D. CHARACTERIZATION, MODELING AND ARTIFICIAL INTELLIGENCE

Scope

The symposium on Challenges and Advances in Microscopy will explore recent breakthroughs and persistent challenges in electron microscopy. Topics will include atomic resolution imaging with aberration corrected instruments, 4D-STEM, including ptychography and other phase-related methods, as well as orientation imaging and 3D analysis in SEM/TEM. In-situ and operando experiments, analytical methods including EDXS, EELS, EBSD and S(P)ED, advances in cryo-EM and sample preparation techniques, will also be covered. Additionally, the symposium will highlight the role of machine learning, automation, data interpretation, time-resolved microscopy, and multimodal/correlative techniques. All of these will focus on real-world applications for scientific and industrial problem solving.

Description

Innovations in electron microscopy play a fundamental role in advancing materials science. This symposium will bring together cutting-edge topics across the field of electron microscopy, offering an outlook on the state of the art in acquisition and data analysis across systems from hard to soft matter. It will cover topics from atomic-resolution imaging with aberration correction and electron ptychography, to the nano and micro-scales using a variety of TEM and SEM techniques. 4D-STEM and phase-related techniques such as holography and DPC will be discussed, along with electron spectroscopies such as EDXS, EELS, and CL. In addition, the symposium will look at automated orientation imaging in SEM and TEM using EBSD and scanning (precession) electron diffraction, along with 3D analysis techniques such as electron tomography and FIB-SEM.

A major focus will be on in-situ and operando experiments that allow real-time observation of dynamic processes; crucial for understanding materials under realistic conditions. Techniques for imaging beam-sensitive/soft matter at low-dose under cryo conditions will also be considered. The role of machine learning, automation, and advanced data interpretation techniques will be highlighted for handling the increasing complexity of EM datasets. Lastly, time-resolved microscopy, multimodal and correlative techniques, and real-world applications will be discussed.

Targeted topics

• Atomic resolution imaging (aberration correction and ptychography)
• Orientation imaging in SEM/TEM (SPED, EBSD, ACOM, TKD)
• Tomography and 3D analysis in FIB-SEM/TEM/APT
• In-situ and operando experiments and EM automation
• Analytical spectroscopies in SEM and TEM (EDXS, EELS, CL)
• Machine learning applications and EM data analysis
• Advancements in sample preparation and FIB
• Soft matter and cryo EM for materials science
• Multimodal and correlative microscopy
• Time-resolved microscopy
• Imaging and advances in instruments, detectors, and modelling
• Electron diffraction tomography and crystallography
• 4D-STEM and phase-based techniques for field mapping (holography, DPC, COM)
• Defect analysis in TEM and SEM (point defects, dislocations, sigma boundaries, grain boundaries, stacking faults, disclinations, complexions, …)
• Industrial applications/real problems solving

• 
  Prof. Angus Kirkland
  Department of Materials University of Oxford (UK)
• 
  Dr. Duncan Alexander
  École Polytechnique Fédérale de Lausanne – EPFL (CH)
• 
  Dr. Grzegorz Cios
  AGH University of Krakow (PL)

Scope

The symposium aims to bring together experts in diffraction using various probes (X-rays, neutrons, and electrons) to present recent advancements in experimental diffraction techniques and sophisticated tools for analyzing the collected data. Exchange of ideas across the different sub-disciplines is triggered, facilitating the transfer of successful strategies and insights from one diffraction technique to another. Focus will be on characterizing microstructures and their evolution during in-situ or operando conditions, encompassing scenarios such as mechanical loading, heating and cooling, chemical reactions, and exposure to diverse atmospheres.

Description

Diffraction techniques using X-rays, neutrons, or electrons are indispensable tools for contemporary materials characterization. With the help of diffraction, the crystal structure and distortions in the crystalline lattice are revealed. Lattice defects, elastic strains, crystallographic orientations and phase fractions can be resolved globally in an averaged manner or locally on the appropriate length scale. Because diffraction is non-destructive, the microstructural evolution can be tracked in-situ in response to different stimuli in different environments close to real-world conditions, complementary to post-mortem investigations. New experimental methods emerge through rapid advancements in instrumentation or continuous improvements of already existing methods, opening new opportunities for studying a wide range of structures and behaviors. Concurrently, data evaluation tools become increasingly refined, enabling improved global descriptions and revealing local microstructural details, for instance, heterogeneities in elastic strains and orientations or specific neighboring orientation relationships allowing reconstruction of transformed conditions. Powerful numerical tools including machine learning facilitate efficient and swift processing of large and complex datasets.

Targeted topics

• X-ray diffraction (XRD)
• Phase, peak profile, texture and stress analysis
• Diffraction with monochromatic and white beam synchrotron radiation
• Neutron diffraction
• Electron diffraction using transmission electron microscopy (TEM)
• Electron backscatter and transmission Kikuchi diffraction (EBSD and TKD)
• Data analysis, evaluation and reconstruction tools, machine learning
• Ex-situ, in-situ and operando investigations

• 
  Prof. Matthias Bönisch
  KU Leuven (BE)
• 
  Prof. Wolfgang Pantleon
  Technical University of Denmark (DK)
• 
  Dr. Yunhui Chen
  RMIT University, Melbourne (AU)

Scope

Scale bridging mechanical characterization is both a well-established and yet very dynamic field of research. Recent innovations in computer-based and experimental techniques have expanded the range of accessible mechanical parameters and materials systems.

Description

Notably, the integration of advanced small-scale mechanical testing, 3D imaging, in-situ and operando techniques, high-performance computing, data-driven mechanics, multi-scale modeling, machine learning and artificial intelligence algorithms is opening up new avenues for research at the nano to meso scales. These approaches offer unprecedented insights into the physical behavior of hierarchical and functional materials and structures. This symposium seeks to bring together these rapidly growing research communities, fostering interdisciplinary collaboration in micro- and nanomechanics to expand the understanding of small-scale material behavior.

Targeted topics

The scope of the symposium encompasses, but is not limited to:

• Mechanics of nanomaterials, nanostructures, thin films, and multiphase materials
• Advances in instrumentation for mechanical testing at the micro- and nanoscale
• Cutting-edge computational, data-driven, machine learning and AI-supported approaches applied to micro- and nanomechanical topics
• Techniques for measuring stress-strain relationships in micro- and nanostructures
• Characterization of strain-rate sensitive deformation mechanisms
• Fatigue, and creep phenomena across multiple length scales
• Techniques for hierarchical and functional materials characterization across different length scales
• 3D characterization of small structures in relation to mechanical phenomena
• In situ and in operando testing for micro- and nanomechanics
• Micro- and nanomechanics of fracture, as well as adhesive and cohesive failures
• Modeling techniques for small-scale mechanics
• Experimentally informed scale-bridging models

• 
  Prof. Benoit Merle
  University of Kassel (DE)
• 
  Prof. Maciej Szczerba
  Polish Academy of Sciences (PL)
• 
  Dr. Verena Maier-Kiener
  Montanuniversität Leoben (AT)
• 
  Dr Andre Clausner
  Fraunhofer IKTS (DE)

Scope

The scope of this symposium comprises the study of hydrogen effects in materials, through materials testing, microstructural characterization and modelling across lengthscales from atomic to macroscopic. Contributions in the field of functional materials for hydrogen related energy conversion and storage, such as catalytically active materials, components of the MEA, bipolar plate materials and hydrides for storage, are also considered.

Description

Hydrogen related changes in properties remain a persistent challenge for metallic materials, posing a significant barrier to the advancement of a sustainable hydrogen economy. Even though the detrimental effect of hydrogen, especially in bcc steels has been studied for decades, the interaction between hydrogen and the alloy microstructure still needs to be better understood as new challenges in the hydrogen economy arise. This requires the study of the influence of pressurized and cryogenic hydrogen on the mechanical properties of hydrogen compatible materials such as austenitic steels and superalloys as well as the influence of hydrogen containing gases on the materials involved in the combustion process. This symposium invites contributions addressing phenomena relating to hydrogen facing materials from an applied or fundamental perspective through testing, characterization and/or modelling. Furthermore, we warmly encourage contributions that explore various aspects of energy applications, with a particular focus on innovative solutions for energy storage and conversion. This includes, but is not limited to, advancements in hydrogen storage materials and cutting-edge developments in photo-electrocatalysis.

Targeted topics

• Fundamentals of hydrogen effects on metallic materials
• Materials exposed to pressurized hydrogen
• Materials exposed to cryogenic hydrogen
• Materials involved in hydrogen combustion processes
• Development of ML-based interatomic potentials for the Ni-H system
• Multiscale approaches to study hydrogen embrittlement
• Ab-initio description of defect phases and phase transformations, including nanohydride formation
• Hydrogen storage materials
• Energy conversion through hydrogen production – photo-electro-catalysis

• 
  Prof. Peter Felfer
  Friedrich-Alexander-Universität Erlangen (DE)
• 
  Dr. Ageo Meier de Andrade
  Chalmers University of Technology (SE)
• 
  Prof. Moyses Araujo
  Karlstad University (SE)

Scope

Introducing Artificial Intelligence and Machine Learning into next generation biomedical material research, the scope of the session will consider the following related key issues:

• Simulations of the biomechanical properties of the materials dedicated for the regenerative medicine,
• Materials in-vitro testing development,
• Lab on chip and Organ on chip design,
• Multifunctional nanoprobes as tools for diagnosis, imaging, and therapy,
• Preclinical analysis, in-vivo testing,
• Holographic visualisations for medical planning and training,
• Clinical studies/post deplanting assessment.

Description

The biomedical diagnosis and therapy has brought tremendous advances in the development of targeted drug-delivery and bioanalytical systems. One of the most basic issues has been numerical simulation, while tremendous impacts are expected by machine learning (artificial intelligence). Artificial Intelligence (AI) significantly enhances the design and fabrication of biomaterials for biomedical applications by streamlining processes and improving material properties. AI technologies, such as machine learning and deep learning, enable rapid prediction and optimization of biomaterial characteristics, facilitating high-throughput screening and automated material discovery. This leads to more efficient design cycles and the development of advanced materials tailored for specific biomedical uses.

On the basis of numerical models, it is necessary to carry out experimental verification in the form of basic issues of biological suitability analysis, such as cytotoxicity analysis, genotoxicity analysis, microbiological analysis. Machine learning can support analyzing the outcomes of the simulation validation. After in-vitro testing, it is necessary to prepare diagnostics for preclinical studies. In-vivo testing on animals always arouses a lot of controversy but for the preparation of new materials dedicated to implantation. For this purpose, the issues of Lab on chip and Organ on Chip, i.e. their control in best replication of human organisms, are very important. Diagnostic cubes make it possible to predict the specific behavior of the organs in question in direct contact with the tested material and to create individual therapies for the individual. Finally, there is the planning and training stage, which in the near future will be based on holographic visualization. A very important aspect is post-deplantation studies both after basic in-vivo research on small animal models, large animal models once after deplantation from humans.

Targeted topics

• Bioinformatics and numerical modelling – improvements by machine learning – predictive Modeling: AI algorithms can forecast material properties, reducing reliance on traditional trial-and-error methods.
• Machine learning aids in the rapid development of customized biomedical polymers, addressing specific patient needs.
• Multifunctional nanoprobes with diagnostic and therapeutic units.
• New approaches to image biological processes at nanometer scale – introduction of AI methods for improved analysis and interpretation.
• In-vitro assessment of the biomedical application.  
• Dual-functional probes for optical imaging and therapy.
• Lab on Chip/Organ on Chip design and application – how can AI support the establishing a better digital “human” twin. Multi-Attribute Optimization: AI allows simultaneous optimization of various material attributes, enhancing performance in applications like drug delivery and tissue engineering.
• In-vivo testing.
• Post deplanting analysis.

• 
  Dr. Jürgen Lackner
  Joanneum Research Forschungsgesellschaft MBH (AT)
• 
  Prof. Roman Major
  Polish Academy of Sciences (PL)
• 
  Prof. Sachiro Kakinoki
  Kansai University (JP)

Scope

The aim of the symposium is to assess the state of the art in applications of theories and data science tools that allow for modeling and simulation for the knowledge-based design of advanced materials. It can cover different length scales, arranging from atomic to microscopic and to mesoscopic scales. Advances and challenges in applications of ab initio calculations, molecular dynamics, Monte-Carlo techniques, phase-field methods, and continuum theories will be discussed.

Description

Complimentary to experimental efforts, theories and numerical simulations can provide fundamental sights and provide (in-situ) data argumentation in a more expeditious and economical way. Furthermore, simulations can predict variables in spatio-temporal resolutions yet far ahead of the best in-situ imaging systems. Moreover, large-scale combinational databases and impactful repositories emerges, assisted by high-throughput synthesis and high-throughput simulations and materials digitalization. Groundbreaking opportunities can be enabled by AI-assisted materials discovery of advanced structural and functional materials. Thus, significant improvement in the predictive power of theoretical and/or data-driven models is envisaged, which is leading to a shift from a pure empirical paradigm in materials design to the knowledge-based materials design concept.

During the symposium, theoretical methodologies will be discussed, starting from the basic concepts of quantum simulations, atomistic simulations, phase-field simulations and continuum theories. Furthermore, the various use of new data-driven techniques, such as Machine Learning (ML) interatomic potentials, ML materials surrogates, and data-driven scale bridging approaches are also highly appreciated.  

Targeted topics

• Multiphysics simulation of process conditions
• Microstructural simulations and phase transformation
• Material scale bridging
• Stress predictions and evaluation across macro- and meso-scales
• Multiphysics phase-field models
• Novel continuum models
• ML interatomic potentials
• ML materials surrogates
• Data-driven scale bridging approaches

• 
  Prof. Mohamad Bayat
  Technical University of Denmark (DK)
• 
  Prof. Bai-Xiang Xu
  Technical University Darmstadt (DE)
• 
  Prof. Dierk Raabe
  MPI for Sustainable Materials, Düsseldorf (DE)

Scope

In this symposium, we call for an open discussion and exchange about the recent technical and scientific challenges involved in developing an interoperable representation of materials and their processing.  We expect recent developments of ontologies, materials data schemas and software solutions that allow representation and integration of workflows, processes and materials in a digitalized manner and their application in accelerating developing new materials and processes.

Description

Materials Science and Engineering is undergoing a major paradigm shift towards more efficient digitalization. Integration and reuse of data and knowledge from synthesis, production, characterization as well as of modelling activities open new perspectives for innovative materials design. Emerging fields of Materials Informatics employing tools such as machine learning, big-data applications, and statistical inference allow accelerating the discovery of new compositions and processes tailored to the production of materials with specific properties and microstructures.

A key to enable the digitalization of materials and to leverage the advantages and opportunities of the digital age is an interoperable digital representation of materials and processes. An appropriate management of materials data requires the use of FAIR principles. Digital workflows ensure the seamless interaction of materials data, AI models and scale bridging simulations. They connect individual software tools, automatize the storage and curation of simulation results as well as intermediate steps and can, therewith, ensure the reproducibility of computational procedures. Ontologies are essential for formally representing universal materials science concepts, their interrelationships, and workflows. Application ontologies enhance identification, data integration and complex simulation workflows. This will improve explainability and validation of real-life and simulated process designs. A unique identification and elucidation of entities and relations is required to meet the FAIR principles.    

Targeted topics

• Rapid Materials Design
• Materials Acceleration Platforms
• High throughput materials characterization
• Digital Workflows
• Data-driven material modelling
• Materials ontologies
• FAIR principles

• 
  Prof. Raymundo Arroyave
  Texas A&M University (US)
• 
  Prof. Christoph Eberl
  Fraunhofer Institute for Mechanics of Materials IWM, Freiburg (DE)
• 
  Prof. Tilmann Hickel
  BAM Berlin (DE)

Scope

This symposium presents an overview of the computational and experimental methods used for the rapid synthesis, characterisation, and discovery of advanced alloys. It will integrate artificial intelligence, computational materials modelling, and state-of-the-art characterisation techniques to advance the understanding and improvement of metallic systems. The focus will be on accelerated improvement of alloy processing, discovery, and enhanced materials characterisation methods by combining computational data-driven methods with experimental techniques.

Description

This symposium will present the most recent high-throughput computational and experimental techniques to advance the understanding, improvement, and discovery of advanced metallic alloys. The topics presented include modern processing technologies, such as wire- and powder-based as well as friction stir additive manufacturing. All metallic alloy systems will be considered, including the development of novel compositions to underpin innovative processing technologies and sustainability demands, all the way to functionally graded materials that aim to deliver spatially tailored performance. Modelling, including data- and AI-driven techniques, as well as physics-based methodologies will be considered, especially when aimed at the improvement of processing schedules and alloy compositions.

Targeted topics

• High-throughput electron microscopy
• High-throughput mechanical testing
• High-throughput synchrotron tomographic and diffraction techniques
• Combinatorial, AI-driven and data-driven alloy discovery
• Rapid materials synthesis
• Additive manufacturing processing
• Materials informatics
• Multiscale materials modelling
• Computational thermodynamics
• Finite element method and crystal plasticity
• Small-scale mechanical testing
• AI-enhanced spectroscopy and imaging techniques
• AI-guided defect detection in additive manufacturing
• Data-driven process optimisation
• ICME

• 
  Prof. Moataz Attallah
  University of Birmingham (UK)
• 
  Prof. Ibrahim Karaman
  Texas A&M University (US)
• 
  Prof. Pedro Rivera
  University of Southampton (UK)

Scope

This symposium explores the convergence of AI, modeling, and data science and advanced experimental characterization in revolutionizing advanced alloy, process design and characterization. It showcases cutting-edge applications of machine learning, integrated computational materials engineering, and big data in metallurgy. The focus is on rapid microstructure characterization, prediction of process parameters and properties, synthetic data generation, the development of interoperable digital representations of materials and processes, and experimental validation of AI and modeling tools. We will examine the implementation of FAIR principles in materials data management and the role of digital workflows in ensuring data coherence and reproducibility in computational metallurgy.

Description

The field of metallurgy is undergoing a profound digital transformation, driven by the integration of AI, advanced modeling and experimental characterization techniques. This symposium brings together experts from academia and industry to explore the latest developments in this rapidly evolving landscape.

Key areas of focus include the application of AI in rapid microstructure characterization, the use of predictive modeling for process parameters and material properties, and related data management and standardization. Additionally, the symposium aims to showcase innovative methods and results that enhance our understanding of the process-microstructure-properties paradigm.

Participants will gain insights into state-of-the-art theoretical and experimental techniques for generating and analyzing high-quality data, leveraging vast and varied datasets spanning physical, chemical, and mechanical domains, as well as the generation and utilization of realistic synthetic data to enhance model training and validation. The symposium will demonstrate how the resulting ICME approach is accelerating the discovery and optimization of novel alloy compositions and processing methods. At the same time, it aims to foster discussions on overcoming technical and scientific challenges in creating standardized, interoperable representations of materials, processes and workflows, as well as big datasets using FAIR principles, paving the way for the next generation of intelligent metallurgical design and manufacturing.

Targeted topics

•  Machine learning for rapid characterization, AI-enhanced microscopy (OM, SEM, EBSD, …) and spectroscopy (Raman, XRD, XPS, FTIR, …), classification of microstructures and defects and related property prediction
• AI-driven alloy and process design with machine learning models, Bayesian optimization, active learning and genetic algorithms
• Interoperable digital representations of materials and processes, FAIR principles, digital workflows, ontologies and standardization in materials data management
• Integrated Computational Materials Engineering (ICME) approaches including multiscale modeling techniques for alloy and process design
• Process control and process-structure-property modeling, ML for sensor data, predictive maintenance and development of digital twins
• Big data analytics, representation learning, synthetic data generation and anomaly detection
• Explainable AI, physics-informed AI and visualization techniques for machine learning for metallurgy
• AI and modeling tools for the design of high-quality products

• 
  Dr. ir. Michael Sluydts
  ePotentia/Ghent University (BE)
• 
  Prof. Ignacio Romero
  IMDEA Materials Institute/Universidad Politécnica de Madrid (ES)
• 
  Dr. Dirk Helm
  Fraunhofer Institute or Mechanics of Materials IWM (DE)