TY - CONF A1 - Ruehle, Bastian T1 - The Role of Analytics for Closing the Loop in Self-Driving Labs N2 - The recent emergence of self-driving laboratories (SDL) and material acceleration plat-forms (MAPs) demonstrates the ability of these systems to change the way chemistry and material syntheses will be performed in the future. Especially in conjunction with nano- and advanced materials which are generally recognized for their great potential in solving current material science challenges, such systems can make disrupting con-tributions. Consequently, new tools that enhance the development and optimization cycle of nano- and advanced materials are crucial. In this contribution, we present our Self-Driving Lab (SDL) for Nano and Advanced Materials [1], that integrates robotics for batched autonomous synthesis – from molecular precursors to fully purified nano-materials – with automated characterization and data analysis, for a complete and reli-able nanomaterial synthesis workflow. By automating the processing and characteriza-tion steps for seven different materials from five representative, completely different classes of nano- and advanced materials (metal, metal oxide, silica, metal organic framework, and core–shell particles) that follow different reaction mechanisms, we demonstrate the great versatility, reproducibility, and flexibility of the platform. The system also incorporates in-line characterization measurement of hydrodynamic diameter, zeta potential, and optical properties (absorbance, fluorescence). In general, the interface with data analysis algorithms from in-line, at-line, and off-line measure-ments is of great importance for closing the design-make-test-analyze cycle and using these platforms efficiently. Here, we will give examples of how automatic image seg-mentation of electron microscopy images with the help of AI [2] can be used for reduc-ing the “data analysis bottleneck” from an off-line measurement. We will also discuss various machine learning (ML) algorithms that are currently implemented in the backend and can be used for ML-guided, closed-loop material optimization in our SDL. Lastly, we will show our recent efforts [3] in making the workflow generation on SDLs more user-friendly by using large language models to generate executable workflows automatically from synthesis procedures given in natural language and user-friendly graphical user interfaces based on node editors that also allow for knowledge graph extraction from the workflows. In this context, we are currently also working on a com-mon description or ontology for representing the process steps and parameters of the workflows, which will greatly facilitate the semantic description and interoperability of workflows between different SDL hardware and software platforms. These features underscore the SDL’s potential as a transformative tool for advancing and accelerating the development of nano- and advanced materials, offering solutions for a sustainable and environmentally responsible future. T2 - Analytica Conference 2026 CY - Munich, Germany DA - 24.03.2026 KW - Self-Driving Labs KW - Materials Acceleration Platforms KW - Advanced Materials KW - Nanomaterials KW - Automation KW - Digitalization KW - Ontologies PY - 2026 AN - OPUS4-65760 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust data generation, heuristics and machine learning for materials design N2 - Machine learning (ML) offers new routes to overcome the limitations of density functional theory (DFT) for advanced materials. We present data-generation strategies and workflows for ML interatomic potentials, including large-scale quantum-chemical bonding analysis.[1,2,3] Incorporating bonding descriptors into ML models enables prediction of phononic properties and validation of correlations between bonding strength, force constants, and thermal conductivity.[3] We introduce autoplex, an automated framework for training ML potentials, supporting general-purpose and phonon-focused workflows.[4] These developments provide a basis for fine-tuning foundation models for thermal transport at reduced cost.[5] For properties such as magnetism or synthesizability, we discuss complementary approaches, comparing ab initio methods with chemical heuristics and experimental data-driven ML models.[6,7]Our work advances scalable, accurate simulations for materials discovery. T2 - DPG Dresden, Condensed Matter CY - Dresden, Germany DA - 10.03.2026 KW - Materials Design KW - Materials Acceleration Platforms KW - Advanced materials simulations KW - Automation KW - High-throughput KW - Thermal Conductivity KW - Chemical Bonding Analysis KW - Phonons KW - Amorphous Materials PY - 2026 AN - OPUS4-65648 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Ryll, Tom William T1 - Multi-modal characterization for transformation of gypsum to anhydrite N2 - We investigate recycling of gypsum waste materials and have developed a method to synthesize phase pure anhydrite in solution, while contaminants could easily be separated from products. In-situ XRD measurements were conducted in a dedicated automation setup at BESSY II to find exact reaction conditions and were supported by in-situ Raman spectroscopy. T2 - Joint BAM-HZB Symposium on Multimodal Synchrotron Experiments for Next-Generation Materials Research CY - Berlin, Germany DA - 25.02.2026 KW - Recycling KW - Automation KW - Synchrotron XRD KW - Raman spectroscopy KW - Gypsum PY - 2026 AN - OPUS4-65607 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Ruehle, Bastian T1 - A Self-Driving Lab for Nano and Advanced Materials Synthesis in a Self-Driving Lab N2 - In this contribution, we present our Self-Driving Lab (SDL) for Nano and Advanced Materials, that integrates robotics for batched autonomous synthesis – from molecular precursors to fully purified nanomaterials – with automated characterization and data analysis, for a complete and reliable nanomaterial synthesis workflow. By fully automating the processing steps for seven different materials from five representative, completely different classes of nano- and advanced materials (metal, metal oxide, silica, metal organic framework, and core–shell particles) that follow different reaction mechanisms, we demonstrate the great versatility and flexibility of the platform. The system also exhibits high modularity and adaptability in terms of reaction scales and incorporates in-line characterization measurement of hydrodynamic diameter, zeta potential, and optical properties (absorbance, fluorescence). We discuss the excellent reproducibility of the various materials synthesized on the platform in terms of particle size and size distribution, and the adaptability and modularity that allows access to a diverse set of nanomaterial classes. We also present several key aspects of the central backend that orchestrates the (parallelized) syntheses workflows. One key feature is the resource management or “traffic control” for scheduling and executing parallel reactions in a multi-threaded environment. Another is the interface with data analysis algorithms from in-line, at-line, and off-line measurements. Here, we will give examples of how automatic image segmentation of electron microscopy images with the help of AI can be used for reducing the “data analysis bottleneck” from an off-line measurement. We will also discuss various machine learning (ML) algorithms that are currently implemented in the backend and can be used for ML-guided, closed-loop material optimization in our SDL. Lastly, we will show our recent efforts in making the workflow generation on SDLs more user-friendly by using large language models to generate executable workflows automatically from synthesis procedures given in natural language and user-friendly graphical user interfaces based on node editors that also allow for knowledge graph extraction from the workflows. In this context, we are currently also working on a common description or ontology for representing the process steps and parameters of the workflows, which will greatly facilitate the semantic description and interoperability of workflows between different SDL hardware and software platforms. T2 - Series on Digitalisation-Meet the Experts | Special Topic: Automation CY - Berlin, Germany DA - 27.02.2026 KW - Self-Driving Labs KW - Materials Acceleration Platforms KW - Advanced Materials KW - Nanomaterials KW - Automation KW - Digitalization PY - 2026 AN - OPUS4-65604 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Crossing Scientific Disciplines with Materials Informatics:� From Atoms to Algorithms N2 - Within this talk, I introduced students from Physics to Materials Informatics. To provide a context for this research, I have introduced the students to BAM and its tasks. I then started to introduce our activity field materials design, including materials acceleration platforms. Then, I explained how simulations speed up the materials searches as parf of materials acceleration platforms. T2 - jDPG Jena Meeting - Poland exchange CY - Jena, Germany DA - 26.02.2026 KW - Automation KW - Machine Learning KW - Materials Design KW - Batteries KW - Workflows PY - 2026 AN - OPUS4-65589 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Liu, Yuanbin A1 - Zhou, Yuxing A1 - Ademuwagun, Richard A1 - Walterbos, Luc A1 - George, Janine A1 - Elliott, Stephen R. A1 - Deringer, Volker L. T1 - Medium-Range Structural Order in Amorphous Arsenic N2 - Medium-range order (MRO) is a key structural feature of amorphous materials, but its origin and nature remain elusive. Here, we reveal the MRO in amorphous arsenic (a-As) using advanced atomistic simulations, based on machine-learned potentials derived using automated workflows. Our simulations accurately reproduce the experimental structure factor of a-As, especially the first sharp diffraction peak (FSDP), which is a signature of MRO. We compare and contrast the structure of a-As with that of its lighter homologue, red amorphous phosphorus (a-P): we find that a-As has a more uniform dihedral-angle distribution, and so we confirm that its structure can be thought of as a 3-fold coordinated continuous random network in first approximation, in contrast to the more molecular-cluster-like structure of a-P. The pressure-dependent structural behaviors of a-As and a-P differ as well, and the origin of the FSDP is closely correlated with the size and spatial distribution of voids in the amorphous networks. Our work provides fundamental insights into MRO in an amorphous elemental system, and more widely it illustrates the usefulness of automation for machine-learning-driven atomistic simulations. KW - Machine Learned Interatomic Potentials KW - Amorphous KW - Advanced Materials KW - Automation KW - Digitalisation KW - Structure-Property- Relationships PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-655952 DO - https://doi.org/10.1021/jacs.5c18688 SN - 0002-7863 SP - 1 EP - 13 PB - American Chemical Society (ACS) AN - OPUS4-65595 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust Data Generation, Heuristics and Machine Learning for Materials Design N2 - This talk first introduces students to the Materials Acceleration Platforms and Advanced Materials Characterization at BAM. Then, it motivates high-throuhgput screening for materials discovery and advanced materials simulations based on these core topics. Then four different research studies are presentend: evaluation of generative models, synthesizability prediction via PU learning, acceleration of materials property predictions with bonding analysis and advanced materials simulations supported by automatically trained machine learning potentials. T2 - Guest Lecture in MSE 403/1003, a Seminar in the Curriculum of the University of Toronto CY - Online meeting DA - 13.02.2026 KW - Automation KW - Materials Acceleration Platforms KW - Machine Learning KW - Workflows KW - Phonons KW - Bonding Analysis KW - Thermal Conductivity PY - 2026 AN - OPUS4-65514 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - George, Janine A1 - Ouellet-Plamondon, Claudiane A1 - Reyes, Kristofer T1 - Introduction to the “Accelerate Conference 2023–2024” themed collection N2 - The collection showcases the ways in which automation, machine learning and robotics are transforming experimental materials science and chemistry into continuous, computationally integrated processes. It features innovations regarding autonomous laboratories, Bayesian optimisation, high-throughput experimentation and computation, and AI-driven literature extraction, which simplify and scale up materials discovery. Together, these works outline a modular, responsible framework for accelerating scientific progress through human-guided, data-driven autonomy. KW - Automation KW - Materials Acceleration Platforms KW - Synthesizability KW - Workflows KW - Large Language Models KW - Ontologies KW - Materials Discovery PY - 2026 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-654658 DO - https://doi.org/10.1039/d5dd90057c SN - 2635-098X SP - 1 EP - 2 PB - Royal Society of Chemistry (RSC) AN - OPUS4-65465 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust Data Generation, Heuristics and Machine Learning for Materials Design N2 - My talk covered, among other things, robust data generation for machine learning. It showed how heuristics can be used within machine learning models and how they might also be extracted from machine learning models. Beyond this, I showed an automated pipeline for training machine learning potentials. T2 - Seminar Gruppe Stephan Roche CY - Barcelona, Spain DA - 22.01.2026 KW - Automation KW - Machine Learning KW - Materials Acceleration Platforms KW - Thermal Conductivity KW - Phonons KW - Bonding Analysis PY - 2026 AN - OPUS4-65428 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust Data Generation, Heuristics and Machine Learning for Materials Design N2 - My talk covered, among other things, robust data generation for machine learning. It showed how heuristics can be used within machine learning models and how they might also be extracted from machine learning models. Beyond this, I showed an automated pipeline for training machine learning potentials. T2 - Workshop on AI in Sustainable Materials Science CY - Düsseldorf, Germany DA - 27.01.2026 KW - Automation KW - Digitalisation KW - Materials Design KW - Thermal Conductivity KW - Chemical bonding KW - Materials Acceleration Platforms PY - 2026 AN - OPUS4-65427 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Ryll, Tom William T1 - In-situ analysis of nucleation processes – case study: calcium sulfate N2 - In this project we investigate nucleation pathways by utilizing synchrotron-XRD and running a case-study on calcium sulfate phases. To accomplish this, we developed a modular automation setup for reactions in solution to run synthesis and control reaction conditions. So far we successfully characterized the recycling process of gypsum and are now investigating the formation of anhydrite. T2 - BESSY@HZB User Meeting CY - Berlin, Germany DA - 02.12.2025 KW - Recycling KW - Gypsum KW - Synchrotron-X-ray-diffraction KW - Raman-spectroscopy KW - Automation PY - 2025 AN - OPUS4-65344 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Fotheringham, U. T1 - Digitalization of Glass Development N2 - Im Vortrag werden erste Ergebnisse aus dem vom BMFTR im Rahmen der MaterialDigital Initiative geförderten Projekt „GlasAgent“ vorgestellt, welches die Glasentwicklung mittels KI vorantreiben soll. In diesem Projekt werden mehrere Entwicklungszyklen inklusive des Recyclingprozesses durchlaufen und die Ergebnisse genutzt, um Datenbanken und Modelle zu verbessern. Mit diesen verknüpft und basierend auf der semantischen GlasDigital-Ontologie soll zukünftig ein Chatbot die Glasentwicklung schneller, präziser und nachhaltiger gestalten. T2 - PMD Vollversammlung CY - Berlin, Germany DA - 26.11.2025 KW - Glass KW - Workflow KW - Automation KW - MAP KW - Ontology KW - Simulation KW - Database PY - 2025 AN - OPUS4-65039 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Resch-Genger, Ute T1 - Quantifying the total and accessible amount of surface functionalities and ligands on nano-materials: Overview and recommended methods N2 - Engineered nanoparticles (NPs) with various chemical compositions and surface functionalities are routinely fabricated for industrial applications such as medical diagnostics, drug delivery, sensing, catalysis, energy conversion and storage, opto-electronics, and information storage. NP function, interaction with biological species, and environmental fate are largely determined by surface functionalities. This calls for reliable, reproducible, and standardized surface characterization methods, which are vital for quality control of NPs, and mandatory to meet increasing concerns regarding their safety. Validated and standardized workflows for surface analysis are also increasingly requested by industry, international standardization organizations, regulatory agencies, and policymakers. To establish comparable measurements of surface functionalities across different labs and ease instrument performance validation, reference test materials and reference materials of known surface chemistry as well as reference data are needed. In the following, different methods for determining surface functionalities on ligand-stabilized core and core/shell NPs include advanced techniques are presented and discussed regarding method-inherent advantages and limitations. Special emphasis is dedicated to traceable quantitative nuclear magnetic resonance (qNMR), X-ray electron spectroscopy (XPS) and time of flight secondary ion mass spectrometry (ToF-SIMS), and simpler optical and electrochemical methods. T2 - LNE Workshop CY - Paris, France DA - 04.11.2025 KW - Quality assurance KW - Fluorescence KW - Nano KW - Particle KW - Synthesis KW - Characterization KW - Advanced material KW - Surface KW - Standardization KW - Reference material KW - Functional group KW - Quantification KW - Coating KW - Automation KW - Potentiometry KW - Method KW - Validation KW - Optical assay KW - Fluram KW - Fluorescamine KW - qNMR KW - Comparison KW - ILC PY - 2025 AN - OPUS4-64726 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Resch-Genger, Ute T1 - The emp project smurfnano – Standardizing the quantification of surface functionalities, ligands, and coatings on nanomaterials N2 - For industrial applications such as medical diagnostics, drug delivery, sensing, catalysis, energy conversion and storage, opto-electronics, and information storage, meanwhile engineered nanoparticles (NPs) with various chemical compositions and surface functionalities are routinely fabricated. NP function, interaction with biological species, and environmental fate are largely determined by surface functionalities. Reliable, reproducible, and standardized surface characterization methods are therefore vital for quality control of NPs, and mandatory to meet increasing concerns regarding their safety. Also, industry, international standardization organizations, regulatory agencies, and policymakers need validated and standardized measurement methods and reference materials. These needs are addressed by the recently started European metrology project SMURFnano involving 12 partners from different National Metrology Institutes, designated institutes, and research institutes, two university groups as well as one large company and one SME producing NPs. This project as well as first results derived from the development of test and reference materials with a well characterized surface chemistry and ongoing interlaboratory comparisons will be presented. T2 - LNE Workshop CY - Paris, France DA - 04.11.2025 KW - Quality assurance KW - Fluorescence KW - Nano KW - Particle KW - Synthesis KW - Characterization KW - Advanced material KW - Surface KW - Standardization KW - Reference material KW - Functional group KW - Quantification KW - Coating KW - Automation KW - Potentiometry KW - Method KW - Validation KW - Optical assay KW - Fluram KW - qNMR KW - Comparison KW - ILC PY - 2025 AN - OPUS4-64725 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust data generation, heuristics and machine learning for designing sustainable materials N2 - Despite advances in computational materials design, simulating large systems—such as defects, interfaces, or amorphous states—with quantum-chemical accuracy remains a major challenge.[1] Machine learning (ML) methods are emerging as powerful tools to overcome these limitations, enabling scalable and accurate modeling beyond traditional quantum-chemical approaches.[2] They also open new avenues for discovering non-toxic, earth-abundant alternatives to existing materials and can be combined with self-driving labs. [3] There are nowadays robust data generation strategies that underpin the development and benchmarking of ML models. [4,5]atomate2 I will focus on such strategies for quantum-chemical bonding analysis and ML interatomic potentials in my talk. Quantum-chemical bonding descriptors can be effectively used in ML models to predict phononic properties. [6] ML interatomic potentials offer a powerful approach for predicting energies, forces, and stresses—but their performance hinges on high-quality training data. Our automated framework, autoplex, enables diverse and scalable training workflows, from random structure searches for general-purpose models to phonon-aware pipelines for high-accuracy predictions.[7] While quantum chemistry excels in many domains, properties like magnetism and synthesizability remain elusive. Here, heuristics or leveraging experimental data for ML offer promising alternatives.[8,9] T2 - Advanced Materials Safety 2025 CY - Dresden, Germany DA - 04.11.2025 KW - Nano Particles KW - Machine Learning KW - Automation KW - Materials Design KW - Sustainability KW - Material Safety PY - 2025 AN - OPUS4-64599 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - INPR A1 - Smales, Glen J. A1 - Appel, Paul Alexander A1 - Breßler, Ingo A1 - Chambers, Aaron A1 - Dumele, Oliver A1 - Ebisch, Maximilian A1 - Frontzek, Julius A1 - del Refugio Monroy, José A1 - Rosalie, Julian M. A1 - Pauw, Brian R. T1 - DACHS and RoWaN: The Automated and Traceable Synthesis of ZIF-8 N2 - Automated synthesis and open-data practices are increasingly seen as key enablers of transparent, traceable, and reproducible science. By combining automation with structured, metadata-rich documentation, it becomes possible to systematically com- pare synthesis strategies and link outcomes to detailed parameters. In this work, we implement such an approach to study the synthesis of ZIF-8, comparing hand and automation-assisted methods under controlled conditions. Using over 100 synthesis experiments, we assess the repeatability of particle size and yield, and explore how variations in mixing and injection influence outcomes. This study demonstrates how traceable synthesis workflows can support repeatability and comparison across synthe- sis strategies. The DACHS (Database for Automation, Characterization and Holistic Synthesis) framework underpins this work, providing a lightweight infrastructure for transparent synthesis data capture. KW - Automation KW - MOFs KW - SAXS PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-645861 DO - https://doi.org/10.26434/chemrxiv-2025-7fgg0 SP - 1 EP - 32 PB - Cambridge AN - OPUS4-64586 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Von der Computerchemie zur Materialinformatik N2 - In dem Vortrag stelle ich meinen Lebenslauf von einem Chemiestudium bis hin zur Materialinformatik vor. Hierbei stelle ich unter anderem Computerchemie und Materialinformatik vor. Das beinhaltet Kurzeinführungen in die Quantenmechanik und das maschinelle Lernen. T2 - Landesseminar Berlin Brandenburg der deutschen Auswahl für die Internationale Chemieolympiade (Schülerwettbewerb) CY - Berlin, Germany DA - 17.10.2025 KW - Materials Design KW - Computational Chemistry KW - Materials Acceleration Platforms KW - Automation KW - Machine Learning Potentials KW - Outreach PY - 2025 AN - OPUS4-64412 LA - deu AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - JOUR A1 - Janssen, Jan A1 - George, Janine A1 - Geiger, Julian A1 - Bercx, Marnik A1 - Wang, Xing A1 - Ertural, Christina A1 - Schaarschmidt, Jörg A1 - Ganose, Alexander Miguel A1 - Pizzi, Giovanni A1 - Hickel, Tilmann A1 - Neugebauer, Jörg T1 - A Python workflow definition for computational materials design N2 - Numerous Workflow Management Systems (WfMS) have been developed in the field of computational materials science with different workflow formats, hindering interoperability and reproducibility of workflows in the field. To address this challenge, we introduce here the Python Workflow Definition (PWD) as a workflow exchange format to share workflows between Python-based WfMS, currently AiiDA, jobflow, and pyiron. This development is motivated by the similarity of these three Python-based WfMS, that represent the different workflow steps and data transferred between them as nodes and edges in a graph. With the PWD, we aim at fostering the interoperability and reproducibility between the different WfMS in the context of Findable, Accessible, Interoperable, Reusable (FAIR) workflows. To separate the scientific from the technical complexity, the PWD consists of three components: (1) a conda environment that specifies the software dependencies, (2) a Python module that contains the Python functions represented as nodes in the workflow graph, and (3) a workflow graph stored in the JavaScript Object Notation (JSON). The first version of the PWD supports directed acyclic graph (DAG)-based workflows. Thus, any DAG-based workflow defined in one of the three WfMS can be exported to the PWD and afterwards imported from the PWD to one of the other WfMS. After the import, the input parameters of the workflow can be adjusted and computing resources can be assigned to the workflow, before it is executed with the selected WfMS. This import from and export to the PWD is enabled by the PWD Python library that implements the PWD in AiiDA, jobflow, and pyiron. KW - Worklows KW - FAIR Workflows KW - Automation KW - Materials Acceleration Platforms PY - 2025 UR - https://nbn-resolving.org/urn:nbn:de:kobv:b43-643325 DO - https://doi.org/10.1039/D5DD00231A SN - 2635-098X SP - 1 EP - 14 PB - Royal Society of Chemistry (RSC) AN - OPUS4-64332 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - George, Janine T1 - Robust Data Generation, Heuristics and Machine Learning for Materials Design N2 - Despite significant progress, computational materials design still faces challenges, especially when simulating large systems needed to describe defects, interfaces, or amorphous states with the accuracy of density functional theory (DFT) or beyond.[1] To address these limitations, machine learning (ML) methods have become increasingly popular in recent years. In this talk, I will present how we’ve developed robust data generation strategies that support the creation and benchmarking of new ML models.[2] I’ll focus on methods for large-scale quantum-chemical bonding analysis and workflows for ML interatomic potentials. We’ve shown that quantum-chemical bonding properties can be used in ML models to predict phononic properties.[3] This enables us to validate several expected correlations— such as the link between bonding strength and force constants—on a large scale. Additionally, we’ve built an automated training framework for machine-learned interatomic potentials (autoplex).[4] Initial workflows include random structure searches, which are well-suited for general-purpose potentials, as well as workflows tailored to ML potentials with accurate phonon properties. While atomistic simulations are highly effective for certain material properties, others— like magnetism or synthesizability—remain difficult. In these cases, it’s promising to benchmark established ab initio methods against chemical heuristics or to develop new ML models based primarily on experimental data.[5,6] T2 - SusML (Sustainable Machine Learning Workshop) CY - Dresden, Germany DA - 29.09.2025 KW - Materials Design KW - Materials Acceleration Platforms KW - Automation KW - Workflows KW - Machine Learned Interatomic Potentials KW - Synthesizability KW - Magnetism PY - 2025 AN - OPUS4-64244 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER - TY - CONF A1 - Ryll, Tom William T1 - Applied and Technical Mineralogy:� high-throughput automated platform for in-situ monitoring of CaSO4 formation N2 - In this project we investigate nucleation pathways by utilizing synchrotron-XRD and running a case-study on calcium sulfate and its polymorphs. To accomplish this, we developed a modular automation setup for reactions in solution to run synthesis and control reaction conditions. So far we successfully characterized the recycling process of gypsum (CaSO4*2H2O) and are now investigating the formation of anhydrite (CaSO4*0H2O) as well as possible applications for the automation setup and analysis. T2 - Geo4Göttingen 2025 CY - Göttingen, Germany DA - 14.09.2025 KW - Recycling KW - Gypsum KW - Synchrotron-X-ray-diffraction KW - Raman-spectroscopy KW - Automation PY - 2025 AN - OPUS4-64139 LA - eng AD - Bundesanstalt fuer Materialforschung und -pruefung (BAM), Berlin, Germany ER -