npj Computational Materials最新文献

{"title":"Elemental numerical descriptions to enhance classification and regression model performance for high-entropy alloys","authors":"Yan Zhang, Cheng Wen, Pengfei Dang, Xue Jiang, Dezhen Xue, Yanjing Su","doi":"10.1038/s41524-025-01560-2","DOIUrl":"https://doi.org/10.1038/s41524-025-01560-2","url":null,"abstract":"<p>The machine learning-assisted design of new alloy compositions often relies on the physical and chemical properties of elements to describe the materials. In the present study, we propose a strategy based on an evolutionary algorithm to generate new elemental numerical descriptions for high-entropy alloys (HEAs). These newly defined descriptions significantly enhance classification accuracy, increasing it from 77% to ~97% for recognizing FCC, BCC, and dual phases, compared to traditional empirical features. Our experimental validation demonstrates that our classification model, utilizing these new elemental numerical descriptions, successfully predicted the phases of 8 out of 9 randomly selected alloys, outperforming the same model based on traditional empirical features, which correctly predicted 4 out of 9. By incorporating these descriptions derived from a simple logistic regression model, the performance of various classifiers improved by at least 15%. Moreover, these new numerical descriptions for phase classification can be directly applied to regression model predictions of HEAs, reducing the error by 22% and improving the <i>R</i>² value from 0.79 to 0.88 in hardness prediction. Testing on six different materials datasets, including ceramics and functional alloys, demonstrated that the obtained numerical descriptions achieved higher prediction precision across various properties, indicating the broad applicability of our strategy.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"55 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143641125","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}

{"title":"Leveraging high-throughput molecular simulations and machine learning for the design of chemical mixtures","authors":"Alex K. Chew, Mohammad Atif Faiz Afzal, Zachary Kaplan, Eric M. Collins, Suraj Gattani, Mayank Misra, Anand Chandrasekaran, Karl Leswing, Mathew D. Halls","doi":"10.1038/s41524-025-01552-2","DOIUrl":"https://doi.org/10.1038/s41524-025-01552-2","url":null,"abstract":"<p>Mixtures of chemical ingredients, such as formulations, are ubiquitous in materials science, but optimizing their properties remains challenging due to the vast design space. Computational approaches offer a promising solution to traverse this space while minimizing trial-and-error experimentation. Using high-throughput classical molecular dynamics simulations, we generated a comprehensive dataset of over 30,000 solvent mixtures to evaluate three machine learning approaches that connect molecular structure and composition to property: formulation descriptor aggregation (FDA), formulation graph (FG), and Set2Set-based method (FDS2S). Our results demonstrate that our new FDS2S approach outperforms other approaches in predicting simulation-derived properties. Formulation-property relationships can reveal important substructures and identify promising formulations at least two to three times faster than random guessing. The models show robust transferability to experimental datasets, accurately predicting properties across energy, pharmaceutical, and petroleum applications. Our research demonstrates the utility of high-throughput simulations and machine learning tools to design formulations with promising properties.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"89 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143635733","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}

{"title":"Spin-splitting above room-temperature in Janus Mn2ClSeH antiferromagnetic semiconductor with a large out-of-plane piezoelectricity","authors":"Haiming Lu, Sitong Bao, Bocheng Lei, Sutao Sun, Linglu Wu, Jian Zhou, Lili Zhang","doi":"10.1038/s41524-025-01566-w","DOIUrl":"https://doi.org/10.1038/s41524-025-01566-w","url":null,"abstract":"<p>Two-dimensional (2D) antiferromagnets have garnered considerable research interest due to their robustness against external magnetic perturbation, ultrafast dynamics, and magneto-transport effects. However, the lack of spin-splitting in antiferromagnetic (AFM) materials severely limits their potential in spintronics applications. Inspired by inherent out-of-plane potential gradient of Janus structure, we predict three stable AFM Janus Mn₂ClXH (X = O, S, and Se) monolayers with spontaneous spin-splitting based on first-principles calculations. Notably, Janus Mn₂ClSeH exhibits a high Néel temperature of up to 510 K, robust perpendicular magnetocrystalline anisotropy, outstanding out-of-plane piezoelectricity of 0.454 × 10⁻¹⁰C/m, and sizeable spontaneous valley polarization of 17.2 meV. Moreover, the spin-splitting can be significantly enhanced through appropriate synergistic regulation of biaxial strain and external electric field. These results demonstrate that the Janus Mn₂ClSeH monolayer is a very potential candidate for designing intriguing antiferromagnet-based devices with fantastic piezoelectric and valleytronic characteristics.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"39 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143635732","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}

{"title":"An NV− center in magnesium oxide as a spin qubit for hybrid quantum technologies","authors":"Vrindaa Somjit, Joel Davidsson, Yu Jin, Giulia Galli","doi":"10.1038/s41524-025-01558-w","DOIUrl":"https://doi.org/10.1038/s41524-025-01558-w","url":null,"abstract":"<p>Recent predictions suggest that oxides, such as MgO and CaO, could serve as hosts of spin defects with long coherence times and thus be promising materials for quantum applications. However, in most cases, specific defects have not yet been identified. Here, by using a high-throughput first-principles framework and advanced electronic structure methods, we identify a negatively charged complex between a nitrogen interstitial and a magnesium vacancy in MgO with favorable electronic and optical properties for hybrid quantum technologies. We show that this NV⁻ center has stable triplet ground and excited states, with singlet shelving states enabling optical initialization and spin-dependent readout. We predict several properties, including absorption, emission, and zero-phonon line energies, as well as zero-field splitting tensor, and hyperfine interaction parameters, which can aid in the experimental identification of this defect. Our calculations show that due to a strong pseudo-Jahn Teller effect and low-frequency phonon modes, the NV⁻ center in MgO is subject to a substantial vibronic coupling. We discuss design strategies to reduce such coupling and increase the Debye-Waller factor, including the effect of strain and the localization of the defect states. We propose that the favorable properties of the NV⁻ defect, along with the technological maturity of MgO, could enable hybrid classical-quantum applications, such as spintronic quantum sensors and single qubit gates.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"25 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143641126","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}

{"title":"Reply to: Comment on “Machine learning enhanced analysis of EBSD data for texture representation”","authors":"J. Wanni, C. A. Bronkhorst, D. J. Thoma","doi":"10.1038/s41524-025-01562-0","DOIUrl":"https://doi.org/10.1038/s41524-025-01562-0","url":null,"abstract":"<p>We respond to Schaeben et al.’s¹ comment on our paper, “Machine Learning Enhanced Analysis of EBSD Data for Texture Representation.” While their observations are factually correct, they do not disprove our results. Our method, TACS, preserves the full distribution of crystallographic orientations and is validated with real-world data. We emphasize the importance of empirical validation over theoretical constructs in assessing machine learning methods’ practical performance.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"214 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143635676","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}

{"title":"Active learning of effective Hamiltonian for super-large-scale atomic structures","authors":"Xingyue Ma, Hongying Chen, Ri He, Zhanbo Yu, Sergei Prokhorenko, Zheng Wen, Zhicheng Zhong, Jorge Íñiguez-González, L. Bellaiche, Di Wu, Yurong Yang","doi":"10.1038/s41524-025-01563-z","DOIUrl":"https://doi.org/10.1038/s41524-025-01563-z","url":null,"abstract":"<p>The first-principles-based effective Hamiltonian scheme provides one of the most accurate modeling techniques for large-scale structures, especially for ferroelectrics. However, the parameterization of the effective Hamiltonian is complicated and can be difficult for some complex systems such as high-entropy perovskites. Here, we propose a general form of effective Hamiltonian and develop an active machine-learning approach to parameterize the effective Hamiltonian based on Bayesian linear regression. The parameterization is employed in molecular dynamics simulations with the prediction of energy, forces, stress and their uncertainties at each step, which decides whether first-principles calculations are executed to retrain the parameters. Structures of BaTiO₃, PbTiO₃, Pb(Zr_0.75Ti_0.25)O₃, and (Pb,Sr)TiO₃ system are taken as examples to show the accuracy of this approach, as compared with conventional parametrization method and experiments. This machine-learning approach provides a universal and automatic way to compute the effective Hamiltonian parameters for any considered complex systems with super-large-scale (more than 10⁷ atoms) atomic structures.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"54 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143618823","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}

{"title":"Amorphization evolution study of CrCoFeNiMn high entropy alloy for mechanical performance optimization by deep potential molecular dynamics","authors":"Wentao Zhou, Jia Song, Lve Lin, Huilong Yang, Shaoqiang Guo, Guang Ran, Yafei Wang","doi":"10.1038/s41524-025-01561-1","DOIUrl":"https://doi.org/10.1038/s41524-025-01561-1","url":null,"abstract":"<p>In the study, we explore the structural evolution of Cantor high-entropy alloy (HEA) under different super-cooling rates and its correlation with mechanical property variations by the developed machine learning-driven deep potential molecular dynamics (DPMD) simulation. Our results reveal the critical super-cooling rate of amorphization-crystallization transition of Cantor alloy and the local structure constitutions at different temperatures during the super-cooling process. The associated mechanical property studies demonstrate the glassy Cantor alloy amorphized at high super-cooling rate exhibits a superior capability of ductility but this capability is unrelated to the amorphization cooling rates. While the high strength of Cantor alloy requires a lower super-cooling rate which might result in the crystallization, amorphizing the Cantor alloy at the critical super-cooling rate of amorphization-crystallization transition could compatibilize both ductility and strength capabilities. Such a discovery sheds new lights on the material development and its mechanical performance optimization for industrial applications.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"92 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143608330","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}

{"title":"Deep-learning real-time phase retrieval of imperfect diffraction patterns from X-ray free-electron lasers","authors":"Sung Yun Lee, Do Hyung Cho, Chulho Jung, Daeho Sung, Daewoong Nam, Sangsoo Kim, Changyong Song","doi":"10.1038/s41524-025-01569-7","DOIUrl":"https://doi.org/10.1038/s41524-025-01569-7","url":null,"abstract":"<p>Machine learning is attracting surging interest across nearly all scientific areas by enabling the analysis of large datasets and the extraction of scientific information from incomplete data. Data-driven science is rapidly growing, especially in X-ray methodologies, where advanced light sources and detection technologies produce vast amounts of data that exceed meticulous human inspection capabilities. Despite the increasing demands, the full application of machine learning has been hindered by the need for data-specific optimizations. In this study, we introduce a new deep-learning-based phase retrieval method for imperfect diffraction data. This method provides robust phase retrieval for simulated data and performs well on partially damaged and noisy single-pulse diffraction data from X-ray free-electron lasers. Moreover, the method significantly reduces data processing time, facilitating real-time image reconstructions that are crucial for high-repetition-rate data acquisition. This approach offers a reliable solution to the phase problem to be widely adopted across various research areas confronting the inverse problem.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"53 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143608326","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}

{"title":"New fast ion conductors discovered through the structural characteristic involving isolated anions","authors":"Qifan Yang, Jing Xu, Yuqi Wang, Xiao Fu, Ruijuan Xiao, Hong Li","doi":"10.1038/s41524-025-01559-9","DOIUrl":"https://doi.org/10.1038/s41524-025-01559-9","url":null,"abstract":"<p>One of the key materials in solid-state lithium batteries is fast ion conductors. However, Li⁺ ion transport in inorganic crystals involves complex factors, making it a mystery to find and design ion conductors with low migration barriers. In this work, a distinctive structural characteristic involving isolated anions has been discovered to enhance high ionic conductivity in crystals. It is an effective way to create a smooth energy potential landscape and construct local pathways for lithium ion migration. By adjusting the spacing and arrangement of the isolated anions, these local pathways can connect with each other, leading to high ion conductivity. By designing different space groups and local environments of the Se²⁻ anions in the Li₈SiSe₆ composition, combined with the ion transport properties obtained from AIMD simulations, we define isolated anions and find that local environments with higher point group symmetry promotes the formation of cage-like local transport channels. Additionally, the appropriate distance between neighboring isolated anions can create coplanar connections between adjacent cage-like channels. Furthermore, different element types of isolated anions can be used to control the distribution of cage-like channels in the lattice. Based on the structural characteristic of isolated anions, we shortlisted compounds with isolated N3−, Cl⁻, I⁻, and S²⁻ features from the crystal structure databases. The confirmation of ion transport in these structures validates the proposed design method of using isolated anions as structural features for fast ion conductors and leads to the discovery of several new fast ion conductor materials.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"54 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143608313","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}

{"title":"PID3Net: a deep learning approach for single-shot coherent X-ray diffraction imaging of dynamic phenomena","authors":"Tien-Sinh Vu, Minh-Quyet Ha, Adam Mukharil Bachtiar, Duc-Anh Dao, Truyen Tran, Hiori Kino, Shuntaro Takazawa, Nozomu Ishiguro, Yuhei Sasaki, Masaki Abe, Hideshi Uematsu, Naru Okawa, Kyosuke Ozaki, Kazuo Kobayashi, Yoshiaki Honjo, Haruki Nishino, Yasumasa Joti, Takaki Hatsui, Yukio Takahashi, Hieu-Chi Dam","doi":"10.1038/s41524-025-01549-x","DOIUrl":"https://doi.org/10.1038/s41524-025-01549-x","url":null,"abstract":"<p>This paper introduces a deep learning (DL)-based method for phase retrieval tailored to single-shot, multiple-frame coherent X-ray diffraction imaging (CXDI), designed specifically for visualizing local nanostructural dynamics within a larger sample. Current phase retrieval methods often struggle with achieving high spatiotemporal resolutions, handling dynamic imaging, and managing computational costs, which limits their applicability in observing nanostructural dynamics. This study addresses these gaps by developing a novel method that leverages a feedforward architecture with a physics-informed strategy utilizing measurement settings, enabling the reconstruction of dynamic “movies\" from time-evolving diffraction images of the illuminated area. The method incorporates key enhancements, such as temporal convolution blocks to capture spatiotemporal correlations and a unified TV regularization applied to the reconstructed object, resulting in improved noise reduction and spatial smoothness. An expanded evaluation framework, including multiple metrics and systematic sensitivity analysis, is employed to comprehensively assess the method’s performance and robustness. Proof-of-concept experiments, including numerical simulations and imaging experiments of a moving Ta test chart and colloidal gold particles (dispersed in aqueous polyvinyl alcohol solutions) with synchrotron hard X-rays, validate the high imaging performance of this method. Experimental results demonstrate that structures in the sample have been successfully reconstructed at short exposure times, significantly outperforming both traditional methods and current DL-based methods. The proposed method provides efficient and reliable reconstruction of dynamic images with low computational costs, making it suitable for exploring fast-evolving phenomena in synchrotron- or free-electron laser-based applications requiring high spatiotemporal resolutions.</p>","PeriodicalId":19342,"journal":{"name":"npj Computational Materials","volume":"27 1","pages":""},"PeriodicalIF":9.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143599874","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}