AI revolutionizes materials science: Students explore new paths!

AI revolutionizes materials science: Students explore new paths!

AI is increasingly being used as a valuable tool in research into materials and their applications. An example of this is the Student Research Group Material Informatics at the Institute of Materials, which produces two committed students, Claas Hardt and Karina Kaidarova. Hardt, who is completing his master's degree in environmental engineering, developed an AI program for image classification of microstructures of metallic powder particles. This innovative work is now being continued by Kaidarova in her bachelor's thesis, which focuses on a different material from a different material family. She uses the results of Hardt's work and the developed AI program to achieve her own research goals. This shows the close connection between theory and practice in materials science.

The Student Research Group, led by Dr. Santiago Benito not only promotes exchange between students, but also enables potential joint scientific publications. The group offers a wide range of research fields, including microstructure classification, which is highlighted as one of the four main focuses of this research. Kaidarova estimates that, thanks to Hardt's preparatory work, she doesn't have to start from scratch, which significantly shortens the development time for her research.

Innovative methods in microstructure characterization

The “Advanced Microstructure Characterization” group, which is part of the research landscape, deals in depth with the quantitative analysis of the three-dimensional material structure. This research uses both classic 2D methods and modern tomographic techniques. The aim is to understand the manufacturing processes of the materials and to identify optimal structures for specific properties. Work with multi-phase steels and cast aluminum alloys are current projects in this group. Particularly noteworthy is the interdisciplinary collaboration with computer science to develop innovative methods of image processing and structure classification.

A decisive advantage of these modern analysis methods is the comprehensive characterization of the material structure, which allows conclusions to be drawn about the effective material properties. This leads to a better correlation between processing and properties, which is relevant for the further development of materials.

Artificial intelligence as a pioneer for progress

Artificial intelligence is playing an increasingly larger role in materials science and is fundamentally changing the way materials are discovered and their properties studied. The development of AI-supported processes promotes innovation and efficiency. AI can massively reduce the time it takes to discover new materials, often by up to 70 percent, while achieving prediction accuracy of over 90 percent. AI is not only used for material discovery, but also for predictive modeling of properties and optimization of design processes. This has significant implications for future developments in materials science as the processes become faster and more precise.

Kaidarova plans to continue researching materials and AI after her bachelor's thesis and sees the Student Research Group as a valuable platform for her scientific development. Hardt also aims to continue working on the topic of microstructure classification in his master's thesis and will remain actively involved in the group. Their collective efforts demonstrate not only the value of interdisciplinary collaboration, but also the significant advances that can be made in applying AI to materials science.

These developments are not only advances for individual students, but could also be crucial to advances across the industry. The Student Research Group is an example of the drive for innovation in materials research, which is inspired by modern technologies and scientific collaboration.