TY - CONF A1 - Huber, Norbert T1 - Perspectives and pitfalls in modeling of structure-property relationships using machine learning N2 - Machine learning (ML) has been increasingly utilized to support microstructure characterization and predict mechanical properties. A successful ML model typically requires a comprehensive understanding of existing knowledge, expertise in translating this knowledge into meaningful input features, an effective ML architecture, and robust validation of the trained model. Despite the rapid growth in publications incorporating ML methods in recent years, there is limited literature specifically addressing nanoporous metals. The talk will give an overview on perspectives and pitfalls in modeling of structureproperty relationships using machine learning with focus on various challenges that arise from the specific nature of nanoporous metals including randomness of microstructure, image segmentation, lack of tomography data, feature engineering for property prediction, and implications for plasticity including anisotropic flow and arbitrary multiaxial loading on the lower scale of hierarchy. An outlook will be given on the perspectives of establishing a culture of open data, specifically towards curated data sets needed for training and validation of ML models. Potential use cases are the comparison of data from different sources, mining of more general relationships, and validation of models trained with computer generated data using experimental data. T2 - 5th International Symposium on Nanoporous Materials by Alloy Corrosion CY - Sendai, Japan DA - 06.10.2025 KW - Nanoporous metals KW - Machine learning KW - Structure-properties relationship KW - Materials design PY - 2025 AN - OPUS4-64536 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Walterbos, Luc T1 - Double trouble: exploring the chemical landscape of halide double perovskites N2 - Halide Double Perovskites (HDPs) are an emerging class of materials with chemical formula A2BB’X6 with possible applications in photovoltaics, X-ray detection, sensing, photocatalysis and spintronics.. However, with more than 40,000 potential HDP compositions, much of the chemical landscape remains unexplored. We have generated a database of spin-polarized, hybrid functional (HSE06) electronic structure data of all HPDs with A=Cs that are predicted to be stable based on a tolerance-factor analysis. Our high-throughput workflow also consists or a chemical bonding and orbital projection analysis based on LOBSTER (ww.cohp.de), leading to a comprehensivedatabase of electronic, magnetic and chemical bonding properties of >2700 HDP compositions, which can serve as a starting point for material design and discovery via interpretable machine learning techniques, which we use to identify unexpected trends and relations in the chemical landscape. T2 - Psi-k 2025 CY - Lausanne, Swiss DA - 25.08.2025 KW - Materials design KW - Halide perovskite KW - Halide double perovskite KW - High-throughput KW - Hybrid functional PY - 2025 AN - OPUS4-64009 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Ganose, Alex M. A1 - Sahasrabuddhe, Hrushikesh A1 - Asta, Mark A1 - Beck, Kevin A1 - Biswas, Tathagata A1 - Bonkowski, Alexander A1 - Bustamante, Joana A1 - Chen, Xin A1 - Chiang, Yuan A1 - Chrzan, Daryl C. A1 - Clary, Jacob A1 - Cohen, Orion A. A1 - Ertural, Christina A1 - Gallant, Max C. A1 - George, Janine A1 - Gerits, Sophie A1 - Goodall, Rhys E. A. A1 - Guha, Rishabh D. A1 - Hautier, Geoffroy A1 - Horton, Matthew A1 - Inizan, T. J. A1 - Kaplan, Aaron D. A1 - Kingsbury, Ryan S. A1 - Kuner, Matthew C. A1 - Li, Bryant A1 - Linn, Xavier A1 - McDermott, Matthew J. A1 - Mohanakrishnan, Rohith Srinivaas A1 - Naik, Aakash A. A1 - Neaton, Jeffrey B. A1 - Parmar, Shehan M. A1 - Persson, Kristin A. A1 - Petretto, Guido A1 - Purcell, Thomas A. R. A1 - Ricci, Francesco A1 - Rich, Benjamin A1 - Riebesell, Janosh A1 - Rignanese, Gian-Marco A1 - Rosen, Andrew S. A1 - Scheffler, Matthias A1 - Schmidt, Jonathan A1 - Shen, Jimmy-Xuan A1 - Sobolev, Andrei A1 - Sundararaman, Ravishankar A1 - Tezak, Cooper A1 - Trinquet, Victor A1 - Varley, Joel B. A1 - Vigil-Fowler, Derek A1 - Wang, Duo A1 - Waroquiers, David A1 - Wen, Mingjian A1 - Yang, Han A1 - Zheng, Hui A1 - Zheng, Jiongzhi A1 - Zhu, Zhuoying A1 - Jain, Anubhav T1 - Atomate2: Modular workflows for materials science N2 - High-throughput density functional theory (DFT) calculations have become a vital element of computational materials science, enabling materials screening, property database generation, and training of “universal” machine learning models. While several software frameworks have emerged to support these computational efforts, new developments such as machine learned force fields have increased demands for more flexible and programmable workflow solutions. This manuscript introduces atomate2, a comprehensive evolution of our original atomate framework, designed to address existing limitations in computational materials research infrastructure. Key features include the support for multiple electronic structure packages and interoperability between them, along with generalizable workflows that can be written in an abstract form irrespective of the DFT package or machine learning force field used within them. Our hope is that atomate2's improved usability and extensibility can reduce technical barriers for high-throughput research workflows and facilitate the rapid adoption of emerging methods in computational material science. KW - Automation KW - Materials design KW - DFT workflows KW - Phonons KW - Thermal conductivity KW - Bonding analysis PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-635759 DO - https://doi.org/10.1039/d5dd00019j SN - 2635-098X SP - 1 EP - 30 PB - Royal Society of Chemistry (RSC) CY - Cambridge AN - OPUS4-63575 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Lobster workflow and applications N2 - This talk introduced bonding analysis to the participants of the school including several examples of its usefulness. Additionally, I talked in detail about the workflows related to bonding analysis within the atomate2 workflow library. I then showed how the workflow was applied to build a very large bonding data database that can now be used for machine learning of materials properties. T2 - "Automated ab initio workflows with Jobflow and Atomate2" CECAM Flagship school CY - Lausanne, Switzerland DA - 17.03.2025 KW - Automation KW - Materials design KW - Machine learning KW - Thermal conductivity KW - Inorganic materials KW - High-throughput PY - 2025 AN - OPUS4-62948 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Albornoz, Ricardo Valencia A1 - Antypov, Dmytro A1 - Blanke, Gerd A1 - Borges, Itamar A1 - Marulanda Bran, Andres A1 - Cheung, Joshua A1 - Collins, Christopher M. A1 - David, Nicholas A1 - Day, Graeme M. A1 - Deringer, Volker L. A1 - Draxl, Claudia A1 - Eardley-Brunt, Annabel A1 - Evans, Matthew L. A1 - Fairlamb, Ian A1 - Fieseler, Kate A1 - Franklin, Barnabas A. A1 - George, Janine A1 - Grundy, Joanna A1 - Johal, Jay A1 - Kalikadien, Adarsh V. A1 - Kapil, Venkat A1 - Kotopanov, Lyubomir A1 - Kumar, Vishank A1 - Kuttner, Christian A1 - Lederbauer, Magdalena A1 - Ojeda-Porras, Andrea Carolina A1 - Pang, Jiayun A1 - Parkes, Michael A1 - Pemberton, Miles A1 - Ruscic, Branko A1 - Ryder, Matthew R. A1 - Sakaushi, Ken A1 - Saleh, Gabriele A1 - Savoie, Brett M. A1 - Schwaller, Philippe A1 - Skjelstad, Bastian Bjerkem A1 - Sun, Wenhao A1 - Taniguchi, Takuya A1 - Taylor, Christopher R. A1 - Torrisi, Steven A1 - Vishnoi, Shubham A1 - Walsh, Aron A1 - Wu, Ruiqi T1 - Discovering trends in big data: General discussion N2 - This article is a discussion of the paper "Specialising and analysing instruction-tuned and byte-level language models for organic reaction prediction" by Jiayun Pang and Ivan Vulić (Faraday discussions, 2025, 256, 413-433). KW - Automation KW - Big data KW - Machine learning KW - Materials design KW - Chemically complex materials PY - 2025 DO - https://doi.org/10.1039/D4FD90063D SN - 1359-6640 SN - 1364-5498 VL - 256 IS - Themed collection: Data-driven discovery in the chemical sciences SP - 520 EP - 550 PB - Royal Society of Chemistry (RSC) CY - Cambridge AN - OPUS4-62652 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Ganose, Alex A1 - Sahasrabuddhe, Hrushikesh A1 - Asta, Mark A1 - Beck, Kevin A1 - Biswas, Tathagata A1 - Bonkowski, Alexander A1 - Bustamante, Joana A1 - Chen, Xin A1 - Chiang, Yuan A1 - Chrzan, Daryl A1 - Clary, Jacob A1 - Cohen, Orion A1 - Ertural, Christina A1 - George, Janine A1 - Gallant, Max A1 - George, Janine A1 - Gerits, Sophie A1 - Goodall, Rhys A1 - Guha, Rishabh A1 - Hautier, Geoffroy A1 - Horton, Matthew A1 - Kaplan, Aaron A1 - Kingsbury, Ryan A1 - Kuner, Matthew A1 - Li, Bryant A1 - Linn, Xavier A1 - McDermott, Matthew A1 - Rohith Srinivaas Mohanakrishnan, A1 - Naik, Aakash A1 - Neaton, Jeffrey A1 - Persson, Kristin A1 - Petretto, Guido A1 - Purcell, Thomas A1 - Ricci, Francesco A1 - Rich, Benjamin A1 - Riebesell, Janosh A1 - Rignanese, Gian-Marco A1 - Rosen, Andrew A1 - Scheffler, Matthias A1 - Schmidt, Jonathan A1 - Shen, Jimmy-Xuan A1 - Sobolev, Andrei A1 - Sundararaman, Ravishankar A1 - Tezak, Cooper A1 - Trinquet, Victor A1 - Varley, Joel A1 - Vigil-Fowler, Derek A1 - Wang, Duo A1 - Waroquiers, David A1 - Wen, Mingjian A1 - Yang, Han A1 - Zheng, Hui A1 - Zheng, Jiongzhi A1 - Zhu, Zhuoying A1 - Jain, Anubhav T1 - Atomate2: Modular workflows for materials science N2 - High-throughput density functional theory (DFT) calculations have become a vital element of computational materials science, enabling materials screening, property database generation, and training of “universal” machine learning models. While several software frameworks have emerged to support these computational efforts, new developments such as machine learned force fields have increased demands for more flexible and programmable workflow solutions. This manuscript introduces atomate2, a comprehensive evolution of our original atomate framework, designed to address existing limitations in computational materials research infrastructure. Key features include the support for multiple electronic structure packages and interoperability between them, along with generalizable workflows that can be written in an abstract form irrespective of the DFT package or machine learning force field used within them. Our hope is that atomate2’s improved usability and extensibility can reduce technical barriers for high-throughput research workflows and facilitate the rapid adoption of emerging methods in computational material science. KW - Automation KW - DFT KW - Digitalisation KW - Materials design KW - Machine learning KW - Machine learned interatomic potentials PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-624487 DO - https://doi.org/10.26434/chemrxiv-2025-tcr5h SN - 2573-2293 SP - 1 EP - 66 PB - American Chemical Society (ACS) CY - Washington, D.C. AN - OPUS4-62448 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Liu, Yuanbin A1 - Morrow, Joe D. A1 - Ertural, Christina A1 - Fragapane, Natascia L. A1 - Gardner, John L. A. A1 - Naik, Aakash A1 - Zhou, Yuxing A1 - George, Janine A1 - Deringer, Volker L. T1 - An automated framework for exploring and learning potential-energy surfaces N2 - Machine learning has become ubiquitous in materials modelling and now routinely enables large-scale atomistic simulations with quantum-mechanical accuracy. However, developing machine-learned interatomic potentials requires high-quality training data, and the manual generation and curation of such data can be a major bottleneck. Here, we introduce an automated framework for the exploration and fitting of potential-energy surfaces, implemented in an openly available software package that we call autoplex ('automatic potential-landscape explorer'). We discuss design choices, particularly the interoperability with existing software architectures, and the ability for the end user to easily use the computational workflows provided. We show wide-ranging capability demonstrations: for the titanium-oxygen system, SiO2, crystalline and liquid water, as well as phase-change memory materials. More generally, our study illustrates how automation can speed up atomistic machine learning -- with a long-term vision of making it a genuine mainstream tool in physics, chemistry, and materials science. KW - Machine learned interatomic potentials KW - Ab initio KW - DFT KW - Automation KW - Workflows KW - Machine learning KW - Materials design KW - Materials discovery PY - 2024 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-623188 DO - https://doi.org/10.48550/arXiv.2412.16736 SN - 2331-8422 SP - 1 EP - 27 PB - Cornell University CY - Ithaca, NY AN - OPUS4-62318 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Anker, Andy S. A1 - Aspuru-Guzik, Alán A1 - Bechtel, Tim A1 - Bigi, Filippo A1 - Briling, Ksenia R. A1 - Das, Basita A1 - David, Nicholas A1 - Day, Graeme M. A1 - Deringer, Volker L. A1 - Dyer, Matthew A1 - Eardley-Brunt, Annabel A1 - Evans, Matthew L. A1 - Evans, Rob A1 - Franklin, Barnabas A. A1 - Ganose, Alex M. A1 - George, Janine A1 - Goulding, Mark A1 - Hickey, Niamh A1 - James, Gillian A1 - Kalikadien, Adarsh V. A1 - Kapil, Venkat A1 - Kulik, Heather J. A1 - Kumar, Vishank A1 - Kuttner, Christian A1 - Lam, Erwin A1 - Lederbauer, Magdalena A1 - Lou, Yuchen A1 - Martin, Jennie A1 - Marulanda Bran, Andres A1 - Mathea, Miriam A1 - Pickard, Chris J. A1 - Ruscic, Branko A1 - Ryder, Matthew R. A1 - Sabanza Gil, Victor A1 - Schwaller, Philippe A1 - Segler, Marwin H. S. A1 - Sun, Wenhao A1 - Tanovic, Sara A1 - Treyde, Wojtek A1 - Walsh, Aron A1 - Wu, Ruiqi T1 - Discovering synthesis targets: General discussion N2 - This article is a discussion of the paper "Analysis of uncertainty of neural fingerprint-based models" by Christian W. Feldmann, Jochen Sieg and Miriam Mathea (Faraday discussions, 2025, DOI: 10.1039/D4FD00095A). KW - Automation KW - Materials acceleration platforms KW - Machine learning KW - Materials design KW - Materials discovery KW - Density functional theory KW - Ab initio PY - 2025 DO - https://doi.org/10.1039/D4FD90064B SN - 1359-6640 SN - 1364-5498 VL - 256 IS - Themed collection: Data-driven discovery in the chemical sciences SP - 639 EP - 663 PB - Royal Society of Chemistry (RSC) CY - Cambridge AN - OPUS4-62317 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Anker, Andy S. A1 - Aspuru-Guzik, Alán A1 - Ben Mahmoud, Chiheb A1 - Bennett, Sophie A1 - Briling, Ksenia R. A1 - Changiarath, Arya A1 - Chong, Sanggyu A1 - Collins, Christopher M. A1 - Cooper, Andrew I. A1 - Crusius, Daniel A1 - Darmawan, Kevion K. A1 - Das, Basita A1 - David, Nicholas A1 - Day, Graeme M. A1 - Deringer, Volker L. A1 - Duarte, Fernanda A1 - Eardley-Brunt, Annabel A1 - Evans, Matthew L. A1 - Evans, Rob A1 - Fairlamb, Ian A1 - Franklin, Barnabas A. A1 - Frey, Jeremy A1 - Ganose, Alex M. A1 - Goulding, Mark A1 - Hafizi, Roohollah A1 - Hakkennes, Matthijs A1 - Hickey, Niamh A1 - James, Gillian A1 - Jelfs, Kim E. A1 - Kalikadien, Adarsh V. A1 - Kapil, Venkat A1 - Koczor-Benda, Zsuzsanna A1 - Krammer, Ferdinand A1 - Kulik, Heather J. A1 - Kumar, Vishank A1 - Kuttner, Christian A1 - Lam, Erwin A1 - Lou, Yuchen A1 - Mante, Eltjo A1 - Martin, Jennie A1 - Mroz, Austin M. A1 - Nematiaram, Tahereh A1 - Pare, Charles W. P. A1 - Patra, Sarbani A1 - Proudfoot, James A1 - Ruscic, Branko A1 - Ryder, Matthew R. A1 - Sakaushi, Ken A1 - Saßmannshausen, Jörg A1 - Savoie, Brett M. A1 - Schneider, Nadine A1 - Schwaller, Philippe A1 - Skjelstad, Bastian Bjerkem A1 - Sun, Wenhao A1 - Szczypiński, Filip T. A1 - Torrisi, Steven A1 - Ueltzen, Katharina A1 - Vishnoi, Shubham A1 - Walsh, Aron A1 - Wang, Xinwei A1 - Wilson, Chloe A1 - Wu, Ruiqi A1 - Zeitler, Jakob T1 - Discovering structure–property correlations: General discussion N2 - This article is a discussion of the paper "Web-BO: Towards increased accessibility of Bayesian optimisation (BO) for chemistry" by Austin M. Mroz, Piotr N. Toka, Ehecatl Antonio del Río Chanona and Kim E. Jelfs (Faraday discussions, 2025, 256, 221-234). KW - Materials design KW - Machine learning KW - Automation KW - Materials discovery PY - 2025 DO - https://doi.org/10.1039/d4fd90062f SN - 1359-6640 SN - 1364-5498 VL - 256 IS - Themed collection: Data-driven discovery in the chemical sciences SP - 373 EP - 412 PB - Royal Society of Chemistry (RSC) CY - Cambridge AN - OPUS4-62208 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - VIDEO A1 - Völker, Christoph T1 - Accelerating the search for sustainable concretes with AI N2 - With 8% of man-made CO2 emissions, cement production is an important driver of the climate crisis. By using alkali-activated binders, part of the energy-intensive clinker production process can be dispensed. However, as numerous raw materials are involved in the manufacturing process here, the complexity of the materials increases by orders of magnitude. Finding a properly balanced binder formulation is like looking for a needle in a haystack. We have shown for the first time that artificial intelligence (AI)-based optimization of alkali-activated binder formulations can significantly accelerate research. The "Sequential Learning App for Materials Discovery" (SLAMD) aims to accelerate practice transfer. With SLAMD, materials scientists have low-threshold access to AI through interactive and intuitive user interfaces. The value added by AI can be determined directly. For example, the CO2 emissions saved per ton of cement can be determined for each development cycle: the more efficient the AI optimization, the greater the savings. Our material database already includes more than 120,000 data points of alternative binders and is constantly being expanded with new parameters. We are currently driving the enrichment of the data with a life cycle analysis of the building materials. Based on a case study we show how intuitive access to AI can drive the adoption of techniques that make a real contribution to the development of resource-efficient and sustainable building materials of the future and make it easy to identify when classical experiments are more efficient. KW - Materials informatics KW - Materials design KW - Alkali activated concrete KW - Machine learning PY - 2022 PB - Bundesanstalt für Materialforschung und -prüfung (BAM) CY - Berlin AN - OPUS4-56639 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -