Research Details

New materials and material composites are at the centre of future challenges in a wide range of application areas, e.g. in healthcare, medical technology and solution strategies related to climate change, resource conservation, energy supply and storage. Knowledge of the properties of materials and their intended areas of application enable the improvement of technical components and the development of new modules.

The research group at INT, RU Microstructure Simulation has been working for 25 years on the development of new material models on the microstructure scale, taking into account multi-physical fields such as phase states, mass and heat transport, thermo- and chemo-mechanics, fluid flow and electrochemistry, and on the creation of a parallel simulation software Pace3D for the calculation of microstructures with high detail resolution in large-scale 3D domains using current high-performance computers. The research work in the field of computer-aided materials research with the special focus on microstructure simulation on high-performance computers and data analysis of material-property relationships is carried out across universities at the Institute for Applied Materials and the Institute for Nanotechnology at the Karlsruhe Institute of Technology and at the Institute for Digital Materials Research at the Karlsruhe University of Applied Sciences. Research groups with specifically oriented research profiles have been established in the various material-physical focal points of microstructure simulation and data analysis.

In the past two decades, microstructure simulation has taken on a key role for materials and component development as well as for process optimisation in industry and science. Calculations with a resolution in the micrometre range make a significant contribution to data-driven material development and represent a link in multiscale material modelling from atomistics to macroscopic treatment.

The virtual laboratory environment, consisting of numerical calculation, comprehensive data analysis and high-resolution visualisation, provides a modern, time-, effort- and resource-saving technology for testing materials and components for quality requirements such as resilience, durability, efficiency and effectiveness. In the research work, digital twins are created for complex microstructures of different material classes.

New simulation methods allow the design of customised materials such as alloys with specific compositions, the analysis of influencing factors and process control conditions on microstructure formation, and the systematic investigation of microstructure-property correlations. Often, quality characteristics of materials and processes can be achieved by a small change in the machining process or by varying the composition. Such a detailed determination of the reaction of the material to an external stress, such as thermal, magnetic field-induced and mechanical stress, is provided by new multiphysics material models and the corresponding software packages for microstructure simulation. The characteristic parameters of the microstructure are decisive for the properties of the material. As a representative example, in many manufacturing processes the grain structure and grain size distribution is a decisive criterion for the hardness, fracture strength and degradation of a material. Simulations allow in-situ insight not only into the final structure, but also into the three-dimensional time-dependent structure formation process. Through targeted process control, the microstructure formation can be influenced in a controlled manner and material with specific properties such as strength, permeability or electrical charge capacity can be designed computer-aided. The calculations replace to a large extent the experimental, metallographic and mechanical microstructure characterisation, which often requires the destruction of the components. The material, the component and the process flow of the future can be designed in a resource- and energy-efficient way on the computer using material simulation methods. Costly experiments can be saved and weak points can be corrected on the computer at the design stage.

By applying advanced data science and machine learning methods of the open source package CIDS (lead: Dr. Arnd Koeppe) to simulation and experimental data, data processing workflows are generated and sensitivity analyses of the microstructure-material property relationships are carried out, which are successfully used to accelerate the development of materials in various material classes. By connecting and integrating the research data, the data science methods and workflows into the open-source research data infrastructure Kadi4Mat of INT (Head: Dr. Michael Selzer), a sustainable use and reproducibility according to FAIR principles (findable, accessable, interoperable and reusable) is ensured. The aim of the coordinated development of the Pace3D and Kadi4Mat software packages is to establish electronic laboratory books, data analysis tools and the resulting workflows, as well as the construction of materials science ontologies.
The knowledge generated from data analysis can be used to develop new materials with customised microstructures and to design process flows. As an integral part of a dynamic adaptive development cycle, material simulations allow for data-driven, accelerated design of new materials and are therefore an indispensable tool for new material development, evaluation and quality improvement.