监督学习增强粒子分割。

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-22 DOI: 10.1038/s41524-025-01802-3

Luca Rettenberger,Nathan J Szymanski,Andrea Giunto,Olympia Dartsi,Anubhav Jain,Gerbrand Ceder,Veit Hagenmeyer,Markus Reischl

{"title":"Leveraging unlabeled SEM datasets with self-supervised learning for enhanced particle segmentation.","authors":"Luca Rettenberger,Nathan J Szymanski,Andrea Giunto,Olympia Dartsi,Anubhav Jain,Gerbrand Ceder,Veit Hagenmeyer,Markus Reischl","doi":"10.1038/s41524-025-01802-3","DOIUrl":"https://doi.org/10.1038/s41524-025-01802-3","url":null,"abstract":"Scanning Electron Microscopes (SEMs) are widely used in experimental science laboratories, often requiring cumbersome and repetitive user analysis. Automating SEM image analysis processes is highly desirable to address this challenge. In particle sample analysis, Machine Learning (ML) has emerged as the most effective approach for particle segmentation. However, the time-intensive process of manually annotating thousands of SEM images limits the applicability of supervised learning approaches. Self-Supervised Learning (SSL) offers a promising alternative by enabling knowledge extraction from raw, unlabeled data. This study presents a framework for evaluating SSL techniques in SEM image analysis, focusing on novel methods leveraging the ConvNeXtV2 architecture for particle detection. A dataset comprising 25,000 SEM images is curated to benchmark these proposed SSL methods. The results demonstrate that ConvNeXtV2 models, with varying parameter counts, consistently outperform other techniques in particle detection across different length scales, achieving up to a 34% reduction in relative error compared to established SSL methods. Furthermore, an ablation study explores the relationship between dataset size and SSL performance, providing actionable insights for practitioners regarding model selection and resource efficiency. This research advances the integration of SSL into autonomous analysis pipelines and supports its application in accelerating materials science discovery.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"100 1","pages":"289"},"PeriodicalIF":9.7,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145134577","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Learning non-local molecular interactions via equivariant local representations and charge equilibration 通过等变局部表示和电荷平衡学习非局部分子相互作用

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-16 DOI: 10.1038/s41524-025-01790-4

Paul Fuchs, Michał Sanocki, Julija Zavadlav

{"title":"Learning non-local molecular interactions via equivariant local representations and charge equilibration","authors":"Paul Fuchs, Michał Sanocki, Julija Zavadlav","doi":"10.1038/s41524-025-01790-4","DOIUrl":"https://doi.org/10.1038/s41524-025-01790-4","url":null,"abstract":"Graph Neural Network (GNN) potentials relying on chemical locality offer near-quantum mechanical accuracy at significantly reduced computational costs. Message-passing GNNs model interactions beyond their immediate neighborhood by propagating local information between neighboring particles while remaining effectively local. However, locality precludes modeling long-range effects critical to many real-world systems, such as charge transfer, electrostatic interactions, and dispersion effects. In this work, we propose the Charge Equilibration Layer for Long-range Interactions (CELLI) to address the challenge of efficiently modeling non-local interactions. This novel architecture generalizes the classical charge equilibration (Qeq) method to a model-agnostic building block for modern equivariant GNN potentials. Therefore, CELLI extends the capability of GNNs to model long-range interactions while providing high interpretability through explicitly modeled charges. On benchmark systems, CELLI achieves state-of-the-art results for strictly local models. CELLI generalizes to diverse datasets and large structures while providing high computational efficiency and robust predictions.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"38 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145067870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Pushing charge equilibration-based machine learning potentials to their limits 将基于电荷平衡的机器学习潜力推向极限

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-16 DOI: 10.1038/s41524-025-01791-3

Martin Vondrák, Karsten Reuter, Johannes T. Margraf

{"title":"Pushing charge equilibration-based machine learning potentials to their limits","authors":"Martin Vondrák, Karsten Reuter, Johannes T. Margraf","doi":"10.1038/s41524-025-01791-3","DOIUrl":"https://doi.org/10.1038/s41524-025-01791-3","url":null,"abstract":"Machine learning (ML) has demonstrated its potential in atomistic simulations to bridge the gap between accurate first-principles methods and computationally efficient empirical potentials. This is achieved by learning mappings between a system’s structure and its physical properties. State-of-the-art models for potential energy surfaces typically represent chemical structures through (semi-)local atomic environments. However, this approach neglects long-range interactions (most notably electrostatics) and non-local phenomena such as charge transfer, leading to significant errors in the description of molecules or materials in polar anisotropic environments. To address these challenges, ML frameworks that predict self-consistent charge distributions in atomistic systems using the Charge Equilibration (QEq) method are currently popular. In this approach, atomic charges are derived from an electrostatic energy expression that incorporates environment-dependent atomic electronegativities. Herein, we explore the limits of this concept at the example of the previously reported Kernel Charge Equilibration (kQEq) approach, combined with local short-ranged potentials. To this end we consider prototypical systems with varying total charge states and applied electric fields. We find that charge equilibration-based models perform well in most situations. However, we also find that some pathologies of conventional QEq carry over to the ML variants in the form of spurious charge transfer and overpolarization in the presence of static electric fields. This indicates a need for new methodological developments.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"9 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145068152","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

High-throughput materials exploration system for the anomalous Hall effect using combinatorial experiments and machine learning 利用组合实验和机器学习的异常霍尔效应的高通量材料探索系统

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-03 DOI: 10.1038/s41524-025-01757-5

Ryo Toyama, Yuma Iwasaki, Prabhanjan D. Kulkarni, Hirofumi Suto, Tomoya Nakatani, Yuya Sakuraba

{"title":"High-throughput materials exploration system for the anomalous Hall effect using combinatorial experiments and machine learning","authors":"Ryo Toyama, Yuma Iwasaki, Prabhanjan D. Kulkarni, Hirofumi Suto, Tomoya Nakatani, Yuya Sakuraba","doi":"10.1038/s41524-025-01757-5","DOIUrl":"https://doi.org/10.1038/s41524-025-01757-5","url":null,"abstract":"The development of new materials exhibiting large anomalous Hall effect (AHE) is essential for realizing highly efficient spintronic devices. However, this development has been a time-consuming process due to the combinatorial explosion for multielement systems and limited experimental throughput. In this study, we identify new materials exhibiting large AHE in heavy-metal-substituted Fe-based alloys using a high-throughput materials exploration method that combines deposition of composition-spread films using combinatorial sputtering, photoresist-free facile multiple-device fabrication using laser patterning, simultaneous AHE measurement of multiple devices using a customized multichannel probe, and prediction of candidate materials using machine learning. Based on experimental AHE data on Fe-based binary system alloyed with various single heavy metals, we perform machine learning analysis to predict the Fe-based ternary system containing two heavy metals for larger AHE. We experimentally confirm larger AHE in the predicted Fe–Ir–Pt system. Using scaling analysis, we reveal that the enhancement of AHE originates from the extrinsic contribution.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"8 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144930726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Machine learning augmented design of 2D magnet with planar cyclo-tetranitrogen: ambient thermal stability from quantum to mesoscale 平面环四氮二维磁体的机器学习增强设计：从量子到中尺度的环境热稳定性

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-02 DOI: 10.1038/s41524-025-01763-7

Dong Fan, Ke Zheng, Hongfang Li, Junjie He, Pengbo Lyu

{"title":"Machine learning augmented design of 2D magnet with planar cyclo-tetranitrogen: ambient thermal stability from quantum to mesoscale","authors":"Dong Fan, Ke Zheng, Hongfang Li, Junjie He, Pengbo Lyu","doi":"10.1038/s41524-025-01763-7","DOIUrl":"https://doi.org/10.1038/s41524-025-01763-7","url":null,"abstract":"Synthesizing polynitrogen compounds that remain stable at ambient conditions is particularly challenging because species beyond the N ≡ N triple bond are inherently unstable. In this study, we combine first-principles calculations with a machine-learning potential (MLP) to investigate the ambient stability of planar cyclo-N4 units embedded in a two-dimensional t-FeN4 monolayer. Our results show that strong Fe–N coordination inhibits N ≡ N reformation, enabling the square cyclo-N4 motif to remain dynamically stable and covalently bonded without high-pressure synthesis. Furthermore, this structure exhibits tunable magnetic anisotropy and a Néel temperature above 600 K, indicating potential for room-temperature spintronic applications. The MLP also enables the simulation of systems comprising over 100,000 atoms, including periodic sheets, nanoribbons, nanomatrices and nanosheets, revealing their structural integrity under thermal fluctuations. These results demonstrate that two-dimensional confinement provides a promising route to stabilize exotic nitrogen topologies, linking quantum-mechanical accuracy with mesoscale modelling for future spin-based technologies.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"141 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144928268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Generalized modeling of carbon film deposition growth via hybrid MD/MC simulations with machine-learning potentials 基于机器学习势的混合MD/MC模拟碳膜沉积生长的广义建模

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-02 DOI: 10.1038/s41524-025-01781-5

Yutao Liu, Tinghong Gao, Qingquan Xiao, Yunjun Ruan, Qian Chen, Bei Wang, Jin Huang

{"title":"Generalized modeling of carbon film deposition growth via hybrid MD/MC simulations with machine-learning potentials","authors":"Yutao Liu, Tinghong Gao, Qingquan Xiao, Yunjun Ruan, Qian Chen, Bei Wang, Jin Huang","doi":"10.1038/s41524-025-01781-5","DOIUrl":"https://doi.org/10.1038/s41524-025-01781-5","url":null,"abstract":"Theoretical investigations into the controlled growth of carbon films are essential for guiding the experimental fabrication of carbon-based devices. However, accurately simulating the deposition process remains a significant challenge. In this work, we developed an active learning workflow to construct a machine learning-based neuroevolution potential (NEP) for investigating carbon atoms deposition growth on various substrates. By integrating molecular dynamics and time-stamped force-biased Monte Carlo simulations, we studied the growth of amorphous carbon films on Si(111) and found that deposition energy strongly influenced bonding topology and film morphology. The NEP reliably captured the surface diffusion of carbon atoms, the formation of carbon chains and rings. We revealed a new growth mechanism of adhesion-driven growth at low energies and peening-induced densification at high energies of carbon atoms on Si(111) substrates. To evaluate the transferability of fitting workflow, we extended the NEP to simulate carbon deposition on Cu(111) and Al2O3(0001) surface. Simulation results demonstrate that the NEP can reproduce the subprocesses of graphene formation during carbon growth on the Cu(111) substrate. In contrast, only disordered carbon chains are observed on the Al2O3(0001) substrate. This work provides atomistic insights into the growth mechanisms of carbon films on representative substrates and establishes a robust computational framework for synthesis of diverse carbon nanostructures.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"14 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144928267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Learning metal microstructural heterogeneity through spatial mapping of diffraction latent space features 通过衍射潜空间特征的空间映射了解金属微观结构的非均质性

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-09-01 DOI: 10.1038/s41524-025-01770-8

Mathieu Calvat, Chris Bean, Dhruv Anjaria, Hyoungryul Park, Haoren Wang, Kenneth Vecchio, J. C. Stinville

{"title":"Learning metal microstructural heterogeneity through spatial mapping of diffraction latent space features","authors":"Mathieu Calvat, Chris Bean, Dhruv Anjaria, Hyoungryul Park, Haoren Wang, Kenneth Vecchio, J. C. Stinville","doi":"10.1038/s41524-025-01770-8","DOIUrl":"https://doi.org/10.1038/s41524-025-01770-8","url":null,"abstract":"To leverage advancements in machine learning for metallic materials design and property prediction, it is crucial to develop a data-reduced representation of metal microstructures that surpasses the limitations of current physics-based discrete microstructure descriptors. This need is particularly relevant for metallic materials processed through additive manufacturing, which exhibit complex hierarchical microstructures that cannot be adequately described using the conventional metrics typically applied to wrought materials. Furthermore, capturing the spatial heterogeneity of microstructures at the different scales is necessary within such framework to accurately predict their properties. To address these challenges, we propose the physical spatial mapping of metal diffraction latent space features. This approach integrates (i) point diffraction data encoding via variational autoencoders or contrastive learning and (ii) the physical mapping of the encoded values. Together, these steps offer a method to comprehensively describe metal microstructures. We demonstrate this approach on a wrought and additively manufactured alloy, showing that it effectively encodes microstructural information and enables direct identification of microstructural heterogeneity not directly possible by physics-based models. This data-reduced microstructure representation opens the application of machine learning models in accelerating metallic material design and accurately predicting their properties.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"31 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144928647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Chemical foundation model-guided design of high ionic conductivity electrolyte formulations 化学基础模型指导高离子电导率电解质配方的设计

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-08-30 DOI: 10.1038/s41524-025-01774-4

Murtaza Zohair, Vidushi Sharma, Eduardo A. Soares, Khanh Nguyen, Maxwell Giammona, Linda Sundberg, Andy Tek, Emilio A. V. Vital, Young-Hye La

{"title":"Chemical foundation model-guided design of high ionic conductivity electrolyte formulations","authors":"Murtaza Zohair, Vidushi Sharma, Eduardo A. Soares, Khanh Nguyen, Maxwell Giammona, Linda Sundberg, Andy Tek, Emilio A. V. Vital, Young-Hye La","doi":"10.1038/s41524-025-01774-4","DOIUrl":"https://doi.org/10.1038/s41524-025-01774-4","url":null,"abstract":"Designing optimal formulations is a major challenge in developing electrolytes for the next generation of rechargeable batteries due to the vast combinatorial design space and complex interplay between multiple constituents. Machine learning (ML) offers a powerful tool to uncover underlying chemical design rules and accelerate the process of formulation discovery. In this work, we present an approach to design new formulations that can achieve target performance, using a generalizable chemical foundation model. The chemical foundation model is fine-tuned on an experimental dataset of 13,666 ionic conductivity values curated from the lithium-ion battery literature. The fine-tuned model is used to discover 7 novel high conductivity electrolyte formulations through generative screening, improving the conductivity of LiFSI- and LiDFOB-based electrolytes by 82% and 172%, respectively. These findings highlight a generalizable workflow that is highly adaptable to the discovery of chemical mixtures with tailored properties to address challenges in energy storage and beyond.","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"27 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-08-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144919410","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

FerroAI: a deep learning model for predicting phase diagrams of ferroelectric materials 铁电材料相图预测的深度学习模型

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-08-29 DOI: 10.1038/s41524-025-01778-0

Chenbo Zhang, Xian Chen

引用次数: 0

Interpretable X-ray diffraction spectra analysis using confidence evaluated deep learning enhanced by template element replacement 使用置信度评估的深度学习模板元素替换增强的可解释x射线衍射光谱分析

IF 9.7 1区材料科学

npj Computational Materials Pub Date : 2025-08-28 DOI: 10.1038/s41524-025-01743-x

Rongchang Xing, Haodong Yao, Zuoxin Xi, Minghui Sun, Qingmeng Li, Jinglong Tian, Hairui Wang, DeTing Xu, Zhaohai Ma, Lina Zhao

{"title":"Interpretable X-ray diffraction spectra analysis using confidence evaluated deep learning enhanced by template element replacement","authors":"Rongchang Xing, Haodong Yao, Zuoxin Xi, Minghui Sun, Qingmeng Li, Jinglong Tian, Hairui Wang, DeTing Xu, Zhaohai Ma, Lina Zhao","doi":"10.1038/s41524-025-01743-x","DOIUrl":"https://doi.org/10.1038/s41524-025-01743-x","url":null,"abstract":"X-ray Diffraction analysis is crucial for understanding material structures but is hindered by complex patterns and the need for expert interpretation. Deep learning offers automation in phase identification but faces challenges such as data scarcity, overconfidence in predictions and lack of interpretability. This study addresses these by employing Template Element Replacement to generate a perovskite chemical space containing physically unstable virtual structures, enhancing model understanding of XRD-crystal structure relationships and improving classification accuracy by ~5%. A Bayesian-VGGNet model was developed, achieving 84% accuracy on simulated spectra and 75% on external experimental data, while simultaneously estimating prediction uncertainty. Evaluation using Bayesian methods revealed low entropy values, indicating high model confidence. Quantifying the importance of input features to crystal symmetry, aligning significant features of seven crystal systems with physical principles. These approaches enhance the model’s robustness and reliability, making it suitable for practical applications.<figure></figure>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"7 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144916146","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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